ITIKI Performance Evaluation 
Bridging Indigenous Knowledge and Science 
Drought Prediction Tool in Kenya
SECURING WATER FOR FOOD
AUGUST 2019
Prepared By: Amanda Satterwhite, Master’s student, The Bush School at Texas A&M University
The Kaizen Company | Contract #: AID-OAA-C-15-00011 
www.securingwaterforfood.org | securingwaterforfood@gmail.com
TABLE OF CONTENTS
ABSTRACT 2
INTRODUCTION 3
BACKGROUND 5
Gender 5
Age and family size 6
Farm size 7
Sources of income 8
METHODOLOGY 9
RESULTS 13
EXPERIENCE WITH INNOVATOR 14
Familiarity with ITIKI 14
Perceptions of ITIKI 16
BENEFITS OF INNOVATION 18
Farming practices 18
Water usage and access 20
Crop yield and survival 22
Income 24
Input usage 27
Gender differences 28
Regional differences 29
DISCUSSION 33
Innovation usage 34
Crop yield and survival 35
Income 36
Gender differences 36
Affordability 37
Other benefits 38
Difficulties using ITIKI 41
User recommendations for improvement 42
CONCLUSION 43
ANNEX I: SURVEY INSTRUMENT 45
2 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
This evaluation assesses the effects of the ITIKI drought prediction tool on farming practices and outcomes 
in Embu County, Kenya. ITIKI combines indigenous weather indicators with meteorological data to generate 
localized predictions on the timing of rain onset and approximate volume of rainfall for each growing season 
(i.e. “less than normal” rainfall, “normal” rainfall) and sends this information to farmers via text message. The 
evaluator surveyed 62 registered ITIKI users, who were selected through a cluster randomized sampling 
design, regarding their experience using this innovation.
User perceptions of the innovation were mostly positive. Despite drought in recent seasons, users felt 
ITIKI intelligence had improved their crop yields compared to what yields would have been without the 
information. The reported effects occurred through a change in farming practices in which, based on the 
weather predictions, users planted early and changed to more drought-resistant crops and varieties. 
However, approximately 20 percent of respondents were unfamiliar with the innovation despite being 
registered users, which highlights the need to address issues with ITIKI’s recruitment and follow-up.
Quantitatively measurable effects are sparse because ITIKI has operated for a limited time in Kenya and 
highly variable rainfall makes it impossible to attribute changes in yields or income across seasons to the 
innovation. Nonetheless, positive user perceptions of ITIKI and reported changes in farming practices as a 
result of ITIKI intelligence suggest the innovation is helping farmers cope with drought. 
ABSTRACT
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 3
INTRODUCTION
4 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
In Kenya, the vast majority of smallholder farmers depend on rainfall to water their crops.1
 Yet, 
variable rainfall patterns mean drought is a common reality, leaving these farmers with harvests 
too small to feed their families. The ITIKI drought prediction tool seeks to help farmers adapt to 
less-than-normal rainfall levels by providing information on the timing and amount of rains. With 
this data, farmers can take advantage of the entire season of rainfall by planting before its onset. 
Furthermore, they can use information on predicted levels of rainfall to decide whether to plant 
more drought-resistant crops or varieties.
ITIKI is unique because it uses both meteorological information from weather sensing stations and 
indigenous indicators to predict rainfall. These indicators are signs used for generations to plan for 
the growing season. They include the behavior of birds and insects, the position of stars, and the 
flowering of trees and plants. For example, people in the Embu region of Kenya noticed dragonflies 
begin flying near the ground when the rains are one to two weeks from onset. By combining these 
information sources, ITIKI seeks to make predictions that utilize rich cultural knowledge and are 
scientifically rigorous. Formulated through a computer algorithm, the predictions are sent to farmers 
via text messages in their local languages.
While ITIKI currently operates in Kenya, Mozambique, and South Africa, this evaluation investigated 
its effects in Embu County, Kenya (its current area of operation in Kenya). The evaluator conducted 
62 interviews with registered ITIKI users in 14 villages. Respondents were selected through a cluster 
randomized design and surveyed regarding their farm activities, use of ITIKI, and perception of its 
benefits, among other topics. The evaluator also conducted key informant interviews with ITIKI founder 
Muthoni Masinde and local project manager Peter Ngoci and held a focus group discussion with ITIKI 
ambassadors. Findings from these interviews helped triangulate and explain quantitative survey 
results. These results illuminate areas where ITIKI has helped users adapt their farming practices 
to less-than-normal rainfall levels. The findings also highlight challenges to ITIKI uptake and impact, 
which are discussed here to help the organization identify areas for improvement.
The report proceeds as follows. The Background section discusses demographic characteristics 
of interviewed farmers. Section II explains sampling methodology and data collection activities. 
Section III presents results, including farmers’ familiarity with and usage of ITIKI; effects in terms of 
farming practices, crop yields, income, and water usage; and, differences in results by gender and 
region. Section IV discusses these results and elaborates on challenges and recommendations for 
improvement. Section V provides a conclusion.
1 The Food and Agriculture Organization of the United Nations reports that on farms below the median size in Kenya, only six 
percent of land is irrigated. (Food and Agriculture Organization of the United Nations. Family Farming Knowledge Platform. 
Accessed from http://www.fao.org/family-farming/data-sources/dataportrait/indicator-details/en/?ind=83468.)
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 5
The field evaluator surveyed 62 farmers in 14 villages in Embu County, Kenya. The survey used 
Fulcrum mobile technology and probed respondents’ personal and farm characteristics and 
experience with ITIKI. The full survey instrument is in Annex I. This section presents demographic 
characteristics of the sample. (The sampling methodology is covered in detail in the next section.)
Gender 
Of the surveyed farmers, 73 percent (45 of 62) were women (Figure 1). This mirrors the registered ITIKI 
user population: among the sample’s 14 villages , women made up an average of 72 percent (and 
a median of 75 percent) of ITIKI users. While the sample gender breakdown is representative of the 
population, this was not the case for each village. In some villages, it was more difficult to schedule 
interviews with male respondents because they appeared more likely than women to be away from 
the home.
FIGURE 1: GENDER DISTRIBUTION OF SURVEY RESPONDENTS
N=62
73%
27%
GENDER DISTRIBUTION OF SURVEY RESPONDENTS
Female
Male
BACKGROUND
6 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Age and family size 
Survey respondents ranged in age from 24 to 80, with a mean age of 49 years (Figure 2). Female 
respondents were, on average, significantly younger than males, with a mean age of 47 years 
compared to males’ mean age of 55 years. Older respondents were more likely to be unfamiliar 
with ITIKI, despite being listed as registered users. The mean age of the 12 non-users was 57 years, 
compared to 47 years for users. This age difference is statistically significant at the 95 percent 
confidence level, a finding which is discussed in Section IV. 
Average household size was 5.9 members, with no statistically significant difference between male 
and female respondents or between ITIKI users and non-users. Household size ranged from one 
person (respondent only) to 15 people. 
FIGURE 2: AGE DISTRIBUTION OF SURVEY RESPONDENTS
AGE DISTRIBUTION OF SURVEY RESPONDENTS N=62
0 24-39 40-50 56-70 71-80
NUMBER OF RESPONDENTS
19
6
15
22
20
25
15
10
5
AGE GROUP
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 7
Farm size
Farm size among respondents ranged from 0.25 acres to 13 acres, with a mean acreage of 3.5 and 
a median of 2.5 (Figure 3). Plot size varied by gender, with an average plot size 3.2 acres for women 
and 4.3 acres for men. However, the largest plot among respondents was 13 acres and was owned 
by a female respondent, even though only one acre was actively farmed. One respondent was unable 
to estimate her farm size. 
FIGURE 3: FARM SIZE OF SURVEY RESPONDENTS
N=61 
FARM SIZE (ACRES)
20
25
30
35
15
10
5
0
<=1 1-3 4-9 10-13
NUMBER OF RESPONDENTS
10
6
16
29
ACRES
Furthermore, 45 of 62 respondents (73 percent) owned all their land, while four respondents owned 
some of their land. Of the 17 respondents who owned none or some of their land, only two paid rent. 
The remainder were granted the right to use land owned by a parent or other relative or simply used 
government land without holding a land title. The percentage of respondents who owned the land 
they farmed was similar among men and women. 
8 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Sources of income
All respondents were engaged in farming. On average, respondents grew four different crops, with 
maize, cowpeas, green grams, sorghum, and millet as the most common crops. Many respondents 
were subsistence farmers: 21 respondents (33 percent) reported they did not sell any crops in the last 
year, while 15 sold only one crop and kept the rest for household consumption.
While many farmers engaged in income-generating activities off the farm, 58 of 62 (94 percent) 
reported farming as their primary occupation while 21 respondents (33 percent) indicated farming 
as their sole occupation (Figure 4). Nine respondents (14 percent) also raised and sold livestock. It 
is probable that some respondents who reported farming as their sole occupation may also raise 
livestock but consider it part of their farming business and did not report the activity. Additionally, 21 
respondents (33 percent) reported a small enterprise, such as operating a motor-taxi service, running 
a small shop or grocery, or sewing clothing. Lastly, 12 respondents (19 percent) engaged in wage 
labor, often in a stone quarry or on a khat farm (khat is a major cash crop in the region). 
FIGURE 4: INCOME SOURCES
N=62
33% 33%
14% 19%
FARMER OCCUPATIONS
Farming ONLY
Small Enterprise
Livestock
Wage Labor
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 9
METHODOLOGY
10 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
The sample was selected through a cluster randomized design, based on a list of over 2,500 registered 
ITIKI users across 245 villages.2 Villages’ mean number of users was 11, while the median was 3. About 
one-third of villages had only one user. To improve fieldwork efficiency, the evaluator restricted the 
village (cluster) selection to villages with users at or above the mean (11 users). This yielded 48 villages. 
ITIKI began operations in the villages surrounding Kiritiri town, in Mbeere South sub-county of Embu 
County (Figure 5). In 2018, users were recruited in the Mwea area and in Mbeere South. The local ITIKI 
manager expressed concern that users in Mwea would be less familiar with ITIKI because they were 
recruited by contacts local to Mwea rather than the ITIKI manager or ITIKI ambassadors. This may 
have lead to less complete sensitization to ITIKI than experienced in Kiritiri area villages.3 Furthermore, 
no ITIKI ambassador was posted in the Mwea villages so Mwea users may not have benefitted 
from the same follow-up as Kiritiri users. The issues surrounding recruitment and ambassador 
presence are discussed later in this report. However, the potential difference in Kiritiri and Mwea user 
experiences did influence the evaluator’s selection of villages. To assess impact among individuals 
who had ample experience with ITIKI while avoiding the bias of only selecting more seasoned users, 
the evaluator selected 12 villages in the Kiritiri area and four in the Mwea area. These were randomly 
picked from the restricted list of 48 villages. 
FIGURE 5: KIRITIRI TOWN, EMBU’S CENTER OF OPERATIONS IN KENYA
2 The list included 2,745 users, but it is noted that several duplicate names were found. The evaluator did not identify every 
instance of duplication, as some names are written differently in different entries (e.g., alternate spelling, middle name used vs. 
omitted), and thus identifying all instances requires a line-by-line review of the list. Similarly, some village names had variant 
spellings and appeared as different villages in the list. While the evaluator attempted to correct this with assistance from ITIKI 
staff, it is likely that the count of 245 villages is somewhat overestimated.
3 During the time of recruitment in Mwea, communities were experiencing tension and violence over land rights. Because of this, 
community members were suspicious of outsiders. The local ITIKI manager thus chose to recruit a local contact to introduce 
ITIKI to these communities, rather than visiting himself or sending an ambassador from Kiritiri area.
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 11
For each of the 16 selected villages, the evaluator stratified users by gender. Six were then randomly 
selected from each village, with the number of males and females determined by the ratio of users in 
the village.4 For each village, three to five females were selected. In many cases, there was no phone 
number for the user , and these users were replaced with new randomly selected users. All six selected 
individuals were contacted for interviews, with the goal of confirming at least four interviews. In cases 
where contacting six people did not yield four scheduled interviews, two to four randomly selected users 
were contacted. Because some confirmed interviewees did not arrive for their interviews during the first 
few days of fieldwork, the evaluator began to schedule additional interviews in each village in an effort 
to conduct at least four. For this reason, the evaluator conducted up to six interviews in some villages.5
Unfortunately, though the overall sample’s gender breakdown (73 percent female) closely matched the 
average percentage of female users among the selected villages (72 percent), the intended gender 
ratios did not hold for each village (Figure 6). In Makima, Gatirari, Gatoori North, Ngarira, and Irabari, the 
difference in percentages of women in the population (all ITIKI users) and in the sample was more than 
20 percentage points. Despite contacting the correct ratio of women and men, a small sample in each 
village made it impossible to control the gender ratio of individuals who actually arrived for the interview.
FIGURE 6: WOMEN AS A PERCENTAGE OF TOTAL USERS 
AND SAMPLED USERS, BY VILLAGE
The sampled users were 62, the “total users” were approximately 2,500
0%
60%
40%
20%
70%
50%
30%
10%
90%
100%
80%
WOMEN AS A PERCENTAGE OF TOTAL USERS AND SAMPLED USERS, BY VILLAGE
PERCENTAGE
Women as % of total users
Women as % of village sample
Kiambere
Ngiori
Makima
Rogogwe
Kathera
Gatirari
Kandongu
Gatoori North
Gatumbiri
Kariari
Ngarira
Namuri
Irabai
Mariguru
VILLAGE
4 The percentage of females in each village was calculated, then multiplied by six and rounded to the nearest whole number 
to calculate the number of females and males in the sample for each village. For example, in Ngiori, 81% of users are female, 
equivalent to 4.8 out of 6. Thus, 5 females – and therefore one male – were selected for the sample in Ngiori.
5 It should be noted that respondents in one of the selected villages, Kiambere, did not live in Kiambere but rather in surrounding 
villages. During recruitment, their locations had been recorded as Kiambere. During the interviews, the respondents’ villages of 
residence were recorded. These were Giatugu, Gituri, Kaithungu, Kanthenge, and Kinyaga. In this report, however, Kiambere is 
considered as a single village cluster in alignment with the ITIKI user list. 
12 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
When visiting the first two villages in Mwea (Makima and Namuri), the evaluator found six out of nine 
respondents were unfamiliar with ITIKI and did not remember receiving a message. Rather than 
waste resources with visits to villages where few individuals were familiar with ITIKI, three scheduled 
respondents from the remaining two villages in Mwea were contacted. None had heard of ITIKI, and 
the remaining two villages in Mwea (Kamweli and Ndune) were not visited. In all, 14 villages were 
visited and 91respondents were interviewed (Figure 7). 
FIGURE 7: SAMPLED VILLAGES
Interviews were conducted using the Fulcrum mobile application. At the beginning of each interview, 
the evaluator explained the evaluation’s purpose and the types of questions that would be asked. She 
advised respondents to notify the evaluator if they did not wish to answer a question, and it would be 
skipped. Each respondent was asked to verbally confirm their desire to participate, and confirmation 
was captured in audio recordings. While the evaluator conducted a few interviews in English or 
Swahili, most respondents felt most comfortable communicating in their local language. For the 
villages in Kiritiri area, this was Kimbeere. For the villages in the Mwea area, this was Kikamba. Thus, for 
most interviews, communication occurred through an interpreter.
Lastly, the evaluator conducted a focus group discussion with three ITIKI ambassadors. They were 
asked to discuss their role, the reaction of users to ITIKI when it is introduced, and any challenges 
in recruitment. The evaluator also carried out a key informant interview with local ITIKI project 
managerPeter Ngoci, to gather information about the user recruitment process, the collection of 
indigenous indicators, and other project operations.
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 13
RESULTS
14 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Familiarity with ITIKI
ITIKI was first introduced in Kenya in late 2016 and was gradually rolled out throughout Embu County. 
While 12 respondents (19%) were not familiar with ITIKI, the majority who do use it have done so for 
approximately 1.5 to 2 years (Figure 8). This is consistent with ITIKI’s heavy recruiting in the second half 
of 2017. Furthermore, because recruitment of ITIKI users occurs through village-level meetings, there 
was a moderately strong correlation of length of use among respondents in a given village (intraclass 
correlation coefficient of 0.52). 
FIGURE 8: DISTRIBUTION OF LENGTH OF ITIKI USE
N=50 DISTRIBUTION OF LENGTH OF ITIKI USE
20%
30%
40%
50%
25%
35%
45%
15%
10%
5%
Does not use About 1 year 1.5–2 years 2.5–3 years
PERCENTAGE
19% 19%
16%
45%
0
Most respondents had difficulty answering the “length of use” survey question. The evaluator assisted 
respondents by asking about specific growing seasons and whether they had used ITIKI in the given 
season. Still, recall was a challenge, and many answers were rough estimates. 
Furthermore, ITIKI did not send any messages during the October to December 2018 season because 
of logistical problems with mobile telephone service providers. These problems persisted with some 
(though not all) mobile telephone companies into the March to May 2019 season, so not all users 
EXPERIENCE WITH INNOVATOR
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 15
received messages during that season. The villages with the greatest percentage of respondents 
unfamiliar with ITIKI were those in the Mwea area. According to ITIKI staff, recruitment occurred in this 
area in mid-2018 before the logistical problems began. The fact that many in this area never received 
messages after signing up helps explain their unfamiliarity with ITIKI. 
Familiarity with ITIKI is correlated with some demographic characteristics. Men were less likely to be 
ITIKI users despite being on the user list: only 65 percent of male respondents were familiar with ITIKI, 
compared to 87 percent of female respondents. Age was also predictive: two-thirds of non-users 
were above the median age of 46, and the mean age of non-users was statistically significantly 
higher than that of users. Interestingly, ITIKI users had smaller farms than non-users on average . 
Among users, the mean farm size was 3.2 acres, compared to 4.9 acres for non-users. This difference 
is statistically significant at the 95 percent level. At the same time, there is a moderately strong, 
positive correlation between age and farm acreage (correlation coefficient of 0.45), so it is possible 
that the difference in use is explained more by age than by farm size.
ITIKI Ambassadors
16 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
The evaluator found that 78 percent of respondents familiar with ITIKI (39 of 50) had heard about 
it from an ITIKI ambassador or from local project manager Peter Ngoci. ITIKI has contracted seven 
ambassadors, who are residents of villages where ITIKI operates and are charged with enrolling 
village residents to receive ITIKI messages. Ambassadors do this through a short presentation at an 
existing village meeting , such as a church service, a village savings and loan association meeting, 
or another common interest group’s meeting, e.g. an association of livestock farmers. After the 
short presentation to“sensitize” people to ITIKI, the ambassador schedules a follow up meeting with 
interested individuals to enroll in ITIKI. Ambassadors are assigned to recruit individuals in their home 
village and in two nearby villages. They are charged with following up with ITIKI users before and 
throughout the growing season to remind them of weather predictions and encourage them to adjust 
their farming plans accordingly. 
With this “hands on” approach, it is not surprising that the villages where the majority of respondents 
reported being unfamiliar with ITIKI were those with no ambassador. These include Ngarira (in 
Kiritiri area) and Makima and Namuri (both in Mwea area). These villages had been introduced to 
ITIKI, but did not receive follow up. The focus group discussion with ITIKI ambassadors revealed 
they often reminded ITIKI users of the presence of “indigenous indicators” that help predict timing 
and amount of rainfall. Thus, even during the season in which no messages were sent, ITIKI users in 
villages with ambassadors received weather information. However, in the villages without an assigned 
ambassador, the break in messages in 2018 (and into 2019 for some users, depending on their mobile 
telephone service provider) meant no information was received. This sheds light on the frequency of 
respondents reporting unfamiliarity with ITIKI in these villages. 
Of the remaining 22 percent of ITIKI users that did not report hearing about the innovation from ITIKI 
staff or ambassadors, nine of 11 respondents noted they had heard about it from a neighbor. This 
may have been an ITIKI ambassador, but the respondent did not recognize the neighbor as such. An 
additional respondent was told about ITIKI from “someone he didn’t know from outside his village,” 
which may have been an ambassador or Peter Ngoci. The remaining ITIKI user did not remember 
where she heard about the technology.
Perceptions of ITIKI
Among respondents familiar with ITIKI, general perceptions of the technology were positive. None 
of the 50 users reported any reluctance to recommend ITIKI to their friends or neighbors, with 40 
percent responding they would recommend it, and 60 percent reporting that they would strongly 
recommend it (Figure 9). One respondent who reported that she would recommend ITIKI seemed 
reluctant to do so, possibly because the respondent had not reported any income or yield benefits 
from ITIKI. When pressed, the respondent maintained that she would recommend the technology. 
While it is possible that some respondents may have reported they would recommend the technology 
out of fear of appearing ungrateful for the project (thus introducing bias in survey results), many 
respondents demonstrated enthusiasm in their recommendations. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 17
FIGURE 9: USER WILLINGNESS TO RECOMMEND ITIKI
N=50
60%
40%
USER WILLINGNESS TO RECOMMEND ITIKI
Would Strongly 
Recommend
Would Recommend
Furthermore, 46 of 50 users reported they shared the knowledge gained through ITIKI updates with their 
families or communities. For example, one noted, “Yes, just today I was sharing my knowledge with my 
neighbors!” Another related that community members ask how her crops perform better than theirs, and 
she tells them it is because of knowledge gained through ITIKI. Additionally, all 50 ITIKI users reported they 
would continue using it because the weather information was helpful for their farming practice. Once ITIKI 
was explained to the non-users, 11 of the 12 reported they would be interested in using it in the future. 
One non-user said, “If I knew what the rains would be like, I could make better decisions on what to plant. 
I wouldn’t waste so much time planting things that will not grow.” Another qualified her answer, explaining 
that she would be willing to use ITIKI information if she first saw that it had benefited existing users. The 
single non-user who said she would not use ITIKI in the future did so because she did not own a phone. 
18 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Farming practices
By providing farmers with predictions on when the rains will arrive and how heavy they will be, ITIKI 
enables farmers to change their farming practices to help cope with drought. Among the most 
important changes is early planting. Traditionally, farmers wait until the rains arrive to plant, but 
when they are informed on when the rains will arrive, they can have their seeds in the field before this 
occurs. In the region served by ITIKI, the rainy period is often too short; one respondent said her village 
only had one day of rain during the most recent season. By planting before the rains come, farmers 
improve crop survival and yields by taking advantage of every opportunity to water their crops. 
Early planting was the most common benefit cited by respondents. Thirty-seven of 50 users (74 
percent) mentioned it in response to an open-ended question about ITIKI benefits. Furthermore, 49 of 
50 users indicated that ITIKI helped them decide when to plant in response to a question about how 
technology has changed their farming practices (Figure 10). 
FIGURE 10: CHANGES IN FARMING PRACTICES
N=50
98%
CHANGES IN FARMING PRACTICES
% of ITIKI Users Who 
Changed Time of Planting
76%
% of ITIKI Users Who Selected 
Different Crops/Varieities
Respondents also commonly reported that ITIKI information about how long the rains would last 
influenced their selection of crops and varieties. Thirty-eight of 50 users (76 percent) stated that 
knowledge of impending drought led them to make one or more of the following changes:
BENEFITS OF INNOVATION
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 19
(1) Add crops that perform better in drought, such as sorghum and cowpeas, or increase their acreage. 
Thirty-three of 50 users (66 percent) reported adding a new drought-resistant crop or increasing 
acreage because of ITIKI information (Figure 11). The most common addition was sorghum, with nearly 
one-half of users having introduced it or increased its acreage after ITIKI. Green grams also appeared 
to have been a popular choice, with 21 of 50 users introducing it or planting more. Lastly, 16 users 
began or increased cultivation of cowpeas and millet, respectively. 
FIGURE 11: USER ADDITION OF DROUGHT-RESISTANT CROPS
N=50 USER ADDITION OF DROUGHT-RESISTANT CROPS
40%
60%
80%
20%
Any crop Cowpeas Green
grams
Sorghum Millet
PERCENTAGE
66%
32%
48%
42%
32%
0
While these were the most common crops added because of ITIKI, others were mentioned by users. 
One respondent began cultivation of sweet yellow (a fruit similar to passionfruit) and papaya. Another 
reported starting cultivation of tomatoes, and a third added acreage to his pigeon pea cultivation 
because of ITIKI information.
(2) Eliminate crops that require more water, such as maize and beans, or reduce acreage of these crops.
Twenty-three ITIKI users (46 percent) stopped or reduced their bean cultivation, while 10 users (20 
percent) stopped or reduced maize cultivation. It is worth noting that even among the 12 non-users, 
two respondents had stopped growing beans and one had stopped growing maize. This raises the 
possibility that these changes are due not to ITIKI, but simply to the region’s increasing frequency of 
drought. Despite the small sample size of non-users and thus the need for caution in interpreting 
results, the rate at which ITIKI users abandoned bean cultivation is higher than the rate among non￾users, with the difference statistically significant at the 95 percent level. While the rate at which users 
abandoned maize also is higher than the rate at which non-users abandoned maize, the difference 
is not statistically significant. Since maize is the most important staple crop in the region, farmers may 
be much more reticent to stop or reduce their maize cultivation.
20 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
(3) Purchasing and planting a drought-resistant seed variety of one (or more) of their existing crops.
Although respondents were not directly asked whether ITIKI information led them to switch to 
drought-resistant seeds, 13 of 50 respondents (26 percent) mentioned this as an ITIKI benefit. The 
evaluator analyzed changes in seed purchase before and after ITIKI. Thirteen respondents (including 
three who mentioned changing seed varieties due to ITIKI) reported purchasing seeds for one or more 
crops in the season after they began using ITIKI, although they did not purchase seeds in the season 
before ITIKI and instead reused seeds from their own harvest. This may be indicative of a change 
to drought-resistant varieties. However, it is unlikely that this change was driven by ITIKI, as almost 
as many respondents (10) reported they had bought seeds in the season before ITIKI, but not in the 
season after. Thus, it is impossible to draw a conclusion about the switch to drought-resistant varieties 
based on farm input data.
Water usage and access
In Embu County, the vast majority of farmers rely only on rain for their agricultural activities. Among 
the 50 ITIKI users interviewed, 43 (86 percent) have no other source of water for crop irrigation 
(Figure 12). The percentage is similar among non-users, with 10 out of 12 reporting rain as their only 
agricultural water source. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 21
The most common alternate source of water was river water. Five respondents (three from Makima 
village, one of whom is an ITIKI user, and 2 from the villages surrounding Kiambere, both of whom are 
ITIKI users) reported using river water for irrigation. Three had a pump to deliver the water from the 
river to their farms, while one carried it by donkey and watered his crops by hand, and another carried 
it by hand. One used river water for khat and tomato crops but not staple cereal crops. Respondents 
in the remaining 12 villages did not have access to river water for irrigation. 
FIGURE 12: SOURCE OF WATER FOR IRRIGATION
N=50
86%
2%
2%
6%
4%
CHANGES IN FARMING PRACTICES
Rain Only
Own Pond
River
Groundwater
Tap Water
Two ITIKI users had their own ponds (plastic-lined trenches near their farms), while another hand￾watered her crops with water from a well. A final ITIKI user watered her kitchen garden with tap water, 
but relied on rain water for her staple crops. 
Since most of the respondents relied on rainwater, they were unable to quantify the amount of water 
used before and after ITIKI. All 43 ITIKI users who relied on rainfall alone reported no change in their 
water use before and after ITIKI, though many made comments about rainfall being sparser during 
the last season. Four additional users of river water, groundwater, or tap water also reported no 
change in water use (Figure 13). One ITIKI user reported that now using less water from her pond 
because she plants early and now relies more on rainfall and less on supplemental water. Another 
user reported pumping less river water to his tomatoes for the same reason. Neither was able to 
quantify the water savings. A final ITIKI user reported using more river water after ITIKI, but not 
because of ITIKI. Because the drought has been more severe since he began using ITIKI, he now uses 
about 800 liters of river water daily, compared to 300 liters daily in the season before ITIKI. Finally, no 
respondents reported any change in water access using ITIKI. 
22 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
FIGURE 13: CHANGE IN WATER USE
N=50
94%
4% 2%
CHANGE IN WATER USE AFTER ITIKI
No Change
More
Less
Crop yield and survival
Since the majority of respondents rely on rain for their yields, highly variable rainfall means large 
variation in yields across seasons. In the most recent season (March to May 2019), rainfall was 
extremely sparse, with 29 of 50 respondents (58 percent) reporting they had harvested nothing. 
An additional seven respondents had only harvested one of the crops they planted, although most 
people planted three or four different crops in the season. While prior seasons had some rainfall, 
respondents did not consider rains to have been “good” for the past two years. 
 To collect data to compare yields before and after ITIKI, the evaluator asked about previous 
seasons, going back to March to May 2016 depending on how long the respondent had been using 
ITIKI. However, respondents frequently mentioned different levels of rainfall in each season and 
often remembered “good rains” when asked about growing seasons prior to ITIKI. In some cases, 
respondents who had used ITIKI for more than two years had difficulty recalling yields from seasons 
before ITIKI. Many respondents reported changing their crop acreage season-to-season or adding or 
eliminating crops. These variations make it impossible to conduct a quantitative comparison of yields 
before and after ITIKI.
However, when asked whether ITIKI had improved their yields, respondents were very positive (Figure 
14). Thirty-seven of 50 respondents (74 percent) felt their yields had either increased or increased 
substantially. Another nine respondents (22 percent) felt yields had remained the same or increased 
only slightly, although four made statements such as, “Because of the drought, my yields didn’t 
improve. But they would have increased if the rains had been better.” Two respondents reported their 
yields had declined, with one clarifying it was because of the lack of rain. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 23
FIGURE 14: CHANGE IN CROP YIELDS
N=50
56%
18%
16%
6%
4%
CHANGE IN CROP YIELDS
Increased
Increased Substantially
Declined
Remained the Same
Increase a Bit
The evaluator sensed some respondents did not feel their yields had increased, but reported an 
increase so as not to appear negative about ITIKI. The evaluator probed these respondents and asked 
if yields had truly increased or were lower because of the lack of rain. Most respondents kept their 
original answer, though some qualified an earlier statement that yields had increased and explained 
that they had only increased “a little bit.” Some appeared to have answered with a comparison of 
their yields to what their yields would have been without ITIKI, with a few making statements such as, 
“My yields are lower than before, but without ITIKI I would not have harvested anything.” Based on this 
comparison, these respondents considered their yields to have “increased.”
Similarly, of the 33 ITIKI users who reported having grown new crops or added acreage to existing 
crops because of ITIKI, 28 reported these crops had done well, while an additional four reported one 
or more of the crops grown because of ITIKI had done well, while other(s) had not (Figure 15). One 
respondent stated that neither crop he grew because of ITIKI (cowpeas and sorghum) had done well, 
but the reason was drought. 
24 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
FIGURE 15: PERCEPTIONS OF CROP SURVIVAL
N=33
28
4
1
PERCEPTIONS OF CROP SURVIVAL
Yes
Mixed
No
Perceptions of crop survival also were positive. Forty-four of 50 users reported crop survival had 
improved with ITIKI. Nine commented the change was due to planting the right crops (namely, 
drought-resistant crops rather than maize and beans), while three users felt the improved survival 
was a product of early planting. Two respondents explicitly qualified that improvements in this area 
were only seen for the drought resistant crops (with several respondents mentioning millet, cowpeas, 
sorghum, and green grams), while survival rates for maize and beans did not improve. The same 
result appeared in analyzing respondents’ answers about the benefits of ITIKI: only two of 50 felt that 
ITIKI had helped them produce more of their “most important crop,” which is maize in this region. 
Income 
It was difficult for respondents to accurately estimate their incomes. For those who sell crops, income 
varies from season to season and is inconsistent across years. Because those respondents with 
small enterprises such as motor-taxi services or sewing businesses have farmers as their clients, their 
income also varies with the agricultural seasons. To assist respondents in estimating their incomes, the 
evaluator asked about monthly or seasonal incomes and estimated annual incomes. Even with this 
aid, 13 of 50 users were not able to provide an estimate and another eight were unwilling to share. 
Some hinted at having a pension or a spouse with a pension, although they were not comfortable 
disclosing the amount.
Among the 29 users who provided an estimated income, 28 fell into the category of “extreme poor” 
(under 240,000 KSh per year, or $2,400 USD), and one reported an annual income of 360,000 KSh, 
qualifying as “poor” (240,001 KSh to 480,000 KSh per year) (Figure 16). ITIKI is thus well-targeted to 
the extreme poor. (Mean income for ITIKI users in the before-ITIKI period was 85,700 KSh annually, 
while the mean income for non-users was 98,300 KSh. However, this difference in mean income 
between users and non-users is not statistically significant.)
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 25
FIGURE 16: ANNUAL INCOME
N=50
56%
26%
2%
16%
ANNUAL INCOME AMONG ITIKI USERS
Extreme Poor
Poor
Does Not Know
Not Willing to Share
Income gains after ITIKI were limited. Only seven users reported a change in income, with an average 
increase of about 30,000 KSh. However, including the users who reported no change in income, the 
average increase was only 8,500 KSh (with a median of 0 KSh). Furthermore, one of the seven users 
who reported increased income clarified that the increase happened after ITIKI was introduced but 
was not because of ITIKI. While ITIKI’s improvement of pecuniary income was limited, many ITIKI users 
reported the innovation had improved their income because it had helped them grow more food. 
Because of this, they bought less food for their families and had more money to spend on other things. 
Thirty-two of 50 users (64 percent) felt ITIKI had improved their income “somewhat,” while another 
nine users (18 percent) felt it had improved their income “significantly” or “very significantly” (Figure 17). 
Of the 41 users who felt ITIKI had improved their income at least somewhat, 20 (49 percent) noted 
that it had improved income through increasing food for the household. 
26 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
FIGURE 17: PERCEIVED INCOME IMPROVEMENTS
N=50
64%
8%
10%
18%
AMOUNT ITIKI IMPROVED INCOME
Somewhat
Significantly
Very Significantly
None
Respondents spent additional income in a variety of ways, but the most common was school fees. 
Thirty-one of the 41 ITIKI users who reported an income improvement after using ITIKI put it toward 
schooling for their children or grandchildren (Figure 18). In Kenya, even public schools charge fees, 
which are quite high considering the average household income. Other popular uses of additional 
income were on home improvements, such as buying a water tank, or farm investments, such as 
buying livestock. Buying additional food not grown on the farm also was common, with 11 of 41 
respondents reporting this use to purchase items such as salt, sugar, or tea. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 27
FIGURE 18: RESPONDENTS’ REPORTED USE OF ADDITIONAL INCOME
N=41 USER ADDITION OF DROUGHT-RESISTANT CROPS
15
25
35
10
20
30
5
School Fees
House Improvements
Farm Investment
Business Investment
(non-farm)
Food for Family
Clothing for Family
Health Costs
Savings
NUMBER OF RESPONDENTS
31
9
2 2
4
2
7
11
0
Within the ITIKI user sample, no respondent reported a change in income that moved them out of the 
“extreme poor” or “low income”categories. In fact, with an average annual income of 94,200 KSh in the 
period after ITIKI (and a median of 80,000 KSh), most users remain far below the extreme poverty line. 
Input usage
Respondents were asked about their use of farm inputs before and after ITIKI, specifically their 
spending on and usage of fertilizer, pesticide, herbicide, seeds, charcoal (as a natural pesticide), water, 
and family and hired labor. No systematic differences were found in use of any of these inputs when 
compared with the period before and after ITIKI. This is not surprising, given that ITIKI’s model does not 
include provision of or training in farm inputs. Three respondents mentioned learning about farm input 
usage as a result of ITIKI, but this knowledge sharing appears to have occurred on an ad-hoc basis 
during meetings where ITIKI was introduced. It also is possible that these respondents confused ITIKI’s 
work with that of another NGO. These issues are explored in detail in the “Discussion” section.
28 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Gender differences
Though respondents were chosen from registered ITIKI user list , some were unfamiliar with ITIKI 
messages. As noted in the previous section, this proved far more likely for men: only 65 percent 
of interviewed men reported having ever received a message from ITIKI, while 87 percent of 
female respondents were familiar with the ITIKI messages (Figure 19). This difference in ITIKI usage 
and gender breakdown among respondents reflects an overall trend in the region of low male 
participation in development projects, as reported by the local project manager. Furthermore, since 
ITIKI often is introduced at meetings of existing village groups where women are more likely to be 
present, women are more prone to hear about the innovation and have an opportunity to enroll.
FIGURE 19: ITIKI UPTAKE, BY GENDER
N=62
0
40
45
30
20
10
35
25
15
5
ITIKI UPTAKE, BY GENDER
Female Male
NUMBER OF RESPONDENTS
6
39
6
11
Users Non-users
While women were more likely than men to use ITIKI, they were less likely to make changes in their crop 
choices. While 10 of 11 male users of ITIKI (91 percent) reported changing what they planted because 
of ITIKI, only 72 percent of female users did so (Figure 20, Panel 1). Though the small sample size for 
men requires caution in interpreting results, this difference between men and women is significant 
at the 90 percent level. Similarly, men were more likely than women to stop growing crops that do 
poorly in drought, with eight of 11 men (73 percent) reporting they had stopped or reduced maize or 
bean cultivation, while only 46 percent of women did so (Figure 20, Panel 2).6 This difference also 
is statistically significant at the 90 percent level. Lastly, men reported adding new crops based on 
ITIKI information at a slightly higher rate than women (Figure 20, Panel 3), but this difference is not 
statistically significant. 
6 The question about previously grown crops did not probe whether respondents had stopped growing a crop because of ITIKI. 
Rather, it asked whether there were any crops they used to grow that they no longer cultivated (or of which they had reduced 
acreage). As noted in the “Farming practices” sub-section, however, the rate of abandoning bean cultivation (as a crop that is 
not drought-resistant) was higher for ITIKI users than for non-users.
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 29
FIGURE 20: CHANGES IN DECISIONS REGARDING CROPS, BY GENDER
WOMEN: N=39 MEN: N=11
72%
28%
CHANGES IN DECISIONS REGARDING CROPS, BY GENDER
Yes
No
WOMEN
91%
9%
Yes
No
MEN
I have changed what I plant because of ITIKI
72%
28%
Yes
No
WOMEN
73%
27%
Yes
No
MEN
I have stopped growing maize or beans
64%
36%
Yes
No
WOMEN
73%
27%
Yes
No
MEN
I have added one or more new crops because of ITIKI
Regional differences 
Respondents from villages with an ITIKI ambassador were far more likely to use ITIKI than those from 
villages without an ambassador. As explained previously, ITIKI contracted seven ambassadors in 
its areas of operation.7
 These ambassadors are charged with recruiting users and following up with 
them before and during the growing seasons to discuss their farming practices relative to weather 
indicators and ITIKI updates. Each ambassador conducts recruitment and follow-up in his or her 
home village and in two additional villages. After interviews, the field evaluator asked the local project 
7 Official contracting of ambassadors occurred in 2019. Before this time, ambassadors provided recruitment and follow-up 
services on a volunteer basis.
30 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
manager to identify the sampled villages with assigned ambassadors and found that seven of the 
selected villages had an assigned ambassador and seven did not.
Only three of 32 respondents in the villages with an assigned ambassador were unfamiliar with ITIKI, 
while nine of 30 respondents (30 percent) in the villages with no ambassador were unfamiliar (Figure 
21). This suggests an ambassador’s presence in a user’s village (or in a nearby village) matters for 
uptake of ITIKI among registered users. A focus group discussion with three ambassadors revealed 
they often discuss weather indicators with members of their villages. One ITIKI user noted the 
ambassador in her village gave her ITIKI updates through word of mouth. In contrast, when asked why 
registered ITIKI users in some villages reported having never heard of ITIKI, one ambassador believed 
it was due to lack of follow-up. She said she had been sent to conduct recruitment through one-time 
meetings in some villages. However, she did not return to discuss weather updates with registered 
users because the villages were far away. This indicates reinforcement of ITIKI updates through in￾person contact does impact usage.
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 31
FIGURE 21: ITIKI USAGE, DISAGGREGATED BY 
AMBASSADOR PRESENCE IN VILLAGE
N=62
0
30
20
10
35
25
15
5
ITIKI USAGE, DISAGGREGATED BY AMBASSADOR 
PRESENCE IN VILLAGE
Ambassador No Ambassador
NUMBER OF RESPONDENTS
3
29
9
21
Users Non-users
Additionally, uptake was notably different between villages in the Kiritiri and Mwea areas. Restricting 
the sample to only those villages without an ambassador showed respondents in Kiritiri villages 
were significantly more likely to be familiar with ITIKI than those in Mwea villages. Among the Kiritiri 
villages without an ambassador, 86 percent of respondents were familiar with ITIKI, compared with 
only 33 percent of respondents in the Mwea villages (though the sample of Mwea respondents was 
only nine individuals). 
This difference likely is rooted in recruitment methodology. While Kiritiri users were almost exclusively 
recruited by the local project manager or an ambassador, Mwea users were recruited by non-ITIKI 
contacts. This occurred because local residents in Mwea were experiencing conflict over land rights 
during recruitment. Because of this, the project manager was concerned that an outsider would be 
received with suspicion. Therefore, he liaised with a local contact to recruit users instead of visiting 
the villages himself or sending an official ITIKI ambassador. The quality of the recruitment efforts may 
have suffered, and new users may not have fully grasped the purpose or benefit of ITIKI. 
Another reason for the variation in recruitment was ITIKI’s partnership with a Nairobi-based 
organization that offers technical and financial support to farmers in Embu. This organization shared 
contact information for farmers who “graduated” from its program so these farmers might benefit 
from ITIKI updates. However, while these individuals were added to the ITIKI user list, the process by 
which ITIKI was explained to them by the partner organization – or whether it was explained at all – 
is unclear. It is possible these individuals were added to the user list without having been sensitized 
to the source of ITIKI updates. Because of this, SMS updates may be unexpected and perceived 
as lacking credibility, leading farmers to ignore the information. In sum, the different recruitment 
processes in Mwea translated into large differences in ITIKI uptake. 
32 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 33
DISCUSSION
34 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Innovation usage
Familiarity with and usage of ITIKI were heavily correlated with demographic characteristics. First, 
the majority of ITIKI users are women. Because ITIKI recruitment usually occurs through meetings of 
existing village development groups (such as savings and loan groups or cooperatives of livestock 
farmers), men are less likely to be exposed to the innovation. As reported by the local project 
manager, men are generally much less involved than women in development projects or village 
associations working toward development goals. This is the likely reason that women made up, on 
average, 72 percent of ITIKI users among the sampled villages . 
While men are less likely to be users of ITIKI, those who registered as users also are less likely 
than women to be familiar with the technology. Despite being on the user list, only 65 percent of 
male respondents were familiar with ITIKI, compared to 87 percent of female respondents. Since 
women were more likely to have been sensitized to ITIKI through a village meeting, they are more 
likely to recognize the source of ITIKI messages, perceive the information as credible, and consider 
the information in their farming decision-making. Men, who are less likely to have undergone this 
sensitization, may have received the ITIKI messages but not recognized their source, thus reporting 
unfamiliarity with ITIKI during the interviews. Regardless, it is unclear how the men who reported 
unfamiliarity with ITIKI – particularly in villages with ITIKI ambassadors and where women were very 
familiar with the technology – became registered as users.
Younger farmers also were more likely to be familiar with and users of ITIKI. Two-thirds of respondents 
who were non-users of ITIKI (despite being on the user list) were above the median age of 46, and 
the mean age of non-users was significantly higher. There are multiple reasons older users may be 
unfamiliar with ITIKI. Many reported not being able to read, which may prevent them from recognizing 
an ITIKI message has arrived. However, some users reporting illiteracy had their children or 
grandchildren read the messages to them, allowing them to use the information without issue. Some 
may not own a phone, such as one elderly woman who reported that she probably gave the recruiter 
her son’s number when asked how she was on the user list. It is possible that older people register 
for ITIKI when presented with the opportunity because others are doing so, even if they do not own a 
phone. In this case, they do not receive any messages.
Most notably, usage of ITIKI was highly dependent on geography. Respondents in villages without 
an assigned ITIKI ambassador were far more likely to report unfamiliarity with ITIKI, highlighting the 
importance of ambassador follow-up. Registered ITIKI users are much more likely to recognize the 
arrival of a message, perceive it as credible, and employ the information in their farming decisions 
if the information has been reinforced by an ambassador as a member of the local community. 
Furthermore, respondents in Kiritiri-area villages reported familiarity with ITIKI at much higher 
rates than those in Mwea-area villages. Recruitment in Mwea did not occur through ITIKI staff or 
ambassadors, but rather through third parties that were either local contacts recruited by the project 
manager to sensitize individuals or representatives of the Nairobi-based partner organization. 
Therefore, the extent to which users understood ITIKI upon registration is dubious. It is possible that 
these users did not recognize ITIKI messages when they arrived and ignored them. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 35
Crop yield and survival
While quantitatively comparing crop yields and survival rates before and after ITIKI was impossible 
for reasons outlined in the previous section, qualitative analysis reveals most respondents felt use of 
ITIKI information had benefited them. Despite severe drought in recent seasons, the majority of users 
reported their yields and crop survival rates, even when poor, were better than they would have been 
without ITIKI. The primary reason for this change was early planting, with 98 percent of ITIKI users 
reporting the information led them to plant before the rains arrived, allowing crops to benefit from 
limited rainfall. These results suggest notification of when rains will arrive is the most valuable piece of 
information for ITIKI users.
A secondary – yet still notable –reported driver of crop benefits was ITIKI information regarding 
the length of the rains, which led 76 percent of ITIKI users to change their crop selection. Based on 
predictions of less-than-normal rainfall, farmers chose to start planting or increase their acreage of 
drought-resistant crops, reduce their acreage of crops requiring more water, and switch to drought￾resistant varieties of their existing crops. Of the 33 ITIKI users who started growing new crops or 
added acreage to existing crops based on ITIKI information, 85 percent reported these crops had 
done well. Thus, ITIKI information on the length of rains also appears to be useful information for 
farmers, motivating decisions that increase crop yields and survival.
36 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
Income
ITIKI had little to no impact on income, with less than 15 percent of respondents reporting an increase 
in income following ITIKI as compared with the period before ITIKI. Even these respondents appeared 
unsure of their income estimates, and they might have reported higher income after ITIKI because 
they felt it was the answer the evaluator desired despite encouragement to report income honestly. It 
is therefore impossible to conclude that ITIKI information improved users’ income.
A primary reason is that the majority of respondents are subsistence farmers and do not sell crops 
or have very limited sales. Thirty-four percent of ITIKI users reported no sales of any crop in the last 
year, while an additional 22 percent reported selling only one crop in the last year, although most 
users cultivate four to five different crops, on average. Thus, even with an increase in crop yields, these 
farmers would not have earned additional income. 
However, over 80 percent of ITIKI users interviewed felt ITIKI had improved their income. When 
probed, many stated it had improved their income by increasing food for the household, which allowed 
them to spend household funds on other items such as school fees, improving the house, or investing 
in the farm. Thus, while ITIKI didn’t improve pecuniary income, it appears to have improved household 
welfare for most users. 
Even so, these gains were limited in size: 64 percent reported their income had only increased 
“somewhat” while 18 percent reported no increase. One reason may be that all users suffered a 
severe drought in the most recent season and yields were especially low. Even if yields were better 
in previous seasons while using ITIKI, the hardship of the most recent season features most saliently 
in respondents’ minds, causing them to feel that their income has not improved. As most farmers do 
not keep income records, it is difficult for them to produce an objective comparison of income before 
and after ITIKI. Thus, their reporting likely depends much more heavily on their perception. Given the 
drought, most respondents perceived their income as lower than usual at the time of the interview.
All respondents except one fell into the “extreme poor” category, indicating the innovation is well 
targeted to the lowest-income households. However, given the lack of pecuniary income gains, there is 
no evidence that use of ITIKI moved any users into a higher income category.
Gender differences 
As discussed earlier, uptake of ITIKI was higher among women. However, changes in farming practices 
among active users differed among women and men, with women being less likely than men to make 
changes in their crop choices. In particular, men were more likely to stop growing or reduce acreage of 
crops that perform poorly in drought, such as maize and beans. One reason may be that women bear 
primary responsibility for feeding their families and are more reticent to take the risk of abandoning 
familiar crops. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 37
In contrast, the rates at which women and men added new, drought-resistant crops or increased 
acreage of these crops based on ITIKI information are not statistically different. This demonstrates that 
women do have openness to adapting their crop selection to weather conditions. However, since adding 
new crops is likely perceived as a lower risk choice than stopping cultivation of familiar crops, it may still 
be that women in the sample are more risk-averse than men. This implication should be considered in 
ITIKI’s strategy for motivating changes in farmers’ practices based on weather information. 
Affordability 
ITIKI’s long-term business plan involves charging users a small fee for updates. To assess the plan’s 
viability , respondents were probed about their willingness to pay (WTP) for ITIKI updates. Thirty of 50 
users (60 percent) reported unwillingness to pay anything for ITIKI updates (Figure 22). Among the 20 
users who reported a positive WTP, amounts varied: while two users were willing to pay up to 100 KSh 
per update, most were not willing to pay more than 40 KSh (Figure 23).8 Including the 60 percent of 
users unwilling to pay anything, the average WTP was 13 KSh per update.
FIGURE 22: USER WILLINGNESS TO PAY FOR UPDATES
N=50
40%
60%
AMOUNT ITIKI IMPROVED INCOME
Willing to Pay
Not Willing
8 Some respondents reported WTP per season, rather than per update. For these respondents, the evaluator halved the 
reported amount under the assumption of two updates per season. Thus, if a respondent reported WTP 200 KSh per season, 
he was assigned a WTP value of 100 KSh per update. While the ITIKI business plan includes four updates per season, this 
target has not been reached in recent season (for reasons described elsewhere in this report). Therefore, the evaluator chose 
to calculate WTP based on two updates per season.
38 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
FIGURE 23: USER WILLINGNESS TO PAY FOR UPDATES (AMOUNT)
N=20
USER WILLINGNESS TO PAY FOR UPDATES 
(AMOUNT)
15
25
10
20
5
100 KSh 50 KSh 40 KSh 20 KSh 10 KSh 5 KSh 2 KSh 1 KSh
NUMBER OF USERS
2
9 10 11
16
18 19 20
0
When asked about WTP, many respondents were confused by the prospect of paying for updates. 
The evaluator clarified that there is no current payment, and the question was whether they would be 
willing to pay in the future. Although several respondents answered yes, they recanted their answer 
when asked to provide an estimate of how much they would pay and instead stated they were 
unwilling to pay anything. The frequency of this occurrence suggests that answers suffer from social 
desirability bias: respondents may report WTP to please the evaluator, when in reality they are not 
willing to pay. Therefore, it is important to recognize that WTP estimates may very well be inflated.
Furthermore, the high rates at which ITIKI users share updates with other community members (46 
of 50 users reported having shared the information) suggest that securing payments from individual 
users may be a challenge. If users can receive the information from a neighbor, they may be reticent 
to pay for it themselves. Furthermore, the ambassadors’ role includes in-person follow up on ITIKI 
updates. The higher usage rates in villages with ambassadors suggest this is an important part of the 
ITIKI model, but it may also discourage payment by providing an alternative, free route for farmers to 
receive ITIKI information.
Other benefits 
A few respondents mentioned gaining knowledge from ITIKI that led to changes in their farming 
practices beyond those addressed elsewhere in this report. Some users remarked that ITIKI helps with 
the timing of activities beyond the planting stage during the growing season : three users mentioned 
that ITIKI helped them know when to spray crops with pesticide, one remarked that ITIKI helped her 
know when to weed, and another noted that ITIKI information helped her know when to harvest. 
ITIKI sends, on average, three messages throughout the season to keep farmers up to date on 
weather predictions. For example, a message toward the end of the season may notify farmers that, 
though the rains have halted, they will return in a week. With this information, farmers can leave their 
crops in the field in anticipation of more rain. On the other hand, if they were to harvest and lay their 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 39
crops out to dry before the rains had ended, the rain would damage the crops. It is interesting that so 
few respondents mentioned that ITIKI helped them with timing of activities beyond planting. This may 
be because ITIKI had issues in recent seasons with mobile telephone service providers, which limited 
the number of messages sent. The delivery of information to farmers beyond the planting period 
– both through text messages and through follow-up by ambassadors – is a topic that ITIKI should 
consider carefully to ensure messages remain impactful throughout the season.
A few farmers also mentioned changes in the way they farm. Two said they formerly intercropped 
but began partitioning crops after ITIKI. Another stated she began using terraces after becoming an 
ITIKI user. Two users began using a plow rather than tilling by hand because of ITIKI, with one stating 
that ITIKI taught her to use a plow to reduce water runoff and another indicating she bought a plow 
because it is difficult to till the land before the rains and a plow helped with early planting.9 In regard 
to farm inputs, two users said ITIKI taught them how to use manure as fertilizer, while one mentioned 
that ITIKI taught her to use ash as a natural pesticide. 
The evaluator probed the local project manager about these results, as provision of technical 
farming knowledge is not a central tenet of the ITIKI model. He explained that, during sessions in 
9 Many users reported using a plow, but only two attributed this to ITIKI. Most users farmed using a plow both before and after 
ITIKI.
40 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
which ITIKI is introduced to potential users, topics such as use of natural pesticides and fertilizers are 
sometimes discussed. This is part of a larger discussion on indigenous knowledge related to farming 
and is intended to help farmers understand indigenous weather indicators and the source of ITIKI 
predictions. However, topics other than weather indicators seem to be discussed on an ad-hoc basis. 
Thus, it is logical that so few respondents reported gaining knowledge on farming practices and 
farming inputs as a benefit of ITIKI. 
Relatedly, some respondents mentioned being provided technical training by other NGOs or 
government agricultural extension services. It is worth noting that some respondents may have been 
unsure of the source of knowledge and may attribute changes to ITIKI that were driven by another 
NGO’s training activities. 
Lastly, one respondent felt the information provided by ITIKI was valuable in itself, independently of 
any farming improvements it motivated. She stated she was happy to have gained the indigenous 
knowledge of her tribe (the Mbeere) and understand how her people traditionally predicted weather 
patterns. The sense of pride over retention of this indigenous knowledge is an intangible, yet not 
insignificant, benefit of ITIKI. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 41
Difficulties using ITIKI
The majority of interviewed ITIKI users (86 percent) reported no difficulties or problems using the 
innovation. Of the nine respondents reporting they are illiterate, seven felt this was not an impediment 
to their ITIKI usage. Four indicated their children or grandchildren read them the messages while three 
received the information orally from an ITIKI ambassador. 
Four users reported problems receiving messages, although two were among the illiterate 
respondents, with one acknowledging she may have received messages and not realized it because 
she cannot read. One cited unreliable cell service and another said that she didn’t always have her 
phone with her so her children may accidentally delete messages before she reads them.
Two users mentioned a period during the last year when they received no messages. This is likely 
due to ITIKI’s contractual and logistical problems with mobile telephone service providers during the 
October to December 2018 season and, to a lesser degree, in the March to May 2019 season. These 
issues are discussed earlier in this report.
Lastly, one ITIKI user related it was hard for him to implement ITIKI’s recommendations about early 
planting because of his farm’s rocky soil. Because of this, he must wait for the rains to soften the soil 
so he can plant. While this difficulty was cited by only one respondent, it is important to note that 
farmers without access to a plow may not have the option of early planting, and thus are unable to 
make use of one of the most important pieces of ITIKI information (the timing of rain onset). 
The evaluator also probed non-users (individuals who are on the ITIKI user list but unfamiliar with the 
innovation) for any information about why they had not received messages. Most non-users were able 
to verify their phone number was correct, although one was incorrect. Another noted that she could 
not remember signing up for ITIKI and did not have a phone, so she likely had given her son’s number. 
One non-user mentioned that his poor vision made it possible that he may have received messages 
without noticing. Another stated he was unable to read and would have been unaware of messages. 
Lastly, two non-users acknowledged receiving a message in June or July 2019, but it was the first 
message they remembered receiving. This likely has to do with the logistical problems with mobile 
service providers which were being resolved at the time of the evaluation. 
None of the respondents reported any negative impacts stemming from ITIKI. However, two 
respondents (from two different villages) noted that some people in their communities had doubts 
about the accuracy of the weather predictions. Both users stated the predictions are sometimes 
inaccurate. Furthermore, both elaborated that some community members were skeptical of the use 
of science to predict weather, believing that God alone controls the weather. This belief fed skepticism 
of the innovation.
42 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
User recommendations for improvement
The evaluator asked ITIKI users for ideas on how to improve the innovation. Sixteen users (32 percent) 
expressed preference for more frequent ITIKI messages throughout the season. When asked how 
often they would like messages, answers varied and included, “several times during the growing 
season,” “at least once a month,” and “four times per year.” One respondent clarified she would like 
three messages during the growing season: one during planting, one during cultivating, and one 
during harvesting. Founder Muthoni Masinde related to the evaluator that, to date, ITIKI has sent an 
average of three messages per season, with plans to send four per season. Based on respondents’ 
answers, however, it appears that users have received less than three messages each season. This 
may be because of mobile service difficulties or because users do not recognize they have received a 
message. ITIKI should make users aware that information is sent throughout the season and not just 
during planting time. This might be achieved by promoting more follow-up by ambassadors.
Three users desired more in-person follow-up from ITIKI staff and ambassadors. Two requested 
seminars on the weather and how to respond to it in farming practices. Another suggested regular 
farm visits from ITIKI staff to assess whether the changes made in response to ITIKI information had 
been beneficial. These results indicate some users prefer receiving information through word of 
mouth, rather than by text message. Furthermore, these results align with findings that respondents 
in villages with ambassadors are more likely to have made use of ITIKI weather predictions, likely 
because the information has been reinforced by those ambassadors. These findings suggest that as 
ITIKI expands, its staff must remain cognizant of how foundational the person-to-person connection is 
to farmer uptake of the innovation.
One ITIKI user expressed a desire for an interactive system to respond to ITIKI messages with 
questions or feedback and receive a response. Another user felt it would be helpful if ITIKI provided 
more information, suggesting that instead of simply predicting less-than-normal rainfall, ITIKI send 
recommendations about the best seed varieties to plant and where to procure them. The local project 
manager explained that ITIKI’s model is to provide weather information, while it is the role of the 
government agricultural extension office to provide advice on farming practices. However, weather 
predictions are shared with that office so its staff may better assist farmers. ITIKI users’ desire for a 
more interactive experience and more detailed advice suggests that ITIKI should explore deepening its 
partnership with local agricultural extension officers. 
Two respondents recommended that ITIKI advertise more. Eleven users asked ITIKI to assist in 
providing seeds, fertilizer, water, or credit for farm inputs. While this is not part of ITIKI’s model, it 
highlights the reality that farmers in its region of operation face severe financial constraints in their 
farming operations. Of these 11, three asked for assistance in irrigation, while one asked for support 
with activities to provide income during drought, such as livestock. These responses shed light on the 
hardship experienced by ITIKI users due to less-than-normal rainfall in recent years.
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 43
CONCLUSION
44 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
The ITIKI drought prediction tool aims to help farmers adjust to less-than-normal rainfall by providing 
information via text message regarding the timing and amount of rains. Through a survey of 62 
registered ITIKI users (50 of whom were active users of the innovation), the evaluator assessed 
ITIKI’s impact in regard to its objective. In particular, the evaluation probed users’ perceptions of the 
innovation and its effect on farming practices, crop yields, and income.
ITIKI users had a very positive perception of the innovation, with all 50 active users reporting they would 
recommend it to others. All except one user described a change in their farming practices driven by 
ITIKI, with early planting as the most common adjustment. The majority of users also reported having 
changed their crop choices in favor of more drought-resistant crops or crop varieties. Even with severe 
drought in the region in recent seasons, many felt ITIKI information had improved their crop yields 
compared to what yields would have been without the intelligence. There were no significant impacts on 
income; however, this is largely because most respondents are subsistence farmers who sell very little of 
their harvest. Regardless, many users felt ITIKI had improved their income because it had increased food 
for their households, thus freeing funds to spend on non-food items. 
Approximately 20 percent of respondents were unfamiliar with the innovation, despite being on 
the user list. This was more common among men, older users, users in Mwea-area villages, and 
users in villages without assigned ambassadors. The geographical differences in familiarity with the 
innovation suggest that the person-to-person recruitment strategy and ambassador follow-up is 
crucial for user uptake. 
Lastly, it is notable that the majority of ITIKI users reported unwillingness to pay for the innovation. 
This, combined with the finding that users share information with one another, may pose challenges 
for ITIKI’s future plans, which include payment by users. 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 45
ANNEX I
46 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
FARMER INFORMATION
NAME 
AGE 
DATE TIME 
GROUP INTERVIEW? o Yes o No
GROUP INTERVIEW NOTES
HOW MANY FAMILY MEMBERS LIVE WITH YOU? 
GENDER o Male o Female
WHAT IS YOUR PRIMARY OCCUPATION?
o Farming
o Wage Labor
o Seasonal Migrant Labor
o Small Enterprise
o Other: 
DO YOU HAVE ANOTHER OCCUPATION?
o Farming
o Wage Labor
o Seasonal Migrant Labor
o Small Enterprise
o Other: 
SIZE OF FARM (ACRES) 
NAME OF VILLAGE 
HOW MUCH LAND DO YOU OWN? 
HOW LARGE IS YOUR FARM/PLOT? 
o Large
o Medium
o Small
o Very Small
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 47
HOW MUCH IS LAND RENT? 
OTHER LAND NOTES
HOW LONG HAVE YOU BEEN USING ITIKI? 
DID YOU PARTICIPATE IN AGRICULTURAL ACTIVITIES THIS YEAR? o Yes o No
HOW MANY MONTHS IS THE PRIMARY GROWING SEASON? 
HOW MANY TIMES DO YOU HARVEST PER YEAR? 
FARM INFORMATION
WHAT CROPS DO YOU GROW AS A RESULT OF THE INNOVATION? LIST FROM MOST IMPORTANT 
TO LEAST IMPORTANT:
1.
2.
3.
DID THE MOST IMPORTANT CROP BENEFIT FROM ITIKI? o Yes o No
DID THE SECOND MOST IMPORTANT CROP BENEFIT FROM ITIKI? o Yes o No
DID THE THIRD MOST IMPORTANT CROP BENEFIT FROM ITIKI? o Yes o No
WHAT IS THE WATER SOURCE FOR YOUR IRRIGATION OF CROPS?
o Own pond
o River
o Groundwater
o Innovation Source
o Other: 
48 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
WHAT IS YOUR METHOD OF IRRIGATION?
o Drip feed
o Flooding
o Hand watering
o Rainfed
o Other: 
HOW MUCH HAS YOUR WATER USAGE CHANGED SINCE USING ITIKI, IF AT ALL?
USING ITIKI HAS YOUR ACCESS TO WATER:
o Had no change
o Improved
o Fundamentally improved (Improved a lot)
o Other: 
PREVIOUSLY GROWN CROPS: DID YOUR FARM PRODUCE DIFFERENT CROPS IN THE PAST THAT 
ARE NO LONGER GROWN HERE? IF SO, WHICH ONES? 
MASS OF PRODUCE: WHAT YIELDS DID YOU HAVE FOR EACH CROP YOU MENTIONED?
MASS OF PRODUCE 2: WHAT YIELDS DID YOU HAVE FOR YOUR CROPS BEFORE USING ITIKI? 
USING ITIKI HAVE YOU, FOR EACH CROP:
o Used more water
o Had no change in water use
o Used less water
o Other: 
USING ITIKI HAVE YOUR CROP YIELDS (ASK FOR EACH CROP):
o Declined
o Remained the same
o Increased
o Substantially increased
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 49
IS THERE A DIFFERENCE IN THE SURVIVAL RATES OF YOUR CROPS DUE TO ITIKI? o Yes o No
HOW MUCH OF YOUR PRODUCE DID YOU CONSUME IN YOUR HOUSEHOLD? (PERCENTAGE – 
NOTE IF DIFFERENT FOR EACH CROP) 
HOW MUCH OF EACH OF THE FOLLOWING INPUTS DID YOU USE BEFORE ITIKI?
FERTILIZER (KG)
PESTICIDE (KG)
HERBICIDE (L)
CHARCOAL (KG)
WATER (TOTAL) 
LABOR (DAYS)
OTHER 
HOW MUCH DID YOU SPEND ON EACH OF THE FOLLOWING INPUTS BEFORE ITIKI? 
FERTILIZER (KG)
PESTICIDE (KG)
HERBICIDE (L)
CHARCOAL (KG)
WATER (TOTAL)
LABOR (DAYS)
OTHER 
HOW MUCH OF EACH OF THE FOLLOWING INPUTS DO YOU USE AFTER ITIKI?
FERTILIZER (KG)
PESTICIDE (KG)
HERBICIDE (L)
CHARCOAL (KG)
WATER (TOTAL)
LABOR (DAYS)
OTHER 
HOW MUCH DID YOU SPEND ON THE FOLLOWING INPUTS AFTER ITIKI? 
FERTILIZER (KG)
PESTICIDE (KG)
HERBICIDE (L)
CHARCOAL (KG)
WATER (TOTAL)
LABOR (DAYS)
OTHER 
50 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
HOW MUCH DID YOU SPEND ON EQUIPMENT BEFORE AND AFTER ITIKI? 
HOW MUCH DID YOU SPEND ON TRANSPORT AND STORAGE BEFORE AND AFTER ITIKI? 
DO YOU HAVE PROBLEMS FINDING A MARKET TO SELL YOUR CROPS IN? o Yes o No
PLEASE EXPLAIN. 
DO YOU HAVE PROBLEMS GETTING YOUR CROPS TO THE MARKET? o Yes o No
PLEASE EXPLAIN. 
OTHER FARM NOTES (OPTIONAL).
INCOME AND EXPENDITURES
WHAT IS YOUR ANNUAL HOUSEHOLD INCOME? 
HOW MUCH INCOME DID YOU MAKE BEFORE ITIKI? 
AFTER ITIKI? 
HAS ITIKI IMPROVED YOUR FAMILY INCOME? 
WHAT PERCENTAGE OF YOUR INCOME DO YOU GET FROM NON-FARM SOURCES? 
HOW MUCH PRODUCE DID YOU SELL FOR EACH OF YOUR CROPS IN THE LAST SEASON AND THE 
LAST YEAR? 
WHAT IS THE PRICE PER KILO YOU RECEIVED FOR EACH OF YOUR CROPS FOR THE LAST SEASON?
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 51
USING ITIKI HAS YOUR ACCESS TO CREDIT:
o Not improved
o Improved
o Improved and have been able to repay over a short period
HOW DO YOU CURRENTLY FINANCE AGRICULTURAL ACTIVITIES?
o Own savings
o Credit and savings scheme
o Other credit
HOW MUCH DO YOU PAY FOR ITIKI? 
HOW MUCH ARE YOU WILLING TO PAY FOR ITIKI?
o Nothing
o ITIKI is free
o The same as what I pay now
o 50% less
o 50% more
o Other: 
HOW HAVE YOU SPENT YOUR NEW INCOME?
o N/A (if no new income)
o Send children to school or keep children in school
o Social functions (like weddings)
o Investment in farming
o Improving house
o Other: 
OTHER INCOME NOTES (OPTIONAL)
52 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
PERCEPTIONS OF ITIKI
WILL YOU USE ITIKI IN THE FUTURE (5 TO 10 YEARS)? o Yes o No
WHY? 
HOW, IF AT ALL, HAVE YOU CHANGED YOUR FARMING PRACTICES DUE TO ITIKI?
o No change
o Introduced new crops
o Changed irrigation system
o Reduced water usage
o It helps me decide when to plant
o It helps me decide which crops to plant
HAVE YOU FACED ANY DIFFICULTIES OR PROBLEMS USING ITIKI? o Yes o No
HOW CAN ITIKI BE IMPROVED? 
HOW DID YOU HEAR ABOUT ITIKI?
o Wealthy farmer
o Neighbor
o Innovation personnel
o Extension worker
o Other: 
WHAT FACTORS INFLUENCED YOU TO TRY ITIKI?
o Demonstration from neighbor’s farm
o Innovation is free from extension services
o No alternative water source
o Other: 
DO YOU SHARE YOUR KNOWLEDGE SKILLS FROM ITIKI WITH OTHERS? o Yes o No 
IF SO, HOW? 
WHAT DO YOU FEEL ARE THE BENEFITS OF ITIKI? 
SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019 53
HAVE YOU HEARD ABOUT CLIMATIC VARIATION? HAVE CHANGES IN RAINFALL OR TEMPERATURE 
AFFECTED YOUR FARMING PRACTICES OR CROP YIELDS COMPARED TO YOUR HISTORICAL 
RAINY/DRY SEASON PERIODS? o Yes o No
PLEASE SPECIFY HOW. 
HOW HAS ITIKI HELPED YOU? PLEASE RANK THE TOP 3 AND EXPLAIN POSITIVES/NEGATIVES.
 Makes water reusable 
 Helps women farmers as well as men 
 They made a special effort to include women farmers 
 Helps in producing more of our most important crop 
 Increases my yield through timely forecasts 
 Helps by lowering cost of inputs 
 Improves health and strength of livestock 
 Helps reduce labor 
 Reduces crop wastage 
 Helps me decide when to plant 
 Helps me decide which crops to plant 
 Other: 
WOULD YOU RECOMMEND ITIKI?
o No
o Yes
o Yes, would strongly recommend
ARE THERE NEGATIVE IMPACTS FROM ITIKI IN THE COMMUNITY? o Yes o No
PLEASE EXPLAIN IF YES. 
54 SECURING WATER FOR FOOD: ITIKI PERFORMANCE EVALUATION | AUGUST 2019
IF THERE HAVE BEEN ANY NEGATIVE IMPACTS, HAVE EFFORTS BEEN MADE TO RESOLVE THEM? 
o Yes o No
EXPLAIN. 
OTHER
INCOME/POVERTY NOTES
GENDER OBSERVATIONS
QUESTIONS/REQUESTS
OTHER NOTES

Securing Water for Food has sourced and invested in a portfolio of innovative 
solutions that aim to help farmers use water more efficiently and effectively, 
improve water storage for lean times, and remove salt from water to make more 
food. Our cohort of innovators are helping people in 35 low-resource countries 
with tools they need to produce more food with less water.
To learn more about Securing Water for Food, 
visit www.securingwaterforfood.org.