1
SC Burma (Myanmar) FY19 EFSP Baseline Report
Table of Contents
FIGURES LIST.......................................................................................................................... 2
TABLE LIST.............................................................................................................................. 2
ACRONYMS.............................................................................................................................. 3
EXECUTIVE SUMMARY ......................................................................................................... 4
2.2 SAMPLING FRAME AND SAMPLE SELECTION ..........................................................................12
2.3 DATA COLLECTION TOOL ........................................................................................................12
2.4 DATA COLLECTION ...................................................................................................................13
2.5 DATA QUALITY ASSURANCE.....................................................................................................13
2.6 DATA ANALYSIS.........................................................................................................................14
2.6.1 INDICATOR DEFINITIONS .......................................................................................................14
Food Consumption Score .............................................................................................................14
Reduced Coping Strategies Index (rCSI) .....................................................................................15
Household Hunger Scale..............................................................................................................16
2.7 LIMITATIONS .............................................................................................................................17
3.1 FINDINGS – SITTWE .............................................................................................................17
RESPONDENT INFORMATION.............................................................................................................17
FOOD CONSUMPTION SCORE ............................................................................................................18
REDUCED COPING STRATEGIES INDEX (RCSI).................................................................................21
HOUSEHOLD HUNGER SCALE (HHS)................................................................................................23
3.2 FINDINGS – PAUKTAW ........................................................................................................24
RESPONDENT INFORMATION.............................................................................................................24
FOOD CONSUMPTION SCORE ............................................................................................................25
REDUCED COPING STRATEGIES INDEX (RCSI).................................................................................27
HOUSEHOLD HUNGER SCALE ...........................................................................................................29
3.3 RELATIONSHIP AMONG FCS, RCSI AND HHS ........................................................................31
Sittwe – Unconditional Cash Transfer Program..........................................................................31
Pauktaw – Conditional Cash Transfer Program - CfW...............................................................33
3.4 CONCLUSIONS .................................................................................................................34
SITTWE – UNCONDITIONAL CASH TRANSFER PROGRAM......................................................................34
PAUKTAW – CONDITIONAL CASH TRANSFER PROGRAM - CFW ...........................................................35
3.5 RECOMMENDATIONS.....................................................................................................36
4.0 ANNEX...............................................................................................................................39
ANNEX 1: BASELINE QUESTIONNAIRE..............................................................................................39
2
Figures List
Figure 1: rCSI score by FCS Status in Sittwe .......................................................................... 31
Figure 2: Food Consumption Score by HH Hunger Status in Sittwe ...................................... 32
Figure 3: rCSI score by HH hunger status in Sittwe ................................................................ 32
Figure 4: rCSI score by FCS Status in Pauktaw ...................................................................... 33
Figure 5: Food Consumption Score by HH Hunger Status in Pauktaw ................................... 33
Figure 6: rCSI score by HH hunger status in Pauktaw ............................................................ 34
Table List
Table 1: Parameters and values applied in sample size determination .................................... 11
Table 2: Questionnaires Structure ............................................................................................ 12
Table 3: FCS food groups and relative weights ....................................................................... 14
Table 4: FCS classification thresholds ..................................................................................... 15
Table 5: Reduced CSI strategies/behaviours and their corresponding severity weights ......... 15
Table 6: Household Food Deprivation Experiences ................................................................ 16
Table 7: HHS Categorical Indicator ........................................................................................ 17
Table 8: Respondent information – Sittwe – Unconditional Cash Transfer ............................ 18
Table 9: Food Consumption Score for Sittwe – Unconditional Cash Transfer ....................... 19
Table 10: Household Food Consumption Practices for Sittwe ................................................ 19
Table 11: Table 11: Household Food Consumption Practices for Sittwe by HH Head Gender 
Status ................................................................................................................................ 20
Table 12: rCIS in Sittwe – Unconditional Cash Transfer ........................................................ 22
Table 13: Household Coping Strategies in Sittwe – Unconditional Cash Transfer ................. 22
Table 14: Household Coping Strategies in Sittwe by HH Head Gender Status ...................... 22
Table 15: HHS in Sittwe – Unconditional Cash Transfer........................................................ 23
Table 16: Table 15: HHS in Sittwe – Unconditional Cash Transfer by HH Head Gender 
Status ................................................................................................................................ 24
Table 17: Respondent information – Pauktaw – Conditional Cash Transfer .......................... 24
Table 18: Food Consumption Score for Pauktaw – Conditional Cash Transfer...................... 25
Table 19: Household Food Consumption Practices for Pauktaw ............................................ 26
Table 20: Household Food Consumption Practices for Pauktaw by HH Head Gender Status
.......................................................................................................................................... 27
Table 21: rCIS in Pauktaw – Conditional Cash Transfer ........................................................ 28
Table 22: Household Usage of Negative Coping Strategies in Pauktaw ................................ 28
Table 23: Household Usage of Negative Coping Strategies in Pauktaw by HH Head Gender 
Status ................................................................................................................................ 29
Table 24: HHS in Pauktaw – Conditional Cash Transfer ........................................................ 30
Table 25: HHS in Pauktaw – Conditional Cash Transfer by HH Head Gender Status ........... 30
Table 26:HHs with IPC phase by three different Food Security Indicators in Sittwe ............. 35
Table 27: HHs with IPC phase by three different Food Security Indicators in Pauktaw ........ 36
3
Acronyms
ACF Action Contre La Faim
CAPI Computer-assisted personal interviewing
CfW Cash-for-Work
CI Confidence Interval
DSW Department of Social Welfare
EFSP Emergency Food Security Program 
FCS Food Consumption Score 
FFP Food for Peace 
HH Households 
HHS Household Hunger Scale 
IDP Internal Displaced Person
LB Lower Bound (95% Confidence Interval) 
MEAL Monitoring, Evaluation, Accountability and Learning 
MCCT Mother and Children Cash Transfer Program 
MEAL Monitoring, Evaluation, Accountability and Learning 
NGOs Non-Governmental Organizations 
NRS Northern Rakhine State 
OTP Outpatient Therapeutic Care Program
PIRS Performance Indicator Reference Sheet
rCSI Reduced Coping Strategies Index 
SC Save the Children
SD Standard Deviation
SRS Simple Random Sampling
UB Upper Bound (95% Confidence Interval) 
UCT Unconditional Cash Transfer 
WFP World Food Program 
4
Executive Summary
As a result of communal violence between Muslim and Buddhist populations in Rakhine State, 
the massive displacement of around 128,000 people (both registered and non-registered) has 
occurred. Among those internally displaced persons (IDPs), majority were Muslims, who are 
currently residing in IDP camps within the Muslim and Rakhine host communities in five 
townships. The majority of the IDP camps locations are in Sittwe and Pauktaw townships. 
Protracted displacement and tightly restricted movement is restraining access to markets, 
livelihoods and basic services (such as health and education) for both IDPs and host 
populations. 
Save the Children (SC) is supporting both host communities and IDPs in the townships of 
Sittwe and Pauktaw, to meet their immediate food security needs, which will help households 
mitigate the need to engage in negative coping strategies or behaviours. SC is implementing 
unconditional cash transfer program (UCT) in Sittwe and conditional cash transfer (cash for 
work) in Pauktaw Township. 
The baseline survey was conducted with a representative sample of families to document the 
initial situation of households prior to receiving cash assistance from Save the Children and to 
determine the current levels of food insecurity among targeted households. The main objective 
of the baseline survey is to set the benchmarks for food security outcome indicators, namely 
food consumption score (FCS), reduced coping strategies index (rCSI), and household hunger 
scale (HHS). The key baseline questions are around these indicators. This survey collected 
data regarding general household characteristics, consumption of household food, and 
utilization of negative coping strategies in the absence of food. 
The sample size of 678 (Sittwe, 339 and Pauktaw, 339) households (HHs) was determined 
using a stratified simple random sampling (SRS) design and the USAID/Food for Peace (FFP) 
“comparative for proportions” formula based on the September 2018 USAID Beneficiary￾Based Sampling Guidance, and utilised a quantitative approach. The final sample size achieved 
is 698,350 HHs from Sittwe and 348 HHs from Pauktaw. While HHs for UCT were selected 
randomly from a participant list (participant-based), the HHs for CfW sample were selected 
randomly in the targeted villages (population-based). 
5
In Sittwe, a total of 350 UCTs beneficiaries participated in the survey and approximately 44% 
of the respondents were from Dar Paing camp. The remaining 56% of respondents were from 
Say Tha Mar Gyi camp. The average household size of the participants’ is 5.2 members, and 
approximately one-third (32.9%) of the respondents reported that their households were headed 
by females. 
The average FCS score for IDPs (UCT beneficiaries) is 35.5; approximately 9% of households 
had a ‘poor’ FCS, 43% had a ‘borderline’ FCS, and 48% had an ‘acceptable’ FCS. Very few 
households consumed dairy and fruits groups in the week prior to the survey, while almost all 
of HHs reported consuming main staples, vegetables and sugar, and three in every four HHs 
reported consuming meat and fish. The male-headed household had better food consumption 
scores compared to female-headed households (male-headed HHs’ FCS: 36.6 vs female￾headed HHs’ FCS: 33.4), and also reported better consumption on fruits, meat/fish and sugar 
food groups. The household size did not seem to impact the condition of household food 
condition score. 
The average rCSI score for HHs in Sittwe, used as coping strategies when HHs do not have 
food or the resources for purchasing food, is 16. Approximately 67% of the HHs reported using 
the two severe strategies in the week prior to the survey; for instance, 76% of HHs borrowed 
food or relied on help from a relative or friend and 58% of HHs restricted consumption of 
adults to prioritize children’s consumption. 
The results of this survey show that approximately half (51%) of IDP HHs experienced severe 
or moderate hunger and 49% experienced little to no hunger in the month prior to the survey. 
However, more, (52%) male-led households experienced less hunger than female-headed HHs
(41%). The analysis shows that there is no relationship between HHs size and household
hunger. 
In Pauktaw, 346 HHs from the conditional cash transfer program (CfW) in the targeted 15 
villages participated in the survey; there were only two HHs who refused to participate in the 
survey. The average household size was 5.1. Approximately 23% of HHs who participated in
the survey are led by females. Evidence1 from existing literature suggests that female-headed 
1 Dreze and Srinivasan, 1997, Meenakshi and Ray, 2002, and Gangopadhyay and Wadhwa, 2003
6
HHs experience more extreme poverty than male-headed HHs; for instance, the baseline shows 
that among individual food group consumption, the female-headed households (66.3%) 
consumed more pulses than male-headed household (48.1%). 
The average FCS score for the sampled HHs from host community is 32.6; approximately 20% 
of households have a ‘poor’ FCS, 39% have a ‘borderline’ FCS, and 41% have an ‘acceptable’ 
FCS. The average number of days for most household consumed food, vegetables, meat/fish, 
and sugar were 3 days a week in prior week, except main staple which was consumed 5 days 
on average. 
The average rCSI in Pauktaw is 10 and the majority of households (79.5%) relied on less 
preferred and less expensive foods. However, as estimated 60% of HHs also reported 
borrowing food (second most severe strategy) and 40% of HHs limited portion sizes. The 
average rCSI score for male-headed HHs was lower than the female-headed HHs. 
Approximately, three in every four households (71.1%) from Pauktaw experienced little to no 
hunger in the month prior to the survey. There was no significant difference between male- or 
female-headed HHs’ hunger levels. 
1.0 BACKGROUND INFORMATION
The situation in Rakhine State is a complicated mix of long-standing tensions between religious 
communities and further compounded by long-term under-development and poverty. Tensions 
between Muslim and Buddhist populations in Rakhine State escalated into violent clashes in 
2012, resulting in the displacement of around 128,000 people (both registered and non￾registered), mainly Muslims, who now reside in internally displaced people (IDP) camps and 
within Muslim host communities in five townships. The majority of these locations are in 
Sittwe and Pauktaw townships. As of October 2018, 101,627 IDPs (18,825 HH) were living in 
15 locations in Sittwe Township and 22,256 IDPs (5,124 HH) in five locations in Pauktaw 
Township. Protracted displacement and tightly restricted movement are restraining access to 
markets, livelihoods and essential services (such as health and education) for both IDP and host 
Muslim populations. The poorest Rakhine within conflict areas also face lost incomes and 
constrained livelihoods owing to the ongoing conflict. Generally, the poorest HHs are landless 
and reliant on seasonal labour incomes. Following the 2012 conflict, many Muslim poor lost 
7
access to Rakhine villages where they accessed agriculture labour incomes. Since August 2017, 
further outbreaks of the conflict in Northern Rakhine State (NRS) have led to the displacement 
of approximately 720,000 people. 
In the Muslim camps of Sittwe, 27% of the displaced are not receiving assistance from WFP2
. 
In most cases, these individuals were not registered for assistance following the 2012 
displacements, as Rakhine authorities determined that only households considered to be 
‘genuinely displaced’ could be eligible for food assistance and registered on beneficiary lists. 
In practice, this has most often meant that only households whose homes were burned and/or 
were displaced immediately with the conflict were registered for food assistance. Those not 
meeting these criteria, even though facing the same shocks and stresses, were not registered to 
receive government-endorsed assistance for food, shelter and other in-kind materials. These 
unregistered households now generally live in makeshift shelters, in separate and distinct parts 
of the camps or villages where the “official” IDPs have settled. 
The households excluded from current food assistance rely on individual strategies to cover 
their most basic food needs. However, these strategies - mainly low paid unskilled casual work 
and small-scale fishing - are not sufficient to cover household needs and lead to repeated use 
of negative coping strategies (accumulating high debt, reduction of food quality and quantities). 
Data collected in 2016/17 during the Sittwe Camps profiling exercise highlighted that 29% and
35% of the displaced in Muslim villages and camps respectively had inadequate diets. 
Comparatively, these rates were 4 and 2%, respectively, in Rakhine villages and relocation 
sites.
Inadequate food consumption was more prevalent in locations (i.e. camps and villages) where 
more displaced households reported not receiving any food assistance, highlighting the 
protective impact of the regular food assistance. For example, in Dar Paing informal 
settlements, only 11% of the displaced reported receiving any food assistance and 58% reported 
inadequate diets (55% borderline, 3% poor). Poor diets are characterized by the consumption 
of rice every day, vegetables every other day, and fish or meat once every two weeks. In 
addition to poor diets, people in makeshift shelters generally have lower socio-economic status 
2 As noted in the Sittwe Camp Profiling Report, 
https://reliefweb.int/sites/reliefweb.int/files/resources/sittwe_camp_profiling_report_lq.pdf 
8
and rate below the Sittwe rural camp populations on many indicators, in particular health and 
education. 
Without access to land, and with depleted productive and financial assets and resources, 
displaced populations have limited options to generate income and frequently resort to using 
negative coping strategies. For Muslim communities, severe restrictions of movement remain 
in place for both camp and host communities as they face checkpoints, border fences and 
restricted access. Additionally, the denial of the rights belonging to citizenship to most 
Muslims means they lack other basic rights, including freedom of movement and the ability to 
access basic services and livelihoods. Construction, fishing/agriculture and mechanic work (in 
small workshops) are the most available livelihoods opportunities. Close to two thirds of 
households rely on casual labor as a primary source of income. The labor market is highly 
saturated, and competition for employment is fierce, resulting in limited incomes: in 2016, half 
of all displaced Muslim households in camps generated 25,000 MMK ($18.50/USD) or 
less/month. 
In Pauktaw Township, poor and very poor households across livelihoods zones purchase 
almost all of their food needs from local markets with up to 60% of their income spent on food, 
half of those on staple foods. Poor and very poor households are landless and rely on small￾scale fishing, seasonal casual daily work, and self-employment activities for income to obtain 
food. Reducing expenditures in bad years/difficult times is a common strategy for the better 
off groups, which affects income options for the poorest. Low social cohesion can often result 
in poor infrastructure maintenance and lower labour offerings at the community level. This 
inordinately affects the poor and impoverished households who rely on casual labour for 
income. To respond to such difficult situations, poor and very poor households resort to several
strategies including removing children from school, labour migration, increased debts, or 
advance selling of labour (at up to 50% lower rates than the prevailing wage). Those strategies 
are detrimental and have lasting impacts on nutrition, education, and health. 
Unconditional Cash Transfers in Sittwe
In Sittwe IDP camps, SC proposes to provide monthly cash transfer to 1,733 HHs (8,781 
individuals) for 12 months. The transfer cash amount of $9.90/USD per month (15,000MMK)/ 
per person ($49.50/month per HH).is equivalent to the value of average combined nutritional 
ration calculated using NutVal 4.0 tool. SC proposes to provide assistance to displaced 
individuals that are not beneficiaries of the WFP General Food Distribution (GFD). 
9
Conditional Cash Transfers in Pauktaw 
SC proposes to target 500 HHs (2,250 individuals) directly, but 1,142 HHs (5,368 individual) 
will be indirectly impacted by the activities and benefit from them in both Rakhine and Muslim 
communities for CfW activities, which cover the following, scope of work at communities: 
rehabilitation of land and water resources; paddy embankments; jetties; and small village paths, 
roads, and bridges. The geographic selection criteria included conflict-affected villages with 
limited land resources and limited access to services. The household with poor and very poor 
socioeconomic status but have labour resources were target and the work schedules were 
organized to avoid overlap with the other existing seasonal labour opportunities. For the 
women beneficiaries, the appropriate work scheme will be considered and which will also need 
to not compromising with women priority on child-care responsibilities and other household 
domestic works. 
1.1 BASELINE PURPOSE & QUESTIONS
The main objective of the baseline survey is to set the benchmarks for food security outcome 
indicators, namely food consumption score (FCS), reduced coping strategies index (rCSI), and 
household hunger scale (HHS). This survey collected data regarding general household 
characteristics, consumption of household food, and utilization of negative coping strategies in 
the absence of food. 
1.1.1 BASELINE QUESTIONS
The baseline assessment utilised a quantitative approach. The assessment data collection was 
based on well-defined and structured questionnaire using Food for Peace (FFP) emergency 
food security program PIRS. Data was collected using Kobo Collect application on Android 
tablets, and processed and analysed using MS Excel and STATA (statistical software). The 
following key outcome indicators are measured as part of the baseline:
1. Food Consumption Score (FCS): Percentage of households with poor, borderline, and 
acceptable food consumption score 
2. Reduced Coping Strategy Index (rCSI): Mean, standard deviation, confidence 
interval (95%), and median 
3. Household Hunger Score (HHS): Prevalence of households with moderate or severe 
hunger 
10
The detail explanation on questionnaires and the calculation of key outcome indicators to 
answer the above questions are outlined in the next sections on methodology. 
11
2.0 BASELINE METHODOLOGY& LIMITATIONS
The baseline study used a quantitative approach; a participant-based sampling methodology 
was used for surveying households for unconditional cash transfer (UCT), while a population￾based approach was used for surveying households for cash for work (CfW) activities. In 
addition to providing baseline values for key food security outcome indicators, the baseline 
also provided benchmarks for these key indicators that will be used to measure the progress of 
the intervention. The same households surveyed as part of the baseline will be surveyed at the 
endline (pre-post panel design) to allow the team to test for difference (statistical inference). 
2.1 Sample Size Determination
The team used a beneficiary-based SRS design based on the proximity of the HHs within the 
camps in Sittwe and used a population-based SRS design for CfW in Pauktaw. The team used 
the sampling methodology detailed in the Sept. 2018 USAID Beneficiary-Based Sampling 
Guidance (see pgs. 57-59). A one-stage SRS design is advantageous because it is an equal 
probability of selection method and data is self-weighted, which is necessary to generate 
unbiased estimates. Data collection in an SRS is simpler to implement, and the resultant data 
is easier to analyse, reducing the chance of process and analytical errors. The team will use the 
same set of beneficiaries for the baseline and final evaluation (panel design), which will help 
SC better compare trends in food security indicators throughout the program. 
The team used the FFP recommended minimum sample size of 339 respondents for its one￾stage SRS sampling design for each strata – UCT and CfW – totalling 678 HHs. This is based 
on the sample sizes FFP has calculated for FCS and HHS, and uses the largest sample size 
between the two. The following table explains the parameters applied in the sample size 
determination for this baseline study. 
Table 1: Parameters and values applied in sample size determination
Parameters Parameters Explanation
P1,est 50% (0.5) The 2019 Myanmar EFSP will be working with new beneficiaries, 
and an estimate is not available for these households so we have 
used the FFP recommended value of 50% (0.5) for 
‘moderate/severe’ HHS and ‘acceptable’ FCS.
P2,est 40% (0.4) The 2019 Myanmar EFSP has set targets for a change of 15% in 
12
‘moderate/severe’ HHS and ‘acceptable’ FCS by endline, but will 
use the FFP 10% change, which would be 40% (0.4)] 
Z1−∝ 95% (1.64) the corresponding value is Z0.95 = 1.64
Z1−β 80% (0.84) the corresponding value is Z0.80 = 0.84
Dest 1 Default value for one State Simple Random Sampling
ninitial 305 Minimum sample size with no consideration on non-response rate
Non-response 
adjustments
10% 10% non-response rate were applied as default rate for survey in 
Myanmar
Nfinal/planned 339 Planned sample size
Actual used 348 and 350 for Sittwe and Pauktaw respectively.
2.2 Sampling Frame and Sample Selection
The sampling frame for the baseline study includes a list of targeted villages and targeted in 
the FFP project design. The study population includes targeted population for conditional cash 
transfer and registered programme participants for the unconditional cash transfer model. 
While a sample of 339 HHs for each site (Sittwe and Pauktaw) was targeted, the final sample 
size achieved was 698 – 350 HHs in Sittwe and 348 HHs in Pauktaw. 
 
2.3 Data Collection Tool 
The survey questionnaires were composed of four main modules as explained in the following 
table.
Table 2: Questionnaires Structure
Module Description
Respondent and 
Household 
Information
Geographical information of household, respondent demographic information and some 
basic information of the household as having mobile phone or not, the household size and 
female- or male-headed households 
Food 
Consumption 
Score
7 days diet recall of 8 food groups for food consumption score calculation
1. Main Staples
2. Pulses
3. Vegetables
4. Fruits
5. Meat/Fish
13
6. Diary
7. Oil
8. Sugar
Reduced Coping 
Strategies Index
The list of five food-related coping strategies that the household used in the seven days 
prior to the survey
1. Rely on less preferred and less expensive foods
2. Borrow food or rely on help from friends or relatives
3. Limit portion size at mealtime
4. Restrict consumption by adults in order for small children to eat
5. Reduce the number of meals eaten in a day
Household 
Hunger Scale
The frequency of three events related to food deprivation were experienced by any 
household member in the last four weeks (30 days)
1. No food at all in the house
2. Went to sleep hungry
3. Went all day and night without eating
2.4 Data Collection 
A quantitative questionnaire was used to collect data from the sampled participants. Data 
collection was done using KoBoCollect application via Android tablets. The surveyors were 
trained on survey objectives, tablet use, and the online KoBO collect programmed 
questionnaire. In addition, the surveyors were oriented on each indicator and the importance of 
data quality, and ethical principles (i.e privacy, confidentiality, professionalism, integrity). The 
data collection was completed at the end of October for Sittwe Township (UCT) and by mid￾January for Pauktaw Township (CfW). 
2.5 Data Quality Assurance
The project coordinator and the project MEAL officer, with additional support and guidance 
from headquarter-based MEAL Specialist, supervised the data collection. The built-in 
validation checks in the KoBO Collect questions were also key in ensuring complete datasets 
were collected. The collection of geo codes allowed for double checking on whether data was 
collected in the sampled villages and camps. The project team closely monitored the progress 
of data collection in the field and, prepared and shared the progress report with MEAL team 
and Yangon office. And, those field progress reported data were validating with the number of 
survey submitted to KoboToolbox servers by MEAL team. 
14
2.6 Data Analysis
The data analysis plan was developed based on review of the existing project monitoring and 
evaluation plan related and the questionnaires applied in baseline study. The baseline value 
indicator estimation was performed based on the FFP Guidance for a Baseline Study and Final 
Evaluation for Emergency Programs and each indicator was calculated for the point estimation 
with the respective standard error value and interval estimation, per the respective PIRS. 
Data cleaning and analysis were performed using Stata 15.1 statistical software. All system 
missing and human error missing were properly coded. Summary statistic for key outcome 
indicator from project M&E plan were calculated as part of descriptive analysis. Both points 
and interval estimation were performed for summary statistic figure and following formulas 
were application of 95% confidence interval estimation. Chi-square test was applied for the 
relationship of main outcome variables (category variable) and to examine the association of 
continuous variable, correlation test was performed3
. 
2.6.1 Indicator Definitions
Food Consumption Score 
The Food Consumption Score (FCS) is a proxy indicator for the food intake and it is a 
composite score based on dietary diversity, food frequency, and the relative nutritional 
importance of different food groups. The nine food groups are applied in calculation of food 
consumption score and the frequency of household consumption in last 7 days were asked in 
the questionnaires. To calculate the FCS, the consumption frequencies are summed and 
multiplied by the standardized food group weight (see table – 3 below). Based on the food 
consumption score, the World Food Programs recommended cut-off points4 were applied to 
determine the household food consumption condition into three groups; poor, borderline, or 
acceptable (see table 4). 
Table 3: FCS food groups and relative weights
Food Group Weight Food Items
3 Bruin, J. 2006. newtest: command to compute new test. UCLA: Statistical Consulting Group. 
https://stats.idre.ucla.edu/stata/ado/analysis/.
4 Vulnerability Assessment and Mapping World Food Programme. 2008. “Food consumption analysis: 
Calculation and use of the food consumption score in food security analysis.”
15
Main Staples 2 Maize, rice, sorghum, other cereals; tubers; plantains
Pulses 3 Beans, peas, groundnuts and cashew nuts
Vegetables 1 Vegetables, and leaves
Fruit 1 Fruits 
Meat/fish 4 Beef, goat, poultry, pork, eggs, and fish
Milk 4 Milk, yogurt and other dairy
Sugar 0.5 Sugar and sugar products, honey
Oil 0.5 Oils, fats, and butter
Condiments 0 Spices, tea, coffee, salt, fish powder, small amounts of milk for tea
Table 4: FCS classification thresholds
Raw Score Category
0 - 21 Poor
21.5 - 35 Borderline
> 35 Acceptable
Reduced Coping Strategies Index (rCSI) 
The reduced coping strategy index (rCSI) measure the frequency and severity of the coping 
behaviours, what people do when they cannot access enough food. It is composed with the 
series of questions about how the households manage to cope with a shortfall in food for 
consumption, and the frequency on each experience is recorded to get the simple numeric score. 
The rCSI is based on a list of five food-related coping strategies that the household used in the 
seven days prior to the survey. The rCSI raw scores are calculated by multiplying the frequency 
with which a behaviour was used by the universal severity weight, then summing the weighted 
scores for each coping strategy (see table 5). The maximum raw score for the rCSI is 56, i.e. a 
household that used all five strategies every day for the last 7 days would have a raw score of 
56. 
Table 5: Reduced CSI strategies/behaviours and their corresponding severity weights
Behaviours/Strategies Severity 
Weight 
Rely on less preferred and less expensive food 1
Borrow food or rely on help from a relative or friend 2
Limit portion size at mealtimes 1
Restrict consumption by adults in order for small children to eat 3
16
Reduce number of meals eaten in a day 1
Household Hunger Scale 
The Household Hunger Scale5 (HHS) is a food deprivation scale that measures the percent of 
households experiencing hunger. The person in the household in charge of food preparation is 
asked about the frequency with which three events were experienced by any household member 
in the last four weeks (30 days): 
1. No food at all in the house 
2. Went to bed hungry 
3. Went all day and night without eating 
If the event is reported as having not been experienced in the last four weeks, the response is 
coded as “never” (value = 0). If the event is reported as having been experienced in the last 
four weeks, a frequency of occurrence question is asked to determine how often the event was 
experienced; see table 6 for each frequency of occurrence question and possible responses. For 
tabulation purposes, the responses are then recoded into three frequency categories (table 7). 
Table 6: Household Food Deprivation Experiences
Question Response Option 
No food at all in the house
How often did this happen in the past [4 weeks/30 days]? 
1 = Rarely (1–2 times)
2 = Sometimes (3–10 times)
3 = Often (more than 10 times) 
Went to bed hungry
How often did this happen in the past [4 weeks/30 days]? 
1 = Rarely (1–2 times)
2 = Sometimes (3–10 times)
3 = Often (more than10 times) 
Went all day and night without eating
How often did this happen in the past [4 weeks/30 days]? 
1 = Rarely (1–2 times)
2 = Sometimes (3–10 times)
3 = Often (more than 10 times) 
5 Ballard, Terri; Coates, Jennifer; Swindale, Anne; and Deitchler, Megan. Household Hunger Scale: Indicator 
Definition and Measurement Guide. Washington, DC: Food and Nutrition Technical Assistance II Project, FHI 
360.
17
Values for the three questions are summed for each household, producing a HHS score ranging 
from 0 to 6. The raw scores are then tabulated into the following categories. 
Table 7: HHS Categorical Indicator
Household Hunger Score Household Hunger Categories 
0–1 Little to no hunger in the household 
2–3 Moderate hunger in the household 
4–6 Severe hunger in the household 
2.7 Limitations
Due to complexities in the implementation area, majorly longer approval processes, some 
limitations were experienced and addressed accordingly to ensure successful and timely 
execution of the baseline. First and foremost, by the time the baseline was commissioned, the 
programme could not complete selection of beneficiaries in Pauktaw area. Therefore, the 
programme adopted a population based sampling methodology. This was not anticipated as 
beneficiaries’ selection in Sittwe was successful and baseline was completed based on targeted 
beneficiaries sampling. This affects regional comparisons of outcome indicators, and the data 
has been analysed and discussed separately for Sittwe and Pauktaw. 
3.0 FINDINGS, CONCLUSIONS & RECOMMENDATIONS
Because the UCT survey and CfW look at different populations – UCT surveyed program 
beneficiaries while CfW conducted a population-based survey – the analyses have been 
separated by activity type and no comparisons has been drawn between the two populations. 
3.1 FINDINGS – SITTWE 
Respondent Information 
There were 350 respondents surveyed from Sittwe, who are beneficiaries of UCTs. 
Approximately 44% of the respondents were from Dar Paing camp and the remaining 56% 
from Say Tha Mar Gyi camp. All respondents from the Sittwe township consented to 
participate in the survey. The average ages of the respondents is 33.4 years and the average 
household size of the respondents’ is 5.2 members. Only 5.7% of participants from Sittwe 
18
reported that their household owned a mobile phone. Approximately one-third (32.9%) of the 
respondents reported that their households were headed by females. 
Table 8: Respondent information – Sittwe – Unconditional Cash Transfer
95% CI
 n % sd lb ub N 
Respondent Information 
Provided consent yes 350 100.0% - 100.0% 100.0% 350 
Female Respondent 245 70.0% 0.5 65.2% 74.8% 350 
Male Respondent 105 30.0% 0.5 25.2% 34.8% 350 
Respondent age (average) 33.4 13.2 32.0 34.8 350 
Respondent is Household Head 180 51.4% 0.5 46.2% 56.7% 350 
Household Information 
Household Size (average) 5.2 1.7 5.0 5.3 350 
Household with mobile phone 20 5.7% 0.2 3.3% 8.2% 350 
Female-headed HH 115 32.9% 0.5 27.9% 37.8% 350 
Male-headed HH 235 67.1% 0.5 62.2% 72.1% 350 
Food Consumption Score 
The household’s food consumption score measures food security through a aggregating food 
intake relative to dietary diversity, food frequency and nutrition values (see section 2.6.1 for 
detailed definition). WFP clarifies households as having food security when the cut-off point 
>35, acceptable which is interpreted to mean that daily per capita calories of around 1950 
kilocalories. While a cut-off point below < 21 is classified as poor food security, a cut of 21-
35 is considered to be on the borderline. The baseline findings indicates that 52% of the HHs 
are experiencing unacceptable food security level with an average FCS score of 35. 
Specifically, approximately 9% of households have poor food security, 43% on the borderline,
and 48% have an ‘acceptable’ FCS (Table 9 below). On average, over 52% of the sampled 
HHs are vulnerable to poor food quality and insufficient quantities. This could mean that some 
HHs are food poor and could be having lower than recommended average dietary energy 
consumptions. Descriptively, very few households (3%) had intake of vitamin A and proteins 
from consumed dairy groups in the week prior to the survey. While majority of the households 
had proteins and carbohydrates (98%) intake from beef and fish (83%), very high number of 
households (approximately 75%) were consuming fats and oils (57%) and sugars (93%) which 
have low nutritional value. The second least consumed food group was fruits group (17%). 
19
The high intake of fish could be related to fishing practices among the Rakhine community due 
to proximity to the Bay of Bengal6
. The high intake of vegetables could be due to seasonality 
as most vegetables are widely available in the winter (October to February). In addition to the 
availability of vegetables, staples like rice and fish are quite abundant in the local markets as 
well 7
. 
Table 9: Food Consumption Score for Sittwe – Unconditional Cash Transfer
95% CI
 n % sd lb ub N 
Food Consumption Score
Household Food consumption score (mean) 35.5 11.82 34.3 36.8 350
HH with Acceptable FCS 168 48.0% 0.50 42.7% 53.3% 350
HH with Borderline FCS 152 43.4% 0.50 38.2% 48.6% 350
HH with Poor FCS 30 8.6% 0.28 5.6% 11.5% 350
IPC on the basis of FCS
HHs in IPC 1/2 - Minimal/Stressed 168 48.0% 0.50 42.7% 53.3% 350
HHs in IPC 3 - Crisis 152 43.4% 0.50 38.2% 48.6% 350
HHs in IPC 4/5 - Emergency/Famine 30 8.6% 0.28 5.6% 11.5% 350
Table 10: Household Food Consumption Practices for Sittwe
95% CI
 n % sd lb ub N 
Household Food Consumption –
proportion of HHs reporting consuming 
each food group in the last week
Main Staples 344 98.3% 0.1 96.9% 99.7% 350
Pulses 155 44.3% 0.5 39.1% 49.5% 350
Vegetables 344 98.3% 0.1 96.9% 99.7% 350
Fruits 60 17.1% 0.4 13.2% 21.1% 350
Meat/Fish 290 82.9% 0.4 78.9% 86.8% 350
Diary 12 3.4% 0.2 1.5% 5.3% 350
Oil 199 56.9% 0.5 51.6% 62.1% 350
Sugar 325 92.9% 0.3 90.1% 95.6% 350
Household Food Consumption Frequency (mean)
Main Staples 6.4 1.6 6.2 6.5 350
Pulses 1.0 1.4 0.9 1.2 350
Vegetables 4.5 2.1 4.3 4.8 350
Fruits 0.4 1.1 0.3 0.5 350
Meat/Fish 2.8 2.2 2.6 3.0 350
6 Tat Lan Sustainable Food Security and Livelihoods Program Baseline Assessment, November 2014
7 Three Livelihood Zones in Tatlan Project Areas of Rakhine State, Myanmar. Assessed Using the Household 
Economy Approach (HEA) November 2013
20
Diary 0.1 0.4 0.0 0.1 350
Oil 1.9 2.1 1.7 2.1 350
Sugar 5.0 2.3 4.8 5.2 350
The same patterns of consumption were also observed for the frequency of each food group. 
While 90% of HHs reported having consumed meat/fish in the week prior to the survey, they 
were only consumed approximately 3 out of 7 days. In Sittwe, HHs reported having consumed 
main staples and sugar almost every day of their household diet and vegetable was only 
consumed 4 days a week. 
Table 11: Table 11: Household Food Consumption Practices for Sittwe by HH Head Gender Status
Women Lead HH Men Lead HH
 n % N n % N 
Household Food Consumption
Main Staples 115 100.0% 115 229 97.4% 235
Pulses 51 44.3% 115 104 44.3% 235
Vegetables 113 98.3% 115 231 98.3% 235
Fruits * 13 11.3% 115 47 20.0% 235
Meat/Fish * 88 76.5% 115 202 86.0% 235
Diary 1 0.9% 115 11 4.7% 235
Oil 58 50.4% 115 141 60.0% 235
Sugar 106 92.2% 115 219 93.2% 235
Household Food Consumption Frequency (mean)
Main Staples 6.6 115 6.3 235
Pulses 1.0 115 1.0 235
Vegetables 4.6 115 4.5 235
Fruits 0.3 115 0.5 235
Meat/Fish ** 2.3 115 3.0 235
Diary 0.0 115 0.1 235
Oil 1.5 115 2.1 235
Sugar * 4.7 115 5.1 235
Household Food Consumption –
proportion of HHs reporting consuming 
each food group in the last week
Household Food consumption score (mean)* 33.4 115 36.6 235
HH with Acceptable FCS 47 40.9% 115 121 51.5% 235
HH with Borderline FCS 57 49.6% 115 95 40.4% 235
HH with Poor FCS 11 9.6% 115 19 8.1% 235
* p < 0.05, ** p < 0.01
The proportions of household in each food consumption score categories were not significant 
by household head gender, but the male lead household had better off food consumption score 
21
compare to female-headed HHs. In general, male headed household had better consumption on 
fruits, meat/fish and sugar food groups. The household size did not explain the condition of 
household food condition score. 
Reduced Coping Strategies Index (rCSI) 
Households use different types of coping strategies when they do not have food or the resources 
for purchasing food. The rCSI reflects both the frequency of each strategy or behaviour and 
the severity. The average rCSI score for HHs in Sittwe is 16; approximately 36% of HHs 
displayed characteristics consistent with IPC8 Phase 3/4/5 (crisis/emergency/famine)
indicating prevalence of food consumption gaps that are reflected by high or above-usual acute 
malnutrition; or are marginally able to meet minimum food needs but only by depleting 
essential livelihood assets or through crisis-coping strategies. 59% of HHs displayed 
characteristics consistent with IPC Phase 2 (stressed) indicating that approximately 6 out of 10 
households surveyed have minimally acceptable food consumption with limited capacity to 
afford essential non-food expenditures (See table 12 for details). Only 5% of the surveyed 
households are able to meet essential food and non-food needs without engaging in atypical 
and unsustainable strategies to access food and income (IPC 1 (minimal). While nearly half of 
the respondents had an ‘acceptable’ FCS, meaning these respondents have access to and 
consume diverse foods, likely in small quantities, because the rCSI highlights the continued 
challenges the HH may be facing in consistently consuming these foods. An average of 74% 
of the sampled HHs were using negative coping mechanisms (Table 13). With approximately 
67% relying on the two most severe negative coping strategies of restricting food consumption 
by adult for children to eat and borrowing food. Specifically, approximately 88% of the 
surveyed HHs reported consuming less expensive and less preferred foods items, while 76% 
either borrowed food from a friend or relative or limiting food portions during meal times. This 
indicates the vulnerability of household to food insecurity given that the available food is 
insufficient to sustain the entire household size. This behaviours could be related to limited 
8 IPC 5: Households have an extreme lack of food and/or other basic needs even after full employment of coping 
strategies. Starvation, death, destitution, and extremely critical acute malnutrition levels are evident. IPC 4: 
Households either have large food consumption gaps, which are reflected in very high acute malnutrition and 
excess mortality; or are able to mitigate large food consumption gaps but only by employing emergency livelihood 
strategies and asset liquidation. IPC 3: Households either have food consumption gaps that are reflected by high 
or above-usual acute malnutrition; or are marginally able to meet minimum food needs but only by depleting 
essential livelihood assets or through crisis-coping strategies. IPC 2: Households have minimally acceptable food 
consumption but are unable to afford some essential non-food expenditures without engaging in stress-coping 
strategies. IPC 1: Households are able to meet essential food and non-food needs without engaging in atypical 
and unsustainable strategies to access food and income. 
22
foods availability at HHs level. FFP cash support might cushion the households from applying 
these coping mechanisms. 
 
Table 12: rCIS in Sittwe – Unconditional Cash Transfer
95% CI
 n % sd lb ub N 
Reduced Coping Strategies Index Score
Average rCSI (mean) 16.0 8.7 15.1 16.9 350
HH with above CSI Score 19 points 116 33.1% 0.5 28.2% 38.1% 350
IPC on the basis of rCSI
HHs in IPC 1 - Minimal 19 5.4% 0.2 3.0% 7.8% 350
HHs in IPC 2 - Stressed 205 58.6% 0.5 53.4% 63.8% 350
HHs in IPC 3/4/5 - Crisis/Emergency/Famine 126 36.0% 0.5 30.9% 41.1% 350
Table 13: Household Coping Strategies in Sittwe – Unconditional Cash Transfer
95% CI
 n % sd lb ub N 
Household Usage of Each Coping Strategy
Rely on less preferred and less expensive foods 309 88.3% 0.3 84.9% 91.7% 350
Borrow food, or rely on help from a friend or relative 268 76.6% 0.4 72.1% 81.0% 350
Limit portion size at mealtimes 267 76.3% 0.4 71.8% 80.8% 350
Restrict consumption by adults in for small children to eat 202 57.7% 0.5 52.5% 62.9% 350
Reduce number of meals eaten in a day 248 70.9% 0.5 66.1% 75.6% 350
Household Usage of Each Coping Strategy Frequency (Mean)
Rely on less preferred and less expensive foods 3.1 1.9 2.9 3.3 350
Borrow food, or rely on help from a friend or relative 2.0 1.6 1.8 2.1 350
Limit portion size at mealtimes 2.1 1.5 1.9 2.2 350
Restrict consumption by adults for small children to eat 1.7 1.8 1.5 1.9 350
Reduce number of meals eaten in a day 1.8 1.5 1.7 2.0 350
The rCSI score and coping strategy behaviours were not significant different between men 
and female-headed HHs. And, the household size also did not determine the rCSI score. 
Table 14: Household Coping Strategies in Sittwe by HH Head Gender Status
Female headed 
HH Male headed HH
 n % N n % N 
Household Usage of Each Coping Strategy
Rely on less preferred and less expensive foods 102 88.7% 115 207 88.1% 235
Borrow food, or rely on help from a friend or relative 86 74.8% 115 182 77.4% 235
Limit portion size at mealtimes 88 76.5% 115 179 76.2% 235
Restrict consumption by adults in for small children to 
eat 60 52.2% 115 142 60.4% 235
23
Reduce number of meals eaten in a day 87 75.7% 115 161 68.5% 235
Reduced Coping Strategies Index Score
Average rCSI (mean) 15.5 115 16.3 235
HH with above CSI Score 19 points 34 29.6% 115 82 34.9% 235
IPC on the basis of rCSI
HHs in IPC 1 - Minimal 5 4.3% 115 14 6.0% 235
HHs in IPC 2 - Stressed 71 61.7% 115 134 57.0% 235
HHs in IPC 3/4/5 - Crisis/Emergency/Famine 39 33.9% 115 87 37.0% 235
* p < 0.05, ** p < 0.01
Household Hunger Scale (HHS) 
The HHS observes the prevalence of hunger at household level. The results of this survey show 
that approximately half (51%) of IDP HHs experienced severe or moderate hunger and 49% 
experienced little to no hunger in the month prior to the survey. Results also show that 
approximately 1% of HHs exhibited characteristics consistent with IPC Phase 5 (famine), 3% 
displayed characterises with IPC 4 (emergency), 47% with IPC 3 (crisis), 31% with IPC 2 
(stressed) and 17% with IPC 1 (minimal).
Table 15: HHS in Sittwe – Unconditional Cash Transfer
95% CI
 n % sd lb ub N 
Household Hunger Experience (In the past 30 days)
No food to eat 202 57.7% 0.5 52.5% 62.9% 350
Go to sleep at night hungry 219 62.6% 0.5 57.5% 67.7% 350
Go a whole day and night without eating anything 105 30.0% 0.5 25.2% 34.8% 350
Household Hunger Scale (mean score) 
Little to no hunger 48.6% 0.5 43.3% 53.8% 350
Moderate hunger 46.9% 0.5 41.6% 52.1% 350
Severe hunger 4.6% 0.2 2.4% 6.8% 350
IPC on the basis of HHS
HHs in IPC 1 - Minimal 61 17.4% 0.4 13.4% 21.4% 350
HHs in IPC 2 - Stressed 109 31.1% 0.5 26.3% 36.0% 350
HHs in IPC 3 - Crisis 164 46.9% 0.5 41.6% 52.1% 350
HHs in IPC 4 - Emergency 12 3.4% 0.2 1.5% 5.3% 350
HHs in IPC 5 - Famine 4 1.1% 0.1 0.0% 2.3% 350
More male-headed HHs less experienced household hunger conditions as 52% of male 
headed household had little to no hunger, but the female-headed HHs had only 41% of 
household with little or no hunger. In term of IPC phase 3 – Crisis, the proportion of male 
lead household were 11.8 percent points less than female-headed HHs. 
24
Table 16: Table 15: HHS in Sittwe – Unconditional Cash Transfer by HH Head Gender Status
Women Lead HH Men Lead HH
 n % N n % N 
Household Hunger Experience (In the past 30 days)
No food to eat 71 61.7% 115 131 55.7% 235
Go to sleep at night hungry 77 67.0% 115 142 60.4% 235
Go a whole day and night without eating anything 38 33.0% 115 67 28.5% 235
Household Hunger Scale (mean score)
Little to no hunger * 47 40.9% 115 123 52.3% 235
Moderate hunger * 63 54.8% 115 101 43.0% 235
Severe hunger 5 4.3% 115 11 4.7% 235
IPC on the basis of HHS
HHs in IPC 1 - Minimal 16 13.9% 115 45 19.1% 235
HHs in IPC 2 - Stressed 31 27.0% 115 78 33.2% 235
HHs in IPC 3 - Crisis * 63 54.8% 115 101 43.0% 235
HHs in IPC 4 - Emergency 4 3.5% 115 8 3.4% 235
HHs in IPC 5 - Famine 1 0.9% 115 3 1.3% 235
 * p < 0.05, ** p < 0.01
There was no associate with household size and household hunger scale. 
3.2 FINDINGS – PAUKTAW 
Respondent Information 
The conditional cash transfer program (CfW) surveyed 348 respondents (a mix of beneficiaries 
and non-beneficiaries) from the township of Pauktaw. Two respondents did not consent to 
participate in the survey from Pauktaw Township. The average age of the respondents was 39.3 
years and the average household size was 5.1. Approximately 15% of participants from 
Pauktaw reported that their household owned a mobile phone, and approximately 23% of the 
respondents reported that their household were headed by females. 
Table 17: Respondent information – Pauktaw – Conditional Cash Transfer
95% CI
 n % sd lb ub N 
Respondent Information 
Provided consent yes 346 99.4% 0.1 98.6% 100.2% 348 
Female Respondent 223 64.5% 0.5 59.4% 69.5% 346 
Male Respondent 123 35.5% 0.5 30.5% 40.6% 346 
Respondent age (average) 39.3 13.5 37.9 40.8 346 
25
Respondent is Head of HH 172 49.7% 0.5 44.4% 55.0% 346 
Household Information 
Household Size (average) 5.1 2.1 4.9 5.3 346 
Household with mobile phone 51 14.7% 0.4 11.0% 18.5% 346 
Female-headed HH 80 23.1% 0.4 18.7% 27.6% 346 
Male-headed HH 266 76.9% 0.4 72.4% 81.3% 346 
Food Consumption Score 
The household’s food consumption score measures food security through a aggregating food 
intake relative to dietary diversity, food frequency and nutrition values (see section 2.6.1 for 
detailed definition). WFP clarifies households as having food security when the cut-off point 
>35, acceptable which is interpreted to mean that daily per capita calories of around 1950 
kilocalories. While a cut-off point below < 21 is classified as poor food security, a cut of 21-
35 is considered to be on the borderline. On average, the sampled host community in Pauktaw 
are on the borderline (32.6 FCS). Approximately 20% of households have poor food security, 
implying limited access to food and insufficient intake of nutritious foods. 39% have a 
‘borderline’ FCS, and 41% have an ‘acceptable’ FCS (Table 18). Over 59% of the surveyed 
households have unacceptable food daily per capita calories intake. Descriptively, the most 
consumed (91% of the HHs) highly nutritious food (protein and hem iron rich) items include
meat/fish. Very few households consumed dairy products (8%) the two week prior to the 
survey, despite this being a main source of vitamin A and protein. The second nutritional value 
ranking consumed food item was main staples (~94%) and the average number of days for 
most household consumed food, vegetables, meat/fish and sugar, were 3 days a week in prior 
week, except main staple which was consumed 5 days in average (Table 19). 
Table 18: Food Consumption Score for Pauktaw – Conditional Cash Transfer
95% CI
 n % sd lb ub N 
Food Consumption Score
Household FCS (mean) 32.6 14.11 31.1 34.1 346
HH with Acceptable FCS 142 0.49 0.03 35.8% 46.2% 346
HH with Borderline FCS 135 0.49 0.03 33.9% 44.2% 346
HH with Poor FCS 69 0.40 0.02 15.7% 24.2% 346
IPC on the basis of FCS
HHs in IPC 1/2 - Minimal/Stressed 142 0.49 0.03 35.8% 46.2% 346
HHs in IPC 3 - Crisis 135 0.49 0.03 33.9% 44.2% 346
26
HHs in IPC 4/5 - Emergency/Famine 69 0.40 0.02 15.7% 24.2% 346
Table 19: Household Food Consumption Practices for Pauktaw
95% CI
 n % sd lb ub N 
Household Food Consumption
Main Staples 325 93.9% 0.2 91.4% 96.5% 346
Pulses 181 52.3% 0.5 47.0% 57.6% 346
Vegetables 341 98.6% 0.1 97.3% 99.8% 346
Fruits 100 28.9% 0.5 24.1% 33.7% 346
Meat/Fish 315 91.0% 0.3 88.0% 94.1% 346
Diary 28 8.1% 0.3 5.2% 11.0% 346
Oil 154 44.5% 0.5 39.2% 49.8% 346
Sugar 314 90.8% 0.3 87.7% 93.8% 346
Household Food Consumption Frequency (mean)
Main Staples 5.2 2.4 5.0 5.5 346
Pulses 1.1 1.3 1.0 1.3 346
Vegetables 3.3 1.5 3.2 3.5 346
Fruits 0.5 0.9 0.4 0.6 346
Meat/Fish 3.0 1.8 2.8 3.2 346
Diary 0.1 0.5 0.1 0.2 346
Oil 1.0 1.5 0.9 1.2 346
Sugar 3.5 2.1 3.3 3.7 346
The proportion of man lead household with Poor FCS (20.7%) was higher than female-headed 
HHs (17.5%). Among individual food group consumption, the female-headed HHs consumed 
pulses more than male lead household. 
27
Table 20: Household Food Consumption Practices for Pauktaw by HH Head Gender Status
Female-Led HH Male-Led HH
 n % N n % N 
Household Food Consumption
Main Staples 76 95.0% 80 249 93.6% 266
Pulses ** 53 66.3% 80 128 48.1% 266
Vegetables 79 98.8% 80 262 98.5% 266
Fruits 18 22.5% 80 82 30.8% 266
Meat/Fish 72 90.0% 80 243 91.4% 266
Diary 5 6.3% 80 23 8.6% 266
Oil 34 42.5% 80 120 45.1% 266
Sugar 73 91.3% 80 241 90.6% 266
Household Food Consumption Frequency (mean)
Main Staples 5.3 80 5.2 266
Pulses ** 1.6 80 1.0 266
Vegetables 3.3 80 3.3 266
Fruits 0.4 80 0.5 266
Meat/Fish 3.0 80 3.0 266
Diary 0.1 80 0.1 266
Oil 0.9 80 1.1 266
Sugar 3.6 80 3.5 266
Food Consumption Score
Household Food consumption score (mean) 34.1 80 32.2 266
HH with Acceptable FCS 42 52.5% 80 100 37.6% 266
HH with Borderline FCS 24 30.0% 80 111 41.7% 266
HH with Poor FCS * 14 17.5% 80 55 20.7% 266
* p < 0.05, ** p < 0.01
The positive medium correlation was identified between household size and household food 
consumption score in female-headed HHs (r = 0.3). However, it cannot literally be translated 
that more household members in female-headed HHs can have better household consumption 
score. Because, the female-headed HHs had less household members than male-headed HHs
in Pauktaw township. 
Reduced Coping Strategies Index (rCSI) 
In Pauktaw, the average rCSI is 10; approximately 17% of HHs displayed characteristics 
consistent with IPC9 Phase 3/4/5 (crisis/emergency/famine), 54% of HHs displayed 
characteristics with IPC Phase 2 (stressed), and 30% with IPC 1 (minimal) (Table 21). 
9 IPC 5: Households have an extreme lack of food and/or other basic needs even after full employment of 
coping strategies. Starvation, death, destitution, and extremely critical acute malnutrition levels are evident. IPC 
4: Households either have large food consumption gaps, which are reflected in very high acute malnutrition and 
excess mortality; or are able to mitigate large food consumption gaps but only by employing emergency 
28
Table 21: rCIS in Pauktaw – Conditional Cash Transfer
95% CI
 n % sd lb ub N 
Reduced Coping Strategies Index Score
Average rCSI (mean) 10.0 9.4 9.0 11.0 346
HH with above CSI Score 19 points 51 14.7% 0.4 11.0% 18.5% 346
IPC on the basis of rCSI
HHs in IPC 1 - Minimal 102 29.5% 0.5 24.7% 34.3% 346
HHs in IPC 2 - Stressed 187 54.0% 0.5 48.8% 59.3% 346
HHs in IPC 3/4/5 - Crisis/Emergency/Famine 57 16.5% 0.4 12.5% 20.4% 346
Although almost half of the household had acceptable FCS, there are struggles with some 
coping behaviours to attained that FCS condition. For instance, majority of households (80%) 
relied on less preferred and less expensive foods, 60% borrowed food, 40% limited portion 
sizes, and approximately one in three households restricted consumption of adults and 
prioritized children’s consumption and/or reduced the frequency or number of meals in a day. 
Table 22: Household Usage of Negative Coping Strategies in Pauktaw
95% CI
 n % sd lb ub N 
Household Usage of Each Coping Strategy
Rely on less preferred and less expensive foods 275 79.5% 0.4 75.2% 83.8% 346
Borrow food, or rely on help from a friend or relative 211 61.0% 0.5 55.8% 66.1% 346
Limit portion size at mealtimes 139 40.2% 0.5 35.0% 45.4% 346
Restrict consumption by adults for small children to eat 129 37.3% 0.5 32.2% 42.4% 346
Reduce number of meals eaten in a day 115 33.2% 0.5 28.2% 38.2% 346
Household Usage of Each Coping Strategy Frequency (Mean)
Rely on less preferred and less expensive foods 2.5 1.8 2.3 2.6 346
Borrow food, or rely on help from a friend or relative 1.3 1.3 1.2 1.5 346
Limit portion size at mealtimes 1.0 1.5 0.9 1.2 346
Restrict consumption by adults for small children to eat 1.0 1.6 0.8 1.2 346
Reduce number of meals eaten in a day 0.9 1.4 0.7 1.0 346
livelihood strategies and asset liquidation. IPC 3: Households either have food consumption gaps that are 
reflected by high or above-usual acute malnutrition; or are marginally able to meet minimum food needs but 
only by depleting essential livelihood assets or through crisis-coping strategies. IPC 2: Households have 
minimally acceptable food consumption but are unable to afford some essential non-food expenditures without 
engaging in stress-coping strategies. IPC 1: Households are able to meet essential food and non-food needs 
without engaging in atypical and unsustainable strategies to access food and income. 
29
The average rCSI score for male-headed HHs (9.2) was lower than female-headed HHs (12.9). 
The female-headed HHs were practicing more on some coping behaviours; relied on less 
preferred and less expensive foods, limited portion sizes, and reduced the frequency or number 
of meals in a day. 
Table 23: Household Usage of Negative Coping Strategies in Pauktaw by HH Head Gender Status
Women Lead HH Men Lead HH
 n % N n % N 
Household Usage of Each Coping Strategy
Rely on less preferred and less expensive foods * 71 88.7% 80 204 76.7% 266
Borrow food, or rely on help from a friend or relative 56 70.0% 80 155 58.3% 266
Limit portion size at mealtimes * 40 50.0% 80 99 37.2% 266
Restrict consumption by adults for small children to eat 35 43.8% 80 94 35.3% 266
Reduce number of meals eaten in a day * 34 42.5% 80 81 30.5% 266
Household Usage of Each Coping Strategy Frequency (Mean)
Rely on less preferred and less expensive foods ** 3.0 80 2.3 266
Borrow food, or rely on help from a friend or relative * 1.7 80 1.3 266
Limit portion size at mealtimes ** 1.5 80 0.9 266
Restrict consumption by adults for small children to eat 1.3 80 0.9 266
Reduce number of meals eaten in a day 1.1 80 0.8 266
Reduced Coping Strategies Index Score
Average rCSI (mean) ** 12.9 80 9.2 266
HH with above CSI Score 19 points 18 22.5% 80 33 12.4% 266
IPC on the basis of rCSI
HHs in IPC 1 - Minimal 17 21.3% 80 85 32.0% 266
HHs in IPC 2 - Stressed 45 56.3% 80 142 53.4% 266
HHs in IPC 3/4/5 - Crisis/Emergency/Famine 18 22.5% 80 39 14.7% 266
* p < 0.05, ** p < 0.01
Household size did not explain the household rCSI score. 
Household Hunger Scale 
The results show that 71% of households from Pauktaw experienced little to no hunger in the 
month prior to the survey. Results also show that approximately 1% of HHs exhibited 
characteristics consistent with IPC Phase 5 (famine), 3% displayed characterises with IPC 4 
(emergency), 25% with IPC 3 (crisis), 37% with IPC 2 (stressed) and 34% with IPC 1 
(minimal).
30
Table 24: HHS in Pauktaw – Conditional Cash Transfer
95% CI
 n % sd lb ub N 
Household Hunger Experience (In the past 30 days)
No food to eat 228 65.9% 0.5 60.9% 70.9% 346
Go to sleep at night hungry 94 27.2% 0.4 22.5% 31.9% 346
Go a whole day and night without eating anything 56 16.2% 0.4 12.3% 20.1% 346
Household Hunger Scale (mean score) 1.2 1.2 1.0 1.3 346
Little to no hunger 246 71.1% 0.5 66.3% 75.9% 346
Moderate hunger 85 24.6% 0.4 20.0% 29.1% 346
Severe hunger 15 4.3% 0.2 2.2% 6.5% 346
IPC on the basis of HHS
HHs in IPC 1 - minimal 117 33.8% 0.5 28.8% 38.8% 346
HHs in IPC 2 - stressed 129 37.3% 0.5 32.2% 42.4% 346
HHs in IPC 3 - crisis 85 24.6% 0.4 20.0% 29.1% 346
HHs in IPC 4 - emergency 11 3.2% 0.2 1.3% 5.0% 346
HHs in IPC 5 - famine 4 1.2% 0.1 0.0% 2.3% 346
There was no significant difference between male- and female-headed households on 
household hunger scale. However, female-headed households were experiencing more on the 
two out of three hunger experiences compared to households headed by men: women-headed 
households were more likely to go to sleep at night hungry and/or went a whole day and night 
without eating anything in the month before survey. 
Table 25: HHS in Pauktaw – Conditional Cash Transfer by HH Head Gender Status
Women Lead HH Men Lead HH
 n % N n % N 
Household Hunger Experience (In the past 30 days)
No food to eat 52 65.0% 80 176 66.2% 266
Go to sleep at night hungry * 30 37.5% 80 64 24.1% 266
Go a whole day and night without eating anything ** 22 27.5% 80 34 12.8% 266
Household Hunger Scale (mean score)
Little to no hunger 52 65.0% 80 194 72.9% 266
Moderate hunger 25 31.3% 80 60 22.6% 266
Severe hunger 3 3.8% 80 12 4.5% 266
IPC on the basis of HHS
HHs in IPC 1 - Minimal 28 35.0% 80 89 33.5% 266
HHs in IPC 2 - Stressed 24 30.0% 80 105 39.5% 266
HHs in IPC 3 - Crisis 25 31.3% 80 60 22.6% 266
HHs in IPC 4 - Emergency 2 2.5% 80 9 3.4% 266
HHs in IPC 5 - Famine 1 1.3% 80 3 1.1% 266
31
* p < 0.05, ** p < 0.01
There was no association of number of household members and household hunger scale. 
3.3 Relationship Among FCS, rCSI and HHS 
Sittwe – Unconditional Cash Transfer Program 
The results explained that households with borderline food consumption status were
significantly higher in practicing coping strategy behaviours compared to the acceptable group
(p < 0.01). There may be some potential explanation that some household with borderline FCS 
score were exercising coping mechanisms to reach the higher levels of food consumption in 
their household. In the meantime, the households with poor FCS scores may have had very 
limited resources or capacity to practice negative coping strategies; for that reason, they had 
lower food consumption and scored lower rCSI score at the same time.
Figure 1: rCSI score by FCS Status in Sittwe
The severe hunger household had lowest household food consumption score compared to 
household which were little to no hunger and/or moderate hunger household. (p < 0.01). The 
household severe hunger had consumed even less than a day in the week prior survey; pulses 
(0.9), fruits (0.3), diary (0.3). Vegetable was consumed approximately 5 days week prior a 
week before data collection, but meat and fish were consumed only 2 days. 
32
Figure 2: Food Consumption Score by HH Hunger Status in Sittwe
The similar pattern was also observed with the relationship between the coping strategies index 
score and household hunger status. Households with ‘severe’ hunger status had the highest 
rCSI scores (p < 0.01), and those household applied approximately 3 days in the week prior 
data collection for all the coping strategy behaviours, expect for the strategy ‘relied on less 
preferred and less expensive foods’ (4 days). 
Figure 3: rCSI score by HH hunger status in Sittwe
33
Pauktaw – Conditional Cash Transfer Program - CfW
The average rCSI score for households with ‘borderline’ FCS score was 12.2, a significantly 
higher rCSI score than the FCS ‘acceptable’ group (p < 0.01). The poor FCS score households 
had average 9.8 rCSI score, but not significantly different from the other remaining two groups. 
Figure 4: rCSI score by FCS Status in Pauktaw
Households from three different HHS groups had no difference on average FCS score. 
Figure 5: Food Consumption Score by HH Hunger Status in Pauktaw
34
The household with little to no hunger had the lowest average rCSI score (6.2), significantly 
lower than the rCSI score of moderate hunger households (18.7) and severe hunger households 
(22.5) (p < 0.091). Households with the severe hunger score applied approximately 3 days in 
the week prior data collection for all coping strategy behaviours. 
Figure 6: rCSI score by HH hunger status in Pauktaw
3.4 CONCLUSIONS 
Sittwe – Unconditional Cash Transfer Program 
Approximately one in every two households were not in the condition of little to no hunger, 
and female-headed HHs had a high possibility of suffering from household hunger status. 
These households had the lowest household food consumption score and more frequently used 
coping strategy behaviours. ‘Severe’ and/or ‘moderate’ hunger household’s diet can be 
determined as having consumed fruit and diary for less than a day, pulses for approximately 
one day, and fish and meat for only 2 days, where there were approximately 5 days a week 
where households ate vegetables and main staple foods. 
In general, a very small proportion of households from Sittwe consumed dairy (3%), and fruits 
followed in second position at 17%. Fish is common meal in this area; approximately one in 
every two households consumed it. Household head gender status also had influence on HHS 
and FCS. The male-headed households were in better off conditions compared to female-
35
headed HHs for both of these indicators, although household size did not seem to have any 
impact on these conditions. 
While nearly half of the respondents had an ‘acceptable’ FCS (meaning these respondents have 
access to and consume diverse foods), the rCSI highlights the continued challenges households 
may be facing in consistently consuming these foods. Additionally, each food group is 
weighted differently, which can skew the scores if higher weighted foods are common in the 
IDP diet – such as fish and vegetables – and, therefore, consumed more regularly. Even if these 
foods are only consumed a couple of days a week, it can result in a higher overall FCS. 
From the different food security indicators (FCS, rCSI and HHS), the results show that 
approximately 50% of HHs exhibited characteristics consistent with IPC Phase 4 (emergency) 
and/or IPC10 Phase 5 (famine), and there is only a small proportion of households with IPC 1 
(minimal). 
Table 26:HHs with IPC phase by three different Food Security Indicators in Sittwe
IPC on the basis of HHS IPC on the basis of rCSI IPC on the basis of FCS
HHs in IPC 1 – minimal 17.4% 5.4% 48.0%
HHs in IPC 2 – stressed 31.1% 58.6%
HHs in IPC 3 - crisis 46.9%
36.0%
43.4%
HHs in IPC 4 - emergency 3.4% 8.6%
HHs in IPC 5 - famine 1.1%
Pauktaw – Conditional Cash Transfer Program - CfW 
Approximately one in every three households were not in the condition of little to no hunger, 
and female-headed HHs experienced higher percentage of individual HHS’s behaviours, such 
as going to sleep at night with hungry and/or going a whole day and night without eating 
anything in the month before survey. Those households also had higher rCSI score. The sever 
and/or moderate hunger household’s diet can be determined as consumed pulses, fruit and diary 
10 IPC 5: Households have an extreme lack of food and/or other basic needs even after full employment of 
coping strategies. Starvation, death, destitution, and extremely critical acute malnutrition levels are evident. IPC 
4: Households either have large food consumption gaps, which are reflected in very high acute malnutrition and 
excess mortality; or are able to mitigate large food consumption gaps but only by employing emergency 
livelihood strategies and asset liquidation. IPC 3: Households either have food consumption gaps that are 
reflected by high or above-usual acute malnutrition; or are marginally able to meet minimum food needs but 
only by depleting essential livelihood assets or through crisis-coping strategies. IPC 2: Households have 
minimally acceptable food consumption but are unable to afford some essential non-food expenditures without 
engaging in stress-coping strategies. IPC 1: Households are able to meet essential food and non-food needs 
without engaging in atypical and unsustainable strategies to access food and income. 
36
for less than a day, vegetable and fish/meat for 3 days and main staple foods for approximately 
5 days a week. 
In general, a very small proportion of households from Pauktaw consumed diary (8%), 
followed by the fruits group at 29%. Household head gender status also had influence on HHS 
and FCS. The man-headed households were in better off conditions for those indicators 
compared to female-headed HHs, and again household size did not have any relationship to
those conditions. 
Although almost half of the household had acceptable FCS, there were struggles with some 
coping behaviours to attain these FCS condition. The majority of households (80%) relied on 
less preferred and less expensive foods, 60% borrowed food, 40% limited portion sizes, and 
approximately one in three households restricted consumption of adults and prioritized 
children’s consumption and/or reduced the frequency or number of meals in a day. Especially 
in female-headed HHs, some coping behaviours were more commonly used, including ‘relied 
on less preferred and less expensive foods’, ‘limited portion sizes’, and ‘reduced the frequency 
or number of meals in a day’. 
From the different food security indicators (FCS, rCSI and HHS), results showed that 
approximately one in every four households was at the IPC Phase 1 (minimal) and the 
remainder were distributed across IPC Phase 2/3/4/5/ (stressed, crisis, emergency and famine). 
Table 27: HHs with IPC phase by three different Food Security Indicators in Pauktaw
IPC on the basis of HHS IPC on the basis of rCSI IPC on the basis of FCS
HHs in IPC 1 - minimal 33.8% 29.5% 41.0%
HHs in IPC 2 - stressed 37.3% 54.0%
HHs in IPC 3 - crisis 24.6% 54.0% 39.0%
HHs in IPC 4 - emergency 3.2% 19.9%
HHs in IPC 5 - famine 1.2%
3.5 RECOMMENDATIONS
INDICATORS TARGETING
Household hunger scale indicator 
The baseline study shows that 51.5% of UTC HHs and 28.9% of CfW HHs experience
moderate or severe hunger. The team initially proposed a target of 15 percentage points lower 
than the baseline value; however, given the actual baseline values, the programme is 
recommending a decrease of 10 percentage point in the proportion of HHs experiencing 
37
moderate or severe hunger among both UCT and CfW populations. Additionally, among the 
HHs experiencing moderate hunger (32.7% in UCT and 24.6% in CfW), most have an average 
score of two (2) points. The programme assumes that food assistance will enable these HHs to
move from moderate to little/no hunger. 
Reduced Coping Strategies Index (rCSI) 
The baseline observed an average rCSI score of 16 points for UCT HHs and 10 points for CfW
HHs. Prior to the baseline results, the team recommended a 10 point decline assuming HHs 
would have higher average scores; however, based on the actual baseline values, the 
programme is recommending a 3-point (average) reduction from baseline to endline for both 
CfW and UCT. The targeting is based on the fact that in the UCT program area, the average 
score is almost identical to the second quantile value, and there was five-digit points difference
between second and first quantile values. In the CfW program area, the average score is located 
around 60 percentiles, and the gap between this average value and second quantile value was 
three-digit points. The standard deviation from both program areas is approximately nine digits 
point. In this context, while the recommended FFP target is a five average point reduction, this 
will be challenging to achieve, and three-digit average points reduction will be the best 
estimation for mean score reduction in the use of coping strategies on average. 
Acceptable Food Consumption Score (FCS) 
The study shows that 48% of UCT HHs and 41% of CfW HHs have an ‘acceptable’ FCS. 
While the FFP-recommended baseline target is a reduction of 15 percentage points from 
baseline value, the distribution of data shows a minimal chance of realising this target. In both 
populations, the cut-off point value for acceptable FCS is around the second quantile, and the 
gap is huge between the average values and first quantile value (approximately eight digit 
points). The standard deviation of FCS indicators of both areas was more than 10. As such, the 
most realistic and feasible target is a 10 percentage point increase in the proportion of 
households with ‘acceptable’ FCS.
PROGRAM BENEFICIARIES TARGETING
The household size was not recommended as the criteria for program beneficiary selection as 
it did not have any effect on the HHS, FCS and rCSI food security indicators. The female- led 
HHs should be prioritized in targeting, especially for conditional case transfer activities. 
NUTRITION PROMOTION KNOWLEDGE
38
Although a high proportion of households reported high vegetable and meat/fish consumption, 
the average frequency was limited. On the other hand, low consumption was reported for pulses 
and legumes. As such, additional nutrition informational sessions could be added, or 
beneficiaries from this project could be linked with other SC Myanmar nutrition promotion 
activities, to promote the awareness and importance of consumption of diverse foods. This 
could be in the form of soft conditions to targeted project participants such as conducting short 
education sessions on healthy diets before distribution or sharing IECs with nutrition￾promoting messages. Building from Save the Children’s experiences in the Mother and Child 
Cash Transfer project, a light Behaviour Change and Communication (BCC) component could 
be added to the design and integrated in cash distribution events. Specifically, the key priority 
messages to can include consumption of foods items from the pulses and legumes group, as 
these have high nutritional values and are hence heavily weighted in the calculation of FCS 
score. Approximately 50% of the sampled households reported that they consumed pulses and 
legumes in the last week, with the average frequency of consumption being only once. 
LINKAGES WITH GOVERNMENT SOCIAL PROTECTION PROGRAM
According to the Department of Social Welfare (DSW) program implementation information, 
the Rakhine state is included in the area of Government Lead Mother and Children Cash 
Transfer Program. Linkages to that program could provide additional benefits for beneficiaries 
on top of having income that is more regular for some vulnerable households (with pregnant 
mother or under two-year-old children). In addition to the MCCT program, the social pension 
program also covers the whole country and supports people ages 85 years or older. This could 
be useful for HHs with elderly people and could reduce the strain on available HHs resources. 
In Rakhine the government-led cash transfer for the elderly is implemented by the State level 
DSW office. The programme should link up with the DSW Office in Sittwe and discuss the 
possibilities of having programme beneficiaries considered for cash distribution. 
39
4.0 Annex 
Annex 1: Baseline Questionnaire