Malawi Strengthening Inclusive Markets 
for Agriculture (MSIKA) Program
Final Evaluation
11/26/2020
DISCLAIMER: This publication was produced at the request of the United States Department of 
Agriculture. It was prepared by an independent third-party evaluation firm. The author’s views 
expressed in this publication do not necessarily reflect the views of the United States Department of 
Agriculture or the United States Government.
MSIKA Final Evaluation
Program: Food for Progress
Agreement Number: FCC-612-2016/006-00
Funding Year: Fiscal Year 2016
Project Duration: September 28, 2016 – September 30, 2021
Implemented by: Land O’Lakes Venture37
Evaluation Authored by: Kadale Consultants (UK)
Jason Agar
Yamikani Chabwera
Patience Mtembezeka
McPherson Chatama
David Namanjasi 
MSIKA Final Evaluation Page iii kadale@africa-online.net
Table of Contents
List of Tables...................................................................................... iv
Acknowledgements............................................................................ vi
Acronyms .......................................................................................... vii
Executive Summary ......................................................................... viii
1 Introduction and Background ................................................ 1
1.1 Objectives of the Final Evaluation ....................................................... 1
1.2 COVID Impact on Evaluation Design and Implementation................. 1
1.3 Background to MSIKA Project.............................................................. 2
1.4 Key Changes to MSIKA ......................................................................... 4
2 Methodology and Implementation ......................................... 5
2.1 Methodology .......................................................................................... 5
2.1.1 Overview........................................................................................................... 5
2.1.2 Household Survey............................................................................................. 6
2.1.3 Farmer Qualitative Interviews..........................................................................10
2.1.4 Key Informant Interviews.................................................................................10
2.1.5 Review of MSIKA Documents and Data ..........................................................11
2.2 Limitations............................................................................................12
3 Results and Findings............................................................ 14
3.1 Farmer Based Organisations..............................................................14
3.1.1 Background and outline of work with FBOs .....................................................14
3.1.2 FBOs summary data .......................................................................................15
3.1.3 Training at FBOs .............................................................................................18
3.1.4 Lead Farmers..................................................................................................25
3.1.5 Sales Agreements and Sales ..........................................................................26
3.1.6 Processing and Storage ..................................................................................27
3.1.7 Access to Market Information ..........................................................................28
3.1.8 COVID Impacts ...............................................................................................29
3.2 Participant Producers..........................................................................30
3.2.1 MSIKA Producer Sample Frame .....................................................................30
3.2.2 Producer Totals...............................................................................................30
3.2.3 Final Evaluation Sample Profile.......................................................................31
3.2.4 Agricultural Practices.......................................................................................33
3.2.5 Crop Inputs .....................................................................................................39
3.2.6 Land Area Under Improved Practices..............................................................40
3.2.7 Production.......................................................................................................40
3.2.8 Post-Harvest Handling Knowledge and Application.........................................43
3.2.9 Post-Harvest Losses .......................................................................................50
3.2.10 Gender........................................................................................................53
3.2.11 Farm Management Knowledge and Application...........................................54
3.2.12 Sales Volume ..............................................................................................55
3.2.13 Sales Value ................................................................................................. 57
3.2.14 Access to Finance.......................................................................................59
3.2.15 Investment...................................................................................................59
3.2.16 Employment ................................................................................................60
3.2.17 COVID effects .............................................................................................60
3.3 Banks, MFIs and VSLs.........................................................................61
3.3.1 Overall Performance .......................................................................................61
3.3.2 Finance for SMEs............................................................................................62
3.3.3 Finance for Producers.....................................................................................63
MSIKA Final Evaluation Page iv kadale@africa-online.net
3.4 Processors ...........................................................................................64
3.4.1 Training and Application of Improved Techniques and Technologies ..............64
3.4.2 MBS Certification ............................................................................................66
3.4.3 Other Support to Processors...........................................................................66
3.4.4 Performance data............................................................................................67
3.5 Government..........................................................................................68
3.5.1 Policy Level.....................................................................................................68
3.5.2 Training District Extension Staff ......................................................................69
3.5.3 Malawi Bureau of Standards ...........................................................................71
3.5.4 Bunda/MSU.....................................................................................................71
4 Conclusions, Lessons and Recommendations .................. 72
4.1 Key Conclusions and Lessons...........................................................72
4.1.1 Methodology ...................................................................................................72
4.1.2 Overall Progress .............................................................................................73
4.1.3 Farmer Based Organisations...........................................................................74
4.1.4 Producer Households......................................................................................76
4.1.5 MFIs/Banks/VSLs............................................................................................77
4.1.6 Processors ......................................................................................................78
4.1.7 Government ....................................................................................................78
4.2 Recommendations...............................................................................79
Annex 1: Table of Program Indicators............................................. 81
Annex 2: MSIKA Results Frameworks............................................. 98
Annex 3: Terms of Reference/Scope of Work ................................. 99
Annex 4: Instruments...................................................................... 108
Annex 5: Comparison of Practices by Crop.................................. 149
List of Tables
Table 1: Minimum and Actual Responses by Crop ............................................................... 8
Table 2: Usable Responses by Crop and by District............................................................. 9
Table 3: KIIs Conducted by Crop, District and Sex .............................................................10
Table 4: KIIs Planned and Conducted by Category ............................................................11
Table 5: MSIKA’s Performance Rating of FBOs, FE, 2020 .................................................15
Table 6: MSIKA’s Performance Rating of FBOs, MTE, 2019 ..............................................15
Table 7: PMM Assessment, MSIKA-Established FBOs at FE, 2020 ...................................16
Table 8: PMM Assessment, Pre-existing FBOs, 2020 ........................................................17
Table 9: MSIKA Activities Reported by FBO Members in the MTE & FE HH Surveys.........17
Table 10: FBOs Trained in Agricultural Production, by Type of Training, at MTE & FE.......19
Table 11: FBOs Trained in Marketing, by Type of Training, at MTE & FE...........................20
Table 12: FBOs Trained in PHH and Storage, by Type of Training, MTE & FE...................21
Table 13: FBOs Trained in Financial Records, Literacy and VSL, MTE & FE .....................22
Table 14: FBOs Trained in Governance, Leadership & Cooperatives, MTE and FE...........23
Table 15: Proportion of Lead Farmers in the FE Survey .....................................................26
Table 16: Lead Farmer Activities by Sex at FE, 2020 .........................................................26
Table 17: Sources of Market Information, FE 2020.............................................................28
Table 18: Indicator 18, Producers Accessing Market Info via their FBO, FE 2020 ..............29
Table 19: Beneficiary Producers, by District, Sex & Crop, FE 2020....................................30
Table 20: Comparison of Baseline vs FE Samples .............................................................32
Table 21: Check Balance of Confounders Using Propensity Score Matching .....................32
Table 22: MSIKA Activities Reported by FBO Members in MTE & FE HH Surveys.............33
Table 23: Application of New Agricultural Practices per Category, FE Survey, 2020 ..........34
Table 24: Application of New Practices by Crop at FE, 2020..............................................35
MSIKA Final Evaluation Page v kadale@africa-online.net
Table 25: Use of Practices for the First Time, FE Survey, 2020..........................................37
Table 26: Why Other Practices are not Used, by Crop, FE 2020 ........................................39
Table 27: Land area in ha under improved practices, MTE and FE HH surveys .................40
Table 28: Producer Yields in kg/ha, FE HH survey, 2020 ...................................................41
Table 29: Producer Yields in kg/ha, Revised Baseline 2017...............................................42
Table 30: Yield Comparison, FE vs baseline by Crop.........................................................42
Table 31: Producers Affected/Not Affected by COVID, Yield Comparison vs Baseline .......43
Table 32: Yield Comparison by Sex, FE vs baseline...........................................................43
Table 33: Use of PHH Practices by Crop, FE HH survey 2020 ...........................................45
Table 34: Use of PHH Practices Across the Four FE crops, at Baseline.............................46
Table 35: First Used New PHH Practice, at FE, 2020.........................................................47
Table 36: First Used New PHH Practice, Across FE Crops at Mid-term, 2019....................48
Table 37: Post-harvest Losses by Crop and by Sex, FE, 2020 ...........................................50
Table 38: Post-harvest Losses by Crop and by Sex, Baseline, 2017 ..................................51
Table 39: Point at which Harvest was Spoiled, FE, 2020....................................................51
Table 40: Point at which Harvest was Spoiled, Baseline, 2017...........................................51
Table 41: Place of Storing Crops, FE and Baseline ............................................................52
Table 42: Building or refurbishing storage calculation for FE population.............................53
Table 43: Work on the Tasks by Sex at MTE Across for the Four FE Crops.......................53
Table 44: Use of Improved Farm Management Practices, Baseline vs FE..........................54
Table 45: Indicator 8: Application of improved farm management practices, FE.................55
Table 46: Volume sold in kg/ha, Revised Baseline vs FE ...................................................56
Table 47: Value of Crop Sold in $/ha, Baseline vs FE.........................................................57
Table 48: Prices in Malawi Kwacha and USD $, Baseline vs FE.........................................57
Table 49: Growers that Sold Nothing, FE, 2020..................................................................58
Table 50: Primary Means to Sell, FE, 2020 ........................................................................58
Table 51: Sales location, FE HH Survey, 2020...................................................................59
MSIKA Final Evaluation Page vi kadale@africa-online.net
Acknowledgements
The Kadale mid-term evaluation team acknowledges the valuable assistance received from 
the Land O’Lakes Venture37 team. This study required the active support and contributions
of the headquarters staff and the Malawi Strengthening Inclusive Markets for Agriculture 
(MSIKA) team based in Lilongwe, Malawi. Our considerable thanks to Boniface Msiska and 
Lyson Kadzakalowa, who assisted us on all aspects of the work. They responded to our many 
questions and provided both data and insights for our evaluation activities. 
We are particularly grateful for the technical inputs of Damiano Chipeta, Benson Kasekera, 
Nyarai Makwiza and Freeman Ngoma that guided us in the understanding of the technical 
aspects of the work. We also recognise the contributions of Meredith Saggers and Joe 
Carvalho who provided overall direction and guided us on a range of issues. 
Kadale benefitted from its dedicated interviewing team who undertook over a thousand phone 
interviews with all the frustrations of out of service numbers, dying phone batteries, poor 
signal and persuading farmers to give of their time. 
The Final Evaluation Team was led by Jason Agar as Team Leader, Yamikani Chabwera as 
Research Manager, Patience Mtembezeka as Assistant Research Manager, McPherson 
Chatama as Fieldwork Co-ordinator, Mercy Butao as Advisory Agronomist with additional 
support from David Namanjasi on data analysis, all of whom worked long hours under great 
pressure to deliver this evaluation. 
We are grateful to our 49 key informants, but most of all we wish to recognise the 979
Malawians who agreed to be interviewed, some of them for qualitative as well as quantitative 
interviews, to tell us about their experiences of working with the MSIKA program. These 
producers are the intended beneficiaries of the program, who have to tackle the day to day 
struggle to improve their livelihoods in difficult and unpredictable circumstances. 
Jason Agar
Director, Kadale Consultants, Lilongwe, Malawi.
November 2020
MSIKA Final Evaluation Page vii kadale@africa-online.net
Acronyms
AEDC Agricultural Extension Development Co-ordinator
AEDO Agricultural Extension Development Officer
COVID Coronavirus disease (2019)
DADO District Agricultural Development Office/Officer
DAES Department of Agricultural Extension Services 
EPA Extension Planning Area
FBO Farmer Based Organisation
GAP Good Agricultural Practices 
GoM Government of Malawi
Ha Hectare
HH Household
HHH Head of Household
Kg(s) Kilogramme(s)
KII Key Informant Interview
LoP Life of Project
LUANAR Lilongwe University of Agriculture and Natural Resources
MBS Malawi Bureau of Standards 
MEL Monitoring, Evaluation and Learning
MFI Micro-finance Institution
MoAIWD Ministry of Agriculture, Irrigation and Water Development
MoIT Ministry of Industry and Trade
MSIKA Malawi Strengthening Inclusive Markets for Agriculture (Program)
MSU Michigan State University 
mT Metric Tonne (1,000 kgs)
MTE Mid-Term Evaluation
NGO Non-governmental Organisation
PHH (Harvest &) Post-Harvest (Storage) and Handling
PMM Performance Measurement and Management
PMP Performance Management Plan
PPI Progress out of Poverty Index 
PPS Proportional to Population Sample
SPV Special Purpose Vehicle 
TNS/PFS TechnoServe/Partners in Food Solutions 
T&T Techniques and Technologies – also known collectively as ‘practices’
ToR Terms of Reference
USD United States Dollar ($)
USDA United States Department for Agriculture
V37 Land O’Lakes Venture 37
VSL(A) Village Savings and Loan (Association)
MSIKA Final Evaluation Page viii kadale@africa-online.net
Executive Summary
The final evaluation (FE) of the Malawi Strengthening Inclusive Markets for Agriculture 
(MSIKA) was conducted from June to October 2020 by Kadale Consultants Ltd.
Overview of the Project
The MSIKA program was originally a five-year value-chain development program targeting 
36,000 producers, 210 Farmer Based Organisations (FBOs) and 24 processors. It operated 
in five districts in Central and Southern Malawi (Dedza, Lilongwe, Mangochi, Mchinji, Ntcheu) 
and originally across seven crop value-chains (tomato, onion, potato, mango, citrus, guava 
and chili). MSIKA was implemented by Land O’Lakes Venture37 (‘V37’). 
MSIKA is a multi-faceted program with integrated components notably: training beneficiary 
producers in agricultural, post-harvest handling (PHH) and farm management practices, 
along with finance and marketing; strengthening FBOs; working with MFIs and banks to 
increase credit for producers and small and medium enterprises (SMEs); working with 
processors to enhance processing and obtain certification; working with Ministry of Agriculture 
on a new Horticulture Policy and training of district extension staff, establishing new product 
standards with Malawi Bureau of Standards (MBS) and researching improved growing 
practices. 
Following a reduction in available funds from monetization in 2019, the length of the 
programme was reduced to four years and the focus was narrowed to four crops (tomato, 
onion, potato and mango). The number of focal FBOs was reduced from 217 to 50. There 
was also a reduction in the number of demonstration plots from 85 to 15.
Evaluation Approach
The evaluation approached changed when it became clear that face to face interviews were 
not feasible due to COVID. The original household (HH) survey of producers was changed 
to a telephone survey with reduced content, but across more farmers (target of 1,060) to 
reach the desired confidence levels and margin of errors overall and for each crop. The 
sample frame for the survey was 30,922 producers trained in agricultural production as at 
May 31st, 2020. In addition, 52 key informant interviews (KIIs) were planned across the range 
of stakeholders from farmer based organisations (FBOs) to government extension staff to 
processors. In addition, the planned16 focus group discussions (FGDs) were changed to 60 
phone qualitative interviews with individual producers. 
Data is reported against the Life of Project (LoP) targets and the impact baseline. Due to the 
reduced scope of the project and progress at the mid-term evaluation (MTE), the LoP targets 
were revised. As noted in the MTE, several baseline targets were recalculated to enable 
consistent treatment for outliers.
The consultant conducted 1,074 interviews with 979 respondents (some were interviewed for 
two crops separately), 49 KIIs, and 63 qualitative interviews, all by remoted means. The 
producer sample exceeded the required confidence levels and margins of error.
Overall Progress
Across its many indicators, the MSIKA program achieved or made considerable progress 
towards many of its LoP targets as summarised in the tables in Annex 1). Several factors 
have impacted on the program in the last two years, in particular:
1. The life of the program was shortened from five to four years and several planned activities 
were dropped due to a shortfall in the expected resources. It meant that the full 
incremental effects of the activities, such as on yields and particularly sales, could not be 
realized due to the shorter period of implementation.
2. COVID began to affect Malawi towards the end of March 2020 and over the next two 
quarters (Apr-Sep 2020). The Government lockdown restricted movement and markets 
MSIKA Final Evaluation Page ix kadale@africa-online.net
and closed tourism and hospitality venues and institutions that are major users of fruit and 
vegetables. This impacted on producer production and sales, and restricted the operation 
of FBOs, processors and the MSIKA team.
3. The Presidential and Parliamentary election in May 2019 was heavily disputed with 
disruption in the cities that limited activities by MSIKA and disrupted progress with 
governmental bodies. 
4. In March 2019, Cyclone Idai hit Mozambique, with the effects reaching Malawi. The 
effects of Idai were seen in the MTE, impacting yields. While farmers could plant in a later 
season, it sapped their resources. 
The above factors are not intended to provide reasons for shortfalls in performance, which 
was good overall; rather they are a reminder that implementation can be disrupted by multiple 
factors beyond the control of the project team. For MSIKA, these have been multiple factors, 
some of them concurrently, such that more could have been achieved without them. 
The lesson for USDA and V37 is that while it is not possible to predict the specific nature and 
timing of disruptions, there can be consecutive or concurrent disruptions in a volatile operating 
environment like Malawi. This should be considered when planning what is realistic to deliver. 
Key Findings 
FBOs: MSIKA has worked with 217 FBOs across the five target districts, which is above the 
LoP target of 210. Of these, 167 (77%) were formed by MSIKA. 
For Year 4 of MSIKA, based on fewer resources and the MTE recommendation to focus on 
better performing FBOs, MSIKA planned to work with 50 better performing FBOs. This plan 
was disrupted by COVID, though the MSIKA team was able to partially implement this 
narrower focus. From the performance data and FBO KIIs, the consultant’s conclusion is that 
it is difficult to improve the performance of low performing FBOs and that working with these 
dilutes project efforts. The lesson for USDA and V37 is that similar projects are likely to 
make more progress and achieve better returns by moving early to focus on better performing 
FBOs. However, it is noted that determining performance can only be done following some 
engagement and time to assess performance. 
FBOs are useful mechanisms for engaging with producers and some may ultimately become 
sustainable organizations. However, the conclusion drawn from the findings is that it is 
unrealistic to expect that many have the potential to become sustainable entities. The lesson
is to be realistic about the potential to turn most FBOs into well-functioning, viable and self￾sustaining registered entities. This is unlikely in most cases, at least based on the MSIKA 
FBOs.
MSIKA made some progress on governance capacity building and in capacity development; 
however, using V37’s PMM assessment tool, suggests that it would take many years and 
considerable investment to develop them, with a very uncertain success rate. For projects 
like MSIKA, FBOs may best be seen as temporal mechanisms to engage efficiently and 
effectively with farmers than to build into a project’s design plans and resources to try to turn 
them into sustainable entities. If there are a few FBOs that have potential, then a more 
focused support package for the few would be the most realistic approach.
MSIKA has been very successful at increasing individual producer yields and has made 
progress on PHH losses. Individual producers have generally increased their sales. 
However, progress with collective selling via the FBOs has been limited. There is potential 
interest in links between producers and two new processors, but this started late in the 
program and was disrupted by COVID, so did not reach fruition. Linking producers to markets 
is a strong theme for MSIKA and something that many projects that work with producers want 
to achieve to convert success in increasing production volumes (and quality) into sales and 
incomes. This aim is important to both V37 and USDA.
MSIKA Final Evaluation Page x kadale@africa-online.net
The lesson from MSIKA is that linking producers to more formal markets with expectations 
on quality, volumes, timeliness and ‘packaging/presentation’ is very difficult and may depend 
heavily on the ongoing facilitation efforts by a project. FBO and member expectations were 
that MSIKA would help them to find attractive formal markets with better prices, however, 
these expectations look unlikely to be met. 
Although this may be disappointing, producers have still found markets for their produce even 
under COVID, mainly because they have to sell and there are retail and wholesale markets 
they can reach, even if these are sub-optimal in demand and prices. With substantially 
increased production, these producers are still likely to be better off even with lower prices 
from these informal markets. Over time, markets may become more structured with the better 
organised FBOs finding ways to aggregate to supply more formal buyers. That process can 
be facilitated by a project, but it takes time and expectations of supply to formal buyers, such 
as processors and bigger retailers/hospitality outlets should be kept low.
Producers: The FE found that MSIKA succeeded in reaching large numbers of producers 
(39,744) with training in improved agricultural and PHH practices that is valued, efficiently 
delivered and that is effective. The adoption rate of practices has been very high, with 99.5% 
of producers applying at least one new practice. As noted, COVID had an effect on the overall 
numbers and the target would have likely been over-achieved, if it had not occurred. 
The extent of application of practices was impressive, with a mean 15.8 practices per 
producer per crop. This widespread application is likely to lead to higher yields. With better 
weather conditions at FE, another year of training, and opportunity to apply practices under 
lead farmer guidance, and with the narrower focus on the four crops, the results on yield at 
FE have been very good with a 54.4% aggregate yield improvement against a target of 26.0%. 
At FE, the post- harvest handling (PHH) applications rates have improved further from the 
MTE level, with tomato scoring over 90% application on most practices. Mango, like all the 
tree crops in the MTE still lags the field crops, reflecting that tree crops are not seen by most 
producers as crops to actively manage and invest in. The overall effect was that PHH losses 
reduced from 14.8% at baseline to 13.1% at FE. Losses are difficult to measure and so 
caution is needed on any PHH loss data.
Somewhat surprisingly given the disruption of COVID on markets, sales volumes increased 
substantially compared to the baseline, with the highest increase in mean sales volume for 
tomato (+70.0%), followed by potato (+28.8%), onion (+21.1%) and mango (+6.1%). The low 
increase in mango is not surprising, but the big increase in tomato seems to be a function of 
the increase in volume produced, meaning production has translated into increased sales. 
The sales value generated interesting results in comparison with the adjusted baseline. The 
sales of the four crops fall within a similar range with tomato at $1,229/ha followed by onion 
at $1,226/ha potato at $1,195/ha and mango at $996/ha. However, comparing these to the 
baseline there were big increases in sales value for mango (+99.7%), followed by potato 
(+92.7%), onion (+63.3%) and tomato (+44.1%). 
In summary, there has been good progress since the baseline and MTE in application of 
agricultural and PHH practices, with large increases in yields, sales volumes and sales value. 
There has been modest progress on PHH losses. 
MFIs/Banks/VSLAs: MSIKA continued with its SME lending mechanism with a Malawi Bank. 
The initial results at mid-term were positive overall. While there was hope for bigger and 
more flexible second loans, with the winding up of MSIKA and a Presidential declaration on 
a moratorium on SME loan repayments, the Bank made the terms much less attractive and 
deterred several SMEs that had been approved. In the end, the total number and value of 
loans disbursed through the facility was much lower than originally intended, highlighting the 
difficulties of establishing a successful SME loan facility. The lesson is that facilities for SME 
lending are not attractive to all banks/lenders and that banks have their own requirements 
that override what the project is seeking. 
MSIKA Final Evaluation Page xi kadale@africa-online.net
MSIKA’s work with a microfinance institution (MFI) to increase agricultural lending has been 
very positive with both parties regarding the initiative as a success. The MFI has been the 
biggest source of lending to MSIKA participants ($413,100) compared to banks ($193,343) 
and VSLs ($74,592). The approach is low cost for MSIKA and proving to be very effective, 
as the MFI is using its own capital. The MFI has been encouraged by the high repayment 
rates from the well-organised and trained MSIKA beneficiary producers. It has seen that 
there is lower risk in horticulture due to the relatively short growing periods meaning there 
can be up to three cycles a year if there is access to irrigation. The MFI is a well-established 
MFI and will continue to offer these loans to these and potentially other MSIKA beneficiary 
producers after the end of MSIKA. This means the result is sustainable for the immediate 
future. The lesson for projects like MSIKA is that MFIs may be a more fruitful partnership 
than banks. 
MSIKA has established 751 VSL associations (VSLAs_, improving access to finance for over 
10,000 beneficiary producers. VSLs are self-sustaining community managed mechanisms 
with a high degree of continuity. Although the value of loans through VSLAs is small compared 
to the MFI and even the bank, the number of loans is higher than these partners indicating 
that many producers (5,444) are accessing small loans in their localities and are able to get 
a share of the profits on the lending when the VSLAs distribute their funds. The lesson is 
that projects like MSIKA should continue to establish VSLAs and train members in VSL. 
Processors: MSIKA has established relationships with 18 processors. There is high uptake 
of good manufacturing practices, increases in storage, certification to Malawi Bureau of 
Standards (MBS) standards, improvements in processing, adoption of marketing and financial 
management practices. However, the outcomes of this work are relatively limited at FE, 
particularly for purchases from FBOs and additional employment. MSIKA believes that with a 
fifth year that it might have been able to increase the purchases from FBOs, and this might 
happen based on the relationships with new processors. In conclusion, development of 
processing in these value-chains is beneficial overall, however, there is not yet a strong 
connection between the processors and the producers and their FBOs. 
At the outset, MSIKA identified 71 processors as potential partners, but most were micro￾enterprises and the pool of larger and medium processors was limited to less than ten. It has 
proven difficult to make much headway with processors due to their relatively small size. The 
lesson is that MSIKA was better focusing on the five processors that have shown willingness 
and capability to make progress, while recognising that there is still limited potential with 
processors in Malawi at this point in time.
Government: MSIKA engaged GoM on policy, training of extension staff, work with MBS and 
research on crop practices/farmer field schools. Updating the Ministry of Agriculture’s 
Horticulture Policy, moved quickly in the early stages due to the support from MSIKA, but 
slowed because the Ministry lacked the resources to progress the process. This was 
compounded by disruption from the election, as well as the major effects on government due 
to COVID. MSIKA has moved the policy process along and there is a good chance that the 
policy will be adopted in the future, but the lesson is that policy change of this nature is often 
very slow, even with strong commitment within government and prone to delays. 
Some AEDOs/AEDCs have been involved and assisting on delivery of producer training, but 
the GoM extension staff are not a key driver in onward training compared to the lead farmers. 
The lesson is that projects like MSIKA have to include GoM staff, but should focus efforts on 
training and supporting lead farmers as the most sustainable, effective and efficient 
mechanism to reach producers. 
MSIKA worked with MBS to develop 10 new product standards. Despite some delay, the 10 
new standards have been adopted. The lesson is that working on regulations or supporting 
measures may yield more results than working on national policies and strategies, and may 
be of more immediate relevance to the private sector.
MSIKA Final Evaluation Page xii kadale@africa-online.net
MSIKA’s engagement with MSU/Bunda on research was linked to the farmer field schools at 
the demonstration sites. Getting the most out of research requires maximising demonstration 
sites for the farmer field schools. It is unclear at this point how much MSIKA will be able to
leverage this research to bring forward practices that can be readily adopted by many 
producers. The lesson is that research may not be able to bring results in the project period 
and it is better to focus on demonstrating under-utilized knowledge on proven practices. 
The following recommendations are made:
#1 That to undertake a phone survey of producers requires access to a sufficient database 
of phone numbers, as representative enough of the population as possible, and may require 
additional efforts to get female producers included. Considerable thought should be given to 
question inclusion, such as those that may rely heavily on face to face interaction with 
respondents.
#2 That in the design of projects of this type, USDA should consider in its expectations and 
ambitions that there is a high likelihood of consecutive or concurrent disruptive events. While 
these can be listed as risks and assumptions, the flexible response shown by USDA on 
MSIKA can avoid projects scrambling to make up ground and hit pre-determined numbers. 
#3 That projects like MSIKA should aim to identify and select out low performing groups and 
FBOs at an early stage using appropriate tools and drawing on experience.
#4 That projects like MSIKA should not aim to turn FBOs into sustainable entities in most 
cases, but be highly selective in deciding which might have potential, and to see most FBOs 
as temporal entities for efficient and effective project delivery that may or may not continue 
post project.
#5 That projects like MSIKA should seek to facilitate sales between better organised FBOs
only where there is a realistic capacity to supply and members are committed to collective 
selling and that there are accessible markets. Where that is not possible, there is still merit 
in working to increase production for sale by individual producers into less formal markets.
#6 That because Banks are regulated and relatively cautious, projects like MSIKA need to 
consider other more flexible and responsive lenders, such as well-established MFIs. 
#7 That VSLs should continue to be a focus for projects like MSIKA as they provide ready 
access to finance and over time, the VSL builds trust and cohesion between producers. 
#8 That projects like MSIKA should find an appropriate balance between working with enough 
processors, but of sufficient size and capacity to invest. Working with many micro- and small￾enterprises is resource intensive with limited potential to deliver results. Work with smaller 
enterprises should be limited to training and general information rather than more intensive 
support relationships.
#9 That projects like MSIKA embarking on support for developing government policies and 
strategies need sufficient time and resource to see such processes through to the end and 
that it should be accepted at the outset that there is a risk that these processes may not yield 
results within the project life or at all.
#10 That projects like MSIKA should focus on lead farmers as a more sustainable, effective 
and efficient mechanism for delivering services, particularly knowledge, than government 
extension staff. However, government structures should be engaged and government staff 
included in training and activities where they can support of lead farmers.
MSIKA Final Evaluation Page 1 kadale@africa-online.net
1 Introduction and Background
This report is the independent Final Evaluation of the Malawi Strengthening Inclusive 
Markets for Agriculture (MSIKA) undertaken by Kadale Consultants Ltd (‘Kadale’) from 
mid-June through to the end of October 2020. 
This evaluation has been conducted for the United States Department of Agriculture
(USDA), with the active co-operation of Land O’Lakes Venture 37 (V37), which is 
implementing the MSIKA program.
This section sets out the objectives of the final evaluation, an important note on the 
impact of COVID on the evaluation design and implementation, a brief background 
about MSIKA and some of the key changes to the project, notably following the mid￾term evaluation (MTE) in late 2019 and the advent of COVID from March 2020.
1.1 Objectives of the Final Evaluation 
The objectives of the final evaluation are as follows: 
1. Assess the relevance of the project strategy and approach to project 
participants;
2. Measure progress the project has made toward key results, including 
effectiveness and efficiency of interventions in achieving established targets; 
3. Identify enablers and constraints to progress (both internal and external factors) 
that have supported or limited success of the project;
4. Assess the sustainability of project outcomes, and the effectiveness of 
sustainability efforts that were undertaken by the project;
5. Assess operational aspects of the project, such as project management and 
monitoring and evaluation; 
6. Document lessons learned, challenges and unanticipated effects;
7. Provide recommendations for best practices that were realized or areas of 
improvement for future USDA and Venture37 program in similar areas and 
value chains. 
1.2 COVID Impact on Evaluation Design and Implementation 
During the implementation of the final evaluation, the impact of COVID on Malawi 
necessitated restrictions on movement, making it impossible to undertake face to face 
interviews. As a result, the methodology and instruments had to be redesigned to be 
based on a phone-based household (HH) survey of producers. The planned focus 
group discussions (FGDs) with producers were replaced by phone based qualitative 
interviews and all the other planned KIIs were also to be conducted by phone. 
This approach created new challenges, a major one was determining the phone 
numbers for producers, which took the MSIKA team considerable time and effort to 
gather from the FBOs. The implications were a delay in the original timescales for 
delivery, and removing questions in the data collection tools to limit the time for the 
interviews. This meant that the final evaluation is not be able to report on three of the 
project indicators. USDA provided approval for this change. 
This change was not ideal, but was necessary to make the data gathering manageable 
for the respondents. Although this was disruptive to the FE research process, the 
consultants believe that the results are still valid and robust. This has been an
opportunity for the consultants to learn from taking a different approach in what has 
been a very challenging operating period.
MSIKA Final Evaluation Page 2 kadale@africa-online.net
1.3 Background to MSIKA Project
MSIKA was originally a five-year, $17.0 million project implemented by Land O’Lakes 
Venture 37 (V37), with funding from the United States Department of Agriculture’s 
(USDA) Food for Progress Program. MSIKA was implemented in partnership with 
TechnoServe and Michigan State University (MSU). 
The goal of MSIKA was to improve the commercialization of fruit and vegetable sectors 
for value addition, and ultimately increase domestic and regional trade. MSIKA had
two strategic objectives, in line with USDA Food for Progress Results Framework. The 
first was to increase agricultural productivity in the fruit and vegetable sector by: 
increasing the availability of improved inputs, improving infrastructure to support on￾farm production, facilitating access to finance, and training farmers on improved 
agricultural and farm management techniques and technologies. The second was to 
expand domestic and then regional trade of fruit and vegetable products by: expanding 
access to finance; improving quality of post-production products, training producers 
and processors on improved post-production processes, facilitating improved linkages 
between buyers and sellers, improving market and trade infrastructure, and facilitating 
improved management of buyer/seller groups. Please see Annex 2 for MSIKA results 
frameworks. 
MSIKA’s theory of change involved strengthening the capacity and linkages between 
farmers, processors and input providers, such as financial institutions, extension 
services, research institutions, and agribusinesses to develop the target horticulture 
crop value chains. MSIKA built the capacity of processors to meet quality standards 
and improve their efficiency, and linked them to financial institutions and horticulture 
producers to be able to expand their production. MSIKA created a special purpose 
vehicle (SPV) with a Malawian bank to provide low costs loans to processors. MSIKA 
also worked with the Government of Malawi (GOM) to create and codify quality 
standards for horticulture to promote compliance amongst processors and improve 
reputation of Malawi products. With increased access to finance and raw materials, 
and improved quality and processes, processors were able to increase their production 
and sales. 
MSIKA also worked to create and strengthen horticulture Farmer Based Organizations 
(FBO) to provide an access point to engage and train tens of thousands of producers
through lead farmer and extension staff training of trainers. MSIKA engaged with the 
research community to test improved horticulture varieties and techniques through 
demonstration plots to create the curricula for this training. MSIKA also linked 
producers to local agribusinesses to secure improved inputs, microfinance and formed
village savings and loans associations to improve their access to finance. Improved 
knowledge, access to inputs, and access to finance to purchase the inputs, producers 
improved their practices and ultimately productivity of the target crops. MSIKA also 
linked the FBOs to market so that they could sell their additional product and ultimately 
increase their sales. 
There were eight specific areas of activity planned (see Scope of Work – Annex 3):
1. Training: In improved agricultural production techniques
Provision of training, technical resources, and cutting-edge research to fruit and
vegetable farmers to improve technical and business skills. Build capacity of lead
farmers and extension agents from the Ministry of Agriculture, Irrigation and Water
Development(MoAIWD). Build capacity through Farmer Field Business Schools (FFBS)
and demonstration plots (Yankho PlotsTM) on topics such as orchard management,
budding and grafting, nursery establishment, pruning, and weeding. Ensure that FFBS
and Yankho PlotsTM build core climate smart agronomic and business skills of target
producers, in partnership with MSU. Engage private sector partners to ensure increased
MSIKA Final Evaluation Page 3 kadale@africa-online.net
utilization of irrigation technologies and increased access to improved inputs. E ngage
private sector partners to co-invest in Yankho PlotsTM and to establish for-profit nursery
enterprises. 
2. Infrastructure: Post-harvest handling and storage
Improve PHH and storage infrastructure by providing training and technical assistance,
facilitate market linkages, and support access to finance to improve on-farm and off-farm
post-harvest infrastructure and to increase the use of improved post-production
processing and handling practices. Facilitate increased adoption of established
standards, such as Global Good Agricultural Practice (Global GAP), by market actors,
such as traders, processors, and producers.
Facilitate Farmer-Based Organizations (FBO) and private enterprises to develop
wholesale markets, and aggregation centers, and to utilize technology and practices
such as optimal packing technology, improved storage practices, and sorting, grading, 
and washing technology. 
3. Training: Post-harvest processing
Improve the efficiency and profitability of value-added post-harvest manufacturing and
promote the utilization of value-preserving practices and technologies throughout the
fruit and vegetable value chains, in partnership with Technoserve. Provide targeted
technical assistance to small, medium- and large-scale fruit and vegetable processors. 
Technical assistance will focus on processing line operational efficiency, improved 
packaging and labelling, modernization of physical equipment and plant facilities, and
improved business and financial management practices. Train fruit and vegetable
processors on improved product quality standards and facilitate Malawi Bureau of 
Standards (MBS) certification. Facilitate FBO access to improved PHH techniques. 
Facilitate improved phytosanitary practices by agricultural enterprises and FBOs
engaged in small-scale processing of fruits and vegetables into products.
4. Capacity Building: Producer groups and cooperatives
Apply cooperative development tools, including V37’s AgPrO products, to improve farm
management and to increase the capacity of FBOs in agricultural production, post￾harvest handling, and processing. Link FBOs to business development service 
providers and finance providers, and collaborate with national-level producer
associations and district-level FBOs (such as the Mchinji Horticulture Association) to
strengthen the delivery of member services and facilitate the transformation of FBOs into
agricultural enterprises.
5. Market Access: Facilitate buyer-seller relationships
Improve market access by facilitating buyer-seller linkages, increasing access to
market information, and strengthening the capacity of key organizations in the trade
sector. Facilitate buyer-seller networking events, linkages to transporters the
establishment and strengthening of equitable outgrower schemes between producers
and processor, stronger linkages to retail supermarkets for standards and specification
trainings and procurement, promotion of village-based input sales agents, and targeted
support for women- and youth-owned small and medium enterprises (SMEs) engaged
in food processing.
6. Financial Services: Facilitate agricultural lending
Build the capacity of a local financial institution to conduct risk assessments and
develop financial products for value addition in the fruit and vegetable sectors, such as
working capital and trade finance for processing, aggregation, irrigation, export, and
transportation. Facilitate increased lending for climate smart irrigation; work with
MSIKA Final Evaluation Page 4 kadale@africa-online.net
financial institutions to develop loan products appropriate for horticulture value-adding
activities; facilitate the development of input credit relationships between processors
and producers or FBOs; improve the financial literacy and credit- worthiness of FBOs;
and facilitate the development of capital investment plans by fruit and vegetable
processors for plant modernization and expansion, including improvement of business
and financial managementsystems.
7. Financial Services: Provide SME finance
Stimulate increased agricultural lending by establishing a Special Purpose Vehicle
(SPV). The SPV will make loans to financial institutions that will on-lend those funds
to enterprises, farmers and farmer organizations. The SPV may catalyze further
investment in value-adding activities and trade, such as loans from the Overseas
Private Investment Corporation (OPIC). Ensure that loans originating from the SPV,
prior to the award’s date of completion, will be used for purposes such as increasing
access to working capital for farmers and SMEs for input purchases, transport for
space arbitrage of fruits and vegetables, market and trade infrastructure, processing
equipment, and equipment upgrades. After the award’s date of completion, ensure
that loans originating from the SPV will be used to expand agricultural production and
trade in African Least Developed Countries, as defined by the World Bank. For the life
of the SPV, principal and interest repaid to the SPV for continued lending and
investment.
8. Government Capacity Building: Improve Enabling Environment
Build the capacity of key government research and extension agencies, including
MoAIWD and the Department of Agricultural Extension Services (DAES). Work closely 
with the Lilongwe University of Agriculture and Natural Resources (LUANAR), in 
partnership with MSU, to conduct research on soil fertility and other value chain specific 
trials. Venture37 will build the capacity of processors, traders, national associations,
and FBOs to advocate for improved regulations and policies. 
Work with the MBS to jointly establish and disseminate international quality standards.
Conduct a gap analysis and recommend improvements to bring MBS’s standards,
certification processes, and quality audit processes up to international levels. 
The key targets were to improve yields and sales through market linkages for 36,000 
farmers through 210 FBOs across five districts.
1.4 Key Changes to MSIKA 
In the course of implementation, there was a budget reduction to $12.2 Million due to 
a shortfall in expected funded through monetization, which necessitated re-strategizing 
and modification of the approach. It was decided that MSIKA would come to an end at 
the end of Year 4, a year earlier than the original plan. Following the recommendations 
of the Mid-Term Evaluation (MTE) and related to the budget reduction, there was a de￾prioritization of guava, chili and citrus; increased focus on continuing activities 1, 4, 5 
and 7; scaling down of activity 6 to focus on the MFI and Village Savings and Loan 
Associations (VSLAs) only; and wrapping up of activities 2, 3 and 8 in Q1 of Year 4. 
The MSIKA team also scaled down the number of participants worked with in Year 4, 
focusing on 50 FBOs representing around 8,000 farmers, and scaling down the farmer 
field schools and demonstration plots from 85 to 15. The modification resulted in
revised indicator targets to reflect the changed focus and resources and these are the 
indicators against which MSIKA is evaluated in this final evaluation. 
Making such major changes part way through a program brings challenges. These
were compounded by the advent of COVID-19 (COVID), which began to impact on 
MSIKA Final Evaluation Page 5 kadale@africa-online.net
Malawi from mid-March 2020, at the end of the project’s Year 4 Q2 and then through 
to the end of the project. From the end of March 2020, the Government of Malawi 
(GoM) introduced measures to reduce the spread of COVID, primarily restrictions on 
movement within Malawi particularly from the cities to the districts, limits on the size of 
meetings/gatherings and closing of institutions, many of which were the main markets 
for fruit and vegetable producers. A further consequence was the restricted movement 
of project staff, hence most of the project activities from March onwards were
coordinated remotely. The need to keep project staff and participants safe took 
precedence. This presented MSIKA with further considerable operational challenges, 
and resulted in further changes to the revised plan of activities. 
This provides the context against which this final evaluation took place. 
2 Methodology and Implementation
This section sets out the methodology and how the final evaluation was conducted. 
2.1 Methodology 
2.1.1 Overview
A key factor for the final evaluation (FE) was the effect of COVID on the original plans. 
With the arrival of COVID in Malawi, it became necessary to move from the planned 
face to face household (HH) surveys, focus group discussions (FGDs) and key 
informant interviews (KIIs) to a revised approach based around remote interviewing 
(by phone, mainly). This also changed the sample size and selection. These points 
are covered in the respective sections below.
Overall, and as planned, this FE adopted a mixed method, non-experimental approach, 
combining quantitative data from a household survey of participant farmers, with 
qualitative interviews (replacing the original intended FGDs) and KIIs with the MSIKA 
team and a wide range of stakeholders. The stakeholder groups interviewed were: 
high-, medium- and low-performing Farmer Based Organisations (FBOs); processors 
and input suppliers small and medium enterprises (SMEs); GoM staff in the districts; 
and financial institutions. The consultants also reviewed MSIKA documents and 
monitoring data, the impact assessment baseline report and data and the mid-term 
evaluation (MTE) report and data.
1
 
For the quantitative work, the team adopted a non-experimental, pre-post design to 
measure the key performance indicators. The pre-intervention data came from the 
impact baseline study that interviewed a random sample of expected MSIKA 
participants, as well as producers outside the target areas as a control group2
. The 
consultant compared the expected MSIKA participants interviewed at baseline with a 
random sample of actual MSIKA participants randomly selected from a database of all 
participants provided from the project. The quantitative design compares estimates of 
the two independent samples of the MSIKA at baseline and at FE. 
It was agreed with USDA and Venture37 that the consultants also should look at the 
effects of the COVID-19 pandemic that had reached Malawi and which led to measures 
by GoM to limit movement and business. Questions around the impact of COVID
replaced the questions around the impact of Cyclone Idai that were asked in the MTE. 
1 Kadale conducted the MTE and provided the data collection to TANGO who undertook the baseline. 2 The initial evaluation plan called for a quasi-experimental design, where the final evaluation would 
have followed up with both participants and a control group, however, due to the budget shortfall, 
the design was updated to a non-experimental design. 
MSIKA Final Evaluation Page 6 kadale@africa-online.net
2.1.2 Household Survey
The key activities for the design and delivery of the FE HH survey were:
Tool Development
The consultants conducted KIIs with MSIKA staff to discuss activities implemented by
the project and how these should be captured through the FE HH survey. The 
consultants revised the HH survey used for the MTE, which was based on the baseline 
instrument. so that accurate comparisons could be made. The main changes in the 
survey were 1) the removal of questions related to the three crops (citrus, guava and 
chilli) that were deprioritized in Year 4; 2) the removal of questions on knowledge of 
practices and hectares per practice, and asking about only one crop per interview to 
reduce the length of the survey; 3) updating the practices to focus on those in the 
MSIKA training packages. Note that not all practices in the baseline were in the MSIKA 
training package, and there are practices that are asked about in the FE that are not 
in the IB. It is not surprising that there were changes, as a baseline is anticipating what 
will be done in the future. The consultants sought to maintain as much comparability 
as possible with the IB and with MSIKA’s own semi-annual surveys. 
Out of the four focal crops in year four, the three field crops (tomato, onion, Irish potato)
have up to three growing/harvesting cycles per year depending on the producer and 
their circumstances. For the remaining prioritized tree crop (mango) there is only one 
growing/harvesting cycle. The HH survey instrument had already been expanded 
during the mid-term evaluation to capture land area, yields, losses and sales for all 
growing cycles for each of the field crops. 
The team translated the instrument into Chichewa and coded it into Open Data Kit
(ODK) for tablet-based data collection. The instrument was piloted through telephone 
interviews to check the flow of the questions and to determine whether the interviewing 
time would be appropriate. The instrument was further refined based on the results 
from the piloting. MSIKA staff were fully consulted on the technical aspects of the 
instrument and provided inputs at each stage. 
Overall, the consultants’ view is that there is good alignment between the FE and 
baseline instruments and data. 
The final instrument is included in Annex 4. 
Training 
Enumerators were trained for three days from 24-26th August 2020. This included a 
telephone-based interviewing practice and on the final afternoon. The training 
consisted of the background to MSIKA, running through the survey section by section 
to clarify questions and responses, mini-quizzes and practicing questions and 
responses. An agronomist provided technical explanations for all the practices so that 
enumerators would understand the agronomy of the crops. There were sessions on 
use of ODK and tablet collection, conduct during the phone interview, sampling,
substitution protocol, quality control/checking procedures and data uploading. The 
telephone based interviewing practice tested that the enumerators could administer 
the instrument and record data on the tablets. It also tested whether the interviewing 
time was within the planned 45 minutes and if there were any issues with telephone￾based interviewing. At the end of the training, Kadale selected 14 of 16 trainees for two 
teams of one supervisor and six enumerators.
Sampling
MSIKA shared an initial database with 38,384 project participants. The MSIKA team 
and the consultants discussed the sample frame for the producer HH survey at the
inception meeting and decided that it should be all producers trained in agricultural 
production as of May 31st, 2020. This was confirmed by the MSIKA team to be 30,922 
MSIKA Final Evaluation Page 7 kadale@africa-online.net
participant producers. The MSIKA team indicated that the difference between the two 
totals is accounted for by some producers that have received other forms of training, 
such as financial literacy, marketing and quality standards, plus non-farmer participant
groups that were trained, such as project staff, government extension staff and other 
stakeholders. The sample for the producer HH survey was therefore based on the 
30,922 farmers. 
MSIKA requested that the sample of producers represent the sample frame with 95% 
confidence overall and be representative at the value chain level with 90% confidence. 
Due to the restructuring of the evaluation from field based to telephone interviewing, 
which resulted in conducting an interview for one crop per farmer, the focus of the 
sample size was directed towards achieving the required minimum number of 
interviews for each crop, as that meant the overall sample would exceed the required 
95% confidence level.
Across the crops, it was agreed that the sampling approach for each crop would be 
calculated at 90% confidence level and 5% margin of error (MoE) according to the 
population of the farmers growing that particular crop. The sample was not designed 
to be able to report results by districts, but by crop as significant differences between 
districts were not expected. Getting a broadly proportional district split was to ensure 
that the sample was indicative of the districts that MSIKA works in.
The consultants used a two-stage approach to sampling. First, Kadale randomly 
selected two FBOs (with a minimum of 12 members) per EPA. If an EPA only had two 
FBOs, they were automatically selected. Second, Kadale randomly sampled 12 
participants from each FBOs using a random number generator. Kadale oversampled 
individuals from FBOs to make sure there would be back-ups if the originally selected 
participant could not be reached. The maximum number of interviews to be conducted 
per FBO was to ensure that the achieved sample would not be skewed more towards 
a few FBOs or EPAs.
With the change from face-to-face to phone interviews, the consultant needed to obtain 
phone numbers for potential respondents. The MSIKA team only had phone numbers 
for about 1/3 of the participants in their database. To reduce the bias that this might 
call, MSIKA made concerted effort to get phone numbers for more participants. Kadale
used the above method to draw a short list of 8,000 participants. The 8,000 participants 
were drawn from a population of 30,922 farmers that had undergone basic agricultural 
training as of 31st of May 2020 (as shared on 2nd July 2020). MSIKA staff worked with 
the FBOs to get numbers for as many of those participants as possible. In the end, 
MSIKA was able to get phone numbers for 2,973 participants from the list of 8,000 
listed participants. This became the final list of participants to attempt to interview.3
 
Lastly, Kadale needed a method to select which crop to ask each participant. During 
the actual data collection, supervisors were given printed random crop selection 
matrices which clearly elaborated on the crop each farmer was to be interviewed for 
depending on the crops the farmer had mentioned to have grown and harvested within 
the specific time period of interest. During each successful call, the enumerators asked 
the farmers to list the crops that s/he had grown. Once these had been listed, the 
enumerators randomly selected which crop to ask about based on the matrix. All 
farmers that only grew one of the focus crops were interviewed on that crop and this 
was considered as a random selection. To ensure an appropriate sample size for each 
of the crops. Kadale used a quota method. Once the random selection method resulted 
in enough interviews on a crop, that crop was removed from the random selection list. 
In summary, for each district, the sampling approach involved identification of EPAs, 
selection of FBOs in each EPA, selection of farmers per FBOs, identification of crops 
3 It was expected that some of the numbers would be unreachable.
MSIKA Final Evaluation Page 8 kadale@africa-online.net
grown by each farmer in the selected FBO, random selection of one crop from the 
stated crops and interviewing.
Sample size
The sample size calculation formula for the specific crops was:
n = N*X / (X + N – 1), where X = Zα/22 *p*(1-p) / MOE2
Through this sample size determination formula, the minimum sample at 90% 
confidence level and 5% MoE for tomato, onion, Irish potato and mango was calculated 
at 268, 264, 266 and 262 respondents respectively. This was adjusted to a target of 
276, 272, 274 and 270 for tomato, onion, Irish potato and mango farmer interviews
respectively, to allow for responses that might be unusable/incomplete. The interviews 
were split by crop based on the explanation above, and set out in Table 1 below.
Survey Implementation
The survey was implemented over a 17-day period (27th August to 17th September
2020), which is similar timing to the IB. MSIKA staff notified village leaders and FBO 
leaders during the phone number collection period to enable participants to be mad 
aware that they could be called, but were otherwise not involved in the survey. 
All the participants for whom there was a phone number were called. Many were not 
reachable, despite at least three attempts to reach each participant. This was 
problematic, especially in Mangochi District partly due to network problems and that 
most participants in the district were registered collectively under a phone number for 
someone else, so if a number could not be reached, it meant all those listed for that
that number could not be reached. Similarly, for those whose number could be 
reached, the owner of the phone sometimes had challenges finding all the individuals 
which had registered under that number. In total after calling all numbers that had 
been listed, 979 participants were reached and interviewed. 
This number of respondents was sufficient for the overall sample to reach the desired 
confidence level and MoE, however it was insufficient to reach the target level for each 
crop. Therefore, h 95 individuals, who reported a second crop of the type which had
a shortfall, were interviewed twice on different crops. This resulted in 1,074 crop 
interviews in total from 979 respondents. Table 1 below sets out the planned interviews 
against the achieved interviews per crop. 
Due to the above issues, the field time was extended to complete the work across the 
value chains, whilst also ensuring that each district had a reasonable number of 
interviews. Overall, the sample is valid and robust. 
Table 1: Minimum and Actual Responses by Crop
Interviews per crop
Crop Minimum Actual Difference
Tomato 268 276 8
Onion 264 264 -
Irish Potato 266 266 -
Mango 262 268 6
Total 1060 1074 14
MSIKA Final Evaluation Page 9 kadale@africa-online.net
Each crop’s planned target was met at the 90% confidence level and 5% MoE. The 
actual sample by district is set out in Table 2 below. 
Table 2: Usable Responses by Crop and by District
District Tomato Onion Irish 
Potato Mango Total: all
crops
Dedza 41 32 76 29 178
Lilongwe 84 58 43 35 220
Mchinji 56 82 79 68 285
Ntcheu 53 53 51 91 248
Mangochi 42 39 17 45 143
Male Producers 158 149 145 129 581
Female Producers 118 115 121 139 493
Total 276 264 266 268 1074
Min sample 90% confidence, 5% MoE 268 264 266 262 1060
Min sample 90% confidence, 5% MoE 68 68 68 68 272
Data Cleaning, Analysis and Comparison with the Baseline
The data was cleaned and tabulated to provide a first-cut analysis. The IB sample was 
drawn from registered farmers who were expected to become MSIKA participants, 
while the FE survey was drawn from actual participants who had at least undergone 
basic agricultural training, so some differences were expected. Kadale tested the 
comparability through a regression analysis based on key characteristics, such as land 
size, education level, and household size. The analysis showed that there was an 
acceptable level of comparability between the samples. The profiles of the baseline 
and FE samples are compared in the findings (section 3.2.1).
When the consultants reviewed the Impact baseline analysis files in the MTE, it 
identified that the baseline analysis had removed high-end outliers for tomato, but not 
for the other six crops. To ensure comparability and to determine the level for outliers, 
the consultants used the Food and Agriculture Organisation’s (FAO) annual maximum 
crop yields based on the application of Good Agricultural Practices (GAP) in 
kilograms/hectare (kgs/ha)4 for vegetables and kilograms/tree for fruits. These maxima 
were applied, involving the removal of outliers above the specified level and resulting 
in re-calculating the baseline. Similarly, during the FE, Kadale applied the same 
maxima as those of the baseline to ensure comparability of the findings and to ensure 
that only realistic yields were taken into account during analysis.
The consultants also noted that the baseline asked only for the “most recent harvest” 
cycle, not all cycles for the previous 12 months. Although the Word version of the 
baseline instrument referred to the previous 12 months, the final Excel/ODK version 
referred to the most recent harvest. It is the latter that was implemented, though it is 
not clear how the change occurred. As a result, the consultants make comparisons 
between the baseline and FE samples based only on the most recent 
growing/harvesting cycle. 
For the FE, all knowledge related questions were taken out so as to shorten the 
interviewing time. Therefore, information to be derived from knowledge related 
questions will make reference to the findings of the MTE. 
In the MTE, the consultants noted that there were price outliers in the baseline that 
had not been removed. These had resulted in some high sales values, beyond what 
4 These are set per crop in the context of Malawi as a maximum yield in kgs/ha or kgs/tree.
MSIKA Final Evaluation Page 10 kadale@africa-online.net
is realistic for these types of households. The consultants adjusted the baseline values 
in the MTE to remove these, so that comparisons are made with this adjusted baseline. 
There is a considerable amount of data in the dataset. Kadale has sought to set out in 
the findings (section 3) what it thinks is most useful, particularly focusing on Life of 
Project (LoP) target achievement for project indicators.
2.1.3 Farmer Qualitative Interviews
The original plan was that the Kadale team would facilitate FGDs with participant
producers to collect qualitative information to support the quantitative survey and 
project data. However, due to the restructuring of the FE from field to remote 
interviewing, it was agreed with the MSIKA team to replace these with 60 qualitative 
KIIs of beneficiary producers growing the four target crops in Year 4. 
Kadale, with inputs from the MSIKA team, developed the guide for the qualitative 
farmer interviews derived from the MTE FGD guide shortened to be suitable for a 
phone interview. The guide is included in Annex 4 
In each district, 12 farmers were to be sampled, ideally with a split of six males and six
females, and three interviews per crop. This was to ensure that the responses would 
cover each district, sex and crop. These farmers were sampled purposively. 
The farmer qualitative interviews were conducted by members of the consultant team 
with qualitative experience. Despite considerable challenges finding mango farmers 
who had received training on tree-crop techniques and technologies, the consultants 
managed to conduct 63 KIIs which surpassed the planned interviews by three. 
Overall, the consultant was satisfied with the quality of discussion in the KIIs and were 
able to gather the diversity of views and qualitative inputs that were necessary.
Table 3: KIIs Conducted by Crop, District and Sex
Crop Sex District
Dedza Lilongwe Mchinji Ntheu Mangochi Total
Tomato Male 1 3 2 2 2 10
Female 2 1 2 1 2 8
Onion Male 2 1 2 1 2 8
Female 1 2 1 2 1 7
Irish Potato Male 2 1 2 1 2 8
Female 1 2 1 2 1 7
Mango Male 1 2 1 3 1 8
Female 2 1 2 1 1 7
Overall Male 6 7 7 7 7 34
Female 6 6 6 6 5 29
Total 12 13 13 13 12 63
The consultants analysed the interviews by themes. This was done by summarizing
the information provided by the farmers using a separate matrix template for each crop. 
Once the write-ups or summaries were finalized, a separate column was added to 
further summarize the common themes that emerged under each question. 
2.1.4 Key Informant Interviews 
The consultants conducted key informant interviews (KIIs) with stakeholders to 
understand their engagement with MSIKA, obtain feedback on the implementation, and 
determine the outcomes that they have seen or achieved resulting from the program. 
MSIKA Final Evaluation Page 11 kadale@africa-online.net
KIIs were planned with the following key stakeholder groups:
1. MSIKA monitoring and evaluation staff, technical staff and management;
2. FBOs: one high-, medium- and low-performing FBO per district;
3. Processors and business partners that have been supported;
4. Agricultural Extension Development Coordinators (AEDCs) and Agricultural 
Extension Development Officers (AEDOs) in the target districts;
5. Financial institutions that have partnered with MSIKA;
6. Govt. Ministry Headquarters and Research Scientists; and
7. USDA.
The consultants developed different KII guides for the discussion with each type of key 
stakeholder. These were tailored to the specific details of the stakeholders as required, 
but shortened so as to conduct these remotely. See Annex 4 for the KII instruments. 
The KIIs were conducted by the team leader and the central team members, as most 
appropriate based on logistical and language reasons. The consultants were able to 
conduct most of the intended KIIs. The exceptions were 1) one processor who kept 
changing appointments, despite making multiple attempts and specific efforts to 
contact them; 2) the Ministry of Agriculture HQ as the information on policy progress 
was obtained from the MSIKA team and 3) Bunda academics, as the research team’s 
assessment was that the report on the research was sufficient. In the latter cases, the 
consultants drew on interviews conducted with those organisations in the mid-term. 
The KIIs with the FBOs and processors were not intended to be representative but to 
gather qualitative insights and validate monitoring data collected by MSIKA. 
Table 4: KIIs Planned and Conducted by Category
Organization Planned Actual Comment
FBO Comments 15 15 All 5 districts over the phone
Ministry of Agriculture, 
District-based
10 11 All 5 districts over the phone
Ministry of Agriculture, HQ 1 0 Not completed - a sufficient progress 
update obtained from MSIKA staff
LUANAR, Research scientists 1 0 Not completed - a sufficient progress 
update obtained from MSIKA staff
Financial institutions 2 1 MFI completed as the priority
Malawi Bureau of Standards 1 0 Unable to complete
Processors 12 11 One processor kept changing the meeting 
time, while another provided part 
information
Other SMEs 2 3 Input suppliers
MSIKA M&E team 1 1 Multiple engagements
MSIKA management 2 2 Multiple engagements
MSIKA technical team 4 4 Single and multiple engagements
USDA 1 1 Completed prior to fieldwork 
commencement
Total 52 49
2.1.5 Review of MSIKA Documents and Data
The MSIKA team provided the consultants with key documents and data at the start of 
the evaluation. These included
• MSIKA program description (original and updated)
MSIKA Final Evaluation Page 12 kadale@africa-online.net
• Evaluation plan (original and updated)
• Performance monitoring plan (PMP) (original and updated)
• Internal Monitoring Evaluation and Learning (MEL) Plan (original and updated)
• Lists of participants and key stakeholders for the sample frames 
• List of FBOs and training conducted at FBOs
• Semi-annual reports to USDA
• Semi-annual data collection tools and data
• MSIKA annual workplans
• the baseline report, database and instruments
• FBO capacity assessment SOW, tools and data 
The evaluation team reviewed these documents and data to inform the evaluation. 
Insights from this information that support the evaluation’s primary data collection are 
reported in the findings section. The role of this documentation and data was to inform 
the development of the instruments and the sampling frame, as well as to provide 
information on other results not directly measured by the survey, farmer qualitative 
interviews and KIIs. Throughout the process, clarifications were requested by the 
consultants and were addressed by the MSIKA team. 
2.2 Limitations 
This was an unusual evaluation process because the COVID pandemic led to the
introduction of restrictions on movement, meetings and operation of institutions that 
affected the data collection. The research was substantially redesigned part way 
through, with the major changes being a shift to a telephone survey for participant
producers and other KIIs and converting FGDs into qualitative interviews.
This contributed to the following limitations:
1. Reduced information on certain topics
The consultants and the MSIKA team had to adapt to a phone-based interview by 
revising all tools, primarily to shorten them after having prepared them based on face 
to face. There had to be several revisions to reduce the length to be manageable in a 
phone interview, particularly with participant producers. The result was that the 
questionnaire only tackled one crop for each participant, and sections on knowledge, 
gender and finance were omitted, while others were reduced in scope.
For some of the topics omitted or compressed, the ultimate effect is limited. For 
example, knowledge is a stepping-stone to application/use, so as long as application 
and use is measured, there is a reasonable conclusion that producers would ‘know’ 
the practice concerned. It would be useful to have the knowledge measured, but 
inferences can be drawn from the MTE results and the qualitative producer interviews. 
For some topics, the missing information would be helpful, like gender and access to 
finance, but these are not measured by specific indicators or the information can be 
drawn from other sources. 
Overall, the consultants felt that there was sufficient information, including drawing on 
MSIKA data and the recent MTE, to inform this final evaluation.
2. Potential producer survey bias
For a phone-based survey, potential respondents need to have a cell-phone or access 
to one, perhaps shared with others. Ownership of a phone might have biased the 
research, perhaps through respondents being better off, and possibly with a male bias.
The exercise to collect phone numbers was challenging, with numbers obtained for 
37% of respondents. The consultants sought to remove any bias within this group by 
MSIKA Final Evaluation Page 13 kadale@africa-online.net
ensuring a sampling protocol that enabled them to work through all the FBOs that were 
in the short list. Ultimately, all those with a number were called, with around 36% of 
them being reached. Reasons for non-response were rarely refusals, but that the
numbers were not reachable. 
The profile of the FE sample was comparable with the baseline, though there is over￾representation of male respondents relative to the sample frame. From the MTE, 
farming was commonly a joint household activity that typically involves both men and 
women in production, even if the amount of input varies, and that men tend to be more 
prominent in sales and use of the income. The over representation of men (54% in 
sample vs 45% in the population) probably reflects the ownership and control of the 
cell phone. Overall, this difference in male:female representation was significant and
is not ideal, but is not regarded as substantively affecting the results, partly because 
men and women tend to operate jointly (see MTE findings on gender). 
There were difficulties getting respondents from Mangochi District and for the mango
value chain, requiring a degree of purposive contacting/contact chasing. In the end, 
the consultants managed to get sufficient responses for each crop, but there were 
fewer responses from Mangochi compared to the plan. Overall, this may have 
introduced some minor limitations on the mango and Mangochi data compared to other 
crops and districts.
3. Challenges of producers estimates
Producers were asked about the size of new and refurbished storage units installed in 
the last 12 months. Following analysis of this data, it became clear to the consultants 
that there was likely to have been an over-estimation of the size of the storage by 
producer participants. Achievement was much higher than the trend from MSIKA 
reports and the MTE. Steps were taken to address this, but the final data still appears 
to be much higher than is likely, though it is expected that the indicator was genuinely 
exceeded. This exercise highlights that there can be difficulties in getting valid 
responses for certain types of questions, such as measurement/estimations through a 
phone interview. If the enumerators had been face-to-face with the respondents, they 
could have viewed the storage that was built or paced out the floor area and asked the 
respondent to show the height visually. This would have reduced the chances of over￾estimation. The consultants’ conclusion is that this did affect the responses on storage 
and possibly some other questions, such as land area, where it might have helped to 
have face to face time to clarify responses. Overall, the results, other than storage
area, were not substantially different than could be expected.
The lesson is that some topics, e.g. that require measurement, may be unreliable in 
this type of interviewing and that some type of further cross-checking at the interview 
stage should be considered.
MSIKA Final Evaluation Page 14 kadale@africa-online.net
3 Results and Findings
This section sets out the findings of the research, set out according to different 
stakeholder groups, starting with FBOs, followed by Producers, Financial Institutions, 
Processors and Government/Government bodies.
A summary of the specific project indicator achievements, including disaggregates is 
included in Annex 1: Table of Program Indicators. 
3.1 Farmer Based Organisations 
This section is based on project information and KIIs conducted with the committees 
of 15 FBOs5
, covering high-, medium- and low-performers across the five districts. The 
KIIs sought to verify information provided by MSIKA about the performance of specific 
FBOs and to generate qualitative data on FBOs engagement with MSIKA, progress, 
successes, challenges and sustainability. The section also draws on an exercise 
conducted by MSIKA in August 2020 to review progress of its FBOs using V37’s
Performance Measurement Management (PMM) tool through phone interviews with 
49 FBOs. Finally, this section also draws out lessons learned. 
3.1.1 Background and outline of work with FBOs
Central to MSIKA’s approach has been to work with, and through, FBOs to strengthen 
their capacity to operate and to deliver services to their members. This approach has 
the merits of providing a mechanism for reaching large numbers of farmers with training
in a relatively efficient manner and as a means for sustaining the project activities and 
impacts beyond the life of the project. 
At the outset of the program, MSIKA undertook a mapping exercise to identify potential 
participants and any established FBOs that the team could work with. This was 
undertaken in collaboration with the Ministry of Agriculture’s district-level structures, 
the District Agricultural Development Offices (DADOs), and with other NGOs operating 
in the target districts. The aim was to identify Extension Planning Areas (EPAs) where 
the target crops were more commonly grown and to determine where there were 
potential participant producers not supported by projects and NGOs. MSIKA worked 
with some existing FBOs, and supported the establishment of new FBOs. 
For the new FBOs, once the target EPAs were agreed, MSIKA called meetings for 
producers of the target crops at Group Village Head (GVH) level, typically covering two 
to three villages. The aim was to introduce the MSIKA project and motivate producers
to form clubs of 15-30 members. From that point, a minimum of five and up to 10 clubs 
were formed into an FBO within approximately a 5km radius (maximum 8km), so that
it would be practical for members to get together. The aim was to have from 75 to 300
producers in one FBO. Each club has five leadership positions (Chairperson, Vice￾Chairperson, Secretary, Vice-Secretary and Treasurer) and the leaders of the clubs 
(Chairpersons) form the committee for their FBO. The FBOs elected officials in the 
same five roles out of this group of leaders, with elections held annually.
The MSIKA team also engaged with existing FBOs in the target EPAs, if they were 
focused on the target crops or had a substantial number of members that grew the 
target crops. Working with existing structures has the advantages of faster
commencement of activities and not requiring all the initial capacity building support. 
However, there can be a disadvantage if a pre-exiting FBOs is not functioning well, as
it is difficult to change embedded behaviours/practices. 
5 FBOs included registered co-operatives and associations plus unregistered groups. Some 
FBOs call themselves co-operatives or associations, but are not formally registered as these. 
For this report, the term ‘FBO’ encompasses any group of producers whether formally 
registered or not that MSIKA has worked with.
MSIKA Final Evaluation Page 15 kadale@africa-online.net
MSIKA encouraged FBOs to form the following four sub-committees: Markets, 
Finance, Production (chaired by a Lead Farmer), and Disciplinary (usually chaired by 
an elder in the community). There were other committees in some longer established 
groups. These sub-committees included FBO leaders, club leaders and members. 
A key point to note was that the focus for Yr 4 was on 50 FBOs, which would have had 
some effect on coverage and activities with FBOs, and was compounded by the impact 
of COVID on FBO activities (see 3.1.8).
3.1.2 FBOs summary data
MSIKA reports that it has worked with 217 FBOs as at September 30th, 2020, 
consisting of 56 in Dedza (25.8% of all FBOs), 36 in Lilongwe (16.6%), 40 in Mangochi 
(18.4%), 41 in Mchinji (18.9%) and 44 in Ntcheu (20.3%). Dedza has the most FBOs 
with the other four districts relatively evenly split. The total was 219 at mid-term, but 
two FBOs were amalgamated into other FBOs covering similar producers. Out of these 
217 FBOs, 167 (77%) were established by MSIKA, and 50 (23.0%) were pre-existing.
For the purposes of the evaluation, MSIKA categorized its FBOs6 into high-performing 
- 59 (27.2%); medium performing - 84 (38.7%) and low performing - 74 (34.1%). 
Overall, 65.9% are medium- or high-performing. 
Compared to the MTE, the number of high and low performing FBOs at FE has 
reduced, resulting in a shift to medium-performing, which now representing 38.7% of 
FBOs compared to 28.3% at mid-term. There has also been an improving trend in 
performance in Mangochi and Lilongwe, while it has weakened in Mchinji and Ntcheu. 
The details for the FE and the mid-term are set out in the two tables below:
Table 5: MSIKA’s Performance Rating of FBOs, FE, 2020
Performance rating as % of each district total
Districts High Medium Low Total FBOs % of All FBOs
# % # % # % # %
Dedza 10 17.9% 31 55.4% 15 26.8% 56 25.8%
Lilongwe 13 36.1% 7 19.4% 16 44.4% 36 16.6%
Mangochi 11 27.5% 17 42.5% 12 30.0% 40 18.4%
Mchinji 12 29.3% 19 46.3% 10 24.4% 41 18.9%
Ntcheu 13 29.5% 10 22.7% 21 47.7% 44 20.3%
All districts 59 27.2% 84 38.7% 74 34.1% 217 100.0%
Table 6: MSIKA’s Performance Rating of FBOs, MTE, 2019
Performance rating as % of each district total
Districts High Medium Low Total FBOs % of All FBOs
# % # % # % # %
Dedza 15 27.3% 15 27.3% 25 45.5% 55 25.3%
Lilongwe 10 28.3% 6 17.1% 19 54.3% 35 16.1%
Mangochi 10 25.6% 14 35.9% 15 38.5% 39 18.0%
Mchinji 19 46.3% 14 34.1% 8 19.5% 41 18.9%
Ntcheu 18 36.7% 13 26.5% 18 36.7% 49 22.6%
All districts 72 32.9% 62 28.3% 85 38.8% 219 100.9%
Source: MSIKA internal data 
6 MSIKA has internally set criteria for assessing performance that draws on the PMM tool. These 
cover questions on governance and leadership, market access, production capacity, adaptive and 
operational capacity and capacity needs. This enables the MISKA team to assess the level of 
performance and prioritization of needs.
MSIKA Final Evaluation Page 16 kadale@africa-online.net
MSIKA undertook a formal assessment of FBO capacity development in August 2020 
using V37’s PMM tool, which it has developed based on its years of working with farmer 
cooperatives. The PMM tool covers six performance areas: leadership (goals, 
strategy); adaptive capacity; management capacity; operational capacity; supply, 
processing and marketing; and productivity and financial performance. Each of these 
has several sub-areas of performance. For each sub-area of performance, there are 
four descriptive statements corresponding to four levels of performance that enable 
the interviewer to place the FBO’s performance in the one that best describes the 
current situation. Level one is the lowest level (‘base’)7 which equates to the starting 
level of capacity with very limited capacity to function as an organisation, make 
decisions, manage finances, etc. up to level four which is the highest-level of capacity 
which equates to well-functioning leadership and governance, robust systems 
throughout, ownership of premises, effective decision-making. 
The tool can be used for assessing progress if there is a baseline to compare to, and 
for helping individual FBOs to consider the areas on which they need to improve as 
the basis to develop an action plan.
The assessment exercise set out the results for 38 FBOs established by MSIKA as 
well as for 11 pre-existing ones. 
For the newly established FBOs, the baseline would be 1.0 since they did not exist 
before. The current mean score for the newly established FBOs across all six areas 
(not weighted) was 1.77.
8
 This equates to FBOs being around three quarters of the 
way between level one and level two. The highest scores were on leadership (1.99) 
and management capacity (1.98), followed by Supply, Processing and Marketing 
(1.81), and adaptive capacity (1.79). The lowest scores were for Operational Capacity 
(1.58) and Productivity and Financial Performance (1.32). The results are set out in the 
table below. 
Table 7: PMM Assessment, MSIKA-Established FBOs at FE, 2020
FBOs established by MSIKA
Performance area Mean Score
Leadership 1.99
Adaptive Capacity 1.79
Management Capacity 1.98
Operational Capacity 1.58
Supply, Processing & Marketing 1.81
Productivity & Financial Performance 1.32
Across all areas 1.77
N=38 newly established FBOs
For the pre-existing FBOs, MSIKA undertook an assessment of the baseline and the 
current scores to determine what change had occurred. The baseline score, 
determined retrospectively, was a mean of 1.45, with a current mean score of 2.04 and 
a change of 0.59 points. The current mean scores for the six performance areas were 
relatively similar and between 1.94 to 2.12.
7 Meaning the lowest possible score is 1.0.
88 The consultants, with agreement of MSIKA, updated the scoring to replace zero scores with ‘n/a’ 
meaning ‘not assessed’. Zero scores meaning not assessed, incorrectly reduced the means.
MSIKA Final Evaluation Page 17 kadale@africa-online.net
Table 8: PMM Assessment, Pre-existing FBOs, 2020
FBOs established prior to MSIKA
Baseline Current Difference
Performance area Mean Score Mean Score
Leadership 1.59 2.12 0.53
Adaptive Capacity 1.55 2.06 0.52
Management Capacity 1.42 2.07 0.65
Operational Capacity 1.73 2.09 0.36
Supply, Processing & Marketing 1.34 1.94 0.60
Productivity & Financial Performance 1.31 1.6 0.65
Across all areas 1.45 2.04 0.59
N=11 pre-existing FBOs
The data from the internal scoring of FBOs by MSIKA and the recent more extensive 
PMM assessment set out above, highlight that there has been progress in 
strengthening the capacity of FBOs. However, the progress has been limited with most 
FBOs not yet at level 2, meaning limited capability. 
In the FE HH survey, participant producers were asked about the MSIKA activities they 
participated in over the last 12 months, all of which are accessed via an FBO. The 
most common were ‘training in VSL’ (87.8%), ‘training in basic agri-production’ 
(86.2%), ‘training in PHH’ (81.5%) and ‘training in gender equality’ (80.0%). The lowest 
scores were ‘participated in international/district fair’ (15.8%) and ‘been linked to 
markets’ (49.8%). 
Of the scores, several were lower than the MTE, notably training in PHH (MTE: 91.8% 
= -10.3% points), ‘linkage to market’ (MTE: 61.9% = -12.1% points) and ‘participated 
in international/district fair’ (MTE: 25.1% = -9.3% points). The lower scores at FE could 
be due to disruption from COVID as it particularly affected training and market 
activities, based on the qualitative interviews with FBOs.
There were increased scores in the FE survey on gender training (MTE: 73.4% = 
+6.6%), suggesting an increased focus on this, in the period when this was possible. 
The FE results are set out in the table below: 
Table 9: MSIKA Activities Reported by FBO Members in the MTE & FE HH Surveys 
Which of the following activities from Land O’Lakes /MSIKA have you participated in?
Activity
MTE FE
Male Female Total Male Female Total
% % % % % % 
a. Training in improved horticulture practices by MSIKA 
staff or a lead farmer 93.8 92.8 93.2 85.8 86.7 86.2 
b. Training in horticulture post-harvest handling by 
MSIKA staff or a lead farmer 91.1 92.3 91.8 79.8 83.6 81.5 
c. Training in marketing 81.7 83.5 82.8 79.1 79.3 79.2 
d. Training in financial literacy 73.5 74.0 73.8 74.4 72.5 73.5 
e. Training in gender equality 77.4 70.7 73.4 81.3 78.4 80.0 
f. Training/membership of a village savings and loan
group
 85.6 89.5 87.9 85.6 90.5 87.8 
g. Been linked to an MFI 62.3 66.1 64.6 63.0 63.3 63.1 
h. Been linked to markets to sell horticulture products 59.5 63.5 61.9 47.3 52.9 49.8 
i. Participated in international/District trade fair 28.4 22.9 25.1 17.0 14.4 15.8 
j. Any other? 0.4 - 0.2 1.9 0.2 1.1 
Total 257 389 646 535 444 979 
N=646 & 979, multiple response possible
MSIKA Final Evaluation Page 18 kadale@africa-online.net
3.1.3 Training at FBOs 
As noted earlier, MSIKA’s strategy is to develop the capacity of FBOs as a 
sustainability mechanism, and because it is efficient and effective to work through a 
community level organisation that can reach many farmers. MSIKA places a high 
priority on training of FBO committees and members. 
The training falls into two broad categories: 
1. Training the FBO leadership to make the FBO an effective organization; and
2. Training farmer members using the FBOs as a grouping mechanism covering 
the key impact areas for MSIKA notably agricultural production, PHH and 
storage, financial management and marketing/sales. 
This section focuses on both of the above categories, as delivered through the FBOs. 
The knowledge and application of training by the producers is covered in the section 
that follows (see section 3.2 below). 
Training is delivered by MSIKA’s technical specialists and by its Farmer Business Co￾ordinators (FBCs), of which there are two per district. It was reported at MTE by MSIKA 
that the FBCs are very stretched in undertaking the training and all their other roles. 
The revised plan for year 4 was to focus on 50 FBOs that had higher potential. As the 
final year/year 4 has unfolded, training conducted up to the end of February was 
recorded by the FBCs, but from March it has been difficult for training to be conducted 
by FBCs/MSIKA technical teams and by FBOs/lead farmers. Although the pandemic 
has waned in Malawi from September, it was too late to pick up further training. 
MSIKA provided a training tracker up to the end of February 2020 from which the 
consultants have analysed the training that MSIKA reported providing to 213 FBOs.9
MSIKA’s training for FBOs and for producers falls into five categories:
1. Agricultural production;
2. Marketing;
3. PHH and storage;
4. Financial Management; and
5. FBO Development.
Each category of training is considered in turn below.
3.1.3.1 Agricultural Production Training
Agricultural production training covers: basic agri-production training, soil fertility 
management, integrated pest management, climate smart agriculture (CSA) and 
irrigation. 
MSIKA intended to train all participant producers in basic agricultural production as a
foundation for all other agricultural training. The aim is to introduce producers to good 
agricultural practices (GAP) that can be applied widely across field and tree crops, 
along with some crop specific training where there are commonly produced crops. This 
is followed by refresher and additional training for more in-depth agricultural production 
topics. Training is intended to be ongoing and progressively building on earlier training 
through a modular approach.
From analysis of the training tracker data, all FBOs have had training for members in 
basic agri-production practices, an increase to 100.0% from 97.2% at MTE. The next 
most common types of training were integrated pest management (FE: 68.1%, 
MTE:49.3%) integrated soil fertility management (FE: 62.0%, MTE:41.3%) and climate 
9 The training tracker has data for 213 out of 217 FBOs. The difference is attributed to 
different dates of preparation of the two documents/sources. 
MSIKA Final Evaluation Page 19 kadale@africa-online.net
smart agriculture (FE 61.5%, MTE: 23.5%). Irrigation was still lagging (FE: 31.9%, 
MTE: 0%), but not all locations are suitable for irrigation. The consultants expect that 
the results would have been higher if COVID had not occurred.
Table 10: FBOs Trained in Agricultural Production, by Type of Training, at MTE & FE
Agricultural production training
% 
trained 
at MTE
% 
trained 
at FE
Basic agri-production 97.2% 100.0%
Integrated soil fertility mgt 41.3% 62.0%
Integrated pest management 49.3% 68.1%
Climate smart agriculture 23.5% 61.5%
Irrigation 0.0% 31.9%
All FBOs 213 213
Multiple response possible
Qualitative interviews with the FBO leaderships supported the above data with 
evidence of widespread adoption of a range of agricultural practices, even in lower 
performing FBOs. The committee members were asked why some practices have 
been adopted and others not. They provided a range of reasons:
i. Some practices are less costly, so easy to adopt; 
ii. Some practices were better understood than others and easier to follow;
iii. One FBO said the training was not “first-hand” or “comprehensive” so not well 
understood; 
iv. Some farmers are more self-motivated and confident, so they adopt practices 
more than those who are not as motivated or confident;
v. Some farmers were practicing these before MSIKA, so they continued;
vi. Some practices were affected by factors such as agrodealers supplying mixed 
variety seed, animals ruining storage sheds and a low supply of manure; and
vii. Lack of markets demotivated some from applying more effort.
FBOs offered a range of suggestions on how to improve the uptake of practices: 
i. Continuation of the training and practical application through demonstrations; 
ii. More training and demonstration time allocation; 
iii. Greater encouragement among the farmer members;
iv. Ready market linkages to motivate them to supply;
v. Frequent extension visits; and
vi. Provision of input loans to farmers
Across all three FBO performance categories, there was evidence of a range of good
results from the use and adoption of the improved techniques and technologies, 
specifically higher yields, higher quality, reduced on-field and transportation losses, 
and higher selling prices: 
"From an acre, farmers used to produce 60 (20L pail), but now are able to 
produce as much as 200 (20L) pails of Irish potato.” (Equates to an increase 
from 960kg to 3,200kg). High performance FBO
“There has been a great rise in production levels and quality in a short period 
because these (Land O Lakes) extension officers follow things up more closely 
than the government ones who are busier and under-resourced sometimes.” 
Medium performance FBO
MSIKA Final Evaluation Page 20 kadale@africa-online.net
“Before MSIKA came, we were not as many who were farming Irish potatoes, but 
now so many farmers are into Irish potatoes, onions and tomatoes." Low 
performance FBO
The conclusion by the consultants is that the agricultural training delivered via the FBOs 
has been effective overall, with an uptake of a range of practices by farmers that have 
positively impacted on their production yields and quality. While there may be some 
gaps in training coverage, and in farmer knowledge and uptake, these were impacted 
by the challenges faced by MSIKA over the second half of the program year.
3.1.3.2 Marketing Training
A key focus for MSIKA is improving links to markets and increasing sales to enable
increased crop volumes to be sold, which in turn motivates and enables farmers to 
invest in increasing production and improving quality.
MSIKA offers a range of marketing training modules: 
1. Introduction to marketing;
2. Marketing principles;
3. Market searching;
4. Costing and pricing; and
5. Customer care management.
The marketing training is mostly delivered as a block of five modules. MSIKA reports 
training that between 81.7% to 83.6% of FBOs were trained in these modules. This 
compares to the range of 49.8% to 53.1% of FBOs at mid-term. It was a clearly 
articulated aim for the MSIKA team to train many more FBOs in marketing and this 
appears to have been achieved. 
Table 11: FBOs Trained in Marketing, by Type of Training, at MTE & FE
Marketing training % trained 
at MTE
% trained 
at FE
Introduction to marketing 53.1% 83.6%
Principles of marketing 52.6% 83.1%
Marketing searching 50.7% 82.2%
Costing and Pricing 49.8% 81.7%
Customer care management 51.6% 82.6%
All FBOs 213 213
Multiple response possible
Out of five FBOs in each category, four high performance, four medium performance 
and four low performance FBOs reported having been trained in Market Access by 
MSIKA. This verifies with what MSIKA had in its records.
The four high performance FBOs reported having benefitted from applying marketing 
principles/practices such as negotiating prices, knowing own ability to supply enough 
volume, understanding the quality sought by buyers, running the farming as a 
business, and offering competitive quality. 
Three medium performance FBOs reported having applied/benefitted from 
activities/practices such as: linked to markets by MSIKA, negotiating prices, ability to 
calculate profits, analyzing markets, advertising, and ability to produce and deliver high 
quality crops. 
Only one low performance FBO reported having applied/benefitted from the training by 
knowing price trends before going on the market. The other two reported not having 
benefitted from the training, because they still have not secured the expected good 
markets.
MSIKA Final Evaluation Page 21 kadale@africa-online.net
Overall, nine of the 15 FBOs reported one or more links to buyers, including retailers 
(five FBOs for fruit and vegetables) and processors (four FBOs for chilies).
"They just linked us to a market, SPAR supermarkets." Medium performance FBO
"At first, members reported their crops to us to inform MSIKA. However, MSIKA did not 
help us find a market and since then we have been resorting to vendors like pre MSIKA 
times...." Low performing FBO
In one case, the FBO had an agreement, but pulled out as they felt they could not 
supply:
"We actually gave up, we were afraid that the volumes being aggregated for supply of 
the very first consignment were very low compared to that in demand..." High 
performance FBO
With the advent of COVID, the marketing training switched to being remote with FBOs 
being provided with Bluetooth speakers. This was a flexible response, though likely to 
have been less effective than face to face training.
From the consultants’ assessment, marketing/access to markets is often the most 
challenging aspect of a program. The training coverage is much higher than in the mid￾term, where it was just over 50% of FBOs and was over 80% when it was stopped in
March. There are expectations by FBOs that MSIKA will find them markets, and this 
may encourage a degree of passivity on the part of the FBOs and members. MSIKA 
is seeking to both improve links, but also for FBOs and farmers to build their capacity 
to find and access markets for themselves. However, a reduction in facilitating linkages 
due to COVID means that some of the expectations remain unfulfilled. There would 
likely have been more benefits and more opportunity to consolidate learning and 
capacity if the marketing training had been completed at an earlier stage, accepting 
that COVID was not predictable and that originally there was to be a fifth year to the 
project.
3.1.3.3 PHH Training
The third set of modules are around reducing losses through improved post-harvest
handling (PHH) and storage. The PHH and storage training modules are:
1. Post-harvest handling;
2. Storage requirements; and
3. International handling standards.
From the FBO training summary, 34.7% of FBOs had been trained in PHH (MTE:
24.4%), 33.3% trained in storage requirements (MTE: 23.5%) and 25.8% in 
international standards (MTE: 18.8%). Detail of the training by module is set out below.
Table 12: FBOs Trained in PHH and Storage, by Type of Training, MTE & FE 
Post-Harvest Handling and Storage training % trained at MTE % trained at FE
Post-Harvest Handling 24.4% 34.7%
Storage Requirements 23.5% 33.3%
International handling Standards 18.8% 25.8%
All FBOs 213 213
Multiple response possible
The plan at mid-term was to increase PHH and storage training. This has happened, 
but falls short of complete coverage. The timing of the training is important as it needs 
to be close to when it will be used, rather than well in advance of it. As March to May 
are peak times for harvesting, then the advent of COVID affected the training plans for 
this set of modules. The overall reduction in the budgets has also played a part. 
MSIKA Final Evaluation Page 22 kadale@africa-online.net
In the KIIs, more of the FBOs reported being trained, than MSIKA reported they had 
training. MSIKA clarified that the PHH training was commonly tied in with the marketing 
training, hence it not appearing in the training record separately. 
International handling standards training is related to indicator 31: Percentage of 
FBOs who are aware of international production and handling standards, as a 
result of USDA assistance. MSIKA conducted a phone survey of FBOs trained as at 
31st December 2018 (MTE), and found 71.8% achievement against the LoP target of 
75.0%. As at 30th September, 2020, MSIKA reports 78% achievement thereby 
exceeding the target for indicator 31. The validity of raising awareness of international 
standards for FBOs that have, or whose members have, very limited capacity and 
potential to export was questioned by the consultant in the mid-term report. The same 
question remains, as there is little scope for most of the target FBOs and farmers to 
contemplate formal exports at this point. The issue of international standards was only 
relevant for FBOs with members that grow chilis, paprika and red cayenne, which are 
primarily for the export market, but which were deprioritized. 
The consultants’ conclusion is that indicator 31 was not appropriate for MSIKA, as 
none of the 16 FBOs visited in the MTE or the 15 FBOs called in the FE were capable 
of directly exporting. 
3.1.3.4 Financial and VSL Training
MSIKA sees financial literacy skills and knowledge for producers as important to their 
development. There are three financial modules:
1. Keeping financial records;
2. Financial literacy; and
3. Village savings and loans (VSL).
Since the MTE, MSIKA has increased the proportion of FBOs with financial and VSL 
training from between 47.4 to 49.3% across the three modules at MTE to 62.4 to 64.3% 
by FE. As with the other training, this had to be cut short due to COVID, meaning that 
fewer FBOs were trained than planned. 
As with marketing, FBOs that are trained in finance tend to have had all three modules. 
There are exceptions due to the needs of FBOs, particularly the well-established ones,
who may already have been trained in one or more aspects. 
Table 13: FBOs Trained in Financial Records, Literacy and VSL, MTE & FE
Financial & VSL training % trained 
at MTE
% trained 
at FE
Keeping financial records 49.3% 64.3%
Financial literacy 47.4% 62.4%
Village savings and loans 49.3% 63.4%
All FBOs 213 213
Multiple response possible 
Of the five FBOs in each category, four high-performance, three medium-performance 
and three low-performance FBOs reported having been trained in Improved financial 
management by MSIKA, verifying what MSIKA had reported. The one other high￾performance FBO reported being trained in Improved Financial Management by a 
consortium of other projects, so it was not necessary.
On Improved Financial management activities, three of five high-performance, two 
medium-performance and no low-performance FBOs reported having MSIKA support 
in implementation of the activities. MSIKA offered support to high-performance FBOs 
in form of facilitating MFI loan applications, coaching, providing stationary and 
MSIKA Final Evaluation Page 23 kadale@africa-online.net
recommending better banks for loans. The medium-performance FBOs reported 
support in form of books for recording VSL information and for other records.
Promoting VSLs has been a key part of MSIKA’s work. The training model relies on 
village agents to mobilize VSLs, who have already been trained in VSL methodologies 
by other organizations. In relation to MSIKA’s FBOs, training in VSL is undertaken by 
a mix of these agents and by MSIKA’s FBCs. 
FBO members are invited to form new VSLs where these do not exist, however it is 
not compulsory for MSIKA members to join a VSL. There is a rule that MSIKA VSLs 
do not allow non-FBO members to join, although FBO members are not stopped from 
joining non-FBO VSLs. VSLs are quite common in rural areas, so restricting access 
would be problematic. 
3.1.3.5 Governance, Leadership and Cooperative Development
The final area of training is governance, leadership and cooperative development. The 
modules are:
1. FBO Governance Structures
2. Leadership Skills
3. Cooperative Education
MSIKA focused on training the FBO leadership in the early stages, however because 
there is a regular change in leaders, there needed to be wider membership awareness 
of governance issues. 
At FE, MSIKA reports that 77.9% of FBOs have been trained in FBO governance (MTE
56.8%, change +21.1%), 64.3% of FBOs trained in leadership skills (MTE 35.2%, 
change = +29.1%) and 1.4% trained in cooperative education (MTE 0%, change 1.4%). 
MSIKA team noted at MTE that more training in governance was required to assist 
sustainability, as well as performance. The FE levels show that there has been a 
substantial increase in training, noting also that some of the pre-existing FBOs would 
not require some of the governance and leadership training. However, the plans for 
further training were affected by the reduction in budgets and the COVID restrictions. 
On cooperative education, MSIKA report that six FBOs were linked to the Ministry, and 
three of these to another V37 project called Co-operative Development Program (CD4) 
for training. This is an additional three to the three that MSIKA reports that have been 
trained in the table below. Details on governance training are set out in the two tables 
below for the FE and the mid-term: 
Table 14: FBOs Trained in Governance, Leadership & Cooperatives, MTE and FE
Governance, leadership & 
cooperative training
% trained 
at MTE
% trained 
at FE
FBO Governance Structures 56.8% 77.9%
Leadership skills 35.2% 64.3%
Cooperative Education 0.0% 1.4%
All FBOs 213 213
Multiple response possible
According to MSIKA, formal cooperative training by the Ministry of Trade and Industry 
is relatively long. The consultant’s finding is that cooperative training is not suited to 
many FBOs, as converting to a registered cooperative requires sufficient levels of 
membership and the capital to sustain it, as well as strong and effective leadership. At 
mid-term, MSIKA staff estimated that around 10% of FBOs have potential to become 
cooperatives. From an analysis of the summary data on FBOs, only 12 were identified 
with membership over 300, while over half have less than 150 members. The larger 
FBOs tend to be the longer established, pre-existing farmer entities. The consultant’s 
MSIKA Final Evaluation Page 24 kadale@africa-online.net
finding is that most of the MSIKA FBOs do not have potential to become functioning
cooperatives. This does not mean FBOs do not have a valuable function as effective 
groupings for accessing training, inputs and potentially for aggregation and selling.
From the qualitative KIIs and discussion with the MSIKA team, the implementation by 
FBOs of the governance training is mixed, as it was at MTE. Two high-performing, two 
medium-performing and one low-performing FBOs found Leadership to be the most
useful topic because good leadership sustains and grows consistent membership. It
also supports confident duty execution and harmonious management of affairs due to 
clear awareness of everyone’s respective roles. From the MTE and wider work, the 
consultant’s finding is that leadership is the most challenging area for FBO functioning 
and development, particularly if the leaders are embedded and are not performing well.
"Knowledge of the things from the training made leadership easy to work with members 
and steer forward wisely"…Low performing FBO
One high performing FBO found By-law formulation the most useful, one medium 
performing found book-keeping to be most useful and one low performance found 
Good governance to be the most useful topic. In the latter case, they said that it helps 
them carry out critical commitments, such as loan repayments without fail.
"Recently we were assisted to access loans from the MFI for capital. People we faithful 
through compliance to the FBO's bylaws. Everyone paid back on time as required..." 
High performing FBO
"Through record keeping, we are also able to have an account of where we are coming 
from and what we are aiming to achieve..."…Medium performing FBO
There were some references in the FBO KIIIs to governance training that had not been 
undertaken, which the consultant presumes relates to the COVID situation.
From the KIIs, there is evidence of activities, such as regular meetings and elections
in many, but not all FBOs. COVID appears to have reduced the number of governance 
related activities by the FBOs, which may be down to the ceasing of regular
engagement with the FBOs by MSIKA’s FBCs. This may imply that ongoing operation 
of some FBOs may partly depend on regular engagement to remind, encourage and 
motivate the FBO leaderships.
FBOs are required by MSIKA (and by the by-laws they have adopted) to keep records, 
such as minutes of meetings, market information (buyer product specifications), 
production and sales records. However, because the research was conducted by 
phone, it was not possible to verify if the records existed or to see the state of the 
records. In some interviews, it was clear that the FBO committees were giving 
information from the records, while in others, the respondents did not appear to have 
the information available.
10 It should be noted that some records may not be kept, e.g. 
sales records, due to lack of transactions. 
In the staff KIIs, MSIKA highlights the need for ongoing and repeat trainings for leaders, 
since there is going to be committee membership turnover, as these are annually 
elected positions. This is a sustainability challenge for some FBOs.
Indicator 33 is the number of FBOs trained in improved financial and 
organizational management. MSIKA reported progress at 183 trained against a
revised LOP target of 135 (136% achievement), as at 30th September 2020. From the 
FBO KIIs, of the five in each category, four high performing, three medium performing
and three low performing FBOs reported having been trained in improved financial 
management by MSIKA. This matched what MSIKA had on record. One high
10 The researcher made an appointment with the committee and requested them to have their 
records to hand.
MSIKA Final Evaluation Page 25 kadale@africa-online.net
performing FBO reported being trained in Improved Financial Management by other 
projects, so there was no need for MSIKA to also train them. 
On improved financial management activities that MSIKA supported and were 
implemented11, three of five high-performing, two medium-performing and no low￾performing FBOs reported having MSIKA support in implementation of the activities. 
Examples were MSIKA support to high performing FBOs in facilitating MFI loan 
application, coaching, providing stationary and recommending better banks for loans
and for medium performing FBOs in the form of books for recording VSL information 
and other records. 
One high performing FBO had implemented three elements of the training, including 
MFI) loan application and management, record keeping in exercise books, and getting 
advice on which banks the FBO could open an account with. 
In contrast to those that have implemented learning, the challenge with some FBOs is 
captured in the quote below:
"MSIKA did not help us get markets, hence we could not implement the systems we 
learnt from them..." Low performing FBO
The implementation by the high performing FBO and the quote from the low performing 
FBO illustrates the range of responses by FBOs about working with MSIKA, and likely 
with other future projects. Some are willing and able to take on elements of training
that is relevant to them, while others have unrealistic expectations of what a project 
can or should do for them, noting that MSIKA is trying to help FBOs find markets 
sustainably. MSIKA staff may have encouraged the expectations over time based on 
some feedback at MTE; however finding a market for every FBO is not possible to fulfil
and some FBOs may see this as only the responsibility of project staff, and not see 
training as something to improve their capacity to do more things for themselves. The 
MTE recommended that MSIKA focus on better performing FBOs, which was a 
recommendation that MSIKA took up by focusing on 50 FBOs, 10 per district. The full 
effect of that change in focus was disrupted by COVID.
In the KIIs, the FBOs gave suggestions on how to improve the training that they had 
received. Training is something that FBOs expect and were clear that they want a 
continuation of. The improvements they proposed were: 
i. Good notice should be given prior to the training
ii. Trainers should be punctual
iii. More time should be allocated to the training
iv. The training length should be extended
v. Some parts should be repeated upon request from participants
vi. Motivational real life stories should be part of the training content
These are not responses from all FBOs, but give a sense of areas that could be 
enhanced in future. Several are to do with the organising of the training (i-iv) and 
others to do with how it is delivered (v-vi).
3.1.4 Lead Farmers
Lead Farmers played an important role in the MSIKA approach with 2-3 designated 
Lead Farmers at each FBO. MSIKA utilized lead farmers as training of trainers, training 
the lead farmers on agricultural practices and PHH, and the lead farmers were 
expected to train other FBO members in these practices, establish demonstration plots 
and provide ongoing advice and extension to members. Many lead Farmers were
11 This relates to Indicator 30, though this indicator is not reported on in Yr 4.
MSIKA Final Evaluation Page 26 kadale@africa-online.net
selected prior to MSIKA by the communities as part of a wider government initiative, 
so it was appropriate for MSIKA to work with these individuals. However, MSIKA also 
picked others that were appropriate based on their experience of the crops and 
performance. In general, MSIKA found that the latter were generally better performing. 
Based on the FE HH survey, 36.0% of respondents said they were lead Farmers. This
high proportion is likely to be a function of doing a phone survey and with the Lead 
Farmers numbers known to MSIKA and the FBOs who provided many of the numbers. 
Of the lead farmers, 41.3% were female and 58.7% were male, though the higher 
number of male lead farmers is in line with the higher proportion of men in the sample. 
Table 15: Proportion of Lead Farmers in the FE Survey
Are you a Lead Farmer trained by Land O Lakes/MSIKA?
Response
Male Female Total
Count % Count % Count %
No 328 61.3 299 67.3 627 64.0
Yes 207 38.7 145 32.7 352 36.0
Total 535 100.0 444 100.0 979 100.0
These Lead Farmers were asked about the activities they had undertaken (prompted) 
for MSIKA. A high 91.3% (mid-term 83.3%) said they had established a demonstration 
plot, 87.1% stated they had trained others (mid-term 94.0%), and 58.4% (mid-term 
79.8%) had submitted training forms. Training and demonstrations are key for MSIKA. 
The lower score for submitting training forms suggests there is more training that has 
occurred than is reported. The responses on activities are set out in Table 16 below
Table 16: Lead Farmer Activities by Sex at FE, 2020 
As a lead farmer, have you done any of the following?
Activity Male Female Total
% % % 
Conducted trainings of other farmers 89.9 83.1 87.1
Submitted training forms to MSIKA about the training 58.1 58.8 58.4
Established a demonstration plot 88.9 94.9 91.3
Any other? 1.5 1.5 1.5
Total 198 136 334
N=352 Lead Farmers
3.1.5 Sales Agreements and Sales
Indicator 29 is the number of sales agreements with FBOs. MSIKA works with 
FBOs to link them to potential buyers and to make sales agreements for FBO members 
to fulfil. At mid-term, the cumulative achievement was 52, but this increased 
substantially in Yr 4 by a reported 209 to reach a cumulative LoP total of 261 against 
a target of 62, representing 421.0% achievement. The additional target for year 4 was 
10 agreements, which looks modest against the actual achievement. 
From the FBO KIIs, five FBOs gave examples covering six sales agreements, with two 
for Tingadalire (chillies), two for North East Foods (chillies) and one FBO with two large 
supermarkets (Spar and Shoprite for vegetables). Two FBOs gave examples of where 
they were linked, but either did not reach an agreement or were not able to supply. 
There were examples where MSIKA said there was an agreement and the FBO did 
not confirm it, but also the other way around where sales agreements were reported 
by the FBOs, but not by MSIKA. This may be a function of changes in leadership where 
earlier agreements were not known or recalled. It could also be a limitation of the 
interview process, as time was short and records could not be checked and were not 
always available despite being requested in advance.
MSIKA Final Evaluation Page 27 kadale@africa-online.net
"That was the only time when we were surprised to enjoy a selling price as high as 
K600 per kilogram of tomato….. The Spar offer was unprecedented…." Medium 
performing FBO.
Sales agreements are not necessarily followed through to make actual sales:
"We actually gave up, we were afraid that the volumes being aggregated for supply of 
the very first consignment were very low compared to that in demand..." High 
performing FBO
MSIKA provides FBOs with sales books and the FBOs ask participant producer 
members to provide sales data to their club, which provides this data to the FBO level
for summarizing and reporting to MSIKA. MSIKA report sales against indicator 21 
volume of commodities sold by project beneficiaries and indicator 23 value of 
sales by project beneficiaries. 
For indicator 21, volume sold, MSIKA reports sales of 62,963 mT against an LOP 
target of 86,330 mT, thereby achieving 72.9% of the target. For indicator 23, value 
sold, MSIKA reports US $18,440,386 against an LoP target of US $21,972, 909, which 
is an achievement of 83.9%. 
Year 4 sales were around 5% of the Year 3 sales, with the impact of COVID affecting 
the result in two ways. Firstly, the COVID pandemic began during the main production
season for the field crops. The closure of governmental institutions, restrictions on 
hotels/lodges, including meetings/conferences, and restrictions on general 
movements, resulted in a significant loss of business from formal buyers for fruit and 
vegetables. Producers only had local markets to supply rather than formal outlets that
could buy large quantities at relatively good prices. See the section 3.1.8 on COVID 
below for more detail. The second effect has been difficulties for MSIKA staff to gather 
information on sales, both from FBOs and from processors, as they were not able to 
go to the field and gather data in person. There also appears to have been less 
reporting of their own sales by farmers to their FBO. This likely resulted in sales for 
the second half of Yr4 being understated. The sales data reported in the FE HH survey 
(section 3.2.12) suggest that farmers were still able to sell, but primarily to local 
markets.
From the MTE and FE KIIs, MSIKA staff highlight the difficulties of getting FBOs to 
keep sales records, particularly as most farmers sell individually rather than 
collectively. Part of the aim of FBOs KIIs was to verify sales agreements and data. 
There was some confirmation, but also some differences. Because the researchers 
could not check the records, it proved difficult to verify sales volumes. 
Overall, there is evidence of sales agreements and sales facilitated by MSIKA through 
the FBOs as a mechanism for their members to sell their produce. Some FBOs seem 
to have grasped the opportunities, while others have expected MSIKA to keep 
providing links. The latent or existing capacity of the FBOs is a key factor in whether 
sales can be made and delivered.
3.1.6 Processing and Storage
From the qualitative interviews, none of the FBOs in the MTE or FE KIIs had 
warehousing or were processing crops. 
The consultant’s finding is that storage and processing are unlikely to happen at the 
present state of development of most of the FBOs. Particularly with the newly 
established FBOs, these are relatively small groups with limited capacity and 
resources. As noted in the PMM assessment, capacity is still weak in many areas.
MSIKA Final Evaluation Page 28 kadale@africa-online.net
3.1.7 Access to Market Information
Indicator 18 is the percentage of producers who have access to current market 
information through their cooperatives or producer associations.
MSIKA aims to increase access to market information for participant producers through 
their FBOs. Market information includes location of buyers, names of possible buyers, 
where to buy inputs, where to access finance, buyers’ requirements, weather, and 
market prices. This access was measured in the FE HH survey by a question about 
the source of market information (unprompted) on each type of market information. 
The findings are set out in Table 17 below.
Across the range of types of market information, the two most commonly sources were 
‘MSIKA extension workers’ (31.3%-52.7%) and ‘radio’ (22.5%-59.2%). MSIKA 
extension workers were the most common source for ‘location of market’ (45.1%), 
‘names of buyers’ (37.1%), ‘where to access finance’ (52.7%), and buyers product 
needs’ (41.6%). Radio was the most common source for ‘where to buy inputs’ (47.9%), 
weather that affects crops (59.2%) and ‘market prices for crops’ (39.4%). MSIKA Lead 
Farmers are also regarded as a significant source, mentioned by 12.6-18.6% of 
respondents across the different types of market information.
For FBOs as a source of market information, the relevant responses were ‘fellow FBO 
members’ (2.8%-4.9%) and from ‘FBO leadership’ (3.4%-5.9%). 
Table 17: Sources of Market Information, FE 2020
Information Sources
Response
Location 
of 
Market 
To Sell
Names 
of 
buyers
Where 
to buy 
inputs
Where 
to 
access 
finance
Buyers’ 
product 
needs
Weather 
that can 
affect 
crops
Market 
prices 
for your 
crops
% % % % % % % 
Neighbor/friend 10.4 10.2 11.8 6.3 5.8 2.8 9.2
Govt ext’n officer 7.8 4.6 9.4 5.5 6.1 9.5 6.5
Land O’Lakes/MSIKA 
extension worker 45.1 37.1 43.7 52.7 41.6 36.0 31.3
LOL/MSIKA lead 
farmer 18.6 16.3 15.7 18.1 16.8 14.8 12.6
Private buyer ext’n 
officer 0.9 0.9 0.4 0.5 0.4 0.1 0.8
NGO/Project ext’n 
officer
2.7 1.2 1.6 3.0 0.9 0.9 1.3
Vendors/traders 12.0 10.3 4.6 0.7 19.1 0.5 20.8
Radio 36.3 27.4 47.9 22.5 23.6 59.2 39.4 
SMS 2.6 1.6 1.2 0.5 0.9 2.6 1.6
FBO Leadership 5.9 5.6 4.7 5.5 5.0 3.4 4.9
Fellow FBO member 4.3 4.2 4.9 4.0 3.6 2.8 4.1
Posters 0.3 0.5 1.2 0.1 0.3 0.1 0.6
Agricultural shows 0.2 0.3 0.4 0.2 0.2 0.7 0.3
Trade fairs 0.1 0.2 0.4 - 0.2 - 0.3
Other sources 3.4 3.7 3.0 2.5 3.3 1.7 3.2
Did not get this info 7.5 17.3 5.1 16.4 12.6 8.3 9.0
All respondents 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Multiple response, n= 979
MSIKA Final Evaluation Page 29 kadale@africa-online.net
The proportion of respondents in the FE HH survey that obtained market information
from these two sources, accounting for any overlap, was 27.6%, with similar results for 
men and women. This shows progress from the baseline at 12.9% and the mid-term 
at 14.9%, but is short of the LoP target of 48.0%. The LoP achievement is 57.4%.
Table 18: Indicator 18, Producers Accessing Market Info via their FBO, FE 2020
% of producers with access to current market information via their 
FOBs
 IB% MTE% FE
FBO Governance Structures 12.9% 14.9% 27.6%
Leadership skills 11.0% 14.1% 27.4%
Cooperative Education 15.0% 16.0% 27.7%
Composite of responses, n= 979
Although there has been an increase in market information that farmers receive 
through their FBOs, they still rely more on MSIKA for information than their FBOs. 
MSIKA is not a sustainable source of information and it is unclear if the FBOs will be 
able to sustain the flow of information post project as much of this has been supplied 
by MSIKA to the FBOs.
In conclusion, there has been progress from the baseline on FBOs as a source of 
information, however this is short of the target. 
3.1.8 COVID Impacts
In the FE KIIs, the FBOs were asked about the effect of COVID on the FBO and its 
members. All FBOs reported COVID effects in their KIIs, notably: 
i. Low market prices locally due to export restrictions
ii. Training suspension by V37, resulting in limited information access
iii. Training suspension by government 
iv. Hampered data reporting activities 
v. Low sales and prices due to reduced market demand
vi. Higher costs of transport and inputs which have raised production costs
vii. More individualistic selling of crops by members
The FBOs said:
"Some buyers buy large volumes and export the vegetables but now they are grounded, 
they can’t buy because they can't export.” High performance FBO
“Although we have had no COVID-19 case here in our community, we are still affected 
in our farming business. If I'm not mistaken, the last time the (MSIKA) extension officer 
came here was in October, 2019.” Medium performance FBO
"Prices dropped for chillies to K 1,000 instead of the agreed K 1,500 per kilogram 
because their demand in Blantyre had also dropped due to lockdown restrictions there.”
Low performance FBO
Most FBOs did nothing in response, as they could not think what they could do. A few 
took some steps to mitigate the effects, as follows: 
i. Individual selling has increased due to fewer opportunities via the FBOs
ii. Using cellphones to share information (two medium performing FBOs) 
iii. Listening to pre-recorded messages on the radio that MSIKA provided (one
high performing FBO)
MSIKA Final Evaluation Page 30 kadale@africa-online.net
iv. Limiting meetings to small groups or leaderships only (one high performing
FBOs, one medium performing FBO, four low performing FBOs) 
v. Waiting until buyers come to the farmgate (two high performing FBOs, one
medium performing FBO)
vi. Continuation of training and practical application through demonstrations
According to one FBO:
"We are now just meeting in more small groups so that we still progress forward." 
Medium Performance FBO
The effects on FBOs have been noted in the reporting of several indicators, notably 
those on sales (indicators 21 and 23), where performance has been much lower than 
expected due to loss of formal buyers. FBOs have made some adaptions, however, 
the loss of the link with MSIKA, mainly through the reduced contact with FBCs, has 
affected planned activities such as training, access to information and market linkages. 
3.2 Participant Producers 
Producer HHs that grow the target fruit and vegetable crops are the main participants
of the MSIKA program. To be included as a participant producer for the FE survey, an 
individual must have been trained in basic agricultural production by MSIKA as of 31st
May 2020. Based on this criterion, MSIKA provided a list of 30,922 trained producers. 
In addition to the HH survey, 63 qualitative KIIs were conducted with farmers, replacing 
the originally intended FGDs. These provide insights into some of the key changes.
3.2.1 MSIKA Producer Sample Frame
This section provides a profile of the producers in the sample frame that were trained 
by MSIKA. Comparisons are made with the baseline when appropriate.
Table 19 summarizes the profile of the producers by district, sex and crop. Producers 
are spread across five districts, with the proportional split per district ranging from the 
lowest at 16.9% in Ntcheu to the highest at 24.6% in Mchinji. At 31st May 2020, the 
beneficiary producers were members of 217 FBOs. 
MSIKA has trained more female producers (55.6%) than male producers (44.4%). 
Table 19: Beneficiary Producers, by District, Sex & Crop, FE 2020
District # of FBOs Individuals trained in basic agri-production 
Total Female Male Tomato Onion Potato Mango 
Dedza 56 5,890 3,666 2,224 3,166 1,454 4,630 763
Lilongwe 36 5,747 2,506 3,241 4,259 1,856 1,474 203
Mangochi 40 6,472 4,592 1,880 5,171 768 868 1,241
Mchinji 41 7,601 3,589 4,012 5,663 3,771 4,167 3,132
Ntcheu 44 5,212 2,827 2,385 3,981 1,240 1,347 2,202
Total 217 30,922 17,180 13,742 22,240 9,089 12,486 7,541
3.2.2 Producer Totals
According to MSIKA data, the program reached 41,185 individuals throughout the four 
years of the MSIKA project implementation, which is 114.4% of the LoP target 
(Indicator 2). Of these, 18,433 were male (91.4% of LoP target) and 22,752 were 
female (143.6% of LoP target). The proportion of female participants is 55.2%. 
Therefore, the overall LoP target for indicator 2 has been met with over achievement 
on female beneficiaries, and a shortfall on male beneficiaries. 
MSIKA Final Evaluation Page 31 kadale@africa-online.net
Indirect beneficiaries (indicator 3) are a function of the direct beneficiaries, so the 
achievement on indicator 3 is a multiple of indicator 2. The reported achievement is 
189,451 against the LoP target of 165,600 (114.4%).
Most of the project participants (39,744) are producers, who received short term 
agricultural productivity training (Indicator 4). MSIKA has overcome a slow start in 
reaching participants and has surpassed the target of 36,000 beneficiary producers 
achieving 110.4% of the target. 
3.2.3 Final Evaluation Sample Profile
This section provides a summary of information about the producers in the FE sample, 
which is compared to the baseline where that information is available and appropriate. 
These are summarised in Table 20 below, followed by an analysis of the differences 
on age and HH size in Table 21. 
Except for the sex of the respondent and relationship to the head of household (HHH), 
the demographics of the baseline and FE samples are very similar. Both baseline and 
FE respondents were on average between 40 and 43 years, with some education 
(93.8% FE versus 88.0% at baseline), and were married (86.2% FE versus 81.9% at 
baseline). Households in the FE sample were nearly all headed by males (96.2%) and 
had 5.8 members, while households interviewed at baseline had 5.6 members. Slightly 
more respondents (74.7%) identified as the HHH, compared to the baseline (70.3). 
For those respondents that were not the HHH, the relationship was mostly that of 
spouse. The majority of the FE sample was male (54.5%). This was a higher proportion 
of males than in the sample frame (44.4%) or in the baseline (50.3%). Overall, the 
datasets were similar for respondents’ marital status, household status and education. 
A further analysis was conducted on the sex of the respondent, where there was a 
superficial difference in the balance (baseline 50.3% male, FE 54.6% male), along with 
age and HH size. The analysis found that the FE (treated) and untreated (Impact 
baseline-treat arm) differed by about 0.02 standard difference (SD) in age, and by 0.15 
SD in household size. Both ANOVA and linear regression analysis of the baseline and
FE datasets indicate that sex was a significant factor [p-values<0.05] in predicting yield 
outcomes with the analysis showing that that there was positive association of yield 
and being male. 
The analysis further showed that only about 3-4% [R-squared] of the variance in both 
the baseline and FE could be explained by the factors listed in the table above strongly 
suggesting that these factors (age, sex and HH size) would not account for the high 
percentage of the changes that may be observed even if they were significant. 
The evaluation team undertook a further review to determine if there were any factors 
that explained why yields were higher for men and lower for women. The first analysis 
was to see if there were differences in the proportion of males and females that 
received various training types, however no significant differences were found. 
The second analysis looked to see if socio-cultural structural factors might have 
contributed to this. It was found that a higher proportion of male respondents (98.3%) 
were married compared to 71.6% of female respondents. Of the female respondents, 
28.4% were divorced or separated, but very few men. Due to the small number of men 
who were not married, it was not possible to compare their relative performance, so 
the yield performance of married men was compared with married women. The gap 
in yield performance was still evident and significant, so cannot be explained by marital 
status.
Overall, there is no explanation of the differences in yield between men and women
from the data in this evaluation. 
MSIKA Final Evaluation Page 32 kadale@africa-online.net
Table 20: Comparison of Baseline vs FE Samples 
Demographic Baseline FE
Male Female All N Male Female All N 
Sex of respondent 50.3% 49.7% 100.0% 590 54.6% 45.4% 100.0% 979
Age(years) 43 40 42 590 42 41 42 979
Household size n/a n/a 5.6 590 n/a n/a 5.8 979
Marital status
Married 93.3% 70.3% 81.9% 590 98.3% 71.6% 86.2% 979
Divorced,
separated,
widowed, never
married 6.7% 29.7% 18.1% 590 1.7% 28.4% 13.8% 979
Education level of respondent
Standard 1-8 63.0% 62.8% 62.9% 590 61.5% 69.8% 65.3% 639
Form 1-4 27.3% 15.0% 21.2% 590 32.1% 16.0% 24.8% 243
Further education 1.3% 0.3% 0.8% 590 0.9% 0.0% 0.5% 5
None 6.7% 17.4% 12.0% 590 2.6% 10.6% 6.2% 61
Adult literacy 1.7% 4.4% 3.1% 590 2.8% 3.6% 3.2% 31
Head of household
Head of household 
is respondent 95.3% 45.1% 70.3% 590 97.0% 47.7% 74.7% 731
Head of household 
is other 4.7% 54.9% 27.7% 590 2.6% 46.6% 22.6% 221
Head of household 
is joint n/a n/a n/a 590 0.4% 5.6% 2.8% 27
relationship to 
H0fH-spouse 7.1% 91.3% 84.6% 590 79.0% 97.0% 97.0% 214
relationship to 
H0fH-other 92.9% 8.7% 15.4% 590 21.0% 1.9% 3.0% 7
N=979
Table 21: Check Balance of Confounders Using Propensity Score Matching
Covariates Mean in treated Mean in untreated Standardized diff
Age 41.84 41.54 0.02
HH size 5.82 5.29 0.15
3.2.3.1 Producers’ Engagement with MSIKA
To provide a picture of their engagement with MSIKA, FE survey respondents were 
asked (prompted) to state which MSIKA activities they have been involved in, in the 
last 12 months. 86.2% respondents reported receiving training in improved horticulture 
practices from MSIKA or a MSIKA lead farmer in the last 12 months This compares to 
93.2% at the MTE. In addition, 81.5% reported receiving training by MSIKA/lead 
farmers in horticulture PHH (MTE 91.8%) and 87.8% were trained in VSL (MTE
87.9%). Financial literacy training was reported at 73.5% (MTE 73.8%), gender 
equality at 80.0% (MTE at 73.4%), and marketing at 79.2% (MTE 82.8%). 
A majority of respondents reported that they were linked to a micro-finance institution 
(MFI) at 63.1% (MTE 64.6%) and close to half (49.8%) were linked to markets (MTE
61.9%). It appears likely that the lower numbers of producers linked with finance and 
market in the final year might be attributed to COVID restrictions in the country and 
ceasing of MSIKA fieldwork. Summary responses are set out in Table 22 below.
MSIKA Final Evaluation Page 33 kadale@africa-online.net
Table 22: MSIKA Activities Reported by FBO Members in MTE & FE HH Surveys 
Which of the following activities from Land O’Lakes /MSIKA have you participated in?
Activity
MTE FE
Male Female Total Male Female Total
% % % % % % 
a. Training in improved horticulture practices by MSIKA 
staff or a lead farmer 93.8 92.8 93.2 85.8 86.7 86.2 
b. Training in horticulture post-harvest handling by 
MSIKA staff or a lead farmer 91.1 92.3 91.8 79.8 83.6 81.5 
c. Training in marketing 81.7 83.5 82.8 79.1 79.3 79.2 
d. Training in financial literacy 73.5 74.0 73.8 74.4 72.5 73.5 
e. Training in gender equality 77.4 70.7 73.4 81.3 78.4 80.0 
f. Training/membership of a village savings and loan
group
 85.6 89.5 87.9 85.6 90.5 87.8 
g. Been linked to an MFI 62.3 66.1 64.6 63.0 63.3 63.1 
h. Been linked to markets to sell horticulture products 59.5 63.5 61.9 47.3 52.9 49.8 
i. Participated in international/District trade fair 28.4 22.9 25.1 17.0 14.4 15.8 
j. Any other? 0.4 - 0.2 1.9 0.2 1.1 
Total 257 389 646 535 444 979 
N=646 & 979, multiple response possible
3.2.4 Agricultural Practices
3.2.4.1 Knowledge
Due to the change in interviewing methodology, it was agreed with USDA that the 
questions on knowledge of agricultural practices would not be asked. Gathering the 
information on use/application of the practices was more important and, applying a 
practice implies that there was knowledge of the practice(s) applied. This decision 
affects Indicator 14 (Percentage of producers who can recite five or more 
improved agricultural techniques and technologies). 
At mid-term when data for indicator 14 was collected, the level of unprompted 
knowledge of five or more practices was 56.7%. The revised target for LoP was 66.0%. 
The qualitative interviews with farmers focused more on training and application than 
knowledge, so the detail is covered in the application of practices section below.
3.2.4.2 Application of Practices
Indicator 7 is the number of individuals who have applied new techniques or 
technologies, as a result of USDA assistance. 
To measure this indicator required a review of the baseline instrument to enable an 
appropriate comparison, as MSIKA adopted a different approach to train producers 
than anticipated at the baseline. This change was to train producers in a wide range of 
practices applicable across many crops, plus some specific practices for particular 
crops, rather than training specific to each crop. There were additional practices 
introduced into the training than were measured at baseline, so a direct comparison of 
these is not possible. 
This question in the baseline and the FE survey was prompted, with respondents asked 
about all the practices for one selected crop (28 practices for onion and tomato and 19 
practices for potato and mango). The full list of practices and percentage of 
respondents applying can be found in Annex 5. 
MSIKA Final Evaluation Page 34 kadale@africa-online.net
The finding on application of practices is that 99.5% (all but five out of 1,074) of 
responses across the four crops had applied one or more new practice in the period 
August 2019-July 2020. This is a very high achievement level equating to 30,778 of 
the project population, which is all except 144 beneficiary producers. The application 
rates of men and women were very similar with 99.7% of men applying one or more 
new agricultural practice, and 99.4% of women applying them.
The LoP target for applying one or more new practice (indicator 7) is 31,680 based on 
an expected 36,000 individuals trained in agricultural practices and food security
(Indicator 4). This is a rate of 88.0% of those trained using at least one new agricultural 
practice. Although the actual application rate is very high (99.5%) for those that were 
trained, the number of producers trained in basic agriculture at 30,922 is lower than 
anticipated, as the reported achievement of indicator 4 includes marketing and 
financial training for producers, as well as training of non-farmers in basic agriculture 
(extension staff). As the FE population was lower than the total of individuals trained 
in basic agricultural production, then the total number of people applying new practices 
(Indicator 7) is also relatively lower, at 30,778. This is 97.2% of the overall LoP target. 
The male target was exceeded (107.1%) and there was a shortfall on female (90.5%).
Based on 99.5% of the sample applying the practices, a slightly higher achievement 
on the number of trained farmers overall would have resulted in the LoP target being 
met. Further training was disrupted by COVID, but training subsequent to finalising the 
sample frame suggests the LoP target for indicator 7 would likely have been exceeded. 
Table 23 below shows the application of practices across the eight disaggregate 
categories. The achievement ranged between 68.8% to 98.3% of respondents
applying one or more. The highest application rates were for disease management 
(98.3%), pest management (97.6%), and soil related fertility and conservation (96.6%). 
The lowest rates were for crop genetics (79.9%) and climate mitigation and adaptation 
(68.8%), though these are still good scores. The results for men and women were very 
similar. Details of the individual practices are in Annex 5. 
The baseline and FE surveys had practices that were common to both surveys, but 
there were some practices in the baseline that were not asked in the FE survey, such 
as green manuring, and some practices asked in the FE survey that were not in the 
baseline, such as staking young trees and intercropping. This means it is not valid to 
compare the FE and baseline disaggregated categories. 
Table 23: Application of New Agricultural Practices per Category, FE Survey, 2020
# of individuals that applied one new techniques & technologies as a result of USDA assistance
Category/Disaggregate FE # applying FE % FE population applying 
Total applying 1,069 99.5 30,778
Male 579 99.7 13,542
Female 490 99.4 17,236
New 821 76.4 23,638
Continuing 248 23.1 7,140
Crop Genetics 858 79.9 24,703
Pest Management 1,048 97.6 30,173
Disease management 1,056 98.3 30,404
Soil-related fertility and conservation 1,038 96.6 29,886
Irrigation 878 81.8 25,279
Water management (non-irrigation based) 963 89.7 27,726
Climate mitigation or adaptation 739 68.8 21,277
Other 903 84.1 25,999
Total w/one or more improved techniques 
or technologies 1,069 99.5 30,778
MSIKA Final Evaluation Page 35 kadale@africa-online.net
An overall achievement rate of 99.5% of respondents applying one or more new 
practices is very high, though it masks other aspects of MSIKA’s performance. The 
consultants calculated the mean number of new improved practices per beneficiary 
producer was 15.8. This compares well to the MTE with a mean of 9.8 practices, 
suggesting further deepening of the application of practices. This may also be a factor 
in the relatively high increases in yields reported in section 3.2.7 below.
Table 24 indicates the number of new practices applied by crop. This ranged from 
implementing no practices to a maximum score of 28 new practices applied on tomato 
and on onion, as well as a maximum score of 19 new practices for potato and for 
mango. The mean number of new practices applied per producer was highest for 
tomato (20.9), onion (20.6) and potato (13.5). Mango had the lowest number of new 
practices applied per producer (8.1). The low value for mango is attributed to most
farmers not managing mango because they inherited these trees and they produce 
without any effort. Field crops more clearly require their effort and time in to produce.
Table 24: Application of New Practices by Crop at FE, 2020
Crop Total 
Producers
Total 
Practices 
applied by 
all 
producers
Mean 
practices 
applied per 
producer 
Minimum 
practices 
applied per 
producer 
Maximum 
practices 
applied per 
producer 
% 
applying 
none
Tomato 276 5775 20.9 11 28 -
Onion 264 5429 20.6 9 28 -
Irish Potato 266 3600 13.5 4 19 -
Mango 268 2177 8.1 - 19 1.8
Total 1060 1074 15.8 - 28 1.8
The detailed application by crop is set out in Annex 5 and provides the detail for each 
practice, including the percentage that are using it and the comparison with the 
baseline usage rates. This table covers 41 practices, across the seven crops with 
baseline, FE and difference scores. 
Across the field crops, ridge and plant spacing, irrigation, selection of best seedlings 
and use of nursery generally scored the highest. For mangoes, pruning scored the 
highest, followed by composting water capture and digging of planting holes prior to 
planting which all had a same score. Use of fish soup, testing soil acidity, and use of 
ash/lime generally scored poorly across all the four crops. Specifically, for mangoes, 
scores were lower for grafting, budding, draining excess water and deflowering of first 
flowers scored poorly. While these are practices MSIKA was to promote more following 
the mid-term findings which are similar to the FE findings, it is a possible that the low 
scores were affected by the abrupt reduction in activities and the inability of farmers to 
conduct meetings at FBO level due to COVID.
There were roughly ten practices where there has been an increase in use of 20% or 
greater between baseline and FE across at least for the three field crops. These 
practices were: composting, manuring, mulching, irrigation, soil and water 
conservation, selecting varieties and spraying. 20% or greater also applied water 
capture and pruning for mango. Overall, there has been an improvement from baseline 
to FE across all the crops except ridging for potatoes with a 5.5% decrease from the 
baseline. The latter may be a function of it already being a very common practice and 
statistical variation. 
Participant producers may have been using practices before MSIKA or leant them from 
other sources, so that the impact of these is not exclusively attributable to MSIKA. 
MSIKA Final Evaluation Page 36 kadale@africa-online.net
Asking about practices used for the first time in the last year, combined with the 
information on source, enables clearer attribution to MSIKA for the subsequent results. 
Respondents were asked in the survey what practices they had used for the first time 
in the last year (August 2019-July 2020). In general, the scores for first use of practices 
were high, across the four field crops. It is not surprising that for some of the practices, 
respondents reported to have not used it for the first time in the last 12 months since 
the sample population took into account all the individuals trained since project 
commencement. This means, it is likely that some individuals have implemented the 
practices for the first time after MSIKA training, but not within the last 12 months. Table 
25 below sets out the first use of practices for the four field crops. Across the field 
crops, composting, soil and water conservation, choosing the variety, uprooting 
infected plants and use of fertiliser scored consistently well, with fish soup and testing 
soil acidity scoring generally lower. 
For tomato, 11 practices have been applied for the first time by over 40% of 
respondents. Composting was applied for the first time by over 50% of respondents. 
There was a high uptake of practices relating to uprooting infected plants, choosing 
the variety and use of recommended plant and ridge spacing. The linked practices of 
testing for acidity and use of lime/ash scored poorly, along with the use of fish soup.
For onion, 11 practices were applied for the first time by over 40% of respondents. 
Over 50% of respondents applied scouting for pests and diseases as well as sterilizing 
nursery beds before planting out. As with tomato, practices relating to composting, 
use of recommended plant and ridge spacing scored highly. Testing for acidity and 
draining excess water scored poorly, as did use of fish soup.
For potato, 5 practices were reportedly applied for the first time by over 40% of 
respondents, with the highest scoring being selection of the best seed potato (44.4%), 
uprooting infected plants and burning (42.9%) and choosing the variety (42.9%). No 
practices were used by over 50% of respondents. The scores for potato were generally 
lower than for tomato and onion. The use of fish soup scored poorly.
For mango, 2 practices were used for the first time by over 40% of respondents, but
no new practices were used for the first time by over 50% of the respondents. The 
lowest uptake by respondents was for budding, fish soup and scouting for pests and 
diseases, all at less than 20%.
MSIKA Final Evaluation Page 37 kadale@africa-online.net
Table 25: Use of Practices for the First Time, FE Survey, 2020
Practice used 
Tomato Onion Potato Mango 
% % % % 
1 Composting 51.1 49.5 37.1 40.4
2 Manuring 42.5 41.5 25.9 31.1
3 Ridging 36.7 33.0 31.7 n/a
4 Mulching 43.0 40.1 26.8 25.7
5 Crop rotation 36.7 38.2 27.3 n/a
6 Minimum tillage 31.7 29.2 n/a n/a
7 Planting seeds in a nursery before planting out 38.5 40.1 n/a n/a
8 Staking 37.6 n/a n/a n/a
9 Succession planting 33.0 n/a 23.4 n/a
10 Using irrigation 30.8 34.0 24.9 24.6
11 Soil and water conservation 39.4 36.8 34.1 42.1
12 Draining excess water 30.3 24.5 n/a 24.0
13 Testing soil acidity 18.1 21.2 n/a 19.7
14 Adding lime or ash to soil before planting to 
reduce acidity 24.0 27.8 n/a 25.7
15 Choosing the variety 47.1 42.5 42.9 n/a
16 De-suckering or removing unwanted side-shoots 41.2 n/a n/a n/a
17 Spraying for pests and disease 35.3 33.5 27.8 25.1
18 Using recommended plant and ridge spacing 46.2 44.8 42.0 n/a
19 Using recommended fertilizer and application 
rates
43.4 44.3 38.0 n/a
20 Sterilizing nursery beds before planting 38.7 81.8 n/a n/a
21 Scouting for pests and diseases 32.1 69.3 40.0 17.5
22 Uprooting infected plants and burning 48.9 38.2 42.9 25.7
23 Sowing seed in row/groove nursery 42.1 42.5 n/a n/a
24 Using fish soup / sugar solution as bait for insects 15.4 17.0 18.0 15.8
25 Planting in sunken beds 36.7 41.0 n/a -
26 Hardening off before planting out 40.3 38.7 n/a n/a
27 Selecting the best seedlings for planting out 43.4 38.7 44.4 n/a
28 Intercropping 30.8 29.2 22.9
29 Green manuring n/a n/a n/a n/a
30 Water capture n/a n/a n/a 39.9
31 Earthing up n/a 34.9 25.4 n/a
32 Clipping plant ends n/a 35.8 n/a n/a
33 Using raised beds n/a 34.0 n/a n/a
34 Chitting n/a n/a 33.7 n/a
35 Pruning n/a n/a n/a 42.6
36 Grafting n/a n/a n/a 21.9
37 Budding n/a n/a n/a 15.8
38 Staking young trees n/a n/a n/a 32.8
39 Digging of planting holes prior to planting n/a n/a n/a 31.7
40 Deflowering of first flowers n/a n/a n/a 26.2
Total 221 212 205 183
From the qualitative interviews, growers of all four crops were able to identify many of 
the practices they were trained in and that they were able to apply. Between them, 
tomato growers identified 25 practices they had applied of which spraying for pests 
and diseases was the most commonly applied and crop rotation, intercropping, 
identifying the type of the seed that buyers are looking for, applying lime and ash to 
control salinity, uprooting infected plants and burning. 
MSIKA Final Evaluation Page 38 kadale@africa-online.net
Onion farmers identified 28 practices in the qualitative interviews. The most commonly 
mentioned practice was application of recommended fertilizer with the least applied 
practice being making beds sunken or raised, uprooting infected plants, choosing the 
variety for the growing characteristics and market that aiming for, clipping plant ends 
to allow seedlings make fresh sprouts, weeding and earthling up. 
The Irish potato growers noted 13 practices from the training that they were able to 
apply. The most commonly applied practice was using recommended fertilizer, while 
soil and water conservation by planting vetiver, selection of best seed potatoes, 
uprooting infected plants, and land preparation by removal of wastes were the least 
commonly applied. 
The mango farmers gave 12 practices that they had applied. The most commonly
applied practice was manuring and the least applied were scouting for pests and 
diseases, fertilizer application, weeding, and grafting. 
It was clear from the KIIs that the farmers had applied a wide range of practices and 
that these have benefited them. 
"I have seen that these practices are of great profit to my farming.” Ntcheu, male tomato 
farmer
"I was able to use less fertilizer because I depend mostly on the use of manure". 
Mangochi male onion farmer. 
"I like the benefits am getting through practicing what I was trained on, I'm making 
profits and can support my family now." Ntcheu, female potato grower
"The training proved that mango production can be taken as a business and the
proceeds from the business can be used to buy food.” Ntcheu farmer, female. 
Respondents were asked why they had not used the ‘other’12 practices between 
August 2019 to July 2020. Due to the shortening of the questionnaire, respondents 
were asked the overall reasons why they did not use some of the other practices, rather 
than on each practice not used, as was asked at mid-term. ‘Other’ therefore refers to 
the practices not used. This was a multiple response question for which the enumerator 
classified the single or multiple responses given rather than reading out the list of 
possible responses. Respondents were asked to give more than one response (“any 
other reasons?”). The question was asked for the overall practices relating to a specific 
crop and are reported below by crop. 
The most common reason across all crops was that the respondent ‘did not know it’, 
with all crops having more than 44% of respondents saying they did not know some of 
the practices. This is an improvement on the mid-term where over 90% of respondents 
said they did not know some of the practices, suggesting that there has been an 
increase in knowledge overall. This was followed by ‘did not fully understand it’ (13.9 
- 30.2%), ‘need did not arise’ (12.0 – 25.4%, ‘inadequate resources’ (6.8 – 20.5%) and 
it was ‘expensive to use’ (9.8% - 15.3%).
The ‘did not know’ response might indicate that some practices were not covered in 
the training, or that these were forgotten by the respondent. It would be surprising if 
respondents knew all the practices, as there were many and respondents may have 
been growing two crops,13 so there was a great deal to ‘know’. Respondents may have 
lost concentration during the training or forgotten the practice, especially as many had
limited education. 
12 Described as ‘other’ as the respondents had been read a list of practices and stated the ones they 
used, so this refers to the ones they did not use. 13 The mean number of crops per farmer was 2.18.
MSIKA Final Evaluation Page 39 kadale@africa-online.net
The second most common response was that ‘did not fully understand it. The third 
most common response was that the ‘need did not arise’ which can be the case for
pest and disease management. The ‘inadequate resources’ and ‘expensive to use’ 
responses suggest that that some practices are too expensive for some respondents
to apply. This is inevitable, given the poverty of many respondents. 
The least commonly given reasons were: ‘inefficient’, ‘too difficult for farmers’, ‘not 
confident to use’ and ‘not confident it will work’. This means that these practices are 
not seen as inefficient, too difficult, or confident that they will work or to use them. In 
other words, the practices are mainly appropriate. Table 26 sets out the reasons given 
for not using practices by crop. Male and female responses were very similar.
Table 26: Why Other Practices are not Used, by Crop, FE 2020
Overall, what are the reasons why you did not use the other improved practices? 
(Unprompted)
Reasons
Tomato Onion Potato Mango 
% % % % 
Did not know it 56.9 56.5 44.0 59.0
Did not understand it 23.2 25.2 13.9 30.2
Need did not arise 22.1 22.1 12.0 25.4
Inadequate resources 17.4 19.8 6.8 20.5
Expensive to use it 13.8 15.3 9.8 13.1
Other 8.7 9.9 3.0 6.3
Time consuming 7.6 7.6 4.5 6.0
Not confident it will work 7.2 6.9 22.2 5.2
Too difficult for farmers to do it 6.2 6.1 21.8 4.5
Not confident I can use it properly 3.6 3.8 24.8 4.1
Inefficient 1.8 3.4 11.3 3.4
Sample size/ difference 276 262 266 268
N= 1,074
3.2.5 Crop Inputs
The FE HH survey did not cover crop inputs so as to reduce the survey length. 
Information on inputs was included in the MTE. 
In summary from the MTE, a high proportion of respondents (60-80% range) were 
using fertilizers, insecticides, watering cans, sprayers, certified seed, compost/manure 
and fungicide, particularly for tomato, onion and potato. These showed improved 
uptake in nearly all areas. The most frequent source of inputs was ‘own/personal’, 
followed by ‘agro-dealers’ at 74-80% of respondents for tomato, onion and potato. 
In terms of ease of getting inputs, for tomato, onion and potato, between 67-70% of 
respondents at mid-term found accessing inputs ‘somewhat hard’ or ‘very hard’. 
Growers of mango reported easier access to inputs, but this was partly because around 
a third did not buy any inputs at all. 
Overall, the mid-term reinforced the importance for MSIKA to work with agro-dealers 
to continue improving access.
MSIKA Final Evaluation Page 40 kadale@africa-online.net
3.2.6 Land Area Under Improved Practices
Indicator 6 is the land area under improved growing techniques and 
technologies (‘growing practices’) as a result of USDA assistance. 
The FE survey asked for the improved practices applied and the area of land on which 
those practices were applied. The land area was measured in hectares (ha) for the 
12-month period from August 2019-July 2020, over up to three separate planting and 
harvest cycles for tomato, onion and potato and one cycle for mango. All cycles were 
totalled to give the total land area. It was agreed after consultation with USDA to 
exclude questions related to the disaggregates for this indicator to reduce interviewing
time. 
The total land area for respondents on which one or more improved practices was 
applied across all growing/harvest cycles14, was 276 ha. The mean land area on which 
improved growing practices were applied was 0.26 ha per respondent from August 
2019 – July 2020. If this mean is applied to the FE population of 30,778, improved 
growing practices were applied on an estimated 7,909 ha. The reduction in total land, 
as well as the mean land area since mid-term due could be due to the reduced number 
of crops in the fourth year. 
To enable a comparison with the FE, the MTE was also revised by taking into account 
the land under improved practices for the four crops, rather than the original seven. 
The mean land area on which improved growing practices were applied was 0.36 ha 
per respondent from July 2018 – June 2019. When this mean is applied to the MTE 
population of 8,442, improved growing practices were applied on an estimated 3,045 
ha. Overall, there has been progress from the MTE to the FE and the indicator has 
surpassed its LOP target. This reflects the decisions made to reduce activity due to the 
shortfall in funding.
Table 27 sets out the land area under improved practices. 
MSIKA reported 6.3 ha of demonstration plots on which practices were applied in Year 
4, down from 12.77 in Year 3. This reflects the decisions made to reduce activity due 
to the shortfall in funding.
Table 27: Land area in ha under improved practices, MTE and FE HH surveys
Number of hectares of land under 
improved techniques or technologies as 
a result of USDA assistance 
Crop MTE FE 
Overall 3,045 7,909
Tomato 1,362 2,506
Onion 276 2,048
Potato 1,265 2,744
Mango 142 610
3.2.7 Production 
A key purpose of the training and other activities is to increase yields for producers.
This is measured through Indicator 1, which is the Percentage change in yield of 
beneficiary producers as a result of USDA assistance. 
The baseline values were tomato at 2,617 kgs/ha, onion at 3,575 kgs/ha, potato at 
3,329 kgs/ha and mango at 4,347 kgs/ha. Following the MTE, the LoP targets were 
14 Land under each cycle was counted separately and summed to give the total.
MSIKA Final Evaluation Page 41 kadale@africa-online.net
revised to overall (+26%), men (+29%), women (+24%), tomato (+25%), potato
(+29%), onion (+30%) and mango (+21%). 
The baseline only had data for one growing cycle, so the FE survey data is reported 
for the comparable production and cycle only. 
One of the recommendations which was made in the MTE was to focus on individual 
crops when reporting the yields, as it is difficult to aggregate them and compare them 
over time due to different weightings relating to number of farmers growing that crop 
at that time. The consultants therefore report on the individual crop comparisons.
As noted in the methodology section, the baseline yield values were revised to use a 
consistent methodology. The baseline calculations for six of the seven crops (all but 
tomato) did not remove outliers, which skewed the yield values. The consultants set 
the maximum yields using FAO maximum expected yield assuming application of GAP. 
These were used in the analysis of the FE, so the baseline and FE were comparable.
The FE HH survey findings on yield are standardised into kgs/ha to allow comparisons 
between crops and with the baseline. Median yields15 have been calculated as well as 
means, as these tend to moderate more extreme values and can be useful to review. 
The crop with the highest yield in kgs/ha was mango with a mean of 7,078 kgs/ha 
followed by potato at 4,931 kg/ha. Onion and tomato had a mean 4,896 kg/ha and 
4,418 kg/ha respectively. All median values were lower than their equivalent means, 
notably tomato at 2,635 kg/ha. The comparison of median values shows the range for 
all crops is between 2,635 kgs/ha and 5,216 kgs/ha 
Comparing male and female mean yields, males generally had higher yields than 
female producers. If median values are used, the gaps are reduced, but still a bit 
higher in favor of men, except mango with women having higher yields. Table 28 sets 
out mean and median yields by crop standardized to kgs/ha and weighted. 
Table 28: Producer Yields in kg/ha, FE HH survey, 2020
Yield 
Kg/ha
FE
Male Female Overall
Mean Median Sample Mean Median Sample Mean Median Sample
Tomato 
(kg/ha) 5,369 3,162 154 3,207 1,897 121 4,418 2,635 275
Onion 
(kg/ha) 5,871 4,320 146 3,647 1,956 114 4,896 3,260 260
Potato 
(kg/ha) 5,639 4,323 133 4,133 2,849 118 4,931 3,751 251
Mango 
(kg/ ha) 7,327 4,890 121 6,853 5,433 134 7,078 5,216 255
N=1,074 
As noted in the methodology, the baseline results were updated during the mid-term 
to remove outliers above the FAO maximum yields per ha. Table 29 sets out the 
revised baseline yields, with a focus on only the four crops which were of focus in year 
four of MSIKA project implementation and sets out the comparison between the FE 
survey and baseline with the outliers removed from both using the same methodology.
15 Medians can be useful measures of the average in a dataset that has zero as the lowest value, but 
an unlimited maximum value, such that higher values can distort the mean average. 
MSIKA Final Evaluation Page 42 kadale@africa-online.net
Table 29: Producer Yields in kg/ha, Revised Baseline 2017
Yield in 
Kg/Ha
Impact Baseline
Male Female Overall
Mean Median Sample Mean Median Sample Mean Median Sample
Tomato 
(kg/ha) 3,305 1,537 173 1,858 880 157 2,617 1,217 330
Onion 
(kg/ha) 4,358 2,748 58 2,061 797 30 3,575 2,005 88
Potato 
(kg/ha) 3,946 2,983 136 2,684 1,637 130 3,329 2,341 266
Mango 
(kg/ ha) 4,017 2,580 112 4,660 2,886 118 4,347 2,737 230
N=686
Compared to the revised baseline, the FE had higher mean yields across all the crops. 
The percentage changes for the crops were as follows; tomato (+69%), onion (+37%), 
potato (+48%) and mango (+63%). Similarly, the percentage changes for median 
values are higher, notably tomato at 116%. These are set out below. 
Table 30: Yield Comparison, FE vs baseline by Crop 
Crop Baseline FE - Overall FE vs Baseline
Mean Median Mean Median Mean Median
Tomato (kg/ha) 2,617 1,217 4,418 2,635 69% 116%
Onion (kg/ha) 3,575 2,005 4,896 3,260 37% 63%
Potato (kg/ha) 3,329 2,341 4,931 3,751 48% 60%
Mango (kg/ ha) 4,347 2,737 7,078 5,216 63% 91%
COVID potentially had an effect on producer yields, so comparison with the baseline 
is also made based on those producers that said they were not affected by COVID. 
Table 31 below sets out the comparison of baseline and FE yields for those not affected 
by COVID. 
The FE mean yields for those not affected by COVID16 compared to the baseline show 
increases in mean yields for tomato (+72%), onion (+42%), potato (+49%) and mango 
(+64%). The change in the yields compared those using the whole sample is quite 
small, though for all four crops the yields for those not affected by COVID were higher 
by up to five percentage points. This is consistent with the expected impact from 
COVID, which occurred in the period when some farmers were planting and growing, 
such that governmental restrictions made access to inputs more challenging due to 
movement restrictions and market closures. Median values for those not affected by 
COVID have a similar pattern to the whole FE sample.
Overall, the comparison of the FE group not affected by COVID, shows progress in 
yield increases for the crops. 
The evaluation team reviewed the yield difference between those affected and not 
affected by COVID-19 and related this to the baseline yields. Overall, the yields were
higher for those not affected by COVID compared to those affected by COVID for 
tomato, potato and mango. The differences are set out in the table below and were
significant.
16 936/1074 (87.2%) were not affected by COVID in their production (excluding outliers). The main 
effect of COVID was on sales.
MSIKA Final Evaluation Page 43 kadale@africa-online.net
Table 31: Producers Affected/Not Affected by COVID, Yield Comparison vs Baseline
Crop 
Comparison of % changes from Impact Baseline 
Not affected by Covid-19 Affected by Covid-19 Overall 
Mean Median Mean Median Mean Median 
Tomato (Kg/ha) 72.1% 118.7% 49.1% 116.5% 68.8% 116.5%
Onion (Kg/ha) 42.3% 73.5% 72.9% 64.1% 36.9% 62.7%
Potato (Kg/ha) 49.4% 60.4% 37.9% 47.7% 48.1% 60.2%
Mango (Kg/ha) 64.3% 86.6% 30.5% 98.5% 62.8% 90.6%
Table 32 below compares male and female yields for those not affected by COVID. 
There was a slight positive yield changes across all the crops for men, while for women, 
tomato and potato had reduced slightly. 
Table 32: Yield Comparison by Sex, FE vs baseline
Crop
FE Not affected by Covid 19 vs Baseline 
Males Females 
Mean Median Mean Median 
Tomato 67% 122% 71% 101%
Onion 39% 60% 91% 151%
Potato 44% 45% 52% 74%
Mango 83% 90% 49% 84%
As there are only minor differences between those not affected by COVID and the 
overall sample, the consultant reports means for the overall population in the summary 
indicator table (see Annex 1) along with the revised baseline with outliers removed.
It is worth noting overall that tomato (31.7%), onion (25.9%) and potato (34.7%), 
account for more than 92% of the production by land area where new practices were 
applied within the reporting time period (August 2019 to July 2020), and so are the 
most important crops for MSIKA. 
With positive increases in mean yields from 37% to 69% across the four crops, MSIKA 
has surpassed the LoP revised targets for each crop.
3.2.8 Post-Harvest Handling Knowledge and Application
PHH knowledge was not covered in the FE HH survey so as to shorten the interview.
Knowledge is a steppingstone to application, and can be assumed to be present from 
a practice(s) being applied, so more reliance is placed on the application than 
knowledge in this final evaluation. Knowledge was explored in the MTE and the details 
below are drawn from the MTE findings. The qualitative interviews with farmers mainly 
covered the application of practices, so are covered in the next section.
The findings on PHH knowledge are set out in the section immediately below, followed 
by those on PHH use in section 3.2.8.2. 
3.2.8.1 Knowledge of PHH
This section summarised key points from the mid-term HH survey.
For tomato, onion and potato, the mid-term unprompted knowledge scores were 
around 20%, suggesting that there was scope to improve knowledge. The mid-term 
unprompted PHH practices scores for mango were generally very low and below those 
for the field crops. 
MSIKA Final Evaluation Page 44 kadale@africa-online.net
At mid-term, ‘MSIKA Extension worker/lead farmer’ was consistently the highest 
source of PHH information for field crops with a wide margin over the next highest 
source, which was either ‘always known’ or a ‘relative/other farmer’. For mango, 
‘always known’ was the most common source for respondents, which is interpreted to 
be passed down knowledge. 
As noted earlier, the baseline question was asked as a prompted question, so it is not 
valid to compare the mid-term findings to it.
3.2.8.2 Use of PHH practices
Respondents were asked in the FE HH survey which practices they used from a 
prompted list of all PHH practices per crop. The findings at FE are set out in Table 33
with the baseline findings in Table 34 below.
The highest scores on PHH use across all the four crops were for ‘grading by size, 
color and shape’ ‘ sorting by variety, size/maturity, damaged, ripeness’ and ‘packing - 
loading into appropriate containers for storage or transportation’. The lowest score was 
‘lifting potatoes with a fork/prong, not a hoe’ which only 11.3% of the sample had used 
in the last year. This is not surprising as most farmers would most likely prioritize buying 
a hoe as it can be used for several crops compared to crop specific tools like a fork or 
prong. As farmers specialise and want to increase quality and yield, these tools would 
likely become more relevant.
There were several changes to the FE questions that were made at mid-term to reflect 
the practices MSIKA was teaching. These did not appear in the baseline or were 
worded differently, making direct comparisons more difficult on these specific 
practices. Having noted this, overall, the use of PHH practices for the four crops has 
increased across most PHH practices compared to the baseline, even though baseline 
scores were quite high to start with. For tomato, the scores are particularly impressive 
as they are all over 83%, with most over 90%. For onions, the lowest score is 73.5%, 
with most scores over 75%. For potato, apart from a very low 11.3% for lifting using a 
fork/prong, four scores are over 90% and the next lowest is 66.2%. The only score for 
mango lower than in the baseline was the use of a special harvesting pole (FE 48.5%, 
baseline 70.1%). Other than storing in crates (a big investment), and catching before 
it falls (which is difficult), all the other practices scored over 82%
MSIKA Final Evaluation Page 45 kadale@africa-online.net
Table 33: Use of PHH Practices by Crop, FE HH survey 2020
Practices used (Prompted)
Tomato Onion Potato Mango 
% % % % 
Timely harvesting not too early so immature and not too 
late so over-ripe 90.2 n/a n/a n/a
Harvesting & selling green/fresh onions with stalks on n/a 76.5 n/a n/a
Harvesting when plant tops wilt and start to wither; 
slashing tops close to soil to cure in soil a day before 
harvesting
n/a n/a 68.4 n/a
Harvesting when ripening just started and color begins to 
change n/a n/a n/a 86.9
Grading by color, size and shape 89.5 87.1 96.2 87.3
Storing in cool places out of the sun prior to sale 96.0 n/a n/a 90.7
Sorting by variety, size/maturity, damaged, ripeness 92.8 87.1 96.6 90.7
Packing - Loading into appropriate containers for storage 
or transportation 83.3 73.5 92.1 88.4
Not over-filling baskets & bags or using very big bags, as 
this damages product at the bottom 93.5 84.1 n/a 82.5
Handling carefully to avoid damage 97.8 n/a 94.7 91.0
Not stacking more than three containers high to avoid 
crushing tomatoes in lower containers.
92.8 n/a n/a n/a
Adding soft dry grass as soft litter 92.0 n/a n/a n/a
Storing in crates to prevent damage in transporting n/a n/a n/a 53.4
Lifting when tops have bent over & dried; n/a 76.1 n/a n/a
Drying (curing) onions in the sun or under a shade before 
storing to improve in keeping quality n/a 74.6 n/a n/a
Trimming dried stems and roots before packing n/a 73.5 n/a n/a
Lifting potatoes with a fork/prong, not a hoe n/a n/a 11.3 n/a
Rubbing off surface dirt after harvest; or after drying & 
curing
n/a n/a 72.9 n/a
Keeping clean potatoes in a cool dry place to cure or heal 
surface damage, before packing, storage & selling n/a n/a 89.8 n/a
Harvesting by climbing the tree n/a n/a n/a 87.3
Harvesting by using a special harvesting pole n/a n/a n/a 48.5
Catching fruit before it falls using bags or spreading sheets n/a n/a n/a 66.8
Using night vented sheds for short term storage for selling 
outside main season n/a n/a 66.2 n/a
Washing to improve appearance n/a n/a n/a 84.0
Sample size 276 264 266 268
Multiple response possible
MSIKA Final Evaluation Page 46 kadale@africa-online.net
Table 34: Use of PHH Practices Across the Four FE crops, at Baseline
Practices used (Prompted)
Tomato Onion Potato Mango 
% % % %
Timely harvesting not too early so immature and not too 
late so over-ripe 94.1 n/a n/a n/a
Harvesting & selling green/fresh onions with stalks on n/a 60.6 n/a n/a
Harvesting when plant tops wilt and start to wither; 
slashing tops close to soil to cure in soil a day before 
harvesting
n/a n/a n/a n/a
Harvesting when ripening just started and color begins to 
change n/a n/a n/a 86.8
Grading by color, size and shape 71.2 58.5 82.7 48.0
Storing in cool places out of the sun prior to sale 82.0 n/a 68.3 44.6
Sorting by variety, size/maturity, damaged, ripeness n/a n/a n/a n/a
Packing - Loading into appropriate containers for storage or 
transportation 63.4 55.3 53.2 34.3
Not over-filling baskets & bags or using very big bags, as 
this damages product at the bottom 73.2 55.3 64.0 43.1
Handling carefully to avoid damage 76.1 n/a 70.1 52.0
Not stacking more than three containers high to avoid 
crushing tomatoes in lower containers. 69.9 n/a n/a n/a 
Adding soft dry grass as soft litter 70.6 n/a n/a n/a
Storing in crates to prevent damage in transporting n/a n/a n/a 9.3
Lifting when tops have bent over & dried; n/a 60.6 66.5 n/a
Drying (curing) onions in the sun or under a shade before 
storing to improve in keeping quality n/a 27.7 n/a n/a
Trimming dried stems and roots before packing n/a 51.1 n/a n/a
Lifting potatoes with a fork/prong, not a hoe n/a n/a 6.8 n/a
Rubbing off surface dirt after harvest; or after drying & 
curing n/a n/a 36.7 n/a
Keeping clean potatoes in a cool dry place to cure or heal 
surface damage, before packing, storage & selling n/a n/a n/a
n/a
Harvesting by climbing the tree n/a n/a n/a 88.7
Harvesting by using a special harvesting pole n/a n/a n/a 70.1
Catching fruit before it falls using bags or spreading sheets n/a n/a n/a 35.3
Using night vented sheds for short term storage for selling 
outside main season n/a n/a n/a n/a
Letting lifted potatoes get a dry surface before packing and 
selling n/a n/a n/a n/a
Washing to improve appearance n/a n/a n/a 43.1
Sample size 306 94 278 204
Multiple response possible
Respondents were asked what PHH practices they used for the first time in the last 
year. The baseline did not have questions related to first time usage of PHH practices, 
as that would not have made sense in the context of new practices yet to be taught. 
Therefore, the mid-term data (Table 36) is compared with the FE data (Table 35).
The findings on first used in the mid-term survey were relatively high with the uptake 
of practices such as ‘timely harvesting’, ‘grading’, ‘sorting’, ‘cool storage’, and ‘careful 
handling’ first used by over 60% of respondents at that point. Even for practices with 
relatively low scores, the uptake was often in the 30-60% range. This means there was 
limited room for further improving.
MSIKA Final Evaluation Page 47 kadale@africa-online.net
‘Storing in crates and appropriate containers’, ‘lifting by fork’ and ‘climbing to harvest’ 
were the lowest scoring. For storage, this could be for cost reasons. For lifting by fork, 
this might be down to the long-standing use of hoes and/or unavailability/cost of forks. 
Prior to 2019, MSIKA team said they had not put as much emphasis on PHH training 
as on growing practices and this likely influenced the scores on mango and the 
generally lower scores on PHH than on improved growing practices. However, for the 
final year of the program, there was more PHH training undertaken.
Across the practices used for the first time in the last 12 months in the FE HH survey, 
the pattern was similar to that of the mid-term. For instance, practices such as ‘timely 
harvesting’, ‘grading’, ‘sorting’, ‘cool storage’, and ‘careful handling’ were first used by 
over 50% of respondents. Even for practices with relatively low scores, the uptake 
was often in the 30-50% range. It should be noted for interpretation that if there was 
already a high first use in earlier years, the scores in later years, as a proportion of the 
whole sample, will necessarily be reduced.
Table 35: First Used New PHH Practice, at FE, 2020
PHH practices used first time Tomato 
%
Onion 
%
Potato 
%
Mango 
%
Timely harvesting not too early so immature and not too 
late so over-ripe
60.4 n/a n/a n/a
Lifting when tops have bent over & dried n/a 56.6 n/a n/a
Lifting potatoes with a fork/prong, not a hoe n/a n/a 12.0 n/a
Harvesting & selling green/fresh onions with stalks on n/a 53.8 n/a n/a
Drying (curing) onions in the sun or under a shade before 
storing to improve in keeping quality n/a 58.2 n/a n/a
Rubbing off surface dirt after harvest; or after drying & 
curing n/a n/a 48.7 n/a
Trimming dried stems and roots before packing n/a 53.3 n/a n/a
Handling carefully to avoid damage 57.3 n/a 58.2 42.4
Storing in cool places out of the sun prior to sale 62.2 n/a n/a 54.0
Grading by color, size and shape 60.4 64.3 58.9 52.5
Sorting by variety, size/maturity, damaged, ripeness 58.5 63.7 62.0 52.5
Packing - Loading into appropriate containers for storage or 
transportation 57.9 53.8 53.2 39.6
Not over-filling baskets & bags or using very big bags, as this 
damages product at the bottom
57.9 63.7 n/a 51.8
Not stacking more than three containers high to avoid 
crushing tomatoes in lower containers. 54.3 n/a n/a n/a
Adding soft dry grass as soft litter 50.0 n/a n/a n/a
Using night vented sheds for short term storage for selling 
outside main season n/a n/a 41.8 n/a
Keeping clean potatoes in a cool dry place to cure or heal 
surface damage, before packing, storage & selling n/a n/a 58.9 n/a
Harvesting when plant tops wilt and start to wither; slashing 
tops close to soil to cure in soil a day before harvesting n/a n/a 53.2 n/a
Washing to improve appearance n/a n/a n/a 37.4
Storing in crates to prevent damage in transporting n/a n/a n/a 36.0
Harvesting when ripening just started and color begins to 
change n/a n/a n/a 49.6
Harvesting by climbing the tree n/a n/a n/a 29.5
Harvesting by using a special harvesting pole n/a n/a n/a 33.8
Catching fruit before it falls using bags or spreading sheets n/a n/a n/a 42.4
Total respondents 164 182 158 139
MSIKA Final Evaluation Page 48 kadale@africa-online.net
Table 36: First Used New PHH Practice, Across FE Crops at Mid-term, 2019
Use of PHH practices for the first time Tomato 
% 
Onion 
%
Potato 
%
Mango 
%
Timely harvesting not too early so immature and not too 
late so over-ripe 60.8 n/a n/a n/a
Lifting when tops have bent over & dried; n/a 61.8 59.3 n/a
Lifting potatoes with a fork/prong, not a hoe n/a n/a 8.7 n/a
Harvesting & selling green/fresh onions with stalks on n/a 18.4 n/a n/a
Drying (curing) onions in the sun or under a shade 
before storing to improve in keeping quality n/a 61.8 n/a n/a
Rubbing off surface dirt after harvest; or after drying & 
curing n/a n/a 42.7 n/a
Trimming dried stems and roots before packing n/a 21.1 n/a n/a
Handling carefully to avoid damage 54.9 n/a 52.0 43.5
Storing in cool places out of the sun prior to sale 64.1 n/a 50.0 47.8
Grading by color, size and shape 66.3 21.1 69.3 54.3
Sorting by variety, size/maturity, damaged, ripeness 63.7 30.3 63.3 50.0
Packing - Loading into appropriate containers for storage 
or transportation 47.6 34.2 48.7 47.8
Not over-filling baskets & bags or using very big bags, as 
this damages product at the bottom 60.8 25.0 n/a 47.8
Not stacking more than three containers high to avoid 
crushing tomatoes in lower containers. 57.5 n/a n/a n/a
Adding soft dry grass as soft litter 55.7 n/a n/a n/a
Using night vented sheds for short term storage for 
selling outside main season n/a n/a 35.3 n/a 
Keeping clean potatoes in a cool dry place to cure or 
heal surface damage, before packing, storage & selling n/a n/a 50.0 n/a
Harvesting when plant tops wilt and start to wither; 
slashing tops close to soil to cure in soil a day before 
harvesting
n/a n/a 59.3 n/a
Washing to improve appearance n/a n/a n/a 32.6
Storing in crates to prevent damage in transporting n/a n/a n/a 13.0
Harvesting when ripening just started and color begins 
to change n/a n/a n/a 28.3
Harvesting by climbing the tree n/a n/a n/a 10.9
Harvesting by using a special harvesting pole n/a n/a n/a 37.0
Catching fruit before it falls using bags or spreading 
sheets n/a n/a n/a 32.6
Total respondents 273 76 150 46
The qualitative interviews conducted at FE gathered feedback on the knowledge and 
application of PHHs by crop, as follows:
Of the 18 tomato farmers interviewed, two said they did not receive any PHH training 
for tomatoes. One farmer who was not growing tomatoes before receiving the training 
from MSIKA has not yet started applying the practices as the tomatoes were in the field 
at the time of the interview. 
For those that were trained, 10 practices were specifically mentioned that have been 
applied. These include: timely harvesting; storing in cool places out of the sun prior to 
sale; grading by color, size and shape; adding soft dry grass as soft litter; not over￾filling baskets and bags or using very big bags as this damages product at the bottom; 
handling carefully to avoid damage; sorting by damaged and ripeness; packing - 
MSIKA Final Evaluation Page 49 kadale@africa-online.net
loading into appropriate containers for storage or transportation; and rubbing off 
surface dirt after harvest.
"One thing I liked about the training is that we were taught to grade, which makes 
selling easier.” FBO member, Dedza.
The most applied practice is the timely harvesting not too early so immature and not 
too over ripe, which was applied by 13 farmers. The least applied practice was rubbing 
off surface dirt after harvest by one farmer. On average each farmer applied four 
practices with the lowest number applied being one and the highest being seven. 
Out of the 15 onion respondents, 12 received training on PHH and three said they did 
not. Seven practices were stated as having been adopted: harvesting and selling green 
or fresh onions with stalks on; making a shed to hang onions to prevent them from 
getting direct sunshine; packing in appropriate bags or baskets; grading by variety,
size and appearance; sorting according to damaged; trimming dried stems and roots 
before packing; and harvesting/lifting when the tops have bent over & dried.
"We realized that marketing of onions was not good because we did not know how 
to grade or sort, but now that we learnt how to grade and sort, marketing and pricing 
is not a problem."… FBO member, Lilongwe.
The most applied practice by 10 farmers is making a shed to hang onions to prevent 
them from getting direct sunshine. The least applied practice is trimming dried stems 
and roots before packing that was applied by one farmer. On average, each farmer 
applied three practices with the lowest applying one practice and the highest applying 
five practices.
Some practices were not applied as the farmers said that they sold the onions while 
they were still fresh:
"I sold out my onions fresh, because I needed quick money for other projects." 
FBO member, Mangochi.
Out of the 15 Irish potato respondents, 12 received training on PHH and three did not. 
For those that were trained, ten practices were adopted: harvesting when plant tops 
wilt and start to wither; rubbing off surface dirt after harvest or after drying and curing; 
keeping clean potatoes in a cool dry place to cure or heal surface damage before 
packing, storage & selling; handling carefully to avoid damage – remove all damaged 
potatoes to avoid rotting; packing - loading commodities into appropriate 
bags/containers for storage or transportation; grading by color, size and shape; sorting 
by variety; removing damaged potatoes; making a shed; adding grass as soft litter; and 
lifting potatoes using forks.
The most adopted practice was harvesting when plant tops wilt and start to wither that 
has been applied by nine farmers and the least adopted practices were rubbing off 
surface dirt after harvest or after drying and curing and adding dry grass as soft litter
which were both applied by one farmer each. On average, one farmer applied four 
practices with the lowest applying one and the highest applying six.
Out of 15 mango respondents, two did not receive any of the PHH training for 
mangoes. One respondent did not apply the practices because he was trained after he 
had already harvested the mangoes. Eight practices have been applied: harvesting 
when ripening just started and the color begins to change; catching fruit before it falls; 
harvesting by climbing with bags; grading; storing in a cool place; harvesting by using 
a special harvesting pole; and cleaning before taking them for sale. 
The most applied practice is storing in a cool place applied by ten Farmers. The least 
adopted practices are catching of fruit before it falls and harvesting using a special 
MSIKA Final Evaluation Page 50 kadale@africa-online.net
harvesting pole which have been applied by one farmer each. On average a farmer 
applied three practices with the lowest applying one and the highest applying five.
"I liked the training because we are having few rotten fruits now." FBO member, 
Mchinji.
Overall, these suggest high up take of PHH practices. 
3.2.9 Post-Harvest Losses
Indicator 26 measures change in post-harvest losses. The LoP target for post-harvest 
losses is 14.0%. This is calculated as the total weight of the losses across all growing/ 
harvesting cycles divided by the total production for those cycles for each crop and 
then aggregated to an overall loss. 
Table 38 sets out the revised weighted baseline for the four crops and the same for 
the FE results are set out in Table 37 below. The consultants removed the results of 
producers who gave a production value above the FAO maximum yield, as the losses 
are a proportion of the yield and so problematic to include when working out the losses. 
The removal of the three crops and introduction of weights has resulted in a revised 
aggregate loss at baseline of 14.8%, with 11.8% for men and 15.4% for women. The 
FE survey found that across the four crops, the losses were 13.1% of production, which 
is 107% of the LoP target set at 14.0%. In aggregate, MSIKA has met and surpassed 
its LoP target, however it should be noted that there are considerable challenges in 
accurately measuring crop losses.
Measuring crop losses is a very difficult task, as losses occur at several points from 
field/harvest through to market. Measuring production is also difficult, but at least the 
producer can record whole units harvested, such as bags or pails. When it comes to
losses, these can be part of a whole unit (a bag of produce all going rotten), but more 
likely losses are part units where the producer sees that some of the vegetables or 
fruits in a bag have rotted, so removes only those affected. This might be repeated
several times. A further difficulty in measuring losses is defining the loss at harvest 
time. It is difficult to assess the loss for a potato plant that is dug up at harvest with a 
mix of usable and not usable potatoes. Rotten or unsellable fruit will be left in/on the 
ground. Ditto with fruits that have fallen before collection, as to how they/the weight 
can be count. Protocols for enumerators can help, but there are genuine difficulties for 
producers to estimate losses. Therefore, data for losses should be treated with caution.
For the different crops at FE, tomato had losses of 12.7%, onion at 9.9%, potato at 
8.9% and mango at 17.5%. The losses for men were 11.8% and for women were 
15.4%. The differences vary by crop, notably in tomato (10.7% for men vs 17.0% for 
women), though the losses are similar for onion, mango and potato. Overall, the losses 
for men were lower than for women at 11.8% and 15.4% respectively.
Table 37: Post-harvest Losses by Crop and by Sex, FE, 2020
FE mean yield and loss (kgs)
Crop
Males Females Total
Yield Loss % Weighted 
calc Yield Loss % Weighted 
calc Yield Loss % Weighted 
calc
Tomato 127,020 13,564 10.7 2.7 60,784 10,308 17.0 4.4 187,804 23,872 12.7 3.25 
Onion 115,796 10,815 9.3 1.9 44,103 5,008 11.4 1.4 159,899 15,823 9.9 1.68 
Potato 147,389 12,764 8.7 2.1 85,414 8,046 9.4 1.9 232,803 20,810 8.9 1.99 
Mango 100,786 16,743 16.6 5.2 86,185 16,010 18.6 7.6 186,971 32,753 17.5 6.15 
Total 490,991 53,886 11.8 276,486 39,372 15.4 767,477 93,257 13.1
MSIKA Final Evaluation Page 51 kadale@africa-online.net
Table 38: Post-harvest Losses by Crop and by Sex, Baseline, 2017
Baseline Mean Yield and Loss (kg)
Crop
Males Females Total
Yield Loss % Weighted
Calc Yield Loss % Weighted
calc Yield Loss % Weighted
Calc
Tomato 94,495 12,734 13.5 3.5 58,762 7,961 13.5 4.1 153,222 20,695 13.5 3.70
Onion 48,227 3,707 7.7 0.6 7,373 1,031 14.0 0.5 55,601 4,738 8.5 0.53
Potato 196,234 14,827 7.6 2.3 81,387 7,202 8.8 2.4 277,620 22.029 7.9 2.31
Mango 72,995 17,723 24.3 8.8 54,055 10,380 19.2 7.5 127,050 28,103 22.1 8.23
Total 411,952 48,991 15.2 201,577 26,573 14.5 613,493 75,564 14.8
Respondents were asked the point at which the crop was spoiled covering pre￾harvest for mid-term, and harvest and post-harvest periods for both evaluations. The 
findings are set out in Table 39 for FE and Table 40 for the baseline.
The most common points at which losses occur were ‘damaged when harvesting’ and 
‘on-farm storage’17, with at least 30% - 60% reporting them both during the baseline 
and FE. Damaged when harvesting includes produce which may get spoiled where it 
is physically difficult to harvest, notably for mango, or where the produce is delicate 
(tomato) or easily damaged at harvesting, such as digging up potatoes. As expected, 
‘could not be sold prior to spoil’ was higher at FE with at least 30-40% stating it across 
the crops, most likely due to COVID which resulted in restrictions on markets. 
Lower scores were reported for damage ‘during transportation’ and ‘damaged at the 
market’ both during the mid-term and FE. Damaged during transportation includes 
losses in the field while transporting back to the farm/storage facility or while 
transporting to the market. 
Table 39: Point at which Harvest was Spoiled, FE, 2020
Point at which crop was lost Tomato Onion Potato Mango
% % % % 
Damaged when harvesting 48.2 39.9 62.6 67.0
Damaged at the farm in storage 31.2 34.0 28.7 23.5
Damaged during transport to market 17.0 5.4 2.6 8.7
Damaged at the market 7.5 5.4 1.7 10.4
Could not be sold prior to spoiling 41.9 45.8 37.8 47.8
Total 253 203 230 230
Multiple response possible
Table 40: Point at which Harvest was Spoiled, Baseline, 2017
Point at which crop was lost
Tomato Onion Potato Mango 
% % % % 
Damage prior to harvest (pest/birds) 51.2 26.1 43.5 33.8
Damage at harvest 31.6 17.4 45.9 39.8
Damaged at farm (storage) 31.0 52.2 45.3 24.4
Damaged during transport 14.1 2.9 3.5 4.0
Damaged at market 7.4 1.4 1.8 4.0
Could not be sold prior to spoil 8.0 4.3 2.4 7.5
Total 326 69 170 97
Multiple response possible
17 On farm storage is the point after harvest where the crop is held on the farm prior to selling.
MSIKA Final Evaluation Page 52 kadale@africa-online.net
Respondents were asked how they store their crops (non-crop specific). The findings 
at FE are set out in Table 41. 
Unsurprisingly, the most common place is inside the house practiced by 49.9% of 
respondents, followed by a separate building/store at 45.8% compared to 26.8% and 
20.8% at mid-term and baseline respectively. This could explain why the volume of 
storage built within the last 12 months has exceeded the revised LoP target. Only 1.2% 
stored at their FBO, which matches the KII responses for FBOs that they do not have 
any storage facilities.
MSIKA reports that it added 450 cu/m of storage through construction and distribution 
of farm storage bins to FBOs.
Similar to the mid-term, women were more likely than men to store in the house, while 
men were more likely to have a separate store. 
Compared to the baseline, more FE respondents were storing in a separate building/
store/shed (45.8% vs 20.8%). There is a clear shift from inside to outside storage.
Table 41: Place of Storing Crops, FE and Baseline
Storage location Baseline Final Evaluation
All % Male % Female % Total % 
Inside my house 55.1 43.7 57.4 49.9
In a separate building/store/shed 20.8 53.5 36.5 45.8
In a Nkokwe/outside store 10.2 1.5 1.4 1.4
In baskets/bags outside or in a pile 8.0 4.3 4.1 4.2
FBO Warehouse 0.2 0.9 1.6 1.2
Do not store (direct to market) 8.0 8.4 10.8 9.5
Total 590 535 444 979
Multiple response possible
Respondents were asked if they had built or restored their place of storage. A total of 
23.3% said they had built a new storage building, room or shed while 13.8% said they 
had restored a storage building, room or shed. As at least 40% mentioned to have 
been storing in the house, it is possible that refurbishment or building relates to the 
house, as a multi-purpose building rather than specifically for storage, but it is not 
possible to say from the data. 
The building and/or restoring of storage contributes to Indicator 27, total increase in 
installed capacity (dry or cold storage). Table 42 below sets out the calculation for 
this indicator.
Respondents gave estimates of the floor area in square metres and height in metres. 
This approach is prone to error, as measurements were estimated by respondents and 
given over the phone such that it was not possible to pace out the estimated storage 
physically. Based on the initial raw data, the consultant’s conclusion was that an upper 
value needed to be set to address unrealistically high values. After consideration, and 
consultation with MSIKA, upper values for the floor area were revised to a maximum 
of 24 square metres, while upper values for height were set at 2.5 metres. This was to 
ensure that the final calculation was closer to realistic measurements. 
The estimated areas for both categories (new and restored/refurbished) were totalled 
and divided by the total number of respondents in the sample, giving a mean average 
volume per sampled participant (979) of 8.7 and 5.5 cubic meters (cu/m) for new and 
restored respectively. This equates to total new storage of 269,680 cubic metres(cu/m) 
and total refurbished storage of 168,950 cubic metres (cu/m) when extrapolated to the
FE population of 30,922 trained farmers. As noted in the MSIKA performance indicator 
table, while the data is probably overstating the extent of the change, the consultant's 
MSIKA Final Evaluation Page 53 kadale@africa-online.net
finding is that the target is very likely to have been achieved even if there had been 
verification of the farmers’ estimated measurements of the storage. 
Table 42: Building or refurbishing storage calculation for FE population
Storage built or refurbished (cu/m)
Sample size 979 
Population 30,922 
Volume new storage/farmer 8.72 
Volume refurbished storage/farmer 5.46 
Total new storage 269,680 
Total refurbished storage 168,950 
Total storage 438,629 
The revised LoP target was a total (new and restored) increase in storage of 45,770 
cubic metres, so with this addition in 2019-20, MSIKA has met and exceeded the 
revised LoP target with an overall achievement of up to 1,059%.
3.2.10 Gender 
Questions on gender were not included in the FE HH survey to reduce its length. This 
section briefly summarises the main points from the mid-term for ease of reference. 
At mid-term, respondents were asked about who (men/women) does the work on 
the crop(s) that were grown by their HH. The most common response was that men 
and women do the work equally, ranging across tasks from 75-81% of respondents. 
This was followed by similar scores for ‘only by men’ ‘only by women’ and ‘more by 
men than by women’. The least common was ‘more by women than by men’. There 
was no discernible pattern between crops.
Respondents were asked to provide the split of key growing tasks by sex, including 
selling and managing the money. The findings have been updated to cover the four 
crops at FE and are set out in Table 43 below. There were relatively even balances, 
but with men more involved in land preparation and managing the proceeds, and 
marginally more involved in planting and managing the crop.
Table 43: Work on the Tasks by Sex at MTE Across for the Four FE Crops
Work done by:
Land 
preparing Planting Managing 
growing Harvesting PHH Selling Managing 
proceeds
% % % % % % % 
Only by Men 9.1 4.9 5.4 2.7 4.9 9.9 13.7
More by men 
than by women
14.8 13.2 13.6 11.5 10.1 10.9 9.8
By men and 
women equally 59.9 65.0 65.8 69.2 65.8 58.5 56.8
More by 
women than by 
men
5.4 5.5 5.4 5.7 7.1 5.7 3.2
Only by women 13.6 13.7 13.4 14.0 17.5 20.3 19.1
Do not know 1.3 1.6 0.6 0.5 0.3 1.1 1.1
Total 634 634 634 634 634 634 634
Multiple response possible
MSIKA Final Evaluation Page 54 kadale@africa-online.net
3.2.11 Farm Management Knowledge and Application
Indicator 15 is the percent of agricultural producers in target region who can 
identify key characteristics of a well-managed farm. The questions related to the 
indicator were taken out to shorten the interviewing time. A summary of the MTE HH 
findings are provided for ease of reference.
MSIKA trained producers in farm management. For unprompted knowledge of 
improved farm management practices, the baseline found that 65.0% of 
respondents could identify at least one characteristic and the mid-term survey found 
that all respondents (100.0%) could identify at least one characteristic. 
Respondents were also asked which management practices they used in the last 
year in the FE HH survey. There was much higher application (prompted), of farm 
management practices than at baseline, which suggests the training was effective in 
stimulating action. Similar to the mid-term, the two lagging practices are ‘sell together 
with other farmers’ and ‘keep good farm records’. From the consultant’s experience in 
Malawi, these are practices that producers have difficulty adopting. For example, 
keeping good farm records requires literacy, which is often very low.
At baseline, 55.9% of respondents were applying at least one farm management 
practice which had increased substantially to 97.7% of participants at FE. 
Table 44: Use of Improved Farm Management Practices, Baseline vs FE
Have you used these management practices in the last 12 months?
Baseline Final Evaluation
Management Practices Male 
%
Female 
%
Total 
%
Male 
%
Female 
%
Total 
%
Keep good farm records 31.7 14.3 23.9 64.6 53 59.4
Sell together with other farmers 10.9 15 12.7 30.2 28.8 29.6
Do costings for growing crops 41 40.8 40.9 75.7 66.5 71.5
Calculate profits after selling 68.3 57.8 3.6 88.6 82.3 85.8
Plan production for upcoming season n/a n/a n/a 90.9 87.2 89.2
Plan production for a specific market 21.9 27.9 24.5 77.2 73.7 75.6
Understand the specifications for 
specific buyers n/a n/a n/a 88.2 83.3 86
Separating commodities into the 
different grades n/a n/a n/a 78.5 81.2 79.7
Timely buying of inputs n/a n/a n/a 90.5 85.1 88.1
Run your farm as business 60.7 44.2 53.3 n/a n/a n/a
Plan your farm and how to rotate 
crops
51.9 55.1 53.3 n/a n/a n/a
Respondent 183 147 330 526 430 956
Multiple response possible.
The effect of MSIKA’s training of producers in farm management practices is measured 
through Indicator 8: is the number of individuals who have applied improved farm 
management practices (i.e. governance, administration, or financial management). 
The calculations for this indicator are set out in Table 45 below.
Based on the FE survey finding above that 97.7% of producers applied one or more 
improved farm management practices, this equates to 30,196 out of 30,922 producers. 
In terms of the disaggregates, the LoP actual for males is 93.5% (14,213 vs 13,286) 
while the LoP actual for females is 139.7% (12,107 vs 16,910). 
MSIKA Final Evaluation Page 55 kadale@africa-online.net
Table 45: Indicator 8: Application of improved farm management practices, FE 
Indicator 8 and disaggregates # who 
applied 
% of sub￾sample 
Total # of producers 
applying
# who have applied improved farm 
management practices 956 97.7 30,196
# who have applied improved farm 
management practices - Male 526 98.3 13,286
# who have applied improved farm 
management practices- Female 430 96.8 16,910
# who have applied improved farm 
management practices - Producers 956 97.7 30,196
3.2.12 Sales Volume 
According to the revised Performance Management Plan (PMP), surveys have not 
given accurate figures for sales when triangulated with other data. Therefore, the
MSIKA team, with USDA agreement, determined that the most accurate way to 
measure sales is through FBO records. 
Nevertheless, this section will report on the findings from the FE HH survey for
triangulation, though the information should be treated with caution. 
The outcome of all the training, organising of farmers, making market linkages and 
providing access to market information (see FBO section 3.1) is that sales are 
expected to increase in both volume (Indicator 21) and value (Indicator 23). 
Indicator 21 is the volume of commodities sold by producers in metric Tonnes 
(mT). Standardising sales data per ha enables an appropriate comparison between 
baseline and FE rather than sales per producer, as land area varies.
18 The original 
impact baseline value was 7.3 mT/ha, however, as noted in the MTE and in the yield 
section, there were outliers in the baseline sample that had not been consistently 
removed. As a result, the unadjusted baseline values considerably overstating the true 
value. The consultant also identified sales outliers in the baseline and FE where the 
volume of sales exceeded the volume of production minus estimated losses. The 
consultant allowed a margin of error, and removed those that still had excess sales.
In the MTE, the consultant noted that the baseline was based on the mean sales per 
ha for vegetables (and chili) added to the mean sales per ha for fruits. Adding the sales 
of all crops together ignored that some crops produce a higher volume per ha than 
others. Therefore, there was need for weighting of the respective samples. For the 
FE, the consultant removed the yield and sales outliers and re-weighted the samples 
for comparison. The baseline and FE findings are set out in Table 46. 
The data for sales volume is reported by crop in kgs19/ha to enable comparisons 
between the crops and with the baseline. The highest mean volume of sales in the FE
is for mango (5,270 kg/ha), followed by onion (4,267 kg/ha), potato (3,780 kg/ha) and 
tomato (3,616 kg/ha). 
Compared to the adjusted baseline, the highest increase in mean sales volume is for 
tomato at +70.0%, followed by potato (+28.8%), onion (+21.1%) and mango (+6.1%). 
The increased sales volume for tomato is important for producers, as it is the most 
commonly grown crop, so increases for tomato suggest widespread benefits. Potato is 
the second most widely grown and onion the third, so the highest increases were for 
the most commonly grow crops and the lowest for the least commonly grown (mango).
18 The baseline producers had relatively large land areas compared to mid-term and FE surveys. 19 Metric tonnes are split into 1,000 kgs so that the differences are clearer.
MSIKA Final Evaluation Page 56 kadale@africa-online.net
Table 46: Volume sold in kg/ha, Revised Baseline vs FE
Crop Kg/Ha Impact Baseline Finals Evaluation-all
Mean Median Mean Difference Median Difference
Tomato 2,126 886 3,616 70.1% 2,213 149.7%
Onion 3,524 2,198 4,267 21.1% 2,717 23.6%
Potato 2,934 1,989 3,779 28.8% 2,766 39.1%
Mango 4,966 3,568 5,270 6.1% 3,532 -1.0%
N=1,074
As noted in earlier sections, the producers have had to cope with COVID, which is 
likely to have been a factor in the lower than target performance for Year 4 and the 
LoP sales volumes. Nevertheless, there have been positive changes from baseline to
FE across the four crops.
Producers were asked in the FE HH survey if sales for their crop were affected by 
COVID. Responses by participants on 607 out of 1,074 (56.5%) crop said sales of the 
crop were affected by COVID. It was therefore surprising to find in the analysis that 
sales of those who say their sales were impacted by COVID on average had higher
sales than those who said they were not affected. There are several possible
explanations. One is that they may be overstating that COVID impacted sales, perhaps 
thinking that there might be some benefit/compensation to them. Another is that they 
produced more, since increased yields would mean they have more to sell and they 
still managed to sell in local markets. A third is that COVID only affected one of the 
three sales seasons, so overall the effect was lessened.
In the qualitative interviews, farmers were asked about their sales experiences on only 
one of their crops. Based on these, the following information was obtained:
Out of 18 tomato farmers, 13 reported that they had sold alone/as individuals to 
vendors. Six had sold vendors, with two selling in groups. 
"Since our produces ripen differently with other farmers, we sell individually by calling 
or taking it to the vendors." Lilongwe, male farmer.
One out of the 18 farmers was linked to a market by MSIKA, with the other 17 saying 
they were not linked or assisted in finding markets.
"We were promised to be linked with markets, but it seems it was a challenge with 
LOL, so we sold our tomatoes to vendors." Dedza male farmer.
Out of 15 onion farmers that were trained in onion production, three farmers reported
selling through their FBOs and 12 farmers reported selling alone to vendors or selling 
themselves in their local area. 
"I sell alone at home to traders that come to me." Dedza, male grower
Out of the 15, four say they were helped or linked with markets, while the other 11 sold 
individually to vendors and through their own selling. 
"LOL linked us to markets and all is working well. We just want to increase production 
so that we fully meet the needs of our customers." Mchinji, female. 
Out of 15 Irish potato farmers trained in Irish potato production, 12 said they were 
selling alone and three said they sold through their FBOs.
"I sold alone, because we usually sell to vendors, so groupings would not work well 
since these people cannot be able to purchase a large stock at once. Only when a 
bigger market is found are we able to sell in groups." Lilongwe, male farmer.
MSIKA Final Evaluation Page 57 kadale@africa-online.net
Five out of the 15 farmers were assisted or linked to markets to sell their produce and 
eight said they were not assisted to access a market. 
"LOL should continue to link us with more markets because their prices vary and we 
might be able to find a market that will benefit (us).” Lilongwe, male farmer.
Out of 15 mango farmers, 13 sold their produce individually to vendors or by their own
retail in their local areas. Two reported that they were able to sell via the FBO.
''We were advised by LOL to sell as a group because you make more money than 
selling alone'' Lilongwe, male farmer.
None of the farmers reported being linked to a market. 
"We were expecting to be linked to markets, but unfortunately we weren't and resulted 
in us selling our produce to vendors." Mangochi, female farmer. 
3.2.13 Sales Value
The method for determining Indicator 23, sales value in US $ was also revised to be 
captured through FBO reported data for the same reasons as for volume (section 
3.2.12). Data on sales value was captured in the FE HH survey to support triangulation. 
To allow comparisons, sales are standardised into US $/ha. As noted with sales 
volume, the baseline had yield, price and sales outliers which were removed to enable 
comparison between the baseline and FE. 
For the overall FE sample, sales of the four crops fall within a similar range with tomato 
at $1,229/ha, onion at $1,226/ha potato at $1,195/ha and mango at $966/ha.
Comparing these to the baseline there were big increases for mango (+99.7%), 
followed by potato (+92.7%), onion (+63.3%) and tomato (+44.1%). See table below.
Table 47: Value of Crop Sold in $/ha, Baseline vs FE
Crop 
Baseline - yield and price 
outliers removed
Final evaluation - yield & price 
outliers removed
Difference FE 
over baseline
Mean 
$/ha
Median 
$/ha N Mean 
$/ha
Median 
$/ha N % 
Tomato 853 408 268 1,229 652 275 44.1%
Onion 751 335 68 1,226 680 259 63.3%
Potato 620 367 200 1,195 816 251 92.7%
Mango 483 208 100 966 550 237 99.7%
Kadale noted that there were different prices in MK in the baseline and the FE. To get 
to the US $ value, the exchange rate used is MK 726.82:$1 as used in the baseline. 
The MK and US $ equivalent prices are compared in Table 48 below. 
While the mean tomato price in US $ was 8.0% lower in the FE, the mean mango price 
was 76.1% higher, onion 24.8% higher and potato 15.6% higher.
Table 48: Prices in Malawi Kwacha and USD $, Baseline vs FE
Crop Impact Baseline Finals Evaluation-all Difference
Price (MK) Price (USD) Price (MK) Price (USD)
Tomato 268 0.37 247 0.34 -0.80%
Onion 175 0.24 218 0.3 24.80%
Potato 157 0.22 182 0.25 15.60%
Mango 76 0.1 102 0.18 76.10%
MSIKA Final Evaluation Page 58 kadale@africa-online.net
In the FE survey, a very small number of growers20 (20 out of 1,074 – 1.9%) did not 
sell any of their crop. Details are set out in Table 49 below. The FE found that 3.4% 
of mango producers, 2.6% of potato producers, 1.1% of tomato producers and 0.4% 
of onion producers did not sell anything. The most common reason, especially by
tomato, onion and potato producers, was that they ‘did not have a surplus to sell’, which 
likely means they lost most or all of their crop. The second most common reason was 
that they ‘could not get a buyer’, given most commonly by mango producers. ‘Poor 
quality’ was the third most common reason and was given mostly by potato producers. 
This may be related to pest and disease damage, and damage at harvest, on the farm 
and at the market. It would be surprising if no growers had faced problems at all, such 
as a crop failure in any season.
Table 49: Growers that Sold Nothing, FE, 2020
Crop Producers # that sold nothing % of that crop
Tomato 276 3 1.1%
Onion 264 1 0.4%
Potato 266 7 2.6%
Mango 268 9 3.4%
Total growers 1,074 20 1.9%
Respondents were asked for the primary way they sold, including selling individually, 
selling with other producers in their club/group or selling with producers outside their 
club/group. The most common method was selling as an individual, ranging from 84-
90% of respondents depending on the crop. Selling with other members of a club/group 
ranged from 9-13%, with the highest responses for onion producers.
Table 50: Primary Means to Sell, FE, 2020
Primary Sell
Response
Tomato Onion Potato Mango
% % % %
With farmers in my farmer club/group 9.1 12.9 8.8 9.3
With farmers not in my farmer club/group 5.1 3 3.8 0.8
On my own as an individual 85.5 84.1 87.4 89.6
Other 0.4 - - 0.4
Total 275 264 262 259
Respondents were asked about the locations where they sold their crop, with 
multiple responses possible (Table 51). The most common response for all crops was 
‘traders who came to me’, ranging from 75-79% across the four crops. ‘Local market’ 
was the second most common response, ranging from 30-44% of respondents. The 
only other substantial response was for ‘neighbours’ at 22-34%. 
20 Note that some were interviewed for two crops.
MSIKA Final Evaluation Page 59 kadale@africa-online.net
Table 51: Sales location, FE HH Survey, 2020 
Where Did You Sell your produce?
Response
Tomato Onion Potato Mango
% % % %
Traders who came to me 79.3 78.0 79.4 74.5
Local market 44.0 30.7 29.8 38.2
Neighbors 34.9 22.0 25.6 23.6
Traders that I delivered to 20.4 14.4 14.9 16.2
Distant market 8.0 7.6 7.3 7.7
My Farmer group 3.6 1.9 2.3 2.7
Direct to Processor/large buyer 0.7 - 1.1 0.8
NGO 0.4 2.7 1.1 0.8
Total 275 264 262 259
Multiple response possible
3.2.14 Access to Finance
This section covers access to finance (A2F) issues relating to producers. A2F 
information is drawn from KIIs with MSIKA staff and the MFI. It was not covered in the 
FE HH survey. 
As at June 2020, MSIKA reported that it had trained 744 VSL groups, each with 15-20 
members. This equated to 10,000-13,000 members.21 In addition, MSIKA reports that
5,444 individuals had received at least one loan through a MSIKA related VSL group.
It was encouraging that two thirds of MSIKA producers were in a MSIKA VSL or 
another VSL group, as these have proven to be effective at enabling rural households 
to save, and as a source of small, short term loans. The advantage for members is that 
although interest is charged on the loans, the accumulated interest earned by the 
group is returned to members, often for buying inputs in October and November for 
use in the main farming season. 
In addition to improved access, producers also needed to understand financial issues. 
As reported earlier (section 3.1.3.4), financial training reached 62-64% of FBOs. 
This topic is also covered in the sections on agricultural lending through MSIKA’s 
relationship with the MFI in section 3.3.3 below and in section 3.1.3 on FBOs.
3.2.15 Investment
This section covers investment made by producers. Section 3.2.9 sets out the volume 
of building and/or refurbishing storage producers invested in. These respondents were 
asked how much they spent on refurbishment/building from August 2019 to July 2020. 
The amount spent on building new storage across the FE sample (979) was the 
calculated at $7.02 per farmer, equating to US $217,275 across the 30,922 FE
population. For refurbishing storage, the amount spent was $3.90/farmer, equating 
to US $120,632 across the FE population.
The total reported spent on equipment and land from August 2019 to July 2020 was a 
mean US $23.3 and $25.2 respectively per farmer, equating to US $ 668,563 and $ 
778,377 across the 30,922 FE population.
21 MSIKA says there may be duplicates in this total, so could not give a precise figure.
MSIKA Final Evaluation Page 60 kadale@africa-online.net
3.2.16 Employment
This section covers employment at farm level based on the FE HH survey. 
Indicator 22 is the “number of jobs attributable to USDA”. Producers contribute to 
the total, along with processors, FBOs and other MSIKA partners.
Respondents were asked if they hired anyone for four weeks or more (converted to full 
time equivalent) in the period August 2019 – July 2020, with 15% saying they did hire. 
The baseline reported 0.09 full time jobs equivalent (FTE) jobs created per participant
household in the previous 12 months, against 0.07 FTE jobs calculated at the FE. 
Comparisons are difficult to make for two reasons. Firstly, the HHs in the baseline had 
larger land areas than the FE HHs, and demand for labour increases with land size as 
the land area and production exceeds what the HH can manage from its own 
resources. Higher land areas in the baseline would tend to require more labour. 
Secondly, the FTE figure is difficult to measure with certainty, as respondents are 
providing estimates of labour. Only those that were hired for more than four weeks 
count to the calculation, but the use of short term labour termed ‘ganyu’ is common for 
tasks like land preparation and weeding, which typically are for less than four weeks 
and may involve different workers over the three growing seasons.22 
The number of people hired by the 15% of participants that hired someone for four 
weeks or more, was split by male and female, and converted to jobs per respondent 
across the sample (979). This was multiplied by the FE population (30,922) to
calculate the total number of people hired for four weeks or more at 2,067 FTE workers
between August 2019 – July 2020
3.2.17 COVID effects
The KIIs with producers sought information on the effects of COVID on producers. 
The prevalence of COVID in rural areas at the time of the research was not clear. The 
main evidence of COVID was in the cities, but that could have been a function of the 
availability of testing and better appreciation of it. 
For the tomato respondents, of the 18 participants, 11 reported having not been 
affected by COVID-19 pandemic and that things were as before:
"Everything is being done normally, just like the time before the Corona Virus." Tomato 
male farmer, Chileka EPA, Lilongwe
"We continued with all our work, but followed the precautionary measures." Tomato 
male farmer, Chowe EPA, Mangochi
For those that reported having been affected by the COVID, things changed especially 
in relation to preventing further spread. The changes included: 
a. Group gatherings were not possible for information sharing within the FBO or 
for marketing
b. They were not able to travel to the city/town markets due to restrictions
c. There were lower farmgate prices due to limited market opportunities as good 
buyers were rare
d. There was a loss of production due to lower demand and higher supply
e. Some FBOs were considering only producing one crop per member to reduce 
flooding the thin market with same crops 
22 Note that these jobs may not necessarily be created, as the demand is seasonal and so people are 
hired each year. 
MSIKA Final Evaluation Page 61 kadale@africa-online.net
"Vendors (in small numbers) coming to buy from us are buying at cheaper prices and 
as a result our tomatoes are getting spoiled due to shortage of buyers.” Tomato female 
farmer, Mkwinda EPA Lilongwe
"(We are) failing to go to markets to sell tomatoes and as a result, vendors coming to 
us are buying at lower prices." Tomato male farmer, Ndemera EPA, Lilongwe. 
3.3 Banks, MFIs and VSLs 
MSIKA aims to stimulate agricultural and small and medium enterprise (SME) lending. 
This section reviews MSIKA’s work with a Malawi Bank SME finance through a Special 
Purpose Vehicle (SPV) and with a Microfinance institution (MFI) for agricultural loans. 
The information for the Bank is drawn from the MTE, as this was not a focal area for 
Year 4. Information for the microfinance institution draws on the MTE, the KIIs with 
the MSIKA team and a KII with the the MFI staff. There is also information on VSL 
from MSIKA KIIs and reports. 
3.3.1 Overall Performance
The relevant indicators are 11 (# of individuals receiving financial services), 12 (# 
of loans disbursed) and 13 (value of loans disbursed). 
For indicator 11, MSIKA reports an additional 5,612 individuals receiving financial 
services in year 4, bringing the LoP total to 13,922 against a target of 9,984 individuals
resulting in an achievement of 139%. Achievement was higher for female beneficiaries 
at 169% than for male at 115%. 
For indicator 12, MSIKA reports an additional 5,074 loans in Year 4 resulting in a LoP 
total of 10,377 against the LoP target of 9,561, an achievement of 109%. These were 
broken down as MFI loans - 4,909, VSL loans - 5,444 and those through the SPV 
facility/ Bank - 24.
For indicator 13, MSIKA reports an additional US$ 413,100 in Year 4, resulting in a 
cumulative LoP total of US$ 699,879, a 55% achievement. The biggest contributor in 
terms of value has been MFI with an LoP total of US$ 431,943 which is 61.7% of the 
loan value. The SPV/Bank loans contributed a LoP total of US$ 193,343 (27.6% of all 
loan value) and VSLs contributed a LoP total of US$ 74,592 (10.7% of all loan value). 
Female beneficiaries received a relatively high proportion of loan value against their 
LoP target (72%) than male beneficiaries (41%).
The key factor in not achieving this target has been the lower than anticipated numbers 
of loans and value of those loans under the SPV with the bank. MSIKA had reported 
delays in getting the facility off the ground (see next section), and eventually settled for 
working with the bank. In essence the loan process was slower than expected and the 
size of the loans limited by the size of the SMEs that applied and the limited collateral 
they had available which restricted the size of loans the bank was willing to lend. This 
is related to the relatively modest scale of operation of many of the SMEs operating in 
fruit and vegetable processing, other than a few larger firms, like Suncrest and Malawi 
Mangoes that have alternative means to secure their finance. While the bank might 
have been overly cautious, it was the small size of the SMEs and their limited capital, 
capacity and collateral that constrained the value of loans.
In contrast, there were larger numbers of loans through the MFI and through VSLs, but 
the average value of these was relatively low and unable to make up the short fall from 
the SME lending. The expectation is that the MFI and the VSLs will continue to lend as 
sustainable savings and loans mechanism. It is also likely that the bank will continue 
to work with the SMEs and FBOs that it has built good relationships with, as long as 
they fulfil the loan terms. The bank may also progressively expand, but this is likely to 
be cautious, as is the nature of banks in Malawi.
MSIKA Final Evaluation Page 62 kadale@africa-online.net
3.3.2 Finance for SMEs
From the MTE findings and FE KIIs, MSIKA intended to have an SPV either as a fund 
managed by MSIKA or through a bank. MSIKA was advised to work through a bank 
or a Malawian entity to manage it. MSIKA went through a lengthy process of 
assessment before finalising a partnership with the selected bank. MSIKA provided 
an initial loan of $250,00023 in March 2019 as a fund for on-lending by the bank. In 
return, the bank agreed to ease its collateral requirements24, and to charge a lower 
interest rate than normal25, on the basis that MSIKA would assist SMEs to prepare 
screened proposals that are then put to the bank, which is described as the ‘pipeline’. 
To support the process, and importantly so from the bank’s perspective, MSIKA 
identified agri-processors and agro-dealers with potential to buy from, or supply, 
MSIKA producers. MSIKA provided due diligence information on these SMEs and 
assisted them to prepare and improve their loan proposals through group sessions. 
The bank conducted its own due diligence and proposal assessment, involving visits 
and gathering evidence of sales and cashflow. Not all proposals were accepted. 
In the MTE KII, the bank’s staff reported that the main issue for the bank is the low 
turnover of money through any bank account, compared to the higher business 
revenues claimed by the SMEs in their proposals and relative to the requested loan 
sizes. The bank decided to offer relatively small first loans compared to what is being 
asked, and making this first loan primarily for working capital, not for fixed asset 
purchase.
26 The aim was to see how the SME manages the initial loan typically for six 
months. If the SME showed good cash turnover through the account and the ability to 
service the loan without problems, then the bank would look to disburse larger second 
loans that could be for fixed asset purchase. 
From the perspective of the SMEs, a reduced loan size was a problem for some SMEs 
who perceived this to be too much effort to get a small loan. For two of the more 
established SMEs, the owners stated that they understood this and so they focused
on servicing the loan to get the bigger second loan they were originally seeking. 
The President of Malawi announced in April 2020 prior to the election, that because of 
COVID, there would be a moratorium on collection/enforcement loans to SMEs. Four 
of the twelve SMEs requested a moratorium. Prior to this, one SME was facing 
problems with repayment. The others felt they could repay to the schedule.27
For Year 4, MSIKA expanded the facility with the bank for lending to selected FBOs, 
which on-lend to their members against business plans. A factor behind this approach 
was that microfinance loans (see next section) are more expensive than the bank loans
at 36% compared to 23% respectively for a six-month period. More loans to FBOs 
were sought by MSIKA, but the bank was more cautious. Loans need to be tied into 
the timing of growing seasons. At this point it is too early to say if this will expand as 
the first loans were given in May 2020. MSIKA reports that the agricultural lending 
team has been responsive and faster at progressing applications than the SME team, 
though the bank in general is slower than that MFI. 
MSIKA continued to try to interest other banks, however with the closing of the project, 
there was not enough time for these to be established and for loans to be repaid before 
project end, so they did not progress for this and other more bank specific reasons. 
23 At 2% of the forex value. 24 This is based on cash collateral of 20-25%, rather than land/vehicles or other ‘hard’ collateral. 25 The bank’s rate was 21% plus $30 processing fee versus other banks at 25% plus 2% processing fee. 26 Most applications have been for fixed asset investment. 27 Interest would still accrue during a moratorium, so there is an incentive for SMEs to repay on time.
MSIKA Final Evaluation Page 63 kadale@africa-online.net
Overall, MSIKA reports that the bank had on-lent US $93,000 to seven processors and 
agro-dealers at mid-term, which increased to US $193,343 by FE. The bank had on￾lent to 12 processors, and 12 FBOs.
28 Two SMEs have had second loans, but it 
appears that the bank has substantially tightened loan conditions requiring an 
unrealistic 75% deposit of cash against a loan on which interest is charged on 100% 
of the loan. This appears to be designed to deter SMEs and may be linked to the 
government announcement of a moratorium on SME interest due to COVID thereby 
deterring banks from lending to SMEs for now. The bank has also told MSIKA that 
they will not add more loans, as those loans would not be cleared before the project 
end. 
3.3.3 Finance for Producers
MSIKA worked with an MFI to finance loans for producers, particularly those with 
access to irrigation.
MSIKA provided technical assistance to the MFI, made links with FBOs, organised 
producers, and trained the producers in financial literacy29. The MFI lent from its own 
funds, which when repaid and recycled have resulted in cumulative loans totalling US$ 
431,943. The MFI’s requirement is that producers have to be members of a VSL with 
a track record of saving and repaying loans. Also, they must have sales records to 
show that they generate sufficient income to repay the loans. 
For the first year of the scheme (ending June 2019), repayment was 98%. For the
2020 loans for the rainfed season, there were arrears of around 8% of the loan amount, 
which may still be collected. Overall, the MFI has been very satisfied with the 
performance of the loans and the programme. 
“Results have been great.” MFI staff, at mid-term.
“The combination with their technical was very valuable”. MFI staff at FE. 
The latter comment is interesting because it points to the value of MSIKA helping to 
screen FBOs as to which might be most appropriate and the training for members that 
has taken place. The staff highlighted that there is a reduced risk for groups where 
there was irrigation: 
“Irrigation makes it safer from drought, which is the most critical problem.” MFI Staff
The MFI mentioned two ways in which working with MSIKA had changed their views. 
The first was that they did not know much about lending into fruit and vegetable value￾chains particularly to farmers. They now see that this can be a viable group of famers 
to lend to. Secondly, they had previously struggled with loans to farmers in the drier 
parts of Mangochi, with problems of default. Working with fruit and vegetable farmers 
who have access to irrigation/water sources or are willing to invest in relatively low cost 
pumps (around US$ 100) has enabled them to develop a new and better loan portfolio, 
which enables them to sustain their operation in Mangochi.
One point of difference between the MFI and MSIKA was that the MFI says it was
trying to collect interest during the tenure of the loan as a way to reduce the liability of 
the farmers and risk. Because the loan was repayable in one payment, MSIKA’s view 
was that the farmers should not be asked for repayment. As the funds at risk are the 
MFI’s and they are experienced in loan management, it would have been better to 
discuss this rather than the MFI and MSIKA before giving farmers contrary information.
28 This has included one default, which FDH thinks was a person who intended to default from the 
start; the remainder of the portfolio is operating at an acceptable level. 29 Budgeting, saving and record keeping. 
MSIKA Final Evaluation Page 64 kadale@africa-online.net
Overall, the partnership with the MFI has operated very well, meeting the needs of both 
parties. The MFI says it will continue to work with the farmers that they have 
relationships with and that are meeting the loan conditions, so the results will likely be 
sustained. Over time, this may lead them to work with other FBOs, which they now 
see as “medium risk”. However, they do value having a link with a project to reduce 
risk and spread some of the preparatory costs, such as training in financial 
management that helps with loan repayments. The MFI also noted that they have
limited funds, so some type of loan facility would have been helpful to expanding their 
reach.
3.4 Processors
The findings in this section are drawn from the KIIs with processors and MSIKA staff, 
and MSIKA data gathered in the FE.
MSIKA supported 18 processors based on their needs, including training in good 
manufacturing practices (GMP), certification by Malawi Bureau of Standards (MBS), 
linking them to financial institutions to invest in their business, and linking them with 
FBOs to source produce.” The access to finance work is covered in the previous 
sections on MSIKA’s work with banks (section 3.3.2).
MSIKA provided monitoring details of 18 processors that it had worked, along with an 
outline of activities and details of sales. There were two other processors that MSIKA 
had been in the process of building relationships, but these had either not progressed
sufficiently before COVID intervened, meaning there was no time left to proceed. 
12 processors that were identified for interviewing, were purposively selected to ensure 
that there was a sufficient spread of products, size and activities. The consultant 
interviewed 11 out of the 12 processors. The larger processors were more difficult to 
schedule interviews with and reluctant to provide sales data to the evaluation team. 
MSIKA requires that the processors or agro-dealers have some link to the FBOs and 
beneficiary producers. From the interviews with processors, there were limited links 
with MSIKA’s FBOs and only relatively small quantities of target crops being 
purchased. Some of the purchases were for chillies, which was not a focal crop for 
Year 4. MSIKA had high hopes for the two additional processors to buy crop from 
FBOs, but there was insufficient time to enable this to happen. 
3.4.1 Training and Application of Improved Techniques and Technologies
A key part of MSIKA’s approach has been to improve quality of processing and the 
standards and products. Initial research in the early phase of MSIKA identified that 
processors often had poor levels of hygiene along with poor manufacturing practices 
and that their products were not certified, so as to be able to sell them, particularly 
through retailers. 
Technoserve (TNS) was commissioned to undertake training in implementing and 
complying with good manufacturing practices (GMP) and standards, such as the 
Malawi Bureau of Standards (MBS) standard ‘MBS21’ on hygiene practices, as well as 
marketing. The training included aspects of innovation, to bring variety to processed 
products, many of which are the same, such as mango achars and chili sauces.
Indicator 32 is the number of processor staff trained in quality standards. The 
LoP target is 187, which was increased following over-achievement of the target at 
mid-term. MSIKA reports achievement of 187 (100%), with both male and female 
targets met. The processors interviewed at the FE confirmed that they had been 
trained in GMP, hygiene and related topics. 
MSIKA also engaged with the processors to assess their needs and develop a tailored
plan to address hygiene improvements needed for MBS certification process, with 
MSIKA Final Evaluation Page 65 kadale@africa-online.net
steps for measures like upgrading plant/equipment/buildings, undertaking product 
development, improving marketing and related financial planning. 
At an early stage, MSIKA decided to split these relationships and the related work for 
efficiency reasons, with some managed by TNS and the rest by the V37 team. As part 
of its work, TNS grouped enterprises for a series of group sessions over six months, 
after which it ended its involvement. This approach is efficient for delivery and has the 
benefit of processors being exposed to what their peers (and sometimes competitors) 
do which can bring new thinking. However, this time-limited engagement process did
not fully respond to the needs of the processors, as the degree and depth of change 
required more time input and to take place over a longer period. 
The processors interviewed were very positive about the way the training was done, 
and that there had been individual follow ups.
Indicator 19 the number of private enterprises, produce organisations, water 
user associations, women’s groups, trade and business associations, and CBOs 
that applied improved techniques and technologies as a result of USDA 
assistance. After the MTE, it was clarified that these related to the application of their 
operations compared to the original wording that referred to the application of 
agricultural techniques and technologies. 
At mid-term, the processors interviewed were able to give multiple examples of 
changes they had made to their businesses relating to GMP, hygiene and process 
improvements. The changes identified from the processor KIIs at FE were greater, with 
a wide range of changes adopted on good manufacturing practices, such as improved, 
clear and accurate labelling, record keeping, factory improvements (improved toilets), 
improved cleaning regimes, established standard operating procedures, conducting 
food analysis and testing, improved storage conditions, accurate records on use of 
ingredients, taste testing, and acidity testing. Only one processor said that they had 
not implemented any GMP changes, as the program had not met their expectations 
(for access to finance); however when interviewed at mid-term, this processor stated 
they had made multiple changes.
"The whole system what we follow now has come from the trainings with LOL. When 
we receive new consignments of agricultural products, we record the amount, check 
on the quality and store them well in a bay before production."
The processors at FE also stated a range of marketing practices that they had adopted 
following the training. Examples included undertaking customer research and 
analysis, improved presentation of the products particularly the quality of labels, setting 
up a website, using Facebook/Instagram/social media and advertising to promote their 
products, attending trade fairs, revised branding, changes to pricing and better stock 
management for distribution.
"At the trade fairs we met many customers some who still buy our products. We had a 
chance to practise some of the marketing skills we learnt in the trainings such as how 
to talk to a customer." Processor KII
Based on the GMP and marketing training, all nine processors that responded on these 
questions gave examples of new business practices they had adopted that can be 
attributed to MSIKA. One other business that said they had not adopted practices, had 
done so at mid-term, but this firm seemed to want to emphasise their disappointment 
at not getting finance from or facilitated by MSIKA. From MSIKA’s reports on the two 
processors that did not provide data for the interviews30, it is clear that these two 
(larger) processors did adopt several practices, notably on GMP. The consultant is 
30 One was partially interviewed and the other kept changing the appointment date.
MSIKA Final Evaluation Page 66 kadale@africa-online.net
satisfied that there was adoption across all processors interviewed and multiple 
practices were adopted in most cases. 
3.4.2 MBS Certification
Data is drawn from KIIs and MSIKA data.
Indicator 25 is the percentage of registered processing firms in target sectors 
that obtain certification with MBS related to product quality. The LoP target is 
30%, with MSIKA reporting achievement of 44.0%, which is an achievement of 147%.
MSIKA commissioned Michigan State University (MSU) to undertake a needs 
assessment and gap analysis on standards and certification. The work with MBS led 
to a process for defining and establishing standards. 
Food processors are required to follow MBS standards, notably on food hygiene. They 
are also required to meet the standards set for particular food products, where those 
standards exist. Many of the processors were operating without certification or only 
with pre-certification status, which potentially creates hazardous situations. There was 
also an absence of MBS standards for several fruit and vegetable-based food 
products. Together this meant that processors could not sell their products through 
formal outlets, thereby reducing their sales. 
“For example, if we found a shop to carry our products for us they would be thrown out 
by the MBS. It has happened before. We found markets yes, but we couldn’t display 
because of the certification.” Cooperative (Mid-term).
At FE stage, three of the six processors that did not already have their products
certified achieved full certification, with three others achieving pre-certification. These 
firms explicitly credited MSIKA with helping them achieve this status. Three other firms 
stated they had maintained accreditation.
3.4.3 Other Support to Processors
MSIKA and TNS also offered other support to processors tailored to each business, 
based on a needs’ assessment. 
All the processors were offered help with loan facilitation with the Bank as covered in 
section 3.3.2. Three of the 11 processors interviewed took Bank loans, while the others
either self-financed their investments, obtained grants or took loans from other banks 
unrelated to the MSIKA program. Two firms said they were offered second loans by 
the bank, but declined them because the conditions had become too stringent:
"The agreement required I deposit three quarters of the amount I wanted. For instance, 
if the loan I wanted was K1 million, I needed to have K0.75 million in my account. I 
thought this was not a fair agreement. It wasn't helpful." Processor KII.
Based on the information in the above comment, it would not make sense for firms to 
proceed with those loans, not least because interest is charged on the whole loan 
amount (not netting off the amount deposited), making the effective interest rate much 
higher for the incremental 25% of the loan value the processor would get over what 
they already had. While the work with the bank was helpful in the earlier phase, it looks 
like these conditions were there to deter borrowers, partly linked to the ending of the 
facility and the bank wanting to protect its position.
Training on GMP/hygiene and support towards reaching MBS standards are discussed 
in the above sections 3.4.1 and 3.4.2. At MTE and FE, there was a range of additional 
support/activities provided. Book-keeping and financial management, as well as 
marketing training were cited by most of the firms. Help with operating procedures, 
planning the factory and identification of new equipment needed was specifically cited
by five firms. Six processors stated they had help with product quality testing. 
MSIKA Final Evaluation Page 67 kadale@africa-online.net
There were some planned activities that did not all come to fruition. There were
attempts and some agreements to link the processors to FBOs who could supply 
product. There were two cases where there were purchases, typically for chilis, but 
these mostly did not result in purchases by processors. There were also efforts to link 
with buyers. While the link was made, it did not result in new business.
"I found a very important supplier for the fruits I use to make my products through one 
of the trade fairs organised by LOL. They also linked us to Malawi Mangoes, but we 
were unable to complete the agreement because there was communication 
breakdown." Processor KII, FE
From the KIIs, the processors had high expectations particularly for new (retail) 
customers/markets and finance. The highest expectation was on finance, some of 
which was satisfied with the initial round of loans via the bank, but the later rounds 
seem to have disappointed as the terms were onerous. One processor was very 
disappointed not to get finance through MSIKA, as that appeared to be their over-riding 
focus and disengaged when they could not qualify for the loan. 
There is always going to be a challenge in any development program over unrealistic
expectations by some parties with perceptions that this is a source of easy/free money. 
3.4.4 Performance data
Although MSIKA identified 71 processors in the early stages of the program, MSIKA 
provided information on 20 processors that it had developed relations with, of which 
two did not have substantive activities resulting in a net 18 processors worked over the 
course of the program. Three of these 18 can be described as large enterprises in 
Malawi’s terms. Two are classed as medium-sized enterprises, with the rest would be 
termed small (5-20 employees) or micro (0-4 employees). 
It has been a challenge for MSIKA to find processors with any scale or potential scale. 
From the FE data, five of the 12 processors31 interviewed had 20 employees or more, 
while six of the 12 had 10 or fewer full-time/permanent employees. 
Work with processors contribute to several indicators. The findings on Indicators 19 
and 25 are reported in the previous section. 
For indicator 22, the number of jobs created, the LoP target was reduced following 
the MTE to 1,657, with many of these expected to come from producers (see section 
3.2.16). The contribution of processors was not measured in the mid-term as the 
number of KIIs was too limited, instead relying on MSIKA reported data. At FE, the 
processors were asked to give information on employment and nine did so. The 
additional jobs created for this group of nine processors since the start of MSIKA was 
reported as 26; however, it is worth noting that the two biggest firms did not provide 
information and at least two of the processors had been reducing employment due to 
operational challenges. It was also noted at mid-term and again at FE that firms found 
it difficult to make a direct attribution of jobs created to MSIKA’s assistance as opposed 
to a range of factors influencing the business. 
For indicator 24 – the value of new public or private investment leveraged by 
USDA assistance – at mid-term, three of the four processors interviewed provided 
evidence of investment, although the amounts were modest at that point at up to 
$20,000 for one processor. At mid-term, reported cumulative public and private 
investment was $550,597 against the overall target of $1,150,000. All of this has been 
private investment, including the loans taken, against a target of for private investment 
of $150,000, thereby exceeding the target. At FE, MSIKA reports US $ 3,320,736 in 
31 These are all of the large and medium in the listing.
MSIKA Final Evaluation Page 68 kadale@africa-online.net
investment, against the revised target of US $ 1,967, 799 thereby exceeding the target 
(169%). This is a combination of investment by producers and by processors.
From the FE KIIs, six processors gave information on additional investments in the 
form of loans received, additional capital from the owners and some grants from other 
sources (not MSIKA). These ranged from US $ 4,000 up to US $ 20,000 per processor. 
Other processors did not provide information, so there may have been additional 
investments that occurred. The investments were wide ranging in scope including
equipment, materials, transport, the factory, and sales kiosks for direct selling. It is 
clear that there has been investment by processors, and much of that is linked to 
improvements that MSIKA has stimulated them to make.
Indicator 27 is an increase in installed storage capacity (dry or cold). It was not 
possible to get details of increases in capacity in the KIIs due to time pressure and 
prioritizing other information, such as employment and investment. It is noted that at 
mid-term, there were examples of processors increasing the quantity, and also the 
quality of their storage.
Indicator 29 relates to numbers of sales agreements with FBOs. Although this 
indicator focuses specifically on FBOs, there were some expectations of links between 
FBOs and the supported processors. At mid-term, there was little evidence of sales 
agreements between the four processors interviewed and FBOs. At FE, two 
processors stated that they had sales agreements with FBOs for chillies. Three other 
processors indicated that they had some engagement with MSIKA FBOs, but these did 
not come to fruition in terms of actual supply of produce. 
At the mid-term, the consultants interviewed a processor, who had around 20 sales 
agreements with FBOs, for around 140 mT of paprika and 70 mT of chili in the 2019
recent season. Due partly to its own financial challenges resulting in buying late and 
to cases of FBO members side selling even though they received seed from the 
processor, these sales agreements were only partially fulfilled. 
The challenge for MSIKA is that even if it facilitates agreements between processors 
and FBOs, these may not be fulfilled due to problems with either or both parties. Other 
inputs on sales by FBOs to other buyers, such as retailers and hotels/lodges is covered 
in section 3.1.5 above.
3.5 Government 
MSIKA has worked with Government of Malawi (GoM) on policy, training of extension 
staff, research in horticultural crop production and PHH, research into improved 
practices (Bunda College and MSU), and MBS to develop product standards. 
3.5.1 Policy Level
MSIKA worked with the Ministry of Agriculture, Irrigation and Water Development (the 
‘Ministry’) to develop a new Horticulture Policy. 
The policy process was initiated by the Ministry and horticulture professionals who 
wanted to see more focus on horticulture and development of a strategy that would 
increase its importance for farmers and the country. Initially, the Ministry group was 
focused on a strategy, but following MSIKA’s interest in a horticulture policy the 
decision was to look at the policy and strategy concurrently. 
As reported at mid-term, MSIKA funded a consultant to support the process and a 
series of planning meetings. MSIKA also supported an online survey of 60 horticulture 
stakeholders to gather insights for the revised policy. Stakeholder meetings were co￾funded by MSIKA and the Ministry. At the start, the process moved quickly, leading to
a draft Horticulture Policy that the Ministry’s Department of Planning needed to align 
with overarching policies. 
MSIKA Final Evaluation Page 69 kadale@africa-online.net
At the mid-term point, the Ministry said that it did not budget for this alignment stage, 
so the process was stalled by the first quarter of 2019. A key factor in this was the 
Presidential and Parliamentary election in May 2019 which created an uncertain 
political situation and divided attention on side of the Ministry. The outcome of that 
election was disputed by the opposition, which embarked on a prolonged series of 
street protests and a judicial process from mid-2019 to early 2020. This uncertainty 
and regular stopping of activities due to protests, made it difficult to make progress on 
the remaining stages of the policy process. The courts eventually declared that the 
election must be re-run. The election was run in late June 2020 and the previous party 
of government was replaced with the former opposition. 
A changeover of parties in this manner has an impact on the Ministries, as there is a 
need to appoint new Ministers and senior staff in the Ministries. Following this 
changeover, MSIKA has reported that there is still a strong motivation in the Ministry 
to move the process forward, and that a further stage of the internal consultation with 
Departmental Directors, has been progressed by the Ministry without MSIKA needing 
to support this. There are now two remaining stages, including the final wider 
consultation with regional stakeholders, which MSIKA reports has been scheduled. 
Once all the Ministry stages have been completed, the Policy is put to Cabinet for 
approval.32 The process is closer to fruition than at mid-term and there do not seem 
to be any fundamental hurdles to its ultimate completion, though, as was always the 
case, the final stages are out of the hands of MSIKA. Beyond finalising the Policy, the 
Ministry will also need to work on its strategy. 
3.5.2 Training District Extension Staff
At the outset of the MSIKA project, it was identified that the provision of extension 
services by GoM staff at district level was a major weakness. To address this, MSIKA 
worked on training for GoM extension staff in horticultural GAP through a master trainer 
approach to promote tested and proven farming and PHH practices. Over the course 
of MSIKA, 267 GoM extension staff were trained in horticulture and PHH practices. 
Some also received training in marketing as well.
For the FE, KIIs were conducted with Agricultural Extension Development 
Coordinators (AEDCs) and/or Agricultural Extension Development Officers (AEDOs), 
mainly the latter. 
In the early phases of the project, MSIKA gave a sub-award to MSU, to conduct a 
training needs assessment in collaboration with the Department of Agricultural 
Research Services (DARS) and the Department of Agricultural Extension Services 
(DAES). From this assessment, MSU developed and adapted training materials that 
would be practical for extension staff and lead farmers to use for onward training of 
farmers through a training of trainers (ToT) program. MSIKA also produced extensive 
materials on growing these crops in both English and Chichewa.
The mid-term reported on the agricultural practices ToTs that were first conducted in 
November 2017. A key learning point for MSIKA was that mixing extension staff with 
lead farmers did not work well, because the level of knowledge is different. This was 
changed from the third ToT onwards. With this adaptation, MSIKA’s assessment is
that the ToTs were effective in passing on horticulture knowledge and skills. 
An additional challenge was getting some of the GoM extension staff to conduct their 
own training of producers. Some GoM extension staff did go ahead and use the 
materials. However, others wanted additional resources to conduct the training, such 
32 A Policy can be developed, but ultimately requires adoption by the Cabinet. Once the technical 
process is complete, a Policy should not face rejection or significant change at Cabinet level, though it 
can be returned with instructions for changes.
MSIKA Final Evaluation Page 70 kadale@africa-online.net
as allowances for themselves and for trainees. MSIKA was neither willing nor able, to
provide financial support. Because of this mixed response by GoM extension staff, 
MSIKA states that it focused on those extension staff that are committed.
MSIKA works through lead farmers, which is discussed in more detail in section 3.1.4. 
MSIKA notes that there is also a mixed response from lead farmers, as with extension 
staff. It has therefore focused on those lead farmers that show sufficient commitment.
From the KIIs with AEDCs/AEDO, there was confirmation that the training of GoM 
extension staff had taken place and that the training had been used for onward training. 
All had attended the agricultural and the PHH training.
"I attended a Training of Trainers then afterwards together with the Extension Workers 
from LOL we went to train the farmers in the FBOs." KII GoM Extn Worker, Ntcheu.
'"The training was done in my section and I was acting as a supervisor in those 
trainings and offering technical advice." KII, GoM Extn. Worker, Lilongwe
Some of the extension workers highlighted the joint nature of the training of farmers. 
There were some extension workers who saw their role as more passive: 
'"Together with the staff from LOL, we were training the farmers. Whenever the staff 
from the District office were present, I was just there to attend the training without any 
active role.” KII, GoM Extn Worker, Mangochi.
The feedback on the training was generally positive, with a few issues, notably two 
who felt the training was rushed and one who felt the examples given by trainers from 
outside Malawi were not relevant enough. 
In terms of implementation of the training, three mentioned that they were not so 
involved in the onward training or knowing what training was being conducted. One 
commented about the insufficiency of MSIKA staff for delivering the onward training:
"The content was good, just that perhaps they employed fewer staff and kept changing 
them and that affected smooth continuity of activities and farther outreach.” KII, GoM 
Extn. Worker, Mchinji
On the PHH training, the general response was that it was good and the timing was 
good, however in some cases the officers said that the timing was not good, so that 
the ‘theory’ was covered, but it was too early for practical demonstrations of harvesting. 
There were also a few comments about not all the EPA was covered with the training.
"Content was good. There were some differences in what the Government and LOL 
Extension Workers know about PHH but after discussions a consensus was reached" 
KII, GoM Extn. Worker, Mangochi.
"It was 100% what was supposed to be taught." KII, GoM Extn. Worker, Lilongwe.
" For example, tomato PHH should really be demonstrated on tomato at that stage, in 
real time. That way, farmers most of whom are illiterate can learn better by observing 
and practicing." KII, Govt Extn Worker, Mchinji
There were some improvements suggested, such as better communication and co￾ordination, clarity on roles for the GoM staff, unfulfilled promise on demo plots and 
related seedling supply, and producer groups being too big to be effective:
"They should communicate beforehand what role I am going to take, say a facilitator. 
That way everyone can thoroughly prepare and work better as a team." KII GoM Extn. 
Worker, Mchinji.
“The major problem was attendance. In a single session, attendees would be 50 or 
more and to me, this affected some farmers especially those that take time to grab 
things. In future trainings, the number of sessions should be increased and the number 
MSIKA Final Evaluation Page 71 kadale@africa-online.net
of those attending a single training session should be reduced so that everyone is able 
to understand." GoM Extn. Worker, Mangochi
"We were surprised that after the training, they did not involve me or any extension 
worker in any of their activities. We were only hearing everything from the farmers and 
we would see some of the activities from afar." KII, GoM Extn. Worker, Lilongwe. 
One reflection from MSIKA was that the younger GoM extension staff were more 
responsive and more likely to implement and support the trainings. 
There was also work by MSIKA FBCs and lead farmers on demonstration plots. The 
feedback from most of the GoM extension staff was that the content of these was 
relevant to the farmers, though there were several comments about issue with 
organisation and timeliness.
Overall, the comments partly reflect perceptions of roles and a view from the GoM staff 
side that they were supposed to be in control. Current GoM policy on extension is for 
plurality, meaning any provider is able to provide it, though with effort to co-ordinate. 
Sometimes there is a desire by GoM staff to control the program and resources that 
may not be matched with their own delivery. MSIKA, like other projects has had to get 
the right balance, though co-ordination, collaboration and good timing are all helpful, if 
difficult to achieve all of the time. 
3.5.3 Malawi Bureau of Standards
A further aspect to the MSIKA project was its work with the Malawi Bureau of Standards 
(MBS) on the development of product standards. MBS has several standards relating 
to food processing, such as MBS21 concerning food hygiene. However, there were 
gaps for several products that can be made from MSIKA’s target crops. As a result, 
nine products were identified that needed standards, as well as one standard that MBS 
wanted to include. 
There is a well-defined process for introducing a new/adapted standard that involves 
initial research to see if there is a standard that can be adopted or adapted. If there is 
no international standard, then there is a defined process to develop a national 
standard. The process involves examining other standards that might provide some 
base information or that can be drawn upon for the new standard. This research/ 
formulation is followed by a formal process of consultation through a technical 
committee consisting of processors, research institutions, the Consumer Association 
of Malawi, the Ministry of Health, the City Councils (who may need to enact by-laws) 
and academic. After there is agreement on the draft standard, MBS circulates the draft 
standard to the World Trade Organisation and on to the international community. 
The major challenge for MSIKA in this process has been the election and its re-run, as 
until there is a Government in place, the Board of parastatals like MBS cannot be 
appointed and this delayed ratification of standards. However, MSIKA reports that 
approval has now been given by the new Board of MBS, and this will be followed in 
the near future by formal Gazetting33, which is in the hands of the Ministry of Justice. 
In summary, the work with MBS moved quickly, was held up by the elections, but 
potentially now can reach fruition.
3.5.4 Bunda/MSU
This section draws mainly on the findings from the MTE. 
MSIKA supported work by MSU and Bunda College on soil fertility through mounting 
soil health trials using different combinations of organic/inorganic fertilisers/composts 
for MSIKA’s target crops. The aim was to determine what physical, chemical and 
33 Gazetting is the process where by regulations are officially published and so become operative.
MSIKA Final Evaluation Page 72 kadale@africa-online.net
biological changes these combinations result in, and the overall impact on plant health 
and yields. The results were to be determined through field trials in the five districts for 
the seven crops. Alongside the research are demonstration plots for farmer field 
schools, at which extension staff and producers can see the results and learn about 
the research.
This was a collaboration between MSU, as the contractor to MSIKA, and Bunda 
College (‘Bunda’) with two academics engaged in this work. The academics at Bunda 
College set up the trials following a methodology agreed with MSU, and monitored the 
trials, collecting the necessary data for analysis, and sending data and samples to 
MSU for further analysis. The trials focused on testing different compost/fertilizer 
combinations, as well as use of drip irrigation 
The Bunda academics reported at mid-term that the work is progressing, and the 
relationship with MSU was working well, but that access to transport has restricted
their work. 
In terms of results, the academics’ conclusions are that combining organic with 
inorganic compost/fertilizer is positive for the biological measurement of the soil, with 
an increase nematode numbers and good crop yields. They have made
recommendations on soil health that will be incorporated into MSIKA’s final training 
materials.
Getting the most out of this investment in research requires the findings to be 
incorporated into the training and demonstration work. With the early curtailment of 
MSIKA, the timing of this is too late to be cascaded out to producers. 
4 Conclusions, Lessons and Recommendations 
This section sets out the conclusions of the final evaluation and recommendations that 
USDA and V37 should consider for future similar projects. 
4.1 Key Conclusions and Lessons 
Conclusions and lessons are set out relating to the main groups of stakeholders 
covered in the evaluation: FBOs, producers, financial institutions, processors and 
government. Prior to these, there is a short section on the methodology conclusions 
and lessons relating to using a phone survey for producers.
4.1.1 Methodology
The use of a phone survey for producers and other KIIs was necessitated by the advent 
of COVID. This change, part way through the design, required the reformulation of the 
instruments to reduce the content to be covered in a maximum 45-minute call. 
Importantly, this change in method also required MSIKA to obtain phone numbers for 
a randomly drawn sample of 8,000 producers. In the end it was only possible to get 
phone numbers for 2,973 of those producers (37%), all of whom the consultants 
attempted to contact. From this list, the consultant was able to reach 979 producers
(33%), some of whom were interviewed on two crops in order to reach the required 
levels of confidence (90%) and margin of error (5%) for each crop. 
The approach and scale of this phone exercise was new to both MSIKA and the 
consultant, so there was inevitably a great deal to learn. A key lesson is that low 
phone ownership means that a sufficient sample is needed to get sufficient 
respondents (979 respondents out of a list of 8,000 names), with a call success of 
around one third due to wrong numbers, phones not available, no signal and 
respondent battery running out mid interview. Sending a modest cash incentive via
airtime to the respondent when the interview was complete helped ensure 
collaboration, though there were some challenges making sure all respondents 
MSIKA Final Evaluation Page 73 kadale@africa-online.net
received their airtime and some questions around how it should be shared if a 
respondent used another person’s phone. Overall, the exercise showed that it was 
feasible to undertake such a survey. 
Another lesson is that there are some questions that are difficult to verify over the 
phone. For example, the findings on the area of new and refurbished storage for 
producers were high, and it is likely that if an enumerator had been present, the 
enumerator could have seen or paced out the floor area and indicated the height so as 
to reduce errors in estimation. This only became apparent when the overall 
calculations were done and were much higher than expected. 
A further lesson is that using the phone may introduce a male bias, as ownership 
and/or control of the phone in Malawi may favor male participation. There may also 
be a bias in that there was a relatively high proportion of lead farmers in the sample, 
possibly because their numbers were known in the FBOs, so they would appear on 
most FBO lists, perhaps disproportionately. The male and lead farmer biases in the 
sample are not easy to resolve in future exercises, but the consultant did undertake 
some interviews with respondents using the phones of other respondents which partly 
offsets these potential biases. 
4.1.2 Overall Progress
Across its many indicators, the MSIKA program achieved or made considerable 
progress towards many of its LoP targets (see summary results, in Annex 1). 
Several factors have impacted on the program in the last two years, in particular: 
1. The life of the program was shortened from five to four years and several planned 
activities were dropped due to a shortfall in the expected resources. The 
shortening of a program was disruptive for two main reasons. First of all, there is 
one year less of implementation of activities, some of which are part of a sequence
and package that might result in the project not getting the most out of each of the 
components of its work. An example of this would be fully using the results of the 
Bunda College research trials to inform training and dissemination of better soil 
management practices. 
The second main effect is that activities that are designed to build up over time, 
have not had sufficient time to realize their full results. For example, MSIKA 
invested heavily in the training of extension staff and lead farmers and the onward 
training of farmers alongside project staff. It takes time for farmers to adopt the full 
set of practices, as they are likely to adopt the easier, lower costs ones first, see the 
results and progressively build their confidence to apply the more difficult and 
costlier ones. MSIKA has been increasing the numbers of farmers trained and the 
range of practices they have been trained in, but the full increase in impact will not 
be seen within the life of the project. An extra year would be expected to result in 
higher performance by producer beneficiaries.
2. A second major factor for MSIKA’s implementation was COVID, which began to 
affect Malawi towards the end of March 2020 and into the next two quarters (Apr￾Sep 2020). While the impact of COVID has not (at least not yet) seen widespread
loss of life in Malawi, the GoM reacted to the pandemic by imposing a series of 
lockdown measures that restricted movement, restricted markets and closed
tourism and hospitality venues and institutions that are major users of fruit and 
vegetables. This impacted on producer production and sales, and it restricted the 
operation of FBOs, processors and the MSIKA team. This reduced planned 
activities and contributed to a shortfall in several targets. The effects of the COVID 
restrictions have also hampered MSIKA’s closing out activities and its ability to 
implement its sustainability strategy.
MSIKA Final Evaluation Page 74 kadale@africa-online.net
3. Malawi had a Presidential and Parliamentary election in May 2019, which was 
heavily disputed. There was considerable disruption in the cities and a risk of 
violence, that limited activities by MSIKA and other partners/stakeholders, notably 
GoM Ministries, MBS and DADOs. The eventual rerun of the election led to a 
change of Government which in turn resulted in changes at senior levels in 
Ministries and governmental bodies. The extended uncertainty in government 
bodies disrupted some of the policy-related work and field work.
4. In March 2019, Cyclone Idai hit Mozambique, with the effects reaching Malawi. 
These occurred on top of a period of prolonged heavy rains, resulting in widespread
flooding and crop losses. The effects of Idai were seen in the MTE, where several 
results were impacted, notably yields. While in some ways the effects are short 
term in that farmers can plant again for the next season, the loss of crop(s)
undermines farmers’ resources and reduces what they have to invest in the next 
planting. It can also have damaging effects on the land quality, if soils are washed 
away, and on buildings for storage, if these are damaged and the farmer cannot 
afford to repair. 
The above factors are not intended to provide reasons for shortfalls in performance, 
which overall was good; rather they are a reminder that implementation of projects can 
be disrupted by multiple factors beyond the control of the project team. In the case of 
MSIKA, there have been multiple factors, some of them concurrently, such that it has 
affected implementation. The restrictions due to COVID and the disruption from the 
election and its aftermath did take a toll on program activities, restricting field 
operations that are central to its operations. 
The lesson for USDA and V37 is that while it is not possible to predict the specific 
nature and timing of disruptions, it is predictable that there may be one, consecutive or 
concurrent disruptions of some kind in a volatile operating environment like Malawi. 
This should be factored in when planning what is realistic to deliver and achieve in a 
project of this nature in a country like Malawi where uncertainty is the norm. 
4.1.3 Farmer Based Organisations
MSIKA has worked with 217 FBOs across the five target districts, which is above the 
LoP target of 210. Of these, 167 (77%) were formed by MSIKA. 
FBOs provide a useful mechanism for organising farmers for engaging with projects, 
receiving project inputs (e.g. training), which is much more efficient than engaging 
directly with farmers. They may also have potential as sustainable organizations for 
collective activities by farmers e.g. for purchasing inputs and sales of produce for 
mutual benefit.
MSIKA assessed that 27% of its FBOs are high performing, 39% are medium 
performing and 34% are low performing. For Year 4 of MSIKA, following the need to 
restructure based on fewer resources and the recommendation in the MTE to focus on 
better performing FBOs, MSIKA planned to focus its work on 50 better performing 
FBOs. This plan was disrupted by COVID, though the MSIKA team was able to 
partially implement this narrower focus. From the performance data and the KIIs 
across the spectrum of FBOs, the consultant’s conclusion is that it is difficult to bring 
low performing FBOs to be medium or even high performing and that working with 
these dilutes the effort of teams that are under pressure to deliver many activities. 
The lesson for USDA and V37 is that MSIKA and similar projects are likely to make 
more progress and achieve better returns by moving early to focus on better performing 
FBOs. Determining performance can only be done following some engagement; 
however, V37 has considerable experience in working with FBOs and has tools for this 
type of assessment, such that early decisions could lead to dropping low performing 
MSIKA Final Evaluation Page 75 kadale@africa-online.net
FBOs at the earliest opportunity. That does not mean that some low performing FBOs 
could not improve with time and effort; rather it recognizes that many do not improve 
much, as they often have weak leadership that is looking for projects to do everything 
for them and to provide resources. MSIKA’s FBCs were stretched and this would have 
enabled them to be better focused. 
As stated above, FBOs are very useful mechanisms for engaging with farmers and 
some may ultimately become sustainable organizations. However, the consultant’s 
conclusion is that it is unrealistic to expect that most have the potential to become 
sustainable entities. It would have been helpful for MSIKA to set out its expectations 
for FBOs more clearly at the outset. The lesson related to this is that it is important to 
be realistic about the potential to turn most FBOs into well-functioning, viable and self￾sustaining registered entities, such as Associations or Co-operatives. This is unlikely 
in most cases, at least based on the MSIKA FBOs.
The FBOs that MSIKA formed are based on multiple small clubs of 10-15 members,
brought together into a newly created larger entity of up to 300 members. While it is 
not feasible to have very large FBOs at least early on, the MSIKA FBOs are too small 
and too weak to become viable entities in their own right. With leaders elected each 
year, and all the challenges associated with self-serving models of leadership in 
Malawi, it is highly unlikely that many can move beyond being mechanisms for 
organising farmers to engage with projects, like MSIKA. The progress made on 
governance capacity building (section 3.1.3.5) and overall in capacity development 
from the PMM assessment, suggests that it would take many years and considerable 
investment to develop them, with a very uncertain success rate. 
MSIKA’s own estimate at mid-term was that up to 10% of FBOs have potential to 
become co-operatives. The consultant’s conclusion from the interviews with FBOs and 
MSIKA’s FBO data is that this is an optimistic assessment, and that fewer than 5% 
might have some potential, perhaps as low as 1-2%. For projects like MSIKA, FBOs 
may best be seen as temporal mechanisms to engage efficiently and effectively with 
farmers, than to build into a project design plans and resources to try to turn them into 
sustainable entities or expect them to continue operation beyond a project life where 
there is no ongoing support. If there are a few FBOs that have potential, such as if 
there are pre-existing FBOs that have been able to progress and have larger 
memberships, then a more focused support package for the few would be a more 
realistic approach. 
FBOs and lead farmers have been a key mechanism for delivering a range of training 
modules including agricultural/horticultural practices, PHH practices, marketing and 
finance. The delivery of some of the training has been disrupted by the wider factors 
that impacted on the project and that the MSIKA technical team and FBCs were 
stretched. This relates back to the point about the level of ambition for what can be 
covered, the focus on fewer FBOs (be more selective earlier) and acceptance of FBOs 
as temporal entities (in most cases). Related to the this, it is not realistic that these 
FBOs would be able to invest or be able to operate storage and processing facilities in 
the foreseeable future. 
As noted in the evaluation, MSIKA has been very successful at increasing individual 
producer yields, and has made progress on PHH losses. Individual producers have
generally increased their sales. However, progress with collective selling via the FBOs 
has been limited. While there have been sales agreements made, these have not 
always been fulfilled, partly because aggregation capacity is weak and there is still a 
strong preference by producers to sell individually. There is potential interest in links 
between producers and HAI and potentially Universal Industries, but this started late 
in the program and was disrupted by COVID, so did not reach fruition. Linking 
producers to markets is a strong theme for MSIKA and something that many projects
MSIKA Final Evaluation Page 76 kadale@africa-online.net
that work with producers want to achieve to convert success in increasing production 
volumes (and quality) into sales and incomes. This am is important to V37 and USDA.
The lesson from MSIKA is that linking producers to more formal markets with 
expectations on quality, volumes, timeliness and ‘packaging/presentation’ is very 
difficult and may depend heavily on the ongoing facilitation efforts by a project and 
private sector businesses that are willing to persevere. In the KIIs, some FBOs said 
they had sales agreements, but they felt that they were not able to fulfil the 
requirements, so they pulled out of those agreements. It is noteworthy that although 
COVID disrupted the supply to formal markets such as lodges/hotels, conference 
venues, and bigger retailers, the producers reported that they still sold what they had 
produced, primarily to local markets and probably at poorer prices than the suspended
formal markets. 
It is a major challenge to facilitate sustainable links between newly formed FBOs (with
members who prefer to sell individually) and processors or other buyers with varying 
and relatively unpredictable demand, and who may switch to other suppliers/producers
when there is a glut of supply and prices can be pushed down. FBO and member 
expectations were that MSIKA would help the FBOs and their members to find 
attractive formal markets with better prices. MSIKA played a part in raising the 
expectations as that was part of the project’s goals; however, these expectations look 
unlikely to be met given all the challenges in accessing such markets. 
Although this may be disappointing, it appears that the producers have still found 
markets for their produce even under COVID, mainly because they have to sell and 
there are retail and wholesale markets they can currently reach, even if these are sub￾optimal in demand and prices. With substantially increased production, these 
producers are still likely to be better off even with lower prices from these informal 
markets. Over time, markets may become more structured with the better organised 
FBOs finding ways to aggregate to supply more formal buyers. That process can be 
facilitated by a project, but it takes time and expectations of supply to formal buyers, 
such as processors and bigger retailers/hospitality outlets should be kept low.
4.1.4 Producer Households
The FE found that MSIKA succeeded in reaching large numbers of producers (39,744) 
with training in improved agricultural and PHH practices that is valued, efficiently 
delivered and that is effective. The adoption rate of practices has been very high, with 
99.5% of producers applying at least one new practice. With a slightly higher population 
of producers trained in agricultural practices, MSIKA would have hit its target for 
indicator 7. As noted, COVID had an effect on the overall numbers and the target would 
have likely been over-achieved, if it had not occurred. 
The extent of application of practices was impressive, with a mean 15.8 practices per 
producer per crop. This widespread application is likely to lead to higher yields. One 
note of caution is that there was a higher proportion of lead farmers in the FE sample 
compared to the FE population and mid-term – lead farmers would likely be expected 
to apply more practices than other producers. 
As noted at mid-term, the baseline for crop yields needed some revision to account for 
outliers. It was noted that progress on crop yields at mid-term was affected by Cyclone 
Idai, yet still showed reasonable progress by MSIKA. With better weather conditions
at FE, another year of training, and opportunity to apply practices under the guidance 
from lead farmers, and with the narrower focus on the four crops, the results on yield 
at FE have been very good with a 54.4% aggregate weighted yield improvement 
against a revised target of 26.0%. 
MSIKA Final Evaluation Page 77 kadale@africa-online.net
In terms of PHH, the MTE found relatively low beneficiary unprompted knowledge, yet 
high application rates for improved PHH practices for field crops. At FE, the 
applications rates have improved further, with tomato scoring over 90% application on
most practices. Mango, like all the tree crops in the MTE still lags the field crops, 
reflecting that tree crops are not seen by most producers as crops to actively manage 
and invest in. The overall effect was that PHH losses reduced from 14.8% at baseline 
to 13.1% by FE. Losses are very difficult to measure and so caution is needed on any 
PHH loss data.
Somewhat surprisingly given the disruption of COVID on markets, sales volumes 
increased substantially compared to the baseline, with the highest increase in mean 
sales volume for tomato at +70.0%, followed by potato (+28.8%), onion (+21.1%) and
mango (+6.1%). The low increase in mango is not surprising, but the big increase in 
tomato seems to be a function of the increase in volume of the crop, with an estimated 
68.8% increase.
The sales value generated interesting results in comparison with the adjusted baseline. 
The sales of the four crops fall within a similar range with tomato at $1,229/ha followed 
by onion at $1,226/ha potato at $1,195/ha and mango at $996/ha. However, comparing 
these to the baseline there were big increases in sales value for mango (+99.7%), 
followed by potato (+92.7%), onion (+63.3%) and tomato (+44.1%). Part of the effect 
is down to higher MK prices (tomato being the exception at minus 8.0% and a relatively 
stable exchange rate of the Malawi Kwacha to the US dollar which overstates the value 
of changes in MK prices over the project period. However, the consultants still found 
relatively large sales increases even after adjusting for MK price changes.
In summary, there has been good progress since the baseline and the mid-term in 
application of agricultural and PHH practices, with large increases in yields, in sales 
volumes and in sales value. There has been modest progress on PHH losses, noting 
that there are challenges in measuring losses. 
4.1.5 MFIs/Banks/VSLs 
MSIKA continued with its SME lending mechanism with the bank. This came relatively 
late in the project due to finding and negotiating the terms of the facility. The initial 
results at mid-term were pointing in the right direction, though the bank was 
unsurprisingly a bit cautious and a bit slow in assessing and disbursing loans. While 
there was hope for bigger and more flexible second loans, with the winding up of 
MSIKA and a Presidential declaration on a moratorium on SME loan repayments, the 
bank made the terms much less attractive and deterred several SMEs that had been 
approved. In the end, the total number and value of loans disbursed through the facility 
was much lower than originally intended, highlighting the difficulties of establishing a 
successful SME loan facility.
The lesson is that facilities for SME lending are not attractive to all banks/lenders and 
that the banks have their own requirements that may override what the project is 
seeking, such as loan values, collateral and payment terms. Banks are cautious, but 
they are also regulated, and caution can reflect regulatory concern over what is seen 
as relatively high risk lending to SMEs.
MSIKA’s work with an MFI to increase agricultural lending has been very positive with 
both parties regarding the initiative as a success. The MFI has been the biggest source 
of lending to MSIKA beneficiaries ($413,100) compared to banks ($193,343) and VSLs 
($74,592). The approach is low cost for MSIKA and proving to be very effective, as 
the MFI is using its own capital. The MFI has been encouraged by the high repayment 
rates from the well-organised and trained MSIKA beneficiary producers. It has seen 
that there is lower risk in horticulture due to the relatively short growing periods 
meaning there can be up to three cycles a year if there is access to irrigation. The
MSIKA Final Evaluation Page 78 kadale@africa-online.net
challenge for the MFI has been that it has limited capital to on-lend, but it is very 
reluctant to borrow at commercial rates, not least because monthly interest rates are 
now capped by the Reserve Bank, limiting what is viable for the MFI. The MFI is a 
well-established MFI and will continue to offer these loans to these and potentially 
other MSIKA beneficiary producers after the end of MSIKA. This means the result is 
sustainable for the immediate future. The lesson for projects like MSIKA is that, while 
there can be value in exploring relationships with banks and sensitising them to the 
needs of agri-business SMEs, MFIs may be a more fruitful partnership than banks. 
The positive effect of MSIKA’s work to establish VSLs for FBO members was 
highlighted at the mid-term. MSIKA has now established 751 VSL associations, 
improving access to finance for over 10,000 beneficiary producers. VSLs are self￾sustaining community managed mechanisms with a high degree of continuity. 
Although the value of loans through VSL is small compared to the MFI and even the 
bank, the number of loans is higher than these partners indicating that many producers 
(5,444) are accessing small loans in their localities and are able to get a share of the 
profits on the lending when the VSLs distribute their funds. The lesson is that projects 
like MSIKA should continue to establish VSLs and train members in VSL. 
4.1.6 Processors
MSIKA has established relationships with 18 processors. This has involved application 
of GMPs/hygiene standards, financial management, storage, processing, access to 
finance and access to markets. There is high uptake of GMPs and hygiene 
improvements, increases in storage, certification to MBS standards, improvements in 
processing with indications that further progress will be made, adopting of marketing 
practices and financial management practices. However, as was the case at mid-term, 
the outcomes of this work are relatively limited at FE, particularly for purchases from 
FBOs and additional employment. MSIKA believes that with a fifth year that it might 
have been able to increase the purchases from FBOs, and this might happen based 
on the relationships with two processors. In conclusion, development of processing in 
these value-chains is beneficial overall, however, there is not yet a strong connection 
between the processors and the producers and their FBOs. 
Five of the 18 processors are larger or medium enterprises, but most are micro, with 
less than 10 employees and with limited potential to increase investment, purchases, 
storage/warehousing and employment. As noted at the mid-term, MSIKA’s efficiency 
and effectiveness was limited from investing time and resource in these micro￾enterprises. At the outset, MSIKA identified 71 processors as potential partners, but 
most were micro-enterprises and the pool of larger and medium processors was limited 
to less than ten. It has proven difficult to make much headway with processors due to 
their relatively small size. The work has led to better understanding and change in the 
smaller processors, and may ultimately lead to wider changes over time.
The lesson is that MSIKA was better focusing on the five processors that have shown 
willingness and capability to make progress, while recognising that there is still limited 
potential with processors in Malawi at this point in time.
4.1.7 Government 
As noted at mid-term, MSIKA engaged GoM on policy, training of extension staff, work 
with MBS and research on crop practices/farmer field schools.
Updating the Ministry of Agriculture’s Horticulture Policy, moved quickly in the early
stages due to the support from MSIKA, however the preparations for fresh Presidential 
election and its aftermath, as well as the major effects on government due to COVID, 
disrupted and slowed down progress. MSIKA has moved the policy process along and 
there is a good chance that the policy will be adopted in the future, but the lesson is 
MSIKA Final Evaluation Page 79 kadale@africa-online.net
that policy change of this nature is often very slow, even with strong commitment within 
government and prone to delays, such that success is out of the hands of projects like 
MSIKA. 
The ToTs for GoM’s district-based extension officers have relevant content based on 
a sound participatory process of development. At mid-term, it was recognised as
necessary to train GoM district extension staff, as there can be difficulties operating in 
the districts if they are excluded or do not feel involved. At FE, the KIIs found that Govt 
staff want to be involved in the co-ordination of activities, but some see this more as a 
matter of being in control, than necessarily to be engaged in delivery. Some 
AEDOs/AEDCs have been involved and assisting on delivery, but the GoM extension 
staff are not likely to be a key driver in onward training. Rather, it is the lead farmers 
that are the main mechanism for onward delivery. The lesson is that projects like 
MSIKA have to include GoM staff, but should focus on training and supporting lead 
farmers as the most sustainable, effective and efficient mechanism to reach producers. 
MSIKA worked with MBS to develop 10 new product standards. This was also held up 
due to the hiatus of the election period and its aftermath and eventual re-running. It is 
likely that the 10 new standards will be adopted. The lesson is that working on 
regulations or supporting measures may yield more results than working on national 
policies and strategies, and may be of more immediate relevance to the private sector.
MSIKA’s engagement with MSU/Bunda on research was linked to the farmer field 
schools at the demonstration sites. Getting the most out of this investment in research 
requires maximising the use of the demonstration sites for the farmers field schools. 
There are challenges working with academia including resourcing (such as transport) 
and the speed of delivering results. It is unclear at this point whether MSIKA will be
able to leverage this research to bring forward practices that can be readily adopted 
by many producers. The lesson is that research may not be able to bring results in 
the project period and it is probably better to focus on demonstrating under-utilized 
knowledge on proven practices. 
4.2 Recommendations 
The following recommendations are made:
#1 That to undertake a phone survey of producers requires access to a sufficient 
database of phone numbers, as representative enough of the population as possible, 
and may require additional efforts to get female producers included. Considerable
thought should be given to question inclusion, such as those that may rely heavily on 
face to face interaction with respondents. 
#2 That in the design of projects of this type, USDA should consider in its expectations 
and ambitions that there is a high likelihood of consecutive or concurrent disruptive 
events. While these can be listed as risks and assumptions, the flexible response 
shown by USDA on MSIKA can avoid projects scrambling to make up ground and hit 
pre-determined numbers. 
#3 That projects like MSIKA should aim to identify and select out low performing groups 
and FBOs at an early stage using appropriate tools and drawing on experience. 
#4 That projects like MSIKA should not aim to turn FBOs into sustainable entities in 
most cases, but be highly selective in deciding which might have potential, and to see 
most FBOs as temporal entities for efficient and effective project delivery that may or 
may not continue post project.
#5 That projects like MSIKA should seek to facilitate sales between better organised 
FBOs only where there is a realistic capacity to supply and members are committed to 
collective selling and that there are accessible markets. Where that is not possible, 
MSIKA Final Evaluation Page 80 kadale@africa-online.net
there is still merit in working to increase production for sale by individual producers into 
less formal markets.
#6 That because Banks are regulated and relatively cautious, projects like MSIKA need 
to consider other more flexible and responsive lenders, such as well-established MFIs.
#7 That VSLs should continue to be a focus for projects like MSIKA as they provide 
ready access to finance and over time, the VSL builds trust and cohesion between 
producers. 
#8 That projects like MSIKA should find an appropriate balance between working with 
enough processors, but of sufficient size and capacity to invest. Working with many 
micro- and small-enterprises is resource intensive with limited potential to deliver 
results. Work with smaller enterprises should be limited to training and general 
information rather than more intensive support relationships.
#9 That projects like MSIKA embarking on support for developing government policies 
and strategies need sufficient time and resource to see such processes through to the 
end and that it should be accepted at the outset that there is a risk that these processes 
may not yield results within the project life or at all. 
#10 That projects like MSIKA should focus on lead farmers as a more sustainable, 
effective and efficient mechanism for delivering services, particularly knowledge, than 
government extension staff. However, government structures should be engaged and 
government staff included in training and activities where they can support of lead 
farmers.
Kadale Consultants Ltd.,
Plot 244, Bwaila Road, Area 15, Lilongwe, MALAWI.
kadale@africa-online.net Tel ++ 265 (0)1 770000
MSIKA Final Evaluation Page 81 kadale@africa-online.net
Annex 1: Table of Program Indicators 
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
1 
Percentage change in yield of 
beneficiary producers as a result of 
USDA assistance 
Baseline 
Excluding 
Outliers 
(kg/ha) 
 
Tomato 2,617 
Onion 3,575 
Potato 3,329 
Mango 4,347 
Guava 3,796 
Citrus 5,517 
Chilli 467 
26.0% 54.4% 26.0% 54.4% 209% The indicator values are from the final evaluation survey,
conducted by Kadale, with 1,074 responses (individual crops) 
drawn from a sample frame of 30,922 trained producers that 
at least received basic agricultural production training before 
31st May 2020. The overall yield for Yr4 (now end of project - 
EOP) improved by 54.4% compared to the baseline, 
exceeding the Yr4 target (26%). Tomato improved the most 
(68.8%) followed by mango (62.8%), potato (48.1%) & onion 
(36.9%). Actual Yr4 yields are: 
Tomato: 4,418 Kg/Ha
Onion 4,896 Kg/Ha
Potato 4,931 Kg/Ha
Mango 7,078 Kg/Ha
Weather conditions for 2019-20 were much better than 
2018-19, where the impacts of Cyclone Idai suppressed yield 
gains. The extra period has also enabled more farmers to 
implement practices and to see the benefits of these. 
MSIKA's other activities on inputs, access to finance, etc., 
have also contributed to the higher results. Kadale set outlier 
values at the FAO maximum yields from applying Good 
Agricultural Practices (GAP) for these crops. 
Percentage change in yield of 
beneficiary producers as a result of 
USDA assistance-Male
28.9% 54.7% 29.0% 54.7% 189% 
Percentage change in yield of 
beneficiary producers as a result of 
USDA assistance-Female
23.6% 61.8% 24.0% 61.8% 258% 
Percentage change in yield of 
beneficiary producers as a result of
USDA assistance- Tomato
25.0% 68.8% 25.0% 68.8% 275% 
Percentage change in yield of 
beneficiary producers as a result of 
USDA assistance- Potato
29.0% 48.1% 29.0% 48.1% 166% 
Percentage change in yield of 
beneficiary producers as a result of 
USDA assistance- Onion
30.0% 36.9% 30.0% 36.9% 123% 
Percentage change in yield of 
beneficiary producers as a result of 
USDA assistance- Mango
21.0% 62.8% 21.0% 62.8% 299% 
2 
Number of individuals benefiting 
directly as a result of USDA 
assistance 
0 
 
8,000 
 
11,433 36,000 
41,185 114% 
MSIKA reached 2,319 individuals amid the COVID-19 
pandemic (April to Sept 2020) mainly through activities that 
sustain impact. The project still reached 4,265 (new) 
individuals in the final year as FBOs continued to grow their 
membership. Throughout the 4 years, the project reached 
41,185 participants, which is 114% of the target. Farmers, 
government staff, processor staff & civil society have 
benefited from MSIKA interventions. Farmers are the biggest 
beneficiary group (95%). 
Number of individuals benefiting 
directly as a result of USDA 
assistance- Male
 
4,480 4,890 
20,160 
 
18,433 91% 
Number of individuals benefiting 
directly as a result of USDA 
assistance- Female
 
3,520 6,543 
15,840 
 
22,752 144% 
Number of individuals benefiting 
directly as a result of USDA 
assistance- New
 
- 
 
4,265 
 
36,000 
 
41,185 114% 
Number of individuals benefiting 
directly as a result of USDA 
assistance- Continuing
 
8,000 
 
7,168 - - 
MSIKA Final Evaluation Page 82 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
3 
Number of individuals benefiting 
indirectly as a result of USDA 
assistance 
0 
36,800 
 
52,592 165,600 
189,451 114% 
10,667 indirect individuals have benefited from MSIKA 
interventions in this reporting period, with a total of 52,592 
in year 4 against a target of 36,800. The indicator is 
calculated by multiplying the number of direct beneficiaries 
by 4.6, which is the average number of individuals per 
household minus 1 (the direct beneficiary). 
4 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance 
0 
 
8,000 
 
10,669 36,000 
39,744 110% 
MSIKA continued fostering agricultural sector productivity in 
the April - Sept period with 1,427 farmers trained in ag￾production, marketing & financial literacy, bringing the Yr4 
total to 10,669 thereby over-achieving the Yr4 target (8,000). 
Cumulatively, MSIKA has trained 39,744 individuals against a 
target of 36,000. 98% of the individuals trained are farmers. 
The indicator counts individuals that have received training in 
ag-production, marketing, financial literacy, and post-harvest 
handling. Participants include farmers, government staff, 
people in civil society and people in firms. 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- Male
 
4,480 
 
4,619 
 
20,160 
 
17,847 89% 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- Female
 
3,520 
 
6,050 
 
15,840 
 
21,877 138% 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- New
 
- 
 
4,256 - 
39,744 - 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- Continuing
 
8,000 
 
6,413 - 
- - 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- Producers
 
8,000 
 
10,507 
 
36,000 
 
39,332 109% 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- People in Firms
 
- 
 
2 - 
97 - 
MSIKA Final Evaluation Page 83 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- People in Government 
 
- 
 
150 - 
267 - 
Number of individuals who have 
received short-term agricultural 
sector productivity or food security 
training as a result of USDA 
assistance- People in Civil Society
0 10 0 48 - 
5 
Percentage of producers utilizing 
improved soil fertility management 
practices 
81% N/A NA N/A N N 
Indicator was removed following modification of PMELP
Percentage of producers utilizing 
improved soil fertility management 
practices - Female
76% N/A NA N/A N N 
Percentage of producers utilizing 
improved soil fertility management 
practices - Male
86% N/A NA N/A N N 
6 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance 
0 
6,581 7,909 10,801 12,129 112% 
Based on the 1,074 sample responses in the final evaluation 
(30,922 agric trained farmer sample frame), a calculated 
7,909 Ha of land was under improved techniques and 
technologies (T&T) in Yr4, surpassing the Yr 4 target by 121%. 
Overall, MSIKA achieved 112% of the revised LOP target. High 
adoption rates contributed to the increased hectarage under 
improved techniques & technologies. 
In terms of the disaggregates, after V37 consultation with 
USDA, we decided to exclude questions specific to measuring 
these disaggregates to shorten the interviewing time because 
the expected in-person interviews were converted to phone 
interviews due to COVID-19. 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - New
5,515 NA 9,052 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - Continuing
1,029 NA 1,689 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - Crop Genetics
4,910 NA 8,059 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - Pest Management
5,536 NA 9,086 NA - 
MSIKA Final Evaluation Page 84 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - Disease management 
5,662 NA 9,293 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - Soil-related fertility and 
conservation 
6,392 NA 10,491 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance – Irrigation
1,837 NA 3,015 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance - Water management 
(non-irrigation based)
1,993 NA 3,271 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance -Climate mitigation or 
adaptation
1,354 NA 2,222 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance – Other
915 NA 1,502 NA - 
Number of hectares of land under 
improved techniques or 
technologies as a result of USDA 
assistance -Total w/one or more 
improved techniques or 
technologies
6,581 NA 10,801 NA - 
MSIKA Final Evaluation Page 85 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
7 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance 
0 
31,680 30,778 31,680 30,778 97% 
The number of individuals applying new T&T was calculated 
based on a population of 30,922 farming beneficiaries who 
have been trained in ag T&T as of 31st May 2020. A very high 
99.5% of the sample was applying the practices. Therefore, if 
the sample frame had included all 39,332 farmers trained it 
would have resulted in hitting the LOP target. Further
training was disrupted by COVID, but subsequent trainings to 
the finalizing of the sample frame strongly suggests the LOP 
target would have been exceeded. 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance- Male
12,643 13,542 12,643 13,542 107% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance- Female
19,037 17,236 19,037 17,236 91% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance- New 
25,188 23,638 25,188 30,778 122% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance- Continuing
6,492 7,140 6,492 0 0% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Crop Genetics
30,990 24,703 30,990 24,703 80% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Pest Management
28,727 30,173 28,727 30,173 105% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Disease management 
31,335 30,404 31,335 30,404 97% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Soil-related fertility and 
conservation 
31,680 29,886 31,680 29,886 94% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Irrigation
31,384 25,279 31,384 25,279 81% 
MSIKA Final Evaluation Page 86 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Water management 
(non-irrigation based)
31,384 27,726 31,384 27,726 88% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance -Climate mitigation or 
adaptation
31,038 21,277 31,038 21,277 69% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance - Other
31,481 25,999 31,481 25,999 83% 
Number of individuals who have 
applied new techniques or 
technologies as result of USDA 
assistance -Total w/one or more 
improved techniques or 
technologies
31,680 30,778 31,680 30,778 97% 
8 
Number of individuals who have 
applied improved farm 
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance 
0 
26,320 30,196 26,320 30,196 115% 
The number of individuals who applied improved farm 
management practices is calculated from the sample of 979 
individuals (sample frame of 30,922 farmers). Overall, a very 
high 97.7% of individuals applied improved farm 
management practices. The number of individual applying is 
30,196, thereby exceeding the target (114.7%). 
Number of individuals who have 
applied improved farm 
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance- Male
14,213 13,286 14,213 13,286 93% 
Number of individuals who have 
applied improved farm 
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance- Female
12,107 16,910 12,107 16,910 140% 
MSIKA Final Evaluation Page 87 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of individuals who have 
applied improved farm 
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance- Producers
26,320 30,196 26,320 30,196 115% 
Number of individuals who have 
applied improved farm
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance- People in Firms
0 0 0 0 - 
Number of individuals who have 
applied improved farm 
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance- People in 
Government
0 0 0 0 - 
Number of individuals who have 
applied improved farm 
management practices (i.e. 
governance, administration, or 
financial management) as a result 
of USDA assistance- People in Civil 
Society
0 0 0 0 - 
9 
Number of agro-dealers and village 
based input agents operating in 
target districts as a result of USDA 
assistance 
0 
77 80 77 
80 104% 
MSIKA has worked with 80 input suppliers. 3 additional input
suppliers were added to 77 existing input suppliers in this 
reporting period. Input suppliers have played a pivotal role by 
increasing access of inputs in the MSIKA districts. The 
indicator counts input suppliers that have been supported 
with MSIKA and continue operating in the MSIKA districts.
Number of agro-dealers and village 
based input agents operating in 
target districts as a result of USDA 
assistance- Male
67 70 67 
70 104% 
Number of agro-dealers and village 
based input agents operating in 
target districts as a result of USDA 
assistance- Female
10 10 10 
10 100% 
MSIKA Final Evaluation Page 88 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
10 Number of irrigation schemes 
facilitated 0 0 0 45 
45 100% 
No additional irrigation schemes were supported or 
facilitated in this reporting period. MSIKA facilitated or 
supported 45 irrigation schemes over the LOP, this is in line 
with the LOP target of 45.
11 
Number of individuals receiving 
financial services as a result of 
USDA assistance 
0 
5,000 5,612 9,984 13,922 139% 
2,876 individuals received financial services in the April-Sept 
reporting period, bringing the year 4 total to 5,612,
exceeding the target. Over the 4 years of the project, 13,922 
unique individuals received financial services against a target 
of 9,984. Most of these individuals attended training and 
mentorship on financial readiness that lead them to access 
financing. Overall, 10,453 loans have been administered to 
the MSIKA beneficiaries, through MFI, VSL, and SPV. 
Number of individuals receiving 
financial services as a result of 
USDA assistance- Male
2,700 2,520 5,391 
6,175 115% 
Number of individuals receiving 
financial services as a result of 
USDA assistance- Female
2,300 3,092 4,593 
7,747 169% 
12 Number of loans disbursed as a 
result of USDA assistance 0 4,108 5,074 9,561 10,377 109% 
2,375 loans have been facilitated through MSIKA access to 
finance activities in this semi-annual reporting period, 
contributing to a total of 5,074 loans, above a target of 4,108 
in year 4, representing 125% achievement. Cumulatively, 
MSIKA has disbursed 10,377 loans in the LOP exceeding the 
LOP target (109%). MSIKA beneficiaries accessed the loans 
from VSL (5,444), MFI (4,909) and SPV (24). 
Adjustments - 50 loans were double counted in the previous 
semi-annual periods, thus subtracted from what was 
reported in the current period. 
13 
Value of loans provided as a result 
of USDA assistance 
0 
 $ 
333,593 
 $ 
413,100 $ 1,263,431 $ 
699,879 55% 
$137,170 value of loans have been disbursed in the period 
April to Sept 2020. The loans are from MFI ($129,495) and 
VSL ($7,675. The April-Sept loans contributes to the total 
value $413,100 in Yr 4, exceeding the Yr 4 target ($333,593) 
by 128.8%. A cumulative total value of $699,879 in loans 
have been disbursed in the LOP, representing 56% 
achievement against the $1,263,431 target. The loans are 
contributed by MFI ($431,943), SPV ($193,343) and VSL 
($74,592). The loans have benefited processors and 
individual farmers. 
The targets that in the previous reporting periods anticipated 
that MSIKA beneficiaries would be getting bigger loans from 
the SPV. SPV loans are pegged at the collateral that the 
beneficiaries have and most processors and farmers that 
Value of loans provided as a result 
of USDA assistance- Female
$ 
118,462 
$ 
154,639 
$ 
379,029 
 $ 
273,504 72% 
Value of loans provided as a result 
of USDA assistance- Male
$ 
177,693 
$ 
175,667 
$ 
568,544 
 $ 
233,032 41% 
MSIKA Final Evaluation Page 89 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Value of loans provided as a result 
of USDA assistance- Joint
$ 
98,718 - $ 
315,858 
 $ 
98,785 31% 
applied have no/limited high value assets, hence low 
achievement in the LOP. There have been more loans 
disbursed through MFI and VSL, but these are usually in 
smaller amounts. 
$94,558 includes loans not qualifying as disaggregates for 
Male, Female and Joint, as these are loans distributed to 
processors and FBOs, as well as including adjusting amounts. 
Therefore, these have bene put under not applicable 
disaggregates. 
The value reported this period reflect the adjustments made 
in the removal of the 50 duplicate loans described in 
Indicator 12. 
Value of loans provided as a result 
of USDA assistance- Not Applicable 82,794 $ 
94,558 - 
14 
Percentage of producers who can 
recite five or more improved 
agricultural techniques and 
technologies 
81% 66% NA 66% NA - 
After V37 consultation with USDA, it was decided to exclude 
questions specific to measuring this indicator to shorten 
interview time. However, it is worth noting that based on the 
very high achievement on the application of new practices in
indicator 7, the level of knowledge of at least one practice 
would be at least be this level of 97.2%. Ultimately, the aim is 
application, with knowledge as a stepping stone towards 
application. Also, the achievement by Yr3 (midterm) was 
56.7% overall, which suggests that with another year's 
training that the LOP target 66% would likely have been met 
if measured. 
Percent of producers who can cite 
five or more improved agricultural 
techniques and technologies￾Female 
76% 64% NA 64% NA - 
Percent of producers who can cite 
five or more improved agricultural 
techniques and technologies- Male
85% 68% NA 68% NA - 
15 
Percent of agricultural producers in 
target region who can identify key 
characteristics of a well-managed 
farm 
65% 90% NA 90% NA - 
After V37 consultation with USDA, it was agreed to exclude 
questions specific to measuring this indicator to shorten 
interview time. However, based on the very high level of 
application in indicator 8 of farm management practices at 
97.7%, the level of knowledge of farm management practices 
is at least this high. Ultimately, the aim is application, with 
knowledge as a stepping stone towards application. It is also 
noted that the mid-term, the achievement level was 100%, 
so it is very likely that the LOP target of 90% would have been 
achieved if measured. 
Percent of agricultural producers in 
target region who can identify key 
characteristics of a well-managed 
farm- Female
63% 88% NA 88% NA - 
Percent of agricultural producers in 
target region who can identify key 
characteristics of a well-managed 
farm- Male
67% 92% NA 92% NA - 
MSIKA Final Evaluation Page 90 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
16 
Number of on-going agricultural 
research initiatives by DARTS 
supported by USDA assistance 
0 NA 0 - 
Removed from list of indicators in modification 4, signed 
September 17, 2020. 
17 
Number of policies, regulations 
and/or administrative procedures in 
each of the following stages of 
development as a result of USDA 
assistance 
0 
2 2 2 
2 100% 
MSIKA has facilitated development of two policies, the 
horticultural policy and product standards for fruits and 
vegetables. The policies are near completion. MSIKA is 
currently facilitating final review meetings of the Draft 
Horticultural Policy. Final Regional validation workshops are 
expected to be conducted in three regions of Malawi to 
ensure buy-in to the policy by the wider stakeholder 
community, before it is passed to the Office of President and 
Cabinet for approval. The 10 fruit & Vegetable Standards 
with MSIKA assisted in developing have been approved by 
the Malawi Bureau of Standards (MBS) Board. The list of 
Standards have been submitted to the Ministry of Industry, 
Trade and Tourism who will prepare "minutes" requesting 
the Ministry of Justice to approve the Gazetting of these 10 
Standards. 
Number of policies, regulations 
and/or administrative procedures in 
each of the following stages of 
development as a result of USDA 
assistance- Stage 1
0 2 1 2 200% 
Number of policies, regulations 
and/or administrative procedures in 
each of the following stages of 
development as a result of USDA 
assistance- Stage 2
0 2 2 2 100% 
Number of policies, regulations 
and/or administrative procedures in 
each of the following stages of 
development as a result of USDA 
assistance- Stage 3
0 2 2 2 100% 
Number of policies, regulations 
and/or administrative procedures in 
each of the following stages of 
development as a result of USDA 
assistance- Stage 4
2 0 2 0 0% 
Number of policies, regulations 
and/or administrative procedures in 
each of the following stages of 
development as a result of USDA 
assistance- Stage 5
0 0 0 0 - 
18 
Percent of producers who have 
access to current market 
information through their 
cooperatives or producer 
associations. 
13% 48% 27.6% 48% 28% 57% 
The indicator is calculated based on the sample of 979 
farmers (sample frame: 30,922). MSIKA has improved 
progressively from the impact baseline (13%) through to Yr4 
(28%). 
MSIKA Final Evaluation Page 91 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Percent of producers who have 
access to current market 
information through their 
cooperatives or producer 
associations- Female
11% 46% 27.4% 46% 27% 60% 
MSIKA undertook activities to accelerate access to marketing 
information through the FBOs e.g. intensified marketing 
activities, continued running of radio programs where 
farmers share marketing information, built aggregation 
centers, and facilitated sales agreements. 
The pandemic resulted in government policies that restricted 
FBO meetings where FBOs share marketing information, and 
also restricted access to markets. 
Percent of producers who have 
access to current market 
information through their 
cooperatives or producer 
associations- Male
15% 50% 27.7% 50% 28% 55% 
19 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance 
0 
129 226 129 226 175% 
MSIKA assessed performance of FBOs and processors in this 
reporting period. A sample of 49 FBOs was randomly drawn 
from 217 total number of FBOs. 10 processors were 
purposively selected from the total processors reached. The 
results reveal that 100% of the sampled FBOs and 9/10 
processors are applying one or more techniques and 
technologies. The total number comes to at least 226 in this 
reporting period, surpassing Yr4 and LOP targets of 129. 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- New
7 107 129 226 175% 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Continuing
122 119 NA - 
MSIKA Final Evaluation Page 92 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Private enterprise
7 9 7 9 129% 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Producer organization
122 217 122 217 178% 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Women’s group
0 NA 0 NA - 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Trade and business 
association 
0 NA 0 NA - 
MSIKA Final Evaluation Page 93 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Water users association
0 NA 0 NA - 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Community-based 
organization
0 NA 0 NA - 
Number of private enterprises, 
producer organizations, water users 
associations, women’s groups, 
trade and business associations, 
and community-based 
organizations (CBOs) that applied 
improved techniques and 
technologies as result of USDA 
assistance- Other
 0 NA 0 NA - 
20 
Number of public-private 
partnerships formed as a result of 
USDA assistance 
0 
0 8 181 
192 106% 
MSIKA formed 3 partnerships in the period April to Sept 
2020, Bayer Crop Science, Newton Investments and Organic 
Fertilizer Production Company. These partnerships have been 
made to enhance showcasing techniques and technologies in 
MSIKA farmer field schools and Yankho plots, hence fostering 
agricultural productivity of MSIKA farmers. Cumulatively, the 
project has made 192 partnerships above the LOP target of 
181. More partnerships entail more investment in agriculture 
which ultimately contributes to agricultural sector growth.
Number of public-private
partnerships formed as a result of 
USDA assistance- Agricultural 
Production
0 5 12 
17 142% 
Number of public-private 
partnerships formed as a result of 
USDA assistance- Agricultural post￾harvest transformation
0 2 9 
11 122% 
MSIKA Final Evaluation Page 94 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Number of public-private 
partnerships formed as a result of 
USDA assistance- Nutrition
0 0 0 
- - 
Number of public-private 
partnerships formed as a result of 
USDA assistance- Multi-focus
0 0 155 
158 102% 
Number of public-private 
partnerships formed as a result of 
USDA assistance- Other
0 1 5 
6 120% 
21 Volume of commodities (metric 
tons) sold by project beneficiaries 
7.3 per 
beneficiary 
 
24,628 
 
1,261 86,330 
62,963 73% 
Sales were significantly down in the latest reporting period. 
FBOs sold 332 MT in April-Sept 2020, so the total sales 
volume in Yr 4 comes to 1,261 MT, representing 5% of Yr4 
target. Cumulative sales were 62,963 MT against the LOP 
target of 86,330 MT (73%). Sales data for processors was not 
available in the reporting period, as project staff could not 
travel to conduct monitoring activities. Processor sales are 
reported in the KIIs to have also been affected by the 
pandemic restrictions. 
The government enforced COVID-19 measures, which led to 
closure of schools, hotels, and other private and government 
bodies. They also restricted meetings and conferences 
nationwide resulting in lost sales for fruit and vegetable 
farmers. Farmers had to rely on local markets. 
22 
Number of jobs attributed to USDA 
assistance 
0 
 
1,657 
 
2,067 1,657 
2,067 125% The Yr 4 and the LOP targets were surpassed other than for 
females. The result was calculated in the Final evaluation 
using data from the beneficiary producers based on the 
sample of 979 farmers (sample frame of 30,922 farmers 
trained in basic ag production as of 31st May 2020). 
The total for Yr4 does not include jobs created by processors 
and by FBOs, and other businesses e.g. input suppliers, as it 
has not been possible to collect this data from all these 
parties due to the COVID restrictions. 
Number of jobs attributed to USDA 
assistance- Female
 
762 
 
507 762 
507 67% 
Number of jobs attributed to USDA 
assistance- Male
 
895 
 
1,560 895 
1,560 174% 
23 Value of sales by project 
beneficiaries 
$2122.36 per 
beneficiary 
 $ 
4,900,310 
 $ 
1,367,776 $ 21,972,909 $ 
18,440,375.71 84% 
FBO sales in April-Sept period were $134,740 to the total 
value of sales in year 4, which is 28% of the target 
($4,900,310). As noted on volumes, sales were particularly 
affected by COVID restrictions. No data was available for 
processors in this period due to limitations on collecting data. 
MSIKA Final Evaluation Page 95 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
24 
Value of new public or private 
investment leveraged by USDA 
assistance 
$0.00 
 $ 
333,593 
 $ 
1,686,530 $ 1,967,799 $ 
3,320,736.32 169% 
MSIKA beneficiaries leveraged the value of $1,413,173 in 
investments in the April-Sept period. This was mainly 
contributed to by loans ($413,100) and Processors 
($1,000,070). The total value for year 4 is $1,686,530, 
representing 506% achievement against a target of 
$333,593. The value of $3,320,736 in investments has been 
made over MSIKA LOP against the LOP target of $1,967,799, 
representing 169% achievement.
There has been an increase in investments in this reporting 
period particularly farmers as they have intensified irrigation 
farming in the MSIKA value chains.
Value of new public or private 
investment leveraged by USDA 
assistance- Public
$ 
- 
$ 
- $ - $0.00 - 
Value of new public or private 
investment leveraged by USDA 
assistance- Private
$ 
333,593 
$ 
1,686,530 
$ 
1,967,799 $3,320,735.93 169% 
25 
Percent of registered processing 
firms in target sectors that obtain 
certification with Malawi Bureau of 
Standards related to product 
quality 
0 30% 44% 30% 44% 147% 
No processing firm obtained certification in the April-Sept 
reporting period. Only essential government services have 
been operating in full due to COVID 19. It's assumed most 
processors have not been visited in this period. Despite this, 
the LOP target had already been achieved (147%). 
26 
Percent post-harvest losses for 
beneficiary producers 12% 14% 13.8% 14% 13.8% 101% 
MSIKA has surpassed the LOP target overall and across the 
disaggregates. The IMPACT Baseline (IB) values have been 
revised to account for just the four crops (Tomato, Onion, 
Potato and Mango) that were of focus in year 4, not the 
original 7 crops. Thus, the revised IB for this indicator is now 
14.77%. In the final evaluation, the losses reported were 
13.8%, meaning that PH losses have reduced marginally 
(from 14.77% to 13.8%) by EOP. Losses are very difficult to 
calculate, as farmers rarely weigh what they throw away. 
There are also losses at various stages (at the farm, transport 
to market and at the market), so it can be difficult to 
estimate/calculate what losses have occurred. At best, the 
figures from farmers are rough estimates. This implies that 
the baseline and measurement data should be taken with 
caution. Also, the target for this indicator (14%) was set at 
the beginning of the project when the focus was on 7 crops 
and not revised following the revisions to the IB or the 
change to 4 focus crops. Hence it is irrelevant given current 
estimates of the IB and the reported losses in the in the FE. 
Percent post-harvest losses for 
beneficiary producers - female 13% 16% 15.4% 16% 15.4% 104% 
Percent post-harvest losses for 
beneficiary producers - male 12% 13% 11.8% 13% 11.8% 110% 
27 
Total increase in installed storage 
capacity (dry or cold storage) as a 
result of USDA assistance 
0 
- 
 
438,788 45,770 
484,558 1059% 
MSIKA has very significantly surpassed the LOP target for this 
indicator across all the disaggregates. The volume in storage 
which was built or refurbished has been calculated based on 
a sample of 979 out of a population of 30,922 individuals. 
MSIKA Final Evaluation Page 96 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
Total increase in installed storage 
capacity (dry or cold storage) as a 
result of USDA assistance- Dry
 
- 
 
438,788 
 
45,770 
 
484,558 1059% 
23.3% farmers built new storage, while 13.78% refurbished 
their storage. 
The data for the calculation has been reviewed several times, 
as the amounts appear very high in comparison with previous 
data. The consultants applied some realistic maxima, but the 
data still appeared high. The consultant attributes the high 
data to farmers estimating over the phone, but the 
enumerators being unable to verify the numbers by getting 
farmer to pace out the size of the storage or indicate its 
height visually. 
While the data is probably overstating the change, the 
consultant's view is that the target is very likely to have been 
achieved even if there had been verification of the farmers 
estimated measurement of the storage. 
Total increase in installed storage 
capacity (dry or cold storage) as a 
result of USDA assistance- Cold
 
- 
 
- - 
- - 
Total increase in installed storage 
capacity (dry or cold storage) as a 
result of USDA assistance￾Refurbished 
 
- 
 
169,109 
 
33,941 
 
203,050 598% 
Total increase in installed storage 
capacity (dry or cold storage) as a 
result of USDA assistance- New
 
8,500 
 
269,680 
 
11,829 
 
281,508 2380% 
28 
Number of national export 
promotion advertising campaigns as 
a result of USDA assistance 
0 0 0 6 
6 100% 
As planned, there was no national export promotion 
advertising in the period April to Sept 2020. Cumulatively the 
project has achieved 100% of 6 LOP target.
29 Number of sales agreements with 
FBOs 0 10 209 62 
261 421% 
No sales agreements were reported in this April - Sept 
period, as COVID 19 forced closure of many institutions such 
as schools, hotels and lodges that are major buyers of fruit 
and vegetables, so farmers were relying on sales in local 
markets not under sales agreements. 
30 
Number of FBOs using improved 
financial management practices and 
systems as a result of USDA 
assistance 
0 - 
Indicator was removed following modification of PMELP
31 
Percentage of FBOs who are aware 
of international production and 
handling standards as a result of 
USDA assistance 
0 71% NA 71% 78% 109.58% 
No data for this indicator was captured during the final 
evaluation, as it was not a priority in the limited interview 
time due to COVID-19. 
Based on the data available prior to that, the LOP target has 
been surpassed (110%). 
32 
Number of processor staff trained 
in quality standards 
0 
0 0 187 
187 100% 
Trainings for processor staff were not planned in this April￾Sept period. The LOP actual is at 187, which is 100% the LOP 
target.
Number of processor staff trained 
in quality standards- Male 0 0 97 
97 100% 
Number of processor staff trained 
in quality standards- Female 0 0 90 
90 100% 
MSIKA Final Evaluation Page 97 kadale@africa-online.net
 Performance Indicator Impact 
Baseline Value 
Year 4 
target 
(NEW) 
Y4 TOTAL LOP TARGET 
(NEW) LOP ACTUAL 
% 
Achievement 
(Y1-Y4) 
COMMENTS 
33 
Number of FBOs trained in 
improved financial and 
organizational management 
0 50 48 135 183 136% 
38 FBOs have been trained in financial and organizational 
management in the April to Sept period, bringing the total for 
year 4 to 48 unique FBOs trained, representing 96% of year 4 
target (50). A total of 183 FBOs have trained over the LOP 
(136% achievement). 
MSIKA Final Evaluation Page 98
kadale@africa-online.net
Annex 2: MSIKA Results Frameworks
MSIKA Final Evaluation Page 99
kadale@africa-online.net
Annex 3: Terms of Reference/Scope of Work
Malawi Strengthening inclusive Markets for Agriculture (MSIKA) Final Evaluation
Modification: August 27, 2020
1. INTRODUCTION AND JUSTIFICATION
This contract is to hire an external evaluation firm to conduct the final evaluation (FE) of the four-year 
MSIKA Food for Progress project, funded by United States Department of Agriculture (USKA) and 
implemented by Land O’Lakes Venture37 (Venture37). The project is implemented in Mchinji, Dedza, 
Ntcheu, Lilongwe and Mangochi Districts of Malawi from October 1, 2018 – December 31, 2020.
The FE will examine the results of the project, and provide recommendations for the implementation 
of similar projects in the future. An externally conducted FE is required by the donor, USDA, in the 
project contract. This document describes the background of the project, the scope of work that the 
evaluation contractor will implement, and the timeline for conducting the scope of work. 
2. BACKGROUND 
MSIKA is a four-year value chain development project that will reach 36,000 smallholder farmers, 217
farmer-based organizations (FBOs), and 18 processors in south central Malawi in the fruit and 
vegetable value chains, specifically tomato, potato, onion, chili, mango, citrus and guava. MSIKA is 
catalyzing increased value addition and income for value chain actors by facilitating improved 
processing, increased crop productivity, improved post-harvest handling (PHH) and storage, expanded 
market linkages between farmers and processors, more efficient domestic trade, and increased 
potential exports of processed products in the long term. MSIKA interventions and market linkages 
will target generating a total of $21,972,909 in sales by project participants across the four years and 
leveraging $1,967,799 in new public or private investment by the end of the project. 
MSIKA focuses on achieving the following objectives:
• Increase agricultural productivity in the fruit and vegetable sector by increasing the availability 
of improved inputs, improving infrastructure to support on-farm production, facilitating 
access to finance, and training farmers on improved agricultural techniques and technologies, 
as well as farm management.
• Expand trade of agricultural products in the fruit and vegetable sector by improving quality of 
postproduction agricultural products, training producers and processors on improved post￾production processes, facilitating improved linkages between buyers and sellers, improving 
market and trade infrastructure, and facilitating improved management of buyer/seller 
groups.
Due to monetization issues, MSIKA had to shorten the project duration from five year to four years.
During the fourth and final year, the project will focus on a subset of the participants (122 FBOs, 
8000 farmers and 7 processors) and the tomato, potato, onion and mango value chains to maximize 
sustainability within the most promising values chains for commercialization. 
2.1 Project Implementation
This section provides a more detailed description of MSIKA activities. 
Training: Improved agricultural production techniques
Venture37will provide tailored training, technicalresources, and cutting-edge research to fruit and
vegetable farmersto improve technical and businessskills. Venture37 will build capacity in tandem
with, and transfer knowledge to, community leaderssuch aslead farmers and extension agents
fromtheMinistryofAgriculture, IrrigationandWaterDevelopment(MoAIWD). Venture37 will also
build capacity through Farmer Field Business Schools(FFBS) and demonstration plots(Yankho
PlotsTM) on topicssuch as orchard management, budding and grafting, nursery establishment,
MSIKA Final Evaluation Page 100
kadale@africa-online.net
pruning, and weeding. Venture37 will also ensure that FFBS and Yankho PlotsTM build core climate
smart agronomic and business skills of target producers.
In addition,Venture37 will engage private sectorpartnersto ensure increased utilization of
irrigation technologies and increased accessto improved inputs. Venture37 will engage private
sector partners to co-invest in Yankho PlotsTM and to establish for-profit nursery enterprises. 
Infrastructure: Post-harvest handling and storage 
Venture37will improve post-harvest handling and storage infrastructure by providing training and
technical assistance,facilitatingmarketlinkages, andsupporting accessto finance to improve on-farm
and off-farm post-harvest infrastructure and to increase the use of improved post-production
processing and handling practices. Venture37 willfacilitate increased adoption of established
standards, such as Global Good Agricultural Practice (Global GAP), by market actors, such as
traders, processors, and producers.
Venture37willfacilitate Farmer-BasedOrganizations(FBO) andprivate enterprisesto develop
wholesale markets, and aggregation centers, and to utilize technology and practicessuch as
optimalpacking technology, improved storage practices, and sorting, grading, and washing
technology. Venture37 willprovide technical assistance to supportthese initiatives.
Training: Post-harvest processing
Venture37willimprovetheefficiencyandprofitabilityofvalue-addedpost-harvest manufacturing and
promote the utilization of value-preserving practices and technologiesthroughoutthe fruit and
vegetable value chains. Venture37 will provide targeted technical assistance to small, medium- and
large-scale fruit and vegetable processors. Technical assistance will focus on processing line
operational efficiency,improved packagingandlabelling,modernizationofphysicalequipment and
plantfacilities, and improved business and financialmanagement practices. Venture37 willtrain fruit
and vegetable processors on improved product quality standards and facilitate Malawi Bureau of 
Standards (MBS) certification. Venture37 will facilitate FBOaccessto improved post-harvest
techniquessuch as use of reusable crates, sorting and grading, and washing. Venture37 will also
facilitate improved phytosanitary practices by agricultural enterprises and FBOs engaged in small￾scale processing of fruits and vegetables into productssuch as jams, juices, sauces and paste making.
Capacity Building: Producer groups and cooperatives
Venture37 will apply cooperative developmenttools, including Venture37’s AgPrO products, to
improve farm management and to increase the capacity of FBOsin agricultural production, post￾harvest handling, and processing. Venture37 will link FBOsto business development service (BDS)
providers and finance providers, andwill collaboratewith national-levelproducer associations and
district-level FBOs(such asthe Mchinji Horticulture Association) to strengthen the delivery of
member services and facilitate the transformation of FBOsinto agricultural enterprises.
Market Access: Facilitate buyer-seller relationships
Venture37will improvemarket accessby facilitatingbuyer-sellerlinkages, increasing access to
marketinformation, and strengthening the capacity of key organizationsin the trade sector. 
Specifically,Venture37willfacilitatebuyer-sellernetworkingevents,linkagestotransporters the
establishment and strengthening of equitable outgrowerschemes between producers and
processors (e.g., Malawi Mangoes, RAB Processors), stronger linkages to retail supermarkets
forstandards and specification trainings and procurement, promotion of village-based input
sales agents, and targeted support for women- and youth-owned small and medium enterprises
(SMEs) engaged in food processing.
Financial Services: Facilitate agricultural lending
Venture37 will build the capacity of a local financial institution to conduct risk assessments and
develop financial productsfor value addition in the fruit and vegetable sectors,such as working
capital and trade finance for processing, aggregation, irrigation, export, and transportation.
MSIKA Final Evaluation Page 101
kadale@africa-online.net
Venture37 willfacilitateincreasedlendingforclimatesmartirrigation;workwithfinancialinstitutions
todevelop loan products appropriateforhorticulturevalue-adding activities;facilitate the
developmentofinput credit relationships between processors and producers or FBOs; improve the
financial literacy and credit- worthiness of FBOs; and facilitate the development of capital investment
plans by fruit and vegetable processorsfor plant modernization and expansion, including
improvement of business and financial management systems. 
Financial Services: Provide SME finance
Venture37will stimulateincreasedagriculturallendingbyestablishing a Special Purpose Vehicle (SPV). 
The SPVwillmake loans to financial institutions that will on-lend those fundsto enterprises, farmers 
and farmer organizations. The SPVmay catalyze further investmentin value-adding activities and
trade,such asloansfromtheOverseas Private Investment Corporation (OPIC). Venture37 will ensure
that loans originating from the SPV, priorto the award’s date of completion, will be used for
purposes such as increasing access to working capital for farmers and SMEsfor input purchases,
transport forspace arbitrage of fruits and vegetables, market and trade infrastructure, processing
equipment, and equipment upgrades. After the award’s date of completion, Venture37will ensure
thatloans originating from the SPV will be used to expand agricultural production and trade in African
Least Developed Countries, as defined by the World Bank. Forthe life of the SPV, Venture37 will use
principal and interest repaid to the SPV for continued lendingandinvestment.
Government Capacity Building: Improve Enabling Environment
Venture37will buildthe capacityof key governmentresearchand extensionagencies, including
MoAIWDand theDepartment ofAgricultural ExtensionServices(DAES). Venture37 will work closely 
with the Lilongwe University of Agriculture and Natural Resources (LUANAR) to conduct research 
on soil fertility and other value chain specific trials. Venture37 will build the capacity of processors,
traders, national associations, and FBOs to advocate for improved regulations and policies.
Venture37 willwork with the Malawi Bureau of Standards(MBS)to jointly establish and disseminate
international quality standards. Venture37 will conduct a gap analysis and recommend concrete
improvementsto bring MBS’s standards, certification processes, and quality audit processes up to
international levels. 
Venture37 will support multi-stakeholder and public- private dialogues on issues of key importance
to the agriculture production, processing, and trade sectors.
3. EVALUATION DESIGN 
Objectives of the Final Evaluation
The objectives of the final evaluation are as follows: 
8. Assess the relevance of the project strategy and approach to project participants;
9. Measure progress the project has made toward key results, including effectiveness and 
efficiency of interventions in achieving established targets (see annex 2 for key outcome 
indicators); 
10. Identify enablers and constraints to progress (both internal and external factors) that have 
supported or limited success of the project;
11. Assess the sustainability of project outcomes, and the effectiveness of sustainability efforts 
that were undertaken by the project; 
12. Assess operational aspects of the project, such as project management and monitoring and 
evaluation; 
13. Document lessons learned, challenges and unanticipated effects;
14. Provide recommendations for best practices that were realized or areas of improvement for 
future USAID and Venture37 program in similar areas and value chains. 
Evaluation Questions 
MSIKA Final Evaluation Page 102
kadale@africa-online.net
The final evaluation will seek to answer the following key questions within the standard evaluation 
criteria: 
Criteria Questions
Relevance • Did the results framework, assumptions, program design and project 
activities meet the needs of the participants and local conditions in the 
five target districts of Malawi? 
• How aligned was the program strategy and activities with Government 
of Malawi (GoM) strategies and with USDA and USG development 
goals, objectives and strategies? 
Effectiveness • What internal and external factors have influenced the ability of the 
project to meet expected results and revised targets? 
• To what extent have the revised program targets and outcomes been
achieved by the end of the project? 
• How have the project participants and their businesses been impacted 
by the COVID-19 pandemic? 
Impact • What impacts are the project activities having on the program 
participants, both positive and negative, especially in relation to the 
expected results and strategic objectives? 
o How has project training and access to finance affected the 
uptake of improved agricultural techniques, farm management 
practices, PHH, and use of improved infrastructure for farmers?
o How has the use of improved agricultural techniques and 
technologies and market linkages affected crop yield and 
farmer sales? 
o How has improved PHH practices affected post-production 
losses and improved the quality of the product? 
o How have producer group trainings affected the capacity of the 
groups? 
o How has the project training and access to finance affected 
uptake of improved processing techniques, quality standards, 
and improved infrastructure for processors? 
• What key successes should be replicated, or key improvements should 
be made in future programming? 
Efficiency • Were the resources and activities provided by the program carried out 
in a timely manner and with effective use of resources? 
• How well has the project been managed and M&E data used to make 
programmatic decisions? 
Sustainability • Which project activities and benefits are likely to be sustained or not, 
why? How?
• What evidence is there that the interventions are likely to scale up 
beyond the project life
Scope of Work
MSIKA Final Evaluation Page 103
kadale@africa-online.net
The contractor will conduct a mixed methods evaluation with a non-experimental design, 
incorporating both primary quantitative and qualitative data collection, supplemented by project 
monitoring data. Due to COVID-19 pandemic, the final evaluation will collect information through 
phone surveys of registered participants, including farmers, FBOs and processors, to compare key 
metrics to the values at the impact baseline. 
The below text details how data will be collected from the key program participants and stakeholders 
to answer the key evaluation questions. A summary of the data collection methods can be found in 
the table below. The current sample frame is 217 FBOs, 18 processors, and about 36,000 farmers, 
whereas the participantsthat will receive focus (continued capacity building and support) in the fourth 
year are 50 FBOs, 8,000 farmers and 7 processors. The final sample frame will be determined in 
collaboration with the contractor. 
Stakeholder Data Collection Method
Participant Farmers
Household Survey
Key Informant Interview
Farmer-based Organizations
Financial Data
Key Informant Interview
Processors
Financial Data
Key Informant Interview
Other Key Stakeholders Key Informant Interview
Farmers: The contractor will collect quantitative data from participant farmers in each of the four 
targeted value chains (tomato, potato, onion, mango) to assess their agricultural and post-production 
practices, crop yields, post-harvest losses, crop sales and use of finance. For each target value chain, 
the final evaluation will collect information from a statistically relevant number of households that 
participate in that value chain at the 90 percent34 confidence level, as possible. Participant 
respondents will be selected randomly from the farmer participant list, the sample size proportionally 
spread across the target districts. With the change to phone interviews, each farmer will only be asked 
about one crop that they grow. This crop will be selected randomly until a quota is reached and then 
that crop will be taken out of the random selection for future farmers. The contractor and project will 
agree at inception how to ensure representation of both the farmers that did and did not receive 
support in year 4. 
The contractor will also collect qualitative data from farmers in the targeted value chains through key 
informant interviews with farmers, in each of the districts. The KIIs will provide context to the 
quantitative data to describe why farmers are or are not changing their agricultural practices, 
successes and challenges in growing and selling their crops, participation in FBOs and feedback on 
how the project can be improved. 
Farmer-based Organizations: The contractor will utilize monitoring data collected from FBOs. This 
data will include group agreements with input providers, processors and retailers, financial 
information on group sales and use of financing. The contractor will also collect quantitative and 
qualitative information from a purposive sample of project-supported FBOs through remote 
structured key informant interviews to verify the monitoring data, understand their relationships with 
input providers, processors, and retailers; the functioning of the group and value they provide to their 
members; their successes and challenges of working as a group, and solicit feedback on how the 
project could improve its activities. The evaluation will sample FBOs across the different districts and 
34 The overall sample size will exceed the 95% confidence level for all horticulture, but meet the 90% 
confidence level for each individual value chain. 
MSIKA Final Evaluation Page 104
kadale@africa-online.net
value chains that are 1) under performing; 2) of average performance; and 3) performing well, 
according to their monitoring data. Samples will be taken both from those that are reached in the 
fourth year and those that have not been reached in the fourth year. 
Processors: The contractor will utilize monitoring data collected from processors, including quantity 
of production and value of sales. The contractor will also collect quantitative and qualitative 
information through remote structured key informant interviews with 12 processors, or those willing 
to participate within the project timeframe, to understand changes they have made in their processing 
practices, ways they are engaging with the FBOs, use of financial resources, successes and challenges 
in their current functioning, and suggestions for project improvement. 
Other Key Stakeholders: The evaluation will conduct remote key informant interviews with other key 
project stakeholders, including input supply distributers, trader/wholesalers, participating 
government staff, local leaders and program staff to understand how they have participated in the 
project, challenges and successes, and suggestions for improvement. 
The specific tasks that the contractor will carry out are detailed below: 
Meeting with USDA: Conduct an introductory phone call with USDA prior to conducting fieldwork for 
the FE. If requested by USDA, conduct a phone call with USDA after the draft report is completed to 
share key findings. 
Review of Documents: Undertake review of the MSIKA program documents and other relevant 
documents that are available at the time, including, but not limited to, the following:
• Project agreement with USDA, including the MSIKA scope of work and all amendments
• MSIKA Performance Management Plan
• MSIKA MEL Plan
• Semi-Annual farmer performance survey data
• Monitoring data, including list of farmers, FBOs, processors, and other key stakeholders
• Semi-annual reports submitted by Venture37 to USDA;
• Initial and impact baseline and midterm report, data collection tools and data; 
• Any other program documents which will enable the evaluator to get acquainted with the 
project progress including value chain and financial studies;
• Relevant Government of Malawi reports and documents for background information and 
establishing the socio-economic and political context in which MSIKA occurred.
Refinement of methodology and data collection tools: The evaluator, in close collaboration with the 
Venture37 Global Monitoring, Evaluation, and Learning team, will do the following: 
• Develop a finalized methodology, including a sampling frame, sampling technique and sample 
sizes for both quantitative and qualitative surveys. The sampling frame will use a minimum 
confidence level of 90 percent for each value chain, as possible. Surveys should be 
comparable to the impact baseline to ensure comparability of data over time, but additional 
questions could be added, or tools created at final evaluation to ensure all evaluation 
questions, described above, are answered. 
• Based upon a reading of the program documents, propose any additional topics or issues for 
analysis prior to conducting the final evaluation.
• Further refine the methodology and tools based on the COVID-19 pandemic to transition 
interviews to remote phone-interviews. 
Field Data Collection
• Plan and coordinate the necessary logistics to collect the data in accordance with the selected 
methodology. 
• Pre-test, edit, translate, finalize and reproduce the survey instruments.
• Train and orient enumerators and data collection team.
MSIKA Final Evaluation Page 105
kadale@africa-online.net
• Carry out the phone interviews using own office space, including: 
o Household survey: Approximately 1,092 phone surveys with farmers; 
o Key informant interviews with farmers: Approximately 12 KIIs
o Interviews with other Key Informants (farmer groups, input services providers, district 
agriculture officers and Venture37 program staff): Approximately 52 KIIs. 
Data entry, analysis and reporting
• Enter, clean, synthesize, analyze, and interpret data from both the quantitative surveys and 
the qualitative protocols using approved statistical packages. The quantitative data from 
farmers will be analyzed using pre-post comparison of key indicators between the impact 
baseline and the final evaluation. The quantitative data from FBOs and processors will be 
compared over time. Qualitative data will be transcribed and analyzed by themes. 
• Prepare a draft evaluation report addressing the objectives and questions of this evaluation 
outlined in this TOR and recommendations on the overall Venture37/FFPr MSIKA project for 
potential similar future project for review by Venture37 staff and stakeholders.
• Develop a PowerPoint presentation of evaluation findings, present and submit to Venture37
and stakeholders.
• Prepare a final evaluation report that includes revisions based on feedback on the draft report 
and presentation.
 
4. TIMEFRAME 
The contractor will carry out the work according to the below timeline, starting on June 15th, 2020. 
Final Evaluation Activity Responsibility Due Date
Review of relevant documents to prepare for 
inception meeting
Evaluator June 15th – 19th, 2019
Inception meeting with Venture37 to discuss 
protocol, methodology, sampling, tools and timeline
Evaluator and 
Venture37
June 22nd, 2020
Develop an inception report and update data 
collection tools 
Evaluator June 23rd – July 8th, 2020
Inception report and tools due to Venture37 Evaluator July 10th, 2020
Venture37 reviews report and tools and provides 
feedback, comments and suggestions to evaluator 
Venture37 July 13th – 17th, 2020
Prepare for field work, finalize tools based on 
Venture37 feedback, test, refine and code
instruments
Evaluator July 13th – 17th, 2020
Adjust methodology and tools for remote data 
collection due to COVID-19
Evaluator July 18th – August 21st
Enumerator training and data collection Evaluator August 24th – September 
11th, 2020
Data cleaning, analysis and report writing Evaluator – September 14th –
October 20th , 2020 
Draft Outline of final evaluation report for review Evaluator September 25th, 2020
Draft final evaluation report submitted to 
Venture37
Evaluator October 20th, 2020
Venture37 reviews draft final report and provides 
evaluator with comments and suggestions for 
revisions
Venture37 October 21st – 27th, 2020 
Presentation of evaluation findings to Venture37 Evaluator Week of October 26th 
MSIKA Final Evaluation Page 106
kadale@africa-online.net
Final Evaluation Activity Responsibility Due Date
Finalize report based on Venture37 feedback and 
prepare all deliverables
Evaluator October 28th – November 
3rd, 2020
All Final Deliverables Due (Final evaluation report, 
clean data, and PPT presentation)
Evaluator November 3rd, 2020
5. REQUIRED DELIVERABLES
The deliverables under this assignment are listed in the table below. 
Deliverable Due Date Description
Inception Report July 10th, 
2020
Report should describe the following:
i- Understanding of the project based on project 
documents and literature review
i- Finalized methodology including detailed sampling plan 
and field procedures
i- Quality control measures
v- Communication protocol
v- Finalized timeline (activities, responsible party, outputs, 
and timing)
i- Draft Data collection tools 
Outline of 
evaluation report
September 
25th , 2020
Submit a high-level outline of the report, including proposed 
sections for results and findings. 
Draft final 
evaluation report 
October 20th, 
2020
The report should be submitted in English addressing all the 
evaluation objectives and questions listed in the scope of work
PowerPoint 
Presentation
November 
3rd , 2020
Presentation should include an abbreviated list of evaluation 
findings that can be presented to relevant internal and 
external stakeholders
Non-technical 
summary of 
evaluation 
findings
November 
3rd 2020
A short (2-4 pages) non-technical summary of the evaluation 
findings, easily understood by non-experts, and include 
graphics such as infographics, charts and/or graphs. Can be the 
same as the executive summary
Final version of 
the final 
evaluation report 
November 
3rd , 2020
Electronic copy of the final evaluation report should be 
submitted in English in both Microsoft- Word and PDF version. 
Report should include the following sections: 
a. Acknowledgements
b. List of Acronyms and abbreviations
c. Table of Contents 
d. Executive Summary (no longer than four pages)
e. Background (Program description and purpose of 
baseline)
f. Methodology and Implementation
g. Results and Findings
h. Recommendations
i. Annex: Table of key program indicators with updated 
values in comparison to baseline values for all required 
disaggregates in the PMP
j. Annex: Scope of Work for the evaluation
k. Annex: Inception Report for the evaluation
l. Annex: Survey Instruments: questionnaire(s), survey(s), 
interview protocol(s), focus group discussion protocol(s)
MSIKA Final Evaluation Page 107
kadale@africa-online.net
Deliverable Due Date Description
Final Data 
Collection Tools
November 
3rd , 2020
Electronic copies of all clean and final English-version of data 
collection tools
Final Cleaned 
Data 
November 
3rd , 2020
Clean and final English versions of:
- quantitative data sets in Microsoft-Excel and any 
other utilized format (SPSS, STATA, etc) 
- qualitative summaries, field and interview notes, 
complete list of key informant interviews and FGDs in 
Microsoft-Word document 
MSIKA Final Evaluation Page 108
kadale@africa-online.net
Annex 4: Instruments 
Household Survey
Malawi Strengthening Inclusive 
Markets for Agriculture (MSIKA)
Final Evaluation 2020
Kadale Consultants Ltd
MSIKA Final Evaluation Page 109
kadale@africa-online.net
2. ID Module
Q# Question Response Logic*
ID1 Enumerator Name 
ID2 Date of Interview
ID4 District Dedza
Lilongwe
Mangochi
Mchinji
Ntcheu
ID5 EPA
ID8 Village
ID6 TA
ID7 GVH
3. Introduction & Consent
Read out and seek consent:
“Hello, my name is [NAME] and I work for Kadale Consultants. As I already explained 
previously, Kadale is conducting a final evaluation for Land O’Lakes’ MSIKA program, which 
includes interviewing MSIKA beneficiary households in your community through the phone. Your 
household has been randomly selected to participate in this survey, which will ask questions
about your household’s farming activities specifically in fruits and vegetable production 
and sales over the phone. The aim is to gather useful information that will be used to 
improve fruit and vegetable farming in this and other communities.
The issues discussed in this interview will remain confidential and what you say will not be 
shared with anyone outside the research team.
The survey will take about 45 minutes. Taking part in this research is voluntary. 
Are you willing to participate in this survey?” 
Consent is given by the person (continue to screening questions)
Does not consent (Terminate interview and record the name and why they did not consent) 
4. Screening Questions
1. What is your first name and family name? (check that it is the selected person, not a 
person sent on their behalf, such as a spouse or other relative)
Yes, it is the selected person (continue) 
No, it is not the selected person (ask if the selected person is available and where 
to find them/how to contact them – then terminate interview and try to find the 
selected person, If not possible, inform the Supervisor).
2. Which of the following crops did you yourself or jointly with other family members 
plant and harvest, in the past 12 months between August 2019 to July 2020?
a. Tomato – yes / no 
b. Onion – yes / no
c. Potato – yes / no
d. At least four mango trees 
e. Did not grow any of these
If Yes to any one or more of the above (continue)
If option e, (If they did not grow any of these crops in the last year (1st August 
2019 to 31st July 2020), then they may not be the person that is registered. Check to 
see if it is the person. If so, find out if they ever grew these and note in what 
period they grew them. If not in the last 12 months end the interview and tell the 
Field Supervisor).
2b. Out of these crops, please say which is the most important crop to you in order, 
starting with the most important. (1 being the most important, 4 the least – ODK will 
only now give the ones they stated in 2 above)
a. Tomato 
b. Onion
c. Potato
MSIKA Final Evaluation Page 110
kadale@africa-online.net
d. Mango
3. Have you yourself received any training from the Land O’Lakes MSIKA project, 
including:
(Read out list)
• Training by a Land O’Lakes /MSIKA staff member or Land O’Lakes /MSIKA lead 
farmer(s)
• or at a Land O’Lakes /MSIKA field day
• or at a Land O’Lakes /MSIKA Yankho plot 
• or at a Land O’Lakes /MSIKA farmer field school? 
Yes (continue to A. General Information - If you continue and it becomes clear that they 
have not been truthful on the screening, check with the Supervisor)
No (If they have not received any training or other services from Land O’Lakes /MSIKA, end 
interview and tell the Field Supervisor)
As already communicated, we will send to you airtime worth MK 500 after the interview is 
completed. If the call is cut off, we will call you back. Does your batter have enough 
charge? If not, when is a good time to call you when it is charged?
Enumerator instruction - Using the random crop selection method, make the selection from 
the crops they grow and tell them that we will be asking them about <CROP>. ODK will link 
to the selected crop.
The crop selected is: 
a. Tomato 
b. Onion
c. Potato
d. Mango
5. A. General Information
Q# Question Response Logic
A2 [Enumerator to note sex of respondent] Male
Female
A3 What is the highest level of education you 
reached? 
[Prompt if necessary]
Adult literacy
Standard 1-8 
Form 1-4 
Further Education
None
A4 What is your marital Status?
[Prompt if necessary]
Married
Widowed
Divorced
Separated
Never Marred
A5 What is your age?
[Estimate age of person if don’t know]
[If less than 17 terminate interview]
___ years
A6 Are you the head of the household (HHH)? Yes 
No
A7 What is your relationship to the HHH? Head of HH
Spouse of Head of HH
Daughter/Son of HHH
Brother/Sister of HHH
Father/Mother of HHH
Other relation of HHH
Friend of HHH
A8 BLANK
[IF NOT HHH]
A9 What is the sex of the HHH? Male
Female [IF A6=No]
A13 How many family members does this 
household have?
A14 –
A23
BLANK
MSIKA Final Evaluation Page 111
kadale@africa-online.net
Q# Question Response Logic
State: Now I want to talk to you about your 
participation in the MSIKA project
A25 Are you a Lead Farmer trained by Land 
O’Lakes /MSIKA?
Yes
No
A25a As a Lead Farmer, have you done any of the 
following? 
Yes
No
If A25=Yes 
(multiple 
response 
possible)
a. Conducted trainings of other farmers
b. Submitted training forms to Land 
O’Lakes /MSIKA about the training
c. Established a demonstration plot
d. Other (please specify)
A26 BLANK
A27 Which of the following activities from 
Land O’Lakes /MSIKA have you participated 
in?
[Read out list]
a. Training in improved horticultural 
farming practices by Land O’Lakes /MSIKA 
staff or Lead Farmer
Yes
No
b. Training in horticultural post-harvest 
handling practices by Land O’Lakes /MSIKA 
staff or Lead Farmer
c. Training in marketing
d. Training in financial literacy
e. Training in gender equality
f. Training or membership of a village 
savings and loan (VSL) group
g. Been linked to an MFI, 
h. Been linked to markets to sell 
horticulture products
i. Participated in an international or a 
district trade fair
j. Any other? (specify)
 
6. S. Crops and Land Area
Q# Question Response Logic
Note I will now ask you questions related to [CROP] As selected in 
intro section
S1a
From how many mango trees did your household harvest 
mango fruit from for the first full harvest in the past 
12 months (August 2019 to July 2020)? 
[If mango 
stated as one 
of their crops]
Cross check 
that must be 
=>4
S1b How many times did you plant and harvest <selected 
field crop> in the last 12 months (August 2019 to July 
2020)?
[This can only be for field crops not tree crops]
One time
Two times
Three times
[If one of the 
field crops is 
selected]
<selected crop>
S2a How many acres of <crop> did your household plant and harvest for the 
first full harvest in the past 12 months (August 2019 to July 2020)? 
[If S1 = yes]
first planting (acres)
S2b How many acres of <crops> did your household plant and harvest a 
second time in the past 12 months (August 2019 to July 2020)?
second planting (acres) [If S1b is two 
or three times]
S2b How many acres of these crops did your household plant and harvest a 
third time in the past 12 months (August 2019 to July 2020)?
third planting (acres) [If S1b is 
three times]
MSIKA Final Evaluation Page 112
kadale@africa-online.net
C. Growing Practices
[base crop relevance on question S2]
Tomatoes (A) 
Q# Question Response Logic
C1_1 BLANK
C1_2 Which of the following improved practices did you use for your TOMATOES in the 
past 12 months (August 2019 to July 2020)? 
[Read out list] 
a. Composting (use of rotted down plant and other organic matter) 
Yes
No
b. Manuring (use of animal urine or excrement) 
c. Ridging (banking up the soil to form a ridge for planting on) 
d. Mulching (cover soil with dead plant/ compost to keep moisture/ 
suppress weed growth and improve soil fertility) 
e. Crop rotation (grow different crops on same plot in successive 
seasons) to control pests and diseases and improve soil fertility
f. Minimum tillage (making planting holes only rather than tilling all 
the soil) 
g. Blank
h. Planting seeds in a nursery before planting out
i. Staking (adding stakes to enable plants to stay upright) 
j. Succession planting (plant part of plot one week, then other parts 
in following weeks to spread harvesting) 
k. Blank
l. Using irrigation (application of water to crops using motorized or 
solar pumps, treadle pumps, watering cans or buckets, river diversion 
or canalization)
m. Soil and water conservation (terracing, vetiver grass) 
n. Draining excess water 
o. Testing soil acidity 
p. Adding lime or ash to soil before planting to reduce acidity 
q. Choosing the variety (choosing different varieties for the growing 
characteristics and market that aiming for) 
r. De-suckering or removing unwanted side-shoots 
s. Spraying for pests and disease 
t. Using recommended plant and ridge spacing 
u. Using recommended fertilizer and application rates 
v. Sterilizing nursery beds before planting (burning crop residues to 
kill weed seeds, disease spores and pest eggs)
w. Scouting for pests and diseases
x. Uprooting infected plants and burning (to avoid spread of disease)
y. Sowing seed in row/groove nursery 
z. Using fish soup / sugar solution as bait for insects
aa. Planting in sunken beds 
bb. Hardening off before planting out (reduction of water to nursery 
seedlings)
cc. Selecting the best seedlings for planting out
dd. Intercropping - growing more than one type of crops (combining 
legumes with non-legume crops) on the same piece of land at the same 
time to improve soil fertility through nitrogen fixation, keep 
moisture, control soil erosion and suppress pest populations
C1_4 On how many acres of TOMATOES have you applied any one or more 
of these improved practices in the past 12 months (August 2019 
to July 2020)?
[Acres]
[Cross 
check 
to S2]
The first planting and harvesting time 
The second planting and harvesting time
The third planting and harvesting time
MSIKA Final Evaluation Page 113
kadale@africa-online.net
C1_5 Have you used this improved practice for the first time in the
last 12 months? (August 2019 to July 2020)
[List from C1_2]
Yes
No
C1_7 Overall, what are the reasons why 
you did not use the other 
improved TOMATO practices?
[Do not read responses]
[enumerator classifies]
[multiple response possible]
Did not know it
Expensive to use it
Time consuming
Inefficient
Not confident it will work
Not confident I can use it properly
Too difficult for farmers to do it
Did not understand it
Inadequate resources
Need did not arise
Other (specify__________)
If 
C1_2= 
no
Onions (B)
Q# Question Response Logic
C2_1 BLANK
C2_2 Which of the following improved practices did you use for your ONIONS in 
the past 12 months (August 2019 to July 2020)? 
[Read out list]
a. Composting (use of rotted down plant and other organic 
matter) 
Yes
No
b. Manuring (use of animal urine or excrement) 
c. Ridging (banking up the soil to form a ridge for planting 
on) 
d. Earthing up (adding soil on growing plants to stimulate 
tuber development) 
e. Mulching (cover soil with dead plant/ compost to keep 
moisture/ suppress weed growth and improve soil fertility
f. Crop rotation (grow different crops on same plot in
successive seasons) to control pests and diseases and improve 
soil fertility 
g. Minimum tillage (making planting holes only rather than 
tilling all the soil,) 
h. Blank
i. Planting seeds in a nursery before planting out
j. Blank
k. Using irrigation (application of water to crops using 
motorized or solar pumps, treadle pumps, watering cans or 
buckets, river diversion or canalization)
l. Soil and water conservation (terracing, plant vetiver 
grass) 
m. Draining excess water 
n. Testing soil acidity 
o. Adding lime or ash to soil before planting to reduce 
acidity 
p. Choosing the variety (choosing different varieties for the 
growing characteristics and market that aiming for) 
q. Spraying for pests and disease 
r. Clipping plant ends to allow seedlings make fresh sprouts
s. Sterilizing nursery beds before planting (burning crop 
residues to kill weed seeds, disease spores and pest eggs)
t. Scouting for pests and diseases
u. Using recommended plant and ridge/row spacing 
v. Using recommended fertilizer and application rates 
w. Uprooting infected plants and burning them (to avoid spread 
of disease)
x. Sowing seed in row/groove nursery 
MSIKA Final Evaluation Page 114
kadale@africa-online.net
Q# Question Response Logic
y. Using raised beds (rainy season)
z. Using fish soup / sugar solution as bait for insects
aa. Planting in sunken beds (dry season)
bb. Hardening off before planting out (reduction of water to 
nursery seedlings)
cc. Selecting the best seedlings for planting out
dd. Intercropping - growing more than one type of crops 
(combining legumes with non-legume crops) on the same piece of 
land at the same time to improve soil fertility through 
nitrogen fixation, keep moisture, control soil erosion and 
suppress pest populations
C2_5 Have you used this improved practice for the first time in the last 12 
months? (August 2019- July 2020)
[List from C2_2]
Yes
No
C2_7 Overall, what are the reasons why you 
did not use the other improved ONION 
practices?
[Do not read responses]
[enumerator classifies]
[multiple response possible]
Did not know it
Expensive to use it
Time consuming
Inefficient
Not confident it will work
Not confident I can use it 
properly
Too difficult for farmers to do it
Did not understand it
Inadequate resources
Need did not arise
Other
If C2_2=no
Irish Potatoes (C)
Q# Question Response Logic
C3_1 BLANK
C3_2 Which of the following improved practices did you use for your IRISH POTATOES in the 
past 12 months (August 2019 to July 2020)? 
[Read out list]
a. Composting (use of rotted down plant and other organic matter) 
Yes
No
b. Manuring (use of animal urine or excrement) 
c. Ridging (banking up the soil to form a ridge for planting on) 
d. Earthing up (adding soil on growing plants to stimulate tuber 
development)
e. Mulching (cover soil with dead plant/ compost to keep moisture/ 
suppress weed growth and improve soil fertility
f. Crop rotation (grow different crops on same plot in successive 
seasons) to control pests and diseases and improve soil fertility
g. Blank
h. Succession planting (plant part of plot one week, then other parts 
later) 
i. Blank
j. Blank
k. Using irrigation (application of water to crops using motorized or 
solar pumps, treadle pumps, watering cans or buckets, river diversion or 
canalization)
l. Soil and water conservation (terracing, plant vetiver grass) 
m. Choosing the variety (choosing different varieties for the growing 
characteristics and market that aiming for) 
C2_4 On how many acres of ONIONS have you applied these improved 
practices in the past 12 months (August 2019 to July 2020)?
[Acres]
[Cross 
check to 
S2]
The first planting time
The second planting time
The third planting time
MSIKA Final Evaluation Page 115
kadale@africa-online.net
Q# Question Response Logic
n. Spraying for pests and disease 
o. Scouting for pests and diseases
p. Using recommended plant and ridge spacing 
q. Using recommended fertilizer and application rates 
r. Uprooting infected plants and burning (to avoid spread of disease) 
s. Using fish soup / sugar solution as bait for insects
t. Selection of seed potatoes
u. Chitting (using diffused storage light to initiate sprouting of 
tubers)
v. Intercropping - growing more than one type of crops (combining legumes 
with non-legume crops) on the same piece of land at the same time to 
improve soil fertility through nitrogen fixation, keep moisture, control 
soil erosion and suppress pest populations
C3_5 Have you used this improved practice for the first time in the last 12 months? 
(August 2019- July 2020)
[List from C3_2]
Yes
No
C3_7
Overall, what are the reasons why you did not 
use the other IRISH POTATO improved
practices?
[Do not read responses – enumerator 
classifies]
Did not know it
Expensive to use it
Time consuming
Inefficient
Not confident it will work
Not confident I can use it properly
Too difficult for farmers to do it
Did not understand it
Inadequate resources
Need did not arise
Other
If 
C3_2= 
no
Mangoes (D)
Q# Question Response Logic
C4_1 BLANK
C4_2 Which of the following improved practices did you use for your MANGOES in the 
past 12 months (August 2019 to July 2020)? 
[Read out list]
a. Composting (use of rotted down plant and other organic matter) 
Yes
No
b. Manuring (use of animal urine or excrement) 
c. Blank
d. Pruning (cut back the canopy to let more light& air in; & to stop 
trees growing too tall to harvest from) 
e. Mulching (cover soil with dead plant/ compost to keep moisture/ 
suppress weed growth and improve soil fertility
f. Spraying for pests and disease (use of crop protection chemicals) 
g. Blank
h. Water capture (creating a basin around fruit trees)
i. Using irrigation (application of water to crops using motorized or 
solar pumps, treadle pumps, watering cans or buckets, river diversion 
or canalization)
j. Soil and water conservation (terracing, plant vetiver grass) 
k. Draining excess water 
l. Testing soil acidity 
m. Adding lime or ash to soil before planting to reduce acidity 
n. Blank
C3_4 Excluding ridging, on how many acres of IRISH POTATOES have you 
applied these farming practices in the last 12 months? (August 
2019 to July 2020)
[Acres]
[Cross 
check 
to S2]
The first planting time
The second planting time
The third planting season
MSIKA Final Evaluation Page 116
kadale@africa-online.net
Q# Question Response Logic
o. Blank
p. Blank
q. Scouting for pests and diseases
r. Blank
s. Uprooting infected plants and burning (to avoid spread of disease) 
t. Using fish soup / sugar solution as bait for insects
u. Grafting (using stems of plants)
v. Budding (using plant buds)
w. Staking young trees
x. Digging of planting holes prior to planting
y. Deflowering of first flowers
C4_7 Overall, what are the reasons why you 
did not use the other improved MANGO
practices?
[Do not read responses – enumerator 
classifies]
Did not know it
Expensive to use it
Time consuming
Inefficient
Not confident it will work
Not confident I can use it 
properly
Too difficult for farmers to 
do it
Did not understand it
Inadequate resources
Need did not arise 
Other
If 
C4_2= 
no
D. Harvest and Post-Harvest Handling Practices
[base crop relevance on question S2]
Tomatoes (A)
Q# Question Response Logic
D1_2 Which improved harvest & post-harvest practices did you use for handling 
TOMATOES the last 12 months (August 2019 to July 2020)? 
[Read list]
a. Timely harvesting not too early so immature and not too 
late so over-ripe
Yes
No 
b. Handling carefully to avoid damage
c. Storing in cool places out of the sun prior to sale
d. Grading by color, size and shape
j. Sorting by variety, size/maturity, damaged, ripeness
e. Packing - Loading into appropriate containers for storage 
or transportation
f. Not over-filling baskets & bags or using very big bags, as 
this damages product at the bottom
g. BLANK
h. Not stacking more than three containers high to avoid 
crushing tomatoes in lower containers.
i. Adding soft dry grass as soft litter
D1_2b Have you used this improved practice for the first time in the last 
12 months? (August 2019 to July 2020)
[List from D1_2]
Yes
No
C4_4 On how many MANGO trees have you applied at least one of these 
improved practices in the past 12 months? (August 2019- July 
2020)
[Trees] 
[Cross 
check 
to S2]
C4_5 Have you used this practice for the first time in the last 12 
months? (August 2019- July 2020)
[list from C4_2]
Yes
No
MSIKA Final Evaluation Page 117
kadale@africa-online.net
Onions (B)
Q# Question Response Logic
D2_2 Which improved harvest & post-harvest practices did you use for handling ONIONS 
in the last 12 months (August 2019 – July 2020)? 
[Read list]
a. Harvesting & selling green/fresh onions with stalks on
Yes
No
b. Lifting when at least 20% of tops have bent over & dried; leave 
until outer is dry & brown (cured)
Drying (curing) onions in the sun or under a shade before storing to 
improve in keeping quality
d. Trimming dried stems and roots before packing
e. Grading by variety, size and appearance
f. Packing - Loading commodities into appropriate bags/containers for 
storage or transportation
g. Not over-filling baskets & bags or using very big bags, as this 
damages product at the bottom
h. Sorting by variety, dryness, removing damaged etc
D2_2b Have you used this improved practice for the first time in the last 
12 months? (August 2019 to July 2020)
[List from D2_2]
Yes 
No
Irish Potatoes (C)
Q# Question Response Logic
D3_2 Which improved harvest & post-harvest practices did you use for 
handling IRISH POTATOES the last 12 months (August 2019 to July 2020?
[Read list]
a. Harvesting when plant tops wilt and start to wither; 
slashing tops close to soil to cure in soil a day before 
harvesting
Yes
No
b. Lifting potatoes with a fork/prong, not a hoe
c. Rubbing off surface dirt after harvest; or after drying 
and curing
d. Blank
e. Keeping clean potatoes in a cool dry place to cure or 
heal surface damage, before packing, storage & selling
f. Handling carefully to avoid damage – remove all damaged 
potatoes to avoid rotting
g. Grading by color, size and shape
k. Sorting by variety, maturity, removing damages, other 
materials etc
h. Packing - Loading commodities into appropriate 
bags/containers for storage or transportation
i. Using night vented sheds for short term storage for 
selling outside main season
D3_2b Have you used this improved practice for the first time in 
the last 12 months? (August 2019 to July 2020)
[list from D3_2]
Yes 
No
Mangoes (D)
Q# Question Response Logic
D4_2 Which improved harvest & post-harvest practices did you use for 
handling MANGOES the last 12 months (August 2019 to July 2020)? 
[Read list]
a. Harvesting when ripening just started and color begins 
to change
Yes
No
b. Harvesting by climbing the tree
c. Harvesting by using a special harvesting pole
d. Catching fruit before it falls using bags or spreading
sheets
e. Handling carefully to avoid damage
MSIKA Final Evaluation Page 118
kadale@africa-online.net
Q# Question Response Logic
f. Storing in cool places and out of the sun prior to sale
g. Grading by color, size, and shape
l. Sorting by variety, size/maturity, ripeness, removing 
damaged 
h. Packing - Loading commodities into appropriate 
containers for storage or transportation
i. Storing in crates to prevent damage in transporting
j. Washing to improve appearance
k. Not over-filling baskets & bags or using very big bags 
as this damages product at the bottom
D4_2b Have you used this improved practice for the first time in 
the last 12 months? (August 2019 to July 2020)
[List from D4_2]
Yes 
No
Q# Question Response Logic
D8 Where do you store this <crop>?
[Read responses]
[select all that apply]
Inside my house
In a separate 
building/store/shed
In a Nkhokwe/outside 
store
In baskets/bags outside 
or in a pile
FBO Warehouse
Other
D8_b Did you build any storage building or 
storeroom or shed in the last 12 months?
Yes
No
D8_bb Did you restore any storage building or 
storeroom or shed in the last 12 months?
Yes
No
If Yes
D9a What is new building/store/shed’s floor 
area? [length x width in square meters] 
[1 large step = 1 pace = 1 meter]
If D8_b= Yes
D9_b What is the new building/store/shed’s 
height? [Meters] 
[1 large step = 1 pace = 1 meter]
If D8_b= Yes
D9_c What was the total amount you spent on the 
new building/store/shed for storing crop?
[MK]
If D8_b= Yes
D10_a What is the restored building/store/shed’s 
floor area? [length x width in square 
meters] 
[1 large step = 1 pace = 1 meter]
If D8_bb= Yes
D10_b What is the restored building/store/shed’s 
height? [Meters] 
[1 large step = 1 pace = 1 meter]
If D8_bb= Yes
D10_c What was the total amount you spent on 
restoring a building/store/shed for storing 
crop?
[MK]
If D8_bb= Yes
F. Production & Sales
[base V.C. relevance on question S2]
Tomatoes (A)
Q# Question Response Logic
MSIKA Final Evaluation Page 119
kadale@africa-online.net
F1_1a In the past 12 months (August 2019 to July 2020), what volume/Weight of 
TOMATOES did your household harvest in the first harvest?
[20 litre pail has the same volume as ndowa]
[other farmers use 40 litre or 60 litre volume pails for onions and 
tomatoes, convert to kgs based on 20 litre pail]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_2a From August 2019 to July 2020, what quantity did you sell of TOMATOES from 
your first harvest? 
[Ask for first harvest in the last 12 months]
[<= 
production]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_3a From August 2019 to July 2020, what volume/weight of your first TOMATO 
harvest was spoiled 
[Ask for first harvest in the last 12 months]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_5a What was the common/typical SALES price for TOMATOES in the 
first harvest?
[Ask for first harvest in the last 12 months]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_1b In the past 12 months (August 2019 to July 2020), what volume/Weight of 
TOMATOES did your household harvest in the second harvest?
[20 litre pail has the same volume as ndowa]
[other farmers use 40 litre or 60 litre volume pails for onions and tomatoes, 
convert to kgs based on 20 litre pail]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_2b From August 2019 to July 2020, what quantity did you sell of TOMATOES from 
your second harvest? 
[Ask for second harvest in the last 12 months]
[<= 
production]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_3b From August 2019 to July 2020, what volume/weight of your second TOMATO 
harvest was spoiled?
[Ask for second harvest in the last 12 months]
 
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_5b What was the common/typical SALES price for TOMATOES in the second harvest?
[Ask for second harvest in the last 12 months]
# of 40Lt Basins (dish) (weight=32kg) 
# of 20lt Pails (weight=16kg)
# of 60lt Pails (Weight=48kg)
# of kgs
MSIKA Final Evaluation Page 120
kadale@africa-online.net
F1_1c In the past 12 months (August 2019 to July 2020), what 
volume/Weight of TOMATOES did your household harvest in the third 
harvest?
[20 litre pail has the same volume as ndowa]
[other farmers use 40 litre or 60 litre volume pails for onions 
and tomatoes, convert to kgs based on 20 litre pail]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_2c From August 2019 to July 2020, what quantity did you sell of
TOMATOES from your third harvest? 
[Ask for third harvest in the last 12 months]
[<= 
production]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_3c From August 2019 to July 2020 what volume/weight of your third 
TOMATO harvest was spoiled?
[Ask for third harvest in the last 12 months]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_5c What was the common/typical SALES price for TOMATOES in the third 
harvest?
[Ask for third harvest in the last 12 months]
# of 20lt Basins (dish) (weight=16kg) 
# of 40lt Pails (weight=32kg)
# of 60lt Pails (Weight=48kg)
# of kgs
F1_4 Why were the TOMATOES spoiled/lost? 
[Select all that apply]
Damaged when harvesting
Damaged at the farm in storage
Damaged during transport to market
Damaged at the market
Could not be sold prior to spoiling
DNK
If F1_3 or 
F1_3c or 
F1_3c>0
F1_4b Why did you not sell any TOMATOES? No surplus to sell
Could not get a buyer
Prices were too poor
Quality of my crop was too poor
Other 
DNK 
If F1_2 = 
0 
F1_6 BLANK 
F1_7 Did you primarily sell TOMATOES:
[Read responses]
With farmers in my club/group
With farmers not in my club/group
On my own as an individual
Other 
DNK 
Only if 
Sales > 0
F1_8 Did you sell TOMATOES at any of the 
following locations?
[Read responses]
[select all that apply]
Neighbors
Local market 
Traders who came to me
Traders that I delivered to
Distant market
Direct to Processor/large buyer
NGO
My Farmer group 
Other 
Only if 
Sales > 0 
F1_10 How was the weather for growing 
TOMATOES in the past 12 months (1st 
August 2019 to 31st July)?
Favorable
Moderate
Bad
DNK
MSIKA Final Evaluation Page 121
kadale@africa-online.net
Onions (B)
Q# Question Response Logic
F2_1a In the past 12 months (August 2019 to July 2020), what
volume/Weight of ONIONS did your household harvest in the first 
harvest?
[20 litre pail has the same volume as ndowa]
[other farmers use 40 litre or 60 litre volume pails for onions and 
tomatoes, convert to kgs based on 20 litre pail]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_2a From August 2019 to July 2020, what quantity did you sell of ONIONS 
from your first harvest? 
[Ask for first harvest in the last 12 months]
[<= 
production]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_3a From August 2019 to July 2020, what volume/weight of your first 
ONIONS harvest was spoiled 
[Ask for first harvest in the last 12 months]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_5a What was the common/typical SALES price for ONIONS in 
the first harvest?
[Ask for first harvest in the last 12 months]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_1b In the past 12 months (August 2019 to July 2020), what 
volume/Weight of ONIONS did your household harvest in the second 
harvest?
[20 litre pail has the same volume as ndowa]
[other farmers use 40 litre or 60 litre volume pails for onions and 
tomatoes, convert to kgs based on 20 litre pail]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_2b From August 2019 to July 2020, what quantity did you sell of ONIONS 
from your second harvest? 
[Ask for second harvest in the last 12 months]
[<= 
production]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_3b From August 2019 to July 2020, what volume/weight of your second 
ONIONS harvest was spoiled?
[Ask for second harvest in the last 12 months]
# of 20lt Pails (weight=22kg) 
MSIKA Final Evaluation Page 122
kadale@africa-online.net
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_5b What was the common/typical SALES price for ONIONS in the second 
harvest?
[Ask for second harvest in the last 12 months]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_1c In the past 12 months (August 2019 to July 2020), what 
volume/Weight of ONIONS did your household harvest in the third 
harvest?
[20 litre pail has the same volume as ndowa]
[other farmers use 40 litre or 60 litre volume pails for onions and 
tomatoes, convert to kgs based on 20 litre pail]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_2c From August 2019 to July 2020, what quantity did you sell of ONIONS 
from your third harvest? 
[Ask for third harvest in the last 12 months]
[<= 
production]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_3c From August 2019 to July 2020, what volume/weight of your third 
ONIONS harvest was spoiled?
[Ask for third harvest in the last 12 months]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_5c What was the common/typical SALES price for ONIONS in the third 
harvest?
[Ask for third harvest in the last 12 months]
# of 20lt Pails (weight=22kg) 
# of 40lt Pails (weight=44kg)
# of 50 kgs bags (Weight=55kg)
# of oxcarts (Weight= 660kgs)
# of kgs
F2_4 Why were the ONIONS spoiled/lost? 
[Select all that apply]
Damaged when harvesting
Damaged at the farm in storage
Damaged during transport to 
market
Damaged at the market
Could not be sold prior to 
spoiling
DNK
F2_4b Why did you not sell any ONIONS? No surplus to sell
Could not get a buyer
Prices were too poor
Quality of my crop was too poor
Other 
DNK 
If F2_2
= 0
MSIKA Final Evaluation Page 123
kadale@africa-online.net
F2_6 BLANK
F2_7 Did you primarily sell ONIONS:
[READ RESPONSES]
With farmers in my club/group
With farmers not in my 
club/group
On my own as an individual
Other 
DNK (do not read out)
Only if 
Sales > 
0 
F2_8 Did you sell ONIONS at any of the 
following locations?
[Read responses]
[select all that apply]
Neighbors
Local market
Traders who came to me
Traders that I delivered to
Distant market
Direct to Processor/large buyer
NGO
Ny farmer group 
Other 
Only if 
Sales > 
0 
F2_10 How was the weather for growing ONIONS 
in the past 12 months (August 2019 to 
July 2020)?
Favorable
Moderate
Bad
DNK
Irish Potatoes (C)
Q# Question Response Logic
F3_1a In the past 12 months (August 2019 to July 2020), what 
volume/Weight of IRISH POTATOES did your household harvest in 
the first harvest?
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_2a From August 2019 to July 2020, what quantity did you sell of 
IRISH POTATOES from your first harvest? 
[Ask for first harvest in the last 12 months]
[<= 
production]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_3a From August 2019 to July 2020, what volume/weight of your 
first IRISH POTATOES harvest was spoiled 
[Ask for first harvest in the last 12 months]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_5a What was the common/typical SALES price for IRISH 
POTATOES in the first harvest?
[Ask for first harvest in the last 12 months]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_1b In the past 12 months (August 2019 to July 2020), what volume/Weight 
of IRISH POTATOES did your household harvest in the second harvest?
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
MSIKA Final Evaluation Page 124
kadale@africa-online.net
# of oxcarts (Weight= 840kgs)
# of kgs
F3_2b From August 2019 to July 2020, what quantity did you sell of IRISH 
POTATOES from your second harvest? 
[Ask for second harvest in the last 12 months]
[<= 
production]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_3b From August 2019 to July 2020, what volume/weight of your second 
IRISH POTATOES harvest was spoiled?
[Ask for second harvest in the last 12 months]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_5b What was the common/typical SALES price for IRISH POTATOES in the 
second harvest?
[Ask for second harvest in the last 12 months]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_1c In the past 12 months (August 2019 to July 2020), what volume/Weight 
of IRISH POTATOES did your household harvest in the third harvest?
# of 20lt Pails (weight=19kg)
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_2c From August 2019 to July 2020, what quantity did you sell of IRISH 
POTATOES from your third harvest? 
[Ask for third harvest in the last 12 months]
[<= 
production]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_3c From August 2019 to July 2020, what volume/weight of your third 
IRISH POTATOES harvest was spoiled?
[Ask for third harvest in the last 12 months]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
F3_5c What was the common/typical SALES price for IRISH POTATOES in the 
third harvest?
[Ask for third harvest in the last 12 months]
# of 20lt Pails (weight=19kg) 
# of 40lt Pails (weight=38kg)
# of 50 kgs bags (Weight=70kg)
# of oxcarts (Weight= 840kgs)
# of kgs
MSIKA Final Evaluation Page 125
kadale@africa-online.net
F3_4 Why were the IRISH POTATOES 
spoiled/lost? 
[Select all that apply]
Damaged when harvesting
Damaged at the farm in 
storage
Damaged during transport to 
market
Damaged at the market
Could not be sold prior to 
spoiling
DNK
F3_4b Why did you not sell any IRISH 
POTATOES? 
No surplus to sell
Could not get a buyer
Prices were too poor
Quality of my crop was too 
poor
Other 
DNK 
If F3_2 = 
0 
F3_6 BLANK
F3_7 Did you primarily sell IRISH
POTATOES:
[Read responses]
With farmers in my 
club/group
With farmers not in my 
club/group
On my own as an individual
Other 
DNK (do not read out)
Only if 
Sales > 0
F3_8 Did you sell IRISH POTATOES at any of 
the following locations?
[Read responses] 
[select all that apply]
Neighbors
Local market
Traders who came to me
Traders that I delivered to
Distant market
Direct to Processor/large 
buyer
NGO
My Farmer group 
Other 
Only if 
Sales > 0
F3_10 How was the weather for growing IRISH 
POTATOES in the past 12 months 
(August 2019 to July 2020)?
Favorable
Moderate 
Bad
DNK
Mangoes (D)
Q# Question Response Logic
F4_1 What volume/Weight of MANGOES did your household harvest in the past 
12 months (August 2019 to July 2020)?
[Do not double count]
# of 15lt Pails (weight=13.9kg) 
# of 20lt Pails (weight=16.3kg)
# of 40lt Pails (weight=32.6kg)
# of kgs
F4_2 From August 2019 to July 2020, what quantity did you sell of 
MANGOES? 
[<= 
production]
# of 15lt Pails (weight=13.9kg) 
# of 20lt Pails (weight=16.3kg)
# of 40lt Pails (weight=32.6kg)
# of kgs
F4_3 From August 2019 to July 2020, what volume/weight of MANGOES was 
spoiled? 
# of 15lt Pails (weight=13.9kg) 
# of 20lt Pails (weight=16.3kg)
# of 40lt Pails (weight=32.6kg)
# of kgs
MSIKA Final Evaluation Page 126
kadale@africa-online.net
F4_5 From August 2019 to July 2020, what was the common/typical SALES price 
for MANGOES?
# of 15lt Pails (weight=13.9kg) 
# of 20lt Pails (weight=16.3kg)
# of 40lt Pails (weight=32.6kg)
# of kgs
F4_4 Why were the MANGOES spoiled/lost? 
[Select all that apply]
Damaged when harvesting
Damaged at the farm in storage
Damaged during transport to market
Damaged at the market
Could not be sold prior to spoiling
DNK
F4_4b Why did you not sell any MANGOES? No surplus to sell
Could not get a buyer
Prices were too poor
Quality of my crop was too poor
Other 
DNK 
If F4_2 = 
0 
F4_6 BLANK
F4_7 Did you primarily sell MANGOES with:
[Read responses]
Farmers in my club/group
Farmers not in my club/group
On my own as an individual
Other 
DNK (do not read out)
Only if 
Sales > 0
F4_8 Did you sell MANGOES at any of the 
following locations?
[Read responses]
[select all that apply]
Neighbors
Local market
Traders who came to me
Traders that I delivered to
Distant market
Direct to Processor/large buyer
NGO
My Farmer group 
Other 
Only if 
Sales > 0
F4_10 How was the weather for growing 
MANGOES in the past 12 months (August 
2019 to July 2020)?
Favorable
Moderate
Bad
DNK
H. Farm Management
[base crop relevance on question S2]
Q# Question Response Logic
H1 BLANK
H2 Have you used these management practices in the last 12 months? 
[READ PRACTICES]
a. Keep good farm records Yes – In all ways
Yes – In some 
ways
No 
DNK
b. (blank)
c. Sell together with other farmers
d. Do costings for growing crops
e. Calculate profits after selling
f. Plan production for upcoming season 
g. Plan production for a specific market
h. Understand the specifications for specific buyers
i. Separating commodities into the different grades
j. Timely buying of inputs
H3 BLANK
H4 Did you employ or hire anyone for more than four 
weeks in the last 12 months (August 2019-July 2020)? 
Yes
No
H5 How many men did you employ or hire? If H4=Yes
MSIKA Final Evaluation Page 127
kadale@africa-online.net
H6 How many women did you employ or hire? If H4=Yes
H7 How many weeks in the last 12 months (August 2019-
July 2020) did you employ or hire men
If H4=Yes
H8 How many weeks in the last 12 months (August 2019-
July 2020) did you employ or hire women
If H4=Yes
H9 –
H13
BLANK
H14 Did you buy any equipment in the last 12 months? Yes
No
H14b Did you buy any land in the last 12 months? Yes
No
H15 What was the total amount you spent on equipment 
in the last 12 months? 
If yes 
in H14
H15b What was the total amount you spent on land in the 
last 12 months?
If yes 
in H14b
I. Information
[base crop relevance on question S2]
Q# Question Responses Logic
I2 Where do you get information about the following:
[Read list] 
[Enumerator to classify responses]
[Select all that apply]
Neighbor/friend
Govt ext’n officer
Land O’Lakes/MSIKA extension 
worker
Land O’Lakes/MSIKA lead farmer
Private buyer ext’n officer 
NGO/Project ext’n officer 
Vendors/traders 
Radio
SMS 
FBO Leadership
Fellow FBO member 
Posters 
Agricultural shows
Trade fairs
Other sources 
Did not get this info 
a. Locations of markets to sell your crops
b. Names of possible buyers
c. Where to buy inputs
d. Where to access to finance
e. Buyers’ product requirements (volumes, 
grades, quality etc)
f. Weather which can affect your crops
g. Market prices for your crops
J. COVID Information
Tomatoes (A)
J1_1a In what ways has COVID-19 
affected your TOMATO farming and 
sales?
[Unprompted]
[Multiple response possible]
[all responses have to be 
related to COVID, not just 
problems they faced which might 
be from other causes. Probe to 
check the response was due to 
COVID not some other factor – if 
it was from COVID and another 
factor, it should be recorded.
Have never heard about Corona 
virus
Not able to get inputs for the 
crop 
Not able to go to the fields to 
plant /look after the crop 
Not able to work because I was 
sick
Not able to hire labour
Not able to harvest
Not able to go to the market
Not able to sell at the market
Other (Specify)
J1_1b To what extent has COVID-19 
affected your TOMATO farming?
[Single response – Prompted]
A lot
Moderately
Not a lot
Not at all
DNK/NR
If J1_1a 
is not 
option a
MSIKA Final Evaluation Page 128
kadale@africa-online.net
Onions (B)
J2_1a In what ways has COVID affected 
your ONION farming and sales?
[Unprompted]
[Multiple response possible]
[all responses have to be 
related to COVID, not just 
problems they faced which might 
be from other causes. Probe to 
check the response was due to 
COVID not some other factor – if 
it was from COVID and another 
factor, it should be recorded.
Have never heard about Corona 
virus
Not able to get inputs for the 
crop
Not able to go to the fields to 
plant /look after the crop
Not able to work because I was 
sick 
Not able to hire labour
Not able to harvest
Not able to go to the market
Not able to sell at the market
Other (Specify)
J2_1b To what extent has COVID 
affected your ONION farming?
[Single response – Prompted]
A lot
Moderately
Not a lot
Not at all
DNK/NR
If J2_1a 
is not 
option a
Irish Potatoes (C)
J2_1a In what ways has COVID affected 
your IRISH POTATO farming and 
sales?
[Unprompted]
[Multiple response possible]
[all responses have to be 
related to COVID, not just 
problems they faced which might 
be from other causes. Probe to 
check the response was due to 
COVID not some other factor – if 
it was from COVID and another
factor, it should be recorded.
Have never heard about Corona 
virus
Not able to get inputs for the 
crop
Not able to go to the fields to 
plant /look after the crop 
Not able to work because I was 
sick
Not able to hire labour
Not able to harvest
Not able to go to the market
Not able to sell at the market
Other (Specify)
J3_1b To what extent has COVID 
affected your IRISH POTATO 
farming?
[Single response – Prompted]
A lot
Moderately
Not a lot
Not at all
DNK/NR
If J3_1a 
is not 
option a
Mangoes (D)
J4_1a In what ways has COVID affected 
your MANGO farming and sales?
[Unprompted]
[Multiple response possible]
[all responses have to be 
related to COVID, not just 
problems they faced which might 
be from other causes. Probe to 
check the response was due to 
COVID not some other factor – if 
it was from COVID and another 
factor, it should be recorded.
Have never heard about Corona 
virus
Not able to get inputs for the 
crop
Not able to go to the fields to 
plant /look after the crop
Not able to work because I was 
sick
Not able to hire labour
Not able to harvest
Not able to go to the market
Not able to sell at the market
Other (Specify)
J4_1b To what extent has COVID 
affected your MANGO farming?
[Single response – Prompted]
A lot
Moderately
Not a lot
Not at all
DNK/NR
If J4_1a 
is not 
option a
MSIKA Final Evaluation Page 129
kadale@africa-online.net
Thank you for your time.
MSIKA Final Evaluation Page 130
kadale@africa-online.net
Qualitative Interview Guide – Farmer Beneficiaries
Name of Interviewer: __________________________ Date held:______________ 2020
Name of Farmer: _____________________________ FBO name: __________________________ 
District:___________________ GVH: __________________ 
EPA:_____________________________ Village: ______________________ 
Respondent is (circle one): Male 01 Female 02
Crop focus (circle one): Tomato Onion Irish potato Mango 
Other target crops grown (circle all that apply): Tomato Onion Irish potato Mango 
(Focus only on the main crop that the person is growing). 
(Check that the respondent is the named person that was selected not a spouse or a proxy, 
otherwise they cannot contribute).
A. Introduction and Consent:
“Hello, my name is <name> and I am working with Kadale Consultants. Kadale has been asked to talk
to fruit and vegetable farmers by MSIKA project, which is run by Land O’Lakes, a US organization that 
works with farmers. Your name has been given to us as someone who have been trained by MSIKA 
project.
We would like to discuss your experience as a <CROP> grower and about the training you received 
from Land O’ Lakes’ MSIKA project. 
The discussion will take about 45 minutes. What you tell us will not be attributed to any of you. Anything 
said by any of you should not be shared outside of the research team. 
Are you willing to take part in the discussion?” (If any say no, ask why and go to the next person) 
Dzina langa ndi [DZINA] ndimagwira ntchito ndi bungwe la Kadale Consultants, ku Lilongwe. Ife 
a Kadale Consultants tapemphedwa ndi a Land O’ Lakes kuti tilankhulane ndi alimi a mbeu za 
masamba komanso zipatso omwe bungweli limagwira nawo ntchito kudzera m’project yawo ya 
MSIKA. Atipatsa maina monga alimi omwe aphunzitsidwa ndi project ya MSIKA. 
Mukucheza kwathu, tikambirana nanu momwe ulimi wanu wa <crop> wakhala ukuyendera 
ndinso maphunziro omwe munalandira kuchokera ku project ya MSIKAyi.
Kucheza kwathu kotenga ola limodzi ndi theka (1 hour 30 minutes). Zomwe tikambirane pano 
sidzidzazindikirika mwanjira ina iliyonse kuti mwanena ndiinuyo ndipo sidzidzamvedwa ndi ena 
omwe sali mbali ya kagulu komwe kakuyendetsa kafukufukuyu.
Muli omasuka kutenga nao mbali pakafukufukuyu? (If any say no, ask why and go to the next 
person)
Reason(s) for not giving consent: 
_________________________________________________________________________________
_________________________________________________________________________________
__________________ 
B. Background Information (to get the person talking)
1. How long have you been growing <the crop>? Were you growing this before being trained by 
LoL/MSIKA?
Mwakhala mukulima mbeu imeneyi kwa nthawi yaitali bwanji? Munkalima mbeu imeneyi 
musanaphunzitsidwe ndi Land O’ Lakes/MSIKA?
MSIKA Final Evaluation Page 131
kadale@africa-online.net
C. Agricultural production training from MSIKA (nb. This can include training by MSIKA directly 
or by Lead Farmers trained by MSIKA. Post-Harvest Handling training is covered in the next
section, so come back to that after discussing agricultural production training)
1. What agricultural production training did you receive from LOL /MSIKA, including by Lead Farmers 
trained by LOL/ MSIKA? What did you like about the training? How could it be improved?
Kodi kuchokera ku Land O’ Lakes/MSIKA komanso ma Lead Farmer ake, munalandira 
maphunziro otani okhudza kasamalidwe ka mbeu? Ndi chiyani chomwe chinakusangalatsani 
pa maphunziro amenewo? Mukuona kuti zingasinthidwe bwanji kuti zikhale bwino kuposerapo?
2. Which of the agricultural production practices that you learnt from LOL/MSIKA have you been able
to apply/use, starting with the most important ones? 
Ndinjira ziti za ulimi zomwe mwakwaniritsa kugwiritsa ntchito mutaphunzira kuchokera ku Land 
O’ Lakes/MSIKA? Potchula, tsogozani zomwe zili zofunika kuposa zina kwa inuyo
Practices that have been applied
1.
2.
3.
4.
5.
6.
Probe: Why apply these? Why not apply others?
What stops you from using other practices that you learned about? This is a key question so probe 
extensively. 
3. What difference has the agricultural production training by Land O’Lakes/MSIKA made to the 
amount of crop you produce and the quality of your crop, if any? If amount and quality have 
changed, how have they also changed the market pricing of the crops? Why do you think that is? 
Probe: This is trying to see what difference it has made to the amount and the quality of their 
production – ask them to compare with the seasons before they applied the practices, and ideally 
some quantification, even if it is “about half more than before.” 
Kodi maphunziro ochokera ku Land O’ Lakes/MSIKA a kasamalidwe ka mbeu asintha bwanji 
kakololedwe komanso quality ya mbeu yanu? Ngati quality ndi kakololedwe zasintha, kodi 
mitengonso yogulitsira mbeu yasintha? Ngati maphunzirowa asintha zimenezi, mukuganiza 
kuti zili choncho chifukwa chiyani?
4. Have you heard of corona virus or Covid-19? - if say yes, continue. If not, skip to 6.
Kodi munamvapo za mulili wa matenda a COVID-19, kapena kuti Corona Virus?
MSIKA Final Evaluation Page 132
kadale@africa-online.net
5. Did COVID-19 (Corona Virus) affect your crop’s production in any way? 
Kodi mulili wa matenda a COVID-19, kapena kuti Corona Virus, udakhudza ulimi wanu mu njira 
iliyonse?
(Probe: If yes, how did it affect your crop production? What did you do to mitigate? Were the trainings 
that were offered by MSIKA useful in mitigating some of the impacts? Ask respondent to be specific 
about which trainings/techniques)
6. Which practices that you have learnt from MSIKA trainings are you likely to keep practicing beyond 
the end of the project? What will encourage you to continue practicing what you have learnt? What 
do you suggest could be done to have you, and other farmers continue with the practices you have 
learnt from MSIKA? What do you think could be done to have agro-dealers still work with farmers 
beyond the end of the project? 
Ndinjira ziti zaulimi zomwe mwaphunzira kuchokera ku MSIKA mukuona kuti mupitilirabe 
kugwiritsa ntchito ngakhale project itatha? Ndichiyani chomwe chikulimbikitseni kupitirizabe 
zomwe mwaphunzirazo? Mukuganiza kuti chofunika kuchitika ndichiyani kuti inu ndi alimi ena 
mupitirizebe kugwiritsa ntchito njira zomwe mwaphunzira ku MSIKAzi? Mukuganiza kuti 
chofunika kuchitika ndichiyani kuti ma agro-dealers (ogulitsa zipangizo zaulimi) apitirizebe 
kugwira ndi alimi ngakhale project itatha?
(Probe: Ask respondents to be specific about which trainings/techniques? Why do you say that?)
D. Post-harvest handling (If the respondent has not had training in PHH, then skip this section)
1. What training on post-harvest handling did you receive from Land O’Lakes/MSIKA including that 
from Lead Farmers trained by LOL/ MSIKA? What did you like about the approach? How can it be 
improved?
Munalandira maphunziro otani kuchokera ku Land O’ Lakes/MSIKA kuphatikizapo ma Lead 
Farmer ake pa kasamalidwe ka zokolola zanu? Ndi chiyani chinakusangalatsani pa 
kaphunzitsidwe kawo? Mukuona kuti zingasinthidwe bwanji kuti zikhale bwino kuposerapo?
(Note: this includes practices about handling the crop at 1. harvesting, 2. storage on the farm and 3. at 
the market, so ensure they consider practices for all three areas)
2. Which of the harvest and post-harvest handling practices that you learnt from Land O’Lakes/MSIKA 
(list from previous question) have you been able to apply/use, starting with the most important ones? 
Ndi maphunziro ati ochokera ku Land O’ Lakes/MSIKA okhudza kakololedwe ndi kasamalidwe 
ka zokolola omwe mwakwanitsa kugwiritsa ntchito? Potchula, tsogozani zomwe zili zofunika 
kuposa zina kwa inuyo?
Practices that have been applied
1.
2.
3.
4.
5.
MSIKA Final Evaluation Page 133
kadale@africa-online.net
Practices that have been applied
Probe: Why did they apply these? Why not apply others? 
What stops you from using other post-harvest and handling practices that you learned about?
Chimakulepheretsani kukwanitsa kugwiritsa ntchito njira zina zomwe munaphunzirazo ndi 
chiyani?
This is a key question so probe extensively.
3. Have you seen any change in losses at harvest, on the farm or at the market?
Probe: If so, what has been the change? 
Mwaonako kusintha kulikonse pa zokolola zowonongeka, ku munda kapena ku msika? 
E. Markets
1. How do you sell your <target crop>?
Mumagulitsa bwanji mbeu yanu ya <mbeu>?
Probe: Do you sell on your own? With others? Where do you sell your crops? Why do you sell it in this 
way? 
Mumagulitsa pa nokha? Pamodzi ndi anthu ena? Mumagulitsa kuti mbeu zanu? Mumagulitsa mu njira 
imeneyi chifukwa chani?
2. How has Land O’ Lakes/MSIKA better helped you to find markets for your crop? What works well 
and what does not work so well? How could MSIKA’s support to help farmers sell their crops be 
improved?
Kodi a Land o’ Lakes akuthandizani bwanji pa nkhani yopeza misika? Ndi ziti zomwe zikuyenda 
bwino ndipo ndi ziti zomwe sizikuyenda bwino? Kodi chithandizo cha MSIKA chothandiza alimi 
kugulitsa mbeu zawo bwino chingapititsidwe bwanji patsogolo kuposerapo?
Probe this to get different responses from group members and see how many get each type (e.g. market 
identification, market prices, market information).
F. Other observations from the facilitator 
Thank you.
MSIKA Final Evaluation Page 134
kadale@africa-online.net
FBO KII Guide 
Prior to the FBO interview, MSIKA to provide the following info for the selected FBO:
Name of FBO: Year Established:
District: Target crops grown:
EPA: GVH location:
Village location: # of MSIKA farmers:
Chairperson name Contact number:
Secretary name Contact number:
Treasurer name Contact number:
How many sales 
agreements has the
FBO made with the 
help of MSIKA?
2020 ___________
2019 ___________
2018 ___________
2017 ___________
2016 ___________
What value of FBO 
member sales are 
claimed by MSIKA?
2020 – MK _________
2019 – MK _________
2018 – MK _________
2017 – MK _________
2016 – MK _________
Has this FBO been 
trained in improved 
financial and 
organizational 
management?
How many improved 
financial mgt 
practices and systems 
has MSIKA helped the 
FBO to implement?
___________
Has this FBO been 
trained in marketing?
Has this FBO been 
trained in good
governance?
Prior to the FBO interview, MSIKA/LOL to provide the information for columns 1-3:
MSIKA to provide Interviewer to complete
Training module /name When held 
(month/yr)
# of 
members 
trained
Verified by 
interviewer Comments
1. Wholly/partially/ 
not verified
2. Wholly/partially 
/not verified
3. Wholly/partially 
/not verified
4. Wholly/partially 
/not verified
5. Wholly/partially 
/not verified
6. Wholly/partially 
/not verified
MSIKA Final Evaluation Page 135
kadale@africa-online.net
MSIKA to provide Interviewer to complete
Training module /name When held 
(month/yr)
# of 
members 
trained
Verified by 
interviewer Comments
7. Wholly/partially
/not verified
8. Wholly/partially 
/not verified
Opening – interview with the committee member(s).
My name is <name>. We are conducting an independent review of Land O’Lakes’ MSIKA program 
which has been training farmer member organisations like yours, and training farmers in agricultural 
techniques and technologies. We want to ask you some questions that will help us to learn more 
about the work of Land O’Lakes/MSIKA. What you tell us will not be shared with Land O’Lakes/MSIKA, 
so they will not find out what you say about them. Are you happy to continue? (Yes/No)
Dzina langa ndine <Name>. Ndimagwira ntchito ku bungwe la Kadale Consults, ku Lilongwe. Ife 
tikuchita kafukufuku, kuunika ntchito za MSIKA program yomwe, pansi pa bungwe la Land O’ Lakes, 
yakhala ikuphunzitsa ndikulimbikitsa pa alimi njira zamakono za ulimi wa mbeu za masamba ndi 
zipatso. Mayankho anu kumafunso athu atithandiza kumvetsetsa za ntchito za program ya MSIKAyi 
ndipo a MSIKA sadzadziwa zomwe inuyo mwatiyankha pakafukufukuyu. Ndinu okondweretsedwa 
kuti tipitilire? (Yes/No)
1. Please introduce the members of the committee:
Mr/Mrs 
/Ms Name Position Phone #
Chair person
Secretary
Treasurer
1. Tell me a bit about this FBO, such as when it was established, how many members it has and the 
crops your members grow. 
Mutandifotokozerako za gulu lanu/kalabu yanu, monga kuti inayamba liti komanso ili ndi 
mamembala angati, ngakhalenso mbeu zomwe mamembalawo akulima, mwa zina
(This question aims to just start the conversation and make them comfortable – some existed before 
MSIKA, and some were formed by MSIKA)
MSIKA Final Evaluation Page 136
kadale@africa-online.net
When FBO started/ Inayamba 
liti
Membership size/ 
Mamembala angati
Crops grown/Mbeu zomwe 
mamembala akulima 
2. What training have you received about running a farmer member-based organisation such as 
<name of the FBO>
Mwalandirapo maphunziro anji pakayendetsedwe ka gulu la alimi longa lanuli la <name of the 
FBO>?
Training module /name When held 
(month/yr)
# of members 
trained Verified by 
interviewer Comments
Phunziro Liti? (Mwezi/ 
Chaka)
Anaphunzitsa 
alimi angati?
1. Wholly/partially 
/not verified
2. Wholly/partially 
/not verified
3. Wholly/partially 
/not verified
4. Wholly/partially 
/not verified
5. Wholly/partially 
/not verified
6. Wholly/partially 
/not verified
7. Wholly/partially 
/not verified
8. Wholly/partially 
/not verified
3. What have been the most useful training the leadership of the FBO has received? Why do you 
say that? What has been the least useful? Why do you say that? In what ways could the training 
be improved?
Ndimaphunziro ati a utsogoleri, omwe gulu lanu linalandira anali aphindu kwambiri? Mukutero 
chifukwa chiyani? Ndimaphunziro ati omwe ndiaphindu lochepa kwambiri? M’maganizo anu, 
chofunika kuchita ndichiyani kuti maphunzirowa akhale abwino kuposera pamenepa?
Has this FBO been 
trained in improved 
financial and 
organizational 
management?
Kodi gulu ili 
linaphunzitsidwapo 
luntha/luso 
lazachuma ndi 
How many improved 
financial mgt practices 
and systems has 
MSIKA helped the FBO 
to implement?
Kodi ndi njira zingati 
zomwe Land O’ Lakes 
yathandiza gulu lanu 
kukhazikitsa 
___________
MSIKA Final Evaluation Page 137
kadale@africa-online.net
kayendetsedwe ka
gulu?
(probe to find out 
what training 
modules/content) 
ndikutsata pa 
luntha/luso 
lazachuma ndi 
kayendetsedwe ka 
gulu?
4. What improved crop practices have you seen many members now adopting? Why these ones 
and not others? What could help to increase the uptake of techniques and technologies by 
members? What have been the results on production volumes and quality from the training?
Kodi ndinjira ziti za ulimi wamakono zomwe mwaonako alimi a gulu lanu akutsatira tsopano? 
Akutsatira zimenezi, kusiya zinazo chifukwa chiyani? Chingachitike ndichiyani kulimbikitsa kuti 
alimi ochuluka mugululi adzitsatira njirazi? Zotsatira zake za maphunzirowa ndizotani kumbali 
ya mulingo ndi kolite yazokolola?
(push for numbers and/or examples of change).
5. What market information has the FBO received and what does it pass on to members? What 
training in accessing markets has the FBO and its members received? How have these trainings 
helped the FBO and its members to access markets? Can you give some examples of these new 
markets?
Kodi gulu lanu lalandirapo mauthenga ati okhudza misika, ndipo ndi ati omwe lapereka kwa alimi 
ake? Gululi ndi alimi ake mwalandirapo maphunziro ati azamisika? Maphunziro amenewa 
athandiza gululi ndi alimi ake motani pakapezedwe kamisika? Mungandipatseko zitsanzo za 
misika yatsopanoyi?
6. How does the FBO help its members to sell their crops? What records do you keep of members 
sales? What was the volume of sales for your members in 2020, 2019, 2018, 2017, 2016?
Kodi gululi limathandizira alimi ake motani kugulitsa mbeu? Mumalembera ndikusunga chiyani 
pa zomwe alimi anu akugulitsa? Alimi anu agulitsa zochuluka bwanji mu chaka cha 2020, 2019, 
2018, 2017 ndi 2016?
How many sales 
agreements has the 
FBO made with the 
help of MSIKA?
Gululi lapeza 
migwirizano ingati 
yogulitsa mbeu 
mothandizidwa ndi 
Land O’ Lakes/MSIKA?
2020 ___________
2019 ___________
2018 ___________
2017 ___________
2016 ___________
What value of FBO 
member sales are 
claimed by MSIKA?
Malonda omwe alimi 
anu anagulitsa mbeu 
mothandizidwa ndi 
Land O’ Lakes/MSIKA 
ndiokwana ndalama 
zingati?
2020-
MK___________
2019 -MK__________
2018 - MK _________
2017 - MK _________
2016 - MK _________
MSIKA Final Evaluation Page 138
kadale@africa-online.net
7. Have you heard of covid-19 or corona virus? Has it affected the FBO activities in any way? What 
were these? What measures were taken to mitigate the effects? Are these effects continuing?
Kodi munamvapo zamuliri wa matenda a COVID-19, kapena kuti Corrona Virus? Muliriwu 
wakhudza ntchito za committee yanu mu njira iliyonse? Munjira ziti? Panachitika chiyani kuti 
muthane ndi zovutazi? Zovutazi zikupitilirabe?
Any other comments you want to share about working with MSIKA/Land O’ Lakes.
Muli ndi ndemanga iliyonse pa momwe mwagwilira ntchito ndi a land O’ lakes/MSIKA?
Thanks 
MSIKA Final Evaluation Page 139
kadale@africa-online.net
KII Topic Guide AEDO or Horticulture Specialist – MSIKA 
Introduction: “My name is <name>, and I am working for Kadale Consultants. Kadale has been asked 
by Land O’Lakes’ MSIKA project to talk to key stakeholders, such as Government Officers, who know 
about the implementation of its programme. 
We would like to discuss your experience of working with Land O Lakes’ MSIKA project. 
The discussion will take about an hour and a half. What you tell us will not be directly attributed to 
you, so Land O’Lakes’ MSIKA team will not know what you have said about them.
Are you willing to be interviewed?” Consent given – Yes / No 
1. When did you first come into contact with Land O’Lakes MSIKA team? What did they tell you 
about the program? What expectations did you have at the start?
2. What are the activities of the MSIKA program that you have seen being implemented in your 
District? Please tell me more about each of these 
Probe: Get the list of activities first, before getting the detail - Ask so that you get them all (“any
other elements?)” – agricultural production training, post-harvest handling training, demonstration 
plots, formation and training of crop-based Farmer Business Organisations (FBOs), working with any 
processors or agro-dealers, etc. 
Probe to get details of each component per the table. We are looking for process issues – e.g. how 
did the co-ordination with govt go; how involved are govt in the identifying areas, making links etc.; 
how well organised were the activities, etc. 
We are also looking for content issues: how appropriate was the training content for these types of 
farmers, were these useful activities, etc.
Probe for suggestions on ways to improve it and why they think that. While not seeking criticisms, 
explore any areas where there is some weakness to address, what contributes to this, how has it 
been addressed since being identified, etc.
District:
Name of interviewee:
Contact # or email:
Position of interviewee:
Conducted by:
Date conducted:
MSIKA Final Evaluation Page 140 kadale@africa-online.net
Activity Interviewee 
confirms
What involvement have 
you had in the trainings 
and other activities run by 
LOL/MSIKA?
What are your comments on 
how the activities were run 
(timely, how organised, 
AEDOs/AEDCs/HCS involved, 
etc.)?
What are your observations on the 
‘content’ of the activities (relevant, 
useful, something smallholders can 
implement, etc.)?
Any improvements or 
suggestions?
Agricultural 
Production 
Training
Yes/No
PHH Training Yes/No
Demo plot 
establishment Yes/No
Formation of, and 
training of Farmer 
Based 
Organisations
Yes/No
MSIKA Final Evaluation Page 141 kadale@africa-online.net
3. What changes have you seen as a result of the MSIKA program, such as changes in farmer knowledge and understanding, adopting agricultural practices, lower harvest 
losses, more sales, etc.
Activity
Include if 
mentioned 
in
Q2 above
What changes have you 
seen knowledge and use of 
practices?
Have you seen any changes for 
farmers, such as higher 
production, higher quality, 
lower losses, more sales?
Have you seen any changes 
in the operations of the 
FBOs? 
Any improvements or 
suggestions?
Agricultural 
Production 
Training
Yes/No
PHH Training Yes/No
Demo plot 
establishment Yes/No
Formation of, and 
training of Farmer 
Based 
Organisations
Yes/No
MSIKA Final Evaluation Page 142 kadale@africa-online.net
4. How has MSIKA improved your and/or other government staff, such as other extn officers, knowledge and skills 
about horticultural crops and fruit trees? 
Probe: In what ways, who benefited, how great was the change in knowledge, skill, etc. 
5. What has been the impact of COVID/Corona Virus on horticulture crop and fruit tree production and 
harvesting? 
Probe for specifics of in what ways (people getting sick, access to inputs, etc.), to what extent/how severe, 
affecting how many farmers in these value-chains (few, some, many, etc), etc.
6. What has been the impact of COVID/Corona Virus on horticulture crops and fruit tree markets? 
Probe for specifics of in what ways (markets closed, no transport, etc.), to what extent/how severe, affecting how 
many farmers in these value-chains (few, some, many, etc), etc.
7. What could be done to help the farmers continue practicing what they have been trained by MSIKA after the 
project’s end?
Thank you.
MSIKA Final Evaluation Page 143 kadale@africa-online.net
MSIKA Final Evaluation Page 144 kadale@africa-online.net
KII Topic Guide Processors – MSIKA Final Evaluation
Prior to each processor interview, obtain key information from the LOL/MSIKA team/key documents on the 
background to the processor and the work that has been done with them. Check if MSIKA worked directly or via 
Technoserve.
Brief background of the processor (what they 
process, products, markets, scale, plans, etc.)
What LOL/MSIKA has done with them:
Training conducted (specify trainings e.g.
quality standards, good manufacturing 
practices, food safety, product marketing, 
Standards regulations – MBS, ISO, HACCP):
Technical support/help facilitated (specify who 
by and the focus of the support)
Facilitated loans (with which bank/MFI and 
how much, for what purpose):
Facilitated MBS certification:
Results claimed: 
Investment leveraged by MSIKA’s support 
(specify MK or USD, what was investment by 
firm from own funds versus what from bank 
loans facilitated by MSIKA):
MBS Certification achieved with MSIKA’s help 
(pre-certify or full certified)
Change in sales (specify if MK or USD) last 
year and over time worked with MSIKA:
2020:
2019:
2018:
2017:
Sales agreements facilitated – number, value 
and with whom:
New markets accessed and new products 
launched:
Investment in current or new warehousing:
Obtain the following data from MSIKA prior to interview.
Type of employee # reported at Baseline # reported latest MSIKA data
Full time – male
Full time – female
Part time – male
Part time – female
Temporary – male
Temporary – female
Introduction: “I am working for Kadale Consultants. Kadale has been asked by Land O’Lakes’s MSIKA project to 
talk to fruit and vegetable processors. Your business name and contacts have been given to us as an organization 
that MSIKA (nb. if relevant “and Technoserve) has been working with. We would like to discuss your experience of 
working with MSIKA. 
The discussion will take about 45 minutes . What you tell us will not be directly attributed to you, so Land O’Lakes 
will not know what you have said about them. Are you willing to take part in the discussion?” 
Consent given – Yes / No If no, ask if there is a particular reason for not giving consent (as this would 
be surprising).
Name of organization:
Name of interviewee:
Contact # or email:
Position of interviewee:
Conducted by:
Date conducted:
MSIKA Final Evaluation Page 145 kadale@africa-online.net
1. “Tell me a little bit about your business, for example, when and how it started/you started it?” 
(This is a gentle entry to make them comfortable to talk, but move on reasonably quickly to) 
2. “What have been the areas of support from, and collaboration with, LOL/MSIKA(/Technoserve)? “ 
(Probe: Check for any areas stated by LOL/MSIKA, but not mentioned by the processor. The respondent may not 
separate the support given, such as help with MBS certification, with the result of that work – record here the nature 
of the support/work done and the result on the next page) 
Support from MSIKA Responses: what done, how useful, challenges/things that 
worked well? 
Training conducted 
(specify all the trainings e.g. quality standards, 
good manufacturing practices, food safety, 
product marketing, Standards regulations – MBS, 
ISO, HACCP etc.):
Technical support help direct from MSIKA or 
Technoserve of facilitated by these 
(specify who by and the nature of the support – 
what did they help with/work on?)
Facilitated loans 
(what was the loan for, with which bank/MFI and 
how much loan obtained/):
Facilitated MBS certification:
Other?
3. Adoption of new business practices/techniques
This is a key section for measuring indicator 19, so must be collected. The focus is on what was 
implemented/adopted, not what was learned/understood. The focus is on change, not continuation. Do not 
prompt on specific practices, but encourage them to keep giving as many changes as they can think of.
New business practices
implemented/adopted
Based on the training and technical support from MSIKA, what 
business practices have you changed/adopted compared to what 
you were doing before working with MSIKA? (Keep asking – any 
others/any more, so as to prompt them to keep giving specific 
things that have changed)
1. Quality Assurance and Good 
Manufacturing Practices:
• Implement Malawi specific 
product standards
• Implement product packing 
standards
• Develop new products based 
on new Mwi standards for 
tomato powder, chili pickles, 
potato starch, potato crisps, 
onion flakes, fruit leather, 
• Develop labelling for new 
products based on standard 
MS 19
• Steps to acquire and maintain 
certification/standards
• Conduct analysis of foods
• Take corrective action to bring 
foods up to standards/maintain 
standards
• Personal hygiene practices 
when handling food – install 
improve washing 
facilities/toilets, staff improving 
their practices, etc.
MSIKA Final Evaluation Page 146 kadale@africa-online.net
New business practices
implemented/adopted
Based on the training and technical support from MSIKA, what 
business practices have you changed/adopted compared to what 
you were doing before working with MSIKA? (Keep asking – any 
others/any more, so as to prompt them to keep giving specific 
things that have changed)
• Staff health – steps taken when 
staff are ill (i.e. not to work) and 
for well-being/health of staff, 
changes to personnel policies, 
safety rules implemented, 
clean clothes, protective 
equipment used
• Food safety – changes to 
improve these, preventing 
contamination
• HACCP – implementing the
principles of HACCP
• Factory layout and equipment 
to improve standards, 
equipment sanitising/cleaning
• Waste disposal
• Record keeping in support of 
quality/standards
2. Product Marketing
Mission and vision
• Customer analysis
• Analysing customer 
attributes
• Differentiating your 
products from competitors
• Changes to way they are 
selling
• Changes to the way they 
price their products
• Changes to how they are 
promoting and 
communicating with 
customers
• Assessing the competition
• Changes to how they brand 
(name) their products
• Inventory management
3. Gender equality
Changes in business practices 
around treatment of women in the 
workplace by mgt and by staff
Development of plans
Implementation of gender plans.
MSIKA Final Evaluation Page 147 kadale@africa-online.net
4. How has the support/collaboration from MSIKA assisted the business? 
(Ask on each result that LOL/MSIKA has indicated as counted for this processor – see page 1 for details):
Results claimed: Details
New investment made by the business (e.g.
new processing equipment, bigger premises, 
more warehousing, more stock, etc. – ask 
what they did…)
Have you increased your 
warehousing/storage capacity (dry or cold) 
with LOL/MSIKA help? Probe for details – 
when, by how much, in what way, why, etc.?
Did the business access a loan(s) with 
MSIKA’s support?
Loans amount (MK/USD):
Loan accessed from (bank): 
Additional investment by the firm: (MK/USD)
(did they put in their own funds or funds other 
than from a loan? How much? Specify where 
funds came from, if possible)
MBS Certification achieved with MSIKA’s help 
(clarify if pre-certify or full certified and what 
their status was before MSIKA)
When did they get this MBS Certification? 
If they have pre-certification, when do they 
hope to get full certification?
Employment change and attribution to MSIKA 
(see table below)
Change in sales (specify if MK or USD) last 
year and over time worked with MSIKA:
2020:
2019:
2018:
2017:
Have there been any sales agreements 
facilitated by MSIKA with farmers that MSIKA 
has supported? How many, which FBOs?
New markets accessed and new products 
launched (get specifics – e.g. new mango 
achar, selling to supermarkets following 
getting MBS certification):
Other results of note? 
How many employees do you currently have? How has that changed in the last year and also since you 
started working with MSIKA?
Type of employee Currently 
employed
Change since last 
year (e.g. +4, -2, 0, 
against each 
category)
Change since working 
with MSIKA (e.g. +4, -
2, 0, against each 
category)
Full time – male
Full time – female
Part time – male
Part time – female
Temporary – male
Temporary – female
MSIKA Final Evaluation Page 148 kadale@africa-online.net
5. How has COVID affected your business operations and plans for investing?
Probe: Any sickness/absence of staff (cautious probe if any employees have died), effect on suppliers 
particularly farmers/FBOs, any problems with getting produce to markets, any issues with falls in demand, 
etc.
Do you have any questions for me?
Thank you.
MSIKA Final Evaluation Page 149 kadale@africa-online.net
Annex 5: Comparison of Practices by Crop
Practice Used Tomato Onion Potato Mango
Baseline EoP Diff Baseline EoP Diff Baseline EoP Diff Baseline EoP Diff
1 Compositing 45.3 78.3 33.0 59.2 82.2 23.0 29.8 65.4 35.6 17.9 61.6 43.7
2 Manuring 64.9 80.1 15.2 73.5 79.9 6.4 42.9 64.3 21.4 14.5 57.0 42.5
3 Ridging 51.4 77.5 26.1 36.7 74.2 37.5 84.1 78.6 -5.5 n/a n/a n/a
4 Mulching 66.7 86.2 19.5 63.3 83.0 19.7 19.0 47.0 28.0 6.0 45.6 39.6
5 Crop rotation 45.6 78.3 32.7 50.0 85.2 35.2 53.6 81.2 27.6 n/a n/a n/a
6 Minimum tillage 14.1 50.0 35.9 7.1 47.3 40.2 5.5 n/a n/a n/a n/a n/a
7 Planting seeds in a nursery before planting out 86.2 96.4 10.2 85.7 95.5 9.8 13.8 n/a n/a n/a n/a n/a
8 Staking 80.8 97.5 16.7 n/a n/a n/a n/a n/a n/a n/a n/a n/a
9 Succession planting 29.1 59.1 30.0 n/a n/a n/a 29.8 43.2 13.4 n/a n/a n/a
10 Using irrigation 88.9 91.3 2.4 81.6 96.2 14.6 68.5 84.6 16.1 17.0 55.9 38.9
11 Soil and water conservation 24.0 71.0 47.0 25.5 67.4 41.9 n/a 71.8 n/a 3.4 60.5 57.1
12 Draining excess water 15.9 50.4 34.5 17.3 43.9 26.6 n/a n/a n/a 2.1 30.0 27.9
13 Testing soil acidity 0.3 19.9 19.6 1.0 25.8 24.8 n/a n/a n/a 0.0 18.6 18.6
14 Adding lime or ash to soil before planting to reduce acidity 0.6 33.7 33.1 1.0 37.9 36.9 n/a n/a n/a 0.0 28.1 28.1
15 Choosing the variety 34.5 79.7 45.2 39.8 73.5 33.7 n/a 81.6 n/a 7.7 n/a n/a
16 De-suckering or removing unwanted side-shoots 63.7 93.8 30.1 n/a n/a n/a n/a n/a n/a 15.3 n/a n/a
17 Spraying for pests and disease 67.3 92.8 25.5 67.3 87.1 19.8 n/a 79.3 n/a 3.0 43.0 40.0
18 Using recommended plant and ridge spacing n/a 98.6 n/a n/a 96.6 n/a n/a 94.4 n/a n/a n/a n/a
19 Using recommended fertilizer and application rates n/a 94.6 n/a n/a 94.3 n/a n/a 91.0 n/a n/a n/a n/a
20 Sterilizing nursery beds before planting n/a 80.4 n/a n/a 81.8 n/a n/a n/a n/a n/a n/a n/a
21 scouting for pests and diseases n/a 77.2 n/a n/a 69.3 n/a n/a 85.0 n/a n/a 44.1 n/a
22 Uprooting infected plants and burning n/a 87.3 n/a n/a 73.9 n/a n/a 74.8 n/a n/a 41.1 n/a
23 Sowing seed in row/groove nursery n/a 81.9 n/a n/a 84.1 n/a n/a n/a n/a n/a n/a n/a
24 Using fish soup/ sugar solution as bait for insects n/a 21.7 n/a n/a 21.2 n/a n/a 21.1 n/a n/a 15.2 n/a
25 Planting in sunken beds n/a 73.2 n/a n/a 80.7 n/a n/a n/a n/a n/a 0.0 n/a
26 Hardening off before planting out n/a 93.8 n/a n/a 93.2 n/a n/a n/a n/a n/a n/a n/a
27 Selecting the best seedlings for planting out n/a 96.0 n/a n/a 93.6 n/a n/a 95.9 n/a n/a n/a n/a
28 Intercropping n/a 51.8 n/a n/a 47.7 n/a n/a 36.1 n/a n/a n/a n/a
29 Green manuring 45.9 n/a n/a 49.0 n/a n/a n/a n/a n/a n/a n/a n/a
30 Water capture 25.8 n/a n/a 19.4 n/a n/a 22.1 n/a n/a 4.3 61.6 57.3
31 Earthing up n/a n/a n/a 79.6 91.7 12.1 83.0 94.4 11.4 n/a n/a n/a
32 Clipping plant ends n/a n/a n/a n/a 71.2 n/a n/a n/a n/a n/a n/a n/a
MSIKA Final Evaluation Page 150 kadale@africa-online.net
Practice Used Tomato Onion Potato Mango
Baseline EoP Diff Baseline EoP Diff Baseline EoP Diff Baseline EoP Diff
33 Using raised beds n/a n/a n/a n/a 78.0 n/a n/a n/a n/a n/a n/a n/a
34 Chitting n/a n/a n/a n/a n/a n/a n/a 63.9 n/a n/a n/a n/a
35 Pruning n/a n/a n/a n/a n/a n/a n/a n/a n/a 62.1 86.8 24.7
36 Grafting n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a 22.8 n/a
37 Budding n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a 15.2 n/a
38 Staking young trees n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a 53.6 n/a
39 Digging of planting holes prior to planting n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a 61.6 n/a
40 Deflowering of first flowers n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a 29.7 n/a
41 Top Working n/a n/a n/a n/a n/a n/a n/a n/a n/a 23.4 n/a n/a
Sample size/difference 333 276 -57 98 264 166 289 266 -23 235 263 28