BASELINE REPORT IRTOUN (RISE AGAIN) III
January 2018
AID-OFDA-G-17-000167
LOA: October 1, 2017-September 30, 2019
Baseline Study IRTOUN III, February 2018
2
Table of Contents 
EXECUTIVE SUMMARY .......................................................................................................................4
INTRODUCTION .................................................................................................................................8
METHODOLOGY.................................................................................................................................9
RESULTS ..........................................................................................................................................13
3.1. HOUSEHOLD CHARACTERISTICS .............................................................................................................13
3.2. INFORMATION ON HOUSEHOLD DISPLACEMENT.......................................................................................16
3.3. HOUSEHOLD INDICATORS .....................................................................................................................17
REFERENCES ....................................................................................................................................43
Baseline Study IRTOUN III, February 2018
3
Acronym List
CSI: Coping Strategy Index
CSIFC: Coping Strategy Index related to Food Consumption
CSIL: Coping Strategy Index related to Livelihoods
HDDS: Household Dietary Diversity Score
HHI: Household Hunger Index
IGA: Income Generating Activity
VSLA: Village Savings and Loan Association
WFP: World Food Program
Baseline Study IRTOUN III, February 2018
4
Executive Summary
Overview of the baseline study
Since February 2014, Mercy Corps has implemented the Irtoun program ("Rise Again") with financial
support from USAID/OFDA and FFP, to enhance food security and economic resilience of communities
recovering from conflicts in the Ansongo circle in the Gao region. The program's activities were extended
in 2016 with Irtoun II in the Timbuktu and Gourma Rharous circles in the Timbuktu region. Despite gains
in the first two programs (Irtoun I/II), the region continues to face high levels of food insecurity and
economic vulnerability due to the continuing threat of violent conflict, banditry and climate shocks.
Irtoun III is designed to improve food security, reduce the risk of conflict and enhance the resilience of
conflict-affected households. More specifically, Irtoun III will work in the same municipalities covered
by Irtoun II, continuing to support agricultural groups and VSLAs to strengthen the local capacities of
agricultural production and livestock breeding as well as to restore their means of subsistence.
The baseline study consists of two components: a quantitative household survey to collect data on key
indicators, and a qualitative survey to answer some of the questions in the study. The results of the
baseline study will be used for the following purposes:
1. Establish benchmarks for key Irtoun III indicators;
2. Assist the project team in establishing realistic target levels for the improvement of these
indicators over a period of twenty-four months;
3. Inform the project team about household livelihoods, the role of livestock, favorable
economic opportunities, and other opportunities to ease recovery from the crisis and equip
communities with the tools they need to sustain their wellbeing.
The study looked at the following aspects: number of months of household food self-sufficiency,
household dietary diversity, domestic hunger index, adaptation strategy index, sources of income, etc.
The survey sample was designed to be statistically representative of households in the project intervention
area. The two-stage cluster design provided a sample of 834 households. For the qualitative part of the
survey, six villages were interviewed through focus groups. Questionnaires were developed and finalized
based on indicators and consultations with the project team. Fieldwork, including training, data collection
and data entry, took place in January 2018.
Key Findings of the Study
Respondents' Characteristics
With regard to the gender of heads of households, just over two-thirds of heads of households are men
(72%) compared to 28% of households headed by women.
The average age of heads of household in the Irtoun III intervention area is 48 years old. This average age
is essentially the same between women and men (the difference is not statistically significant: t = -817
and p = 0.414).
With regard to the level of education, the majority of heads of household in the project have no level of
education (39.2%). The second largest group of heads of household have attended nothing but Koranic
school (31.9%). Relatively few heads of household have reached formal levels of school education or
adult literacy.
Baseline Study IRTOUN III, February 2018
5
Ethnically, the project area is almost completely Sonrhaï, with 83.7% of heads of households belonging to
this ethnic group. In Ansongo, the Sonrhaï are even more dominant (87.4%).
Demographically, the average number of people per household is 8.24 over the entire intervention area,
with a slightly increasing trend in Gourma-Rarhous (8.5%) and Timbuktu (8.49). Households composed
of seven people are the most common within the project area. Overall, and at the level of each specific
area, the least populated household is composed of two (2) people and the most populated has up to
fourteen (14) individuals.
As for the presence of displaced persons in households, 10% of the households surveyed (83 households)
are hosting at least one displaced person. Of these households, 30% (25 households) are in Ansongo
compared to 70% (58 households) in Gourma-Rarhous and Timbuktu, with an average of 2.5 persons per
household and a minimum number of one person and a maximum of 20 people welcomed into the home.
In terms of the cultural environment, before the crisis, communities had very important cultural events on
a regular basis. These events reinforce social cohesion, fraternity, mutual aid and tolerance. The events
are mainly traditional dances that are organized during ceremonies and celebrations (tam-tam, takamba,
songs etc.) and inter and intra-community meetings.
Perceived Increase in the Number of Months of Household Food Self-Sufficiency
Overall, 98% of the households responding to this question said that access and use of improved seeds
and other agricultural inputs would allow for an increase in the number of months of food security in their
households. With support for improved seeds and other modern agricultural inputs, households foresee an
average increase of 3.88 months over the current four-month coverage provided by their agricultural
production. If this perception is accurate, households participating in Irtoun-III will potentially double
their food self-sufficiency (from about 4 months to 7-8 months).
Months of Household Food Supply
In the last twelve months preceding this study, households in the project intervention area had, on
average, six months of adequate food supply (food supply meaning a household's own production,
reserve, purchases, collection, etc.).
To this end, the project must aim to significantly increase the number of months of adequate food supply
for households. To do this, the project is proposed to target the average supply of the upper tercile (one
third), an average target of 8 months of adequate food supply for all beneficiary households.
Household Dietary Diversity Score (HDDS)
The overall average HDDS is 6.2 indicating a dietary diversity that is barely acceptable. The dietary
diversity of households has an average of at least six groups that have been consumed by households
during the 24 hours preceding the present study. Female-headed households have a greater dietary
diversity than those headed by men, with a statistically significant difference (p = 0.002).
As a target, it is recommended that the project plan to raise this average HDDS to at least eight by the end
of the implementation of Irtoun III. The project should accomplish this by increasing households'
awareness that consuming foods such as beans, vegetables, etc. (food that is generally more accessible
than meat and can be/are produced by households) increases the chances of being food secure.
Household Hunger Index (HHI)
Baseline Study IRTOUN III, February 2018
6
Overall, 55% of households suffer from moderate or severe hunger, with a higher prevalence among
female-headed households (68%) and geographically in the Gourma- Rarhous circle (62%). The HHI is
based on perceptions of hunger over the last four weeks and is therefore sensitive to the season in which
the survey is conducted. In the case of this survey, data was collected during the month of January 2018.
Coping Strategy Index related to Food Consumption (CSIFC)
In the period of this study, there is relatively low frequency of use of adverse coping strategies for food 
consumption among households in the project intervention area. The average score is 14.6 for the entire 
area, indicating a low use of coping strategies. Nonetheless, this indicator still shows that during the 
period of this study, female-headed households made use of strategies more frequently than male-headed 
households: Female-headed households showed an average score of 16.7 vs 13.8 for men (statistically 
significant difference with a value of p = 0.004).
Coping Strategy Index related to Livelihoods (CSIL)
The baseline showed a relatively low average value for livelihoods-related coping strategies (12.8), 
indicating that households are not using a large number of coping strategies related to livelihoods. Still, 
the situation was more difficult for female-headed households (average of 16.7) than male-headed 
households (average of 11.2). In the Timbuktu region, households are using more coping strategies than 
in other areas. 33% of households had used at least one stress-level strategy in the past 30-days (selling 
household goods such clothing, jewelry, radio, etc; buying food on credit or using savings; spending 
savings to meet basic needs); while 29% of households reported using crisis-level strategies (selling 
productive assets, consuming seed stocks, removing children from school) or emergency-level strategies 
(selling house or land, begging, selling off female animals, entire household migrating).
Peaceful, Safe and Secure Communities.
26% of respondents describe their community as peaceful, safe and secure. This indicates that on a global
scale, only a small percentage of household heads believe that their community or village is peaceful, safe
and secure. The low level of this indicator reflects the shared trauma that the inhabitants face due to the
fragility of the local security situation.
Respondent Perception of Conflict Management
The Likert scale was used to evaluate the respondents’ perception on this topic. Using this technique, a
series of four statements were posed to each respondent, asking the respondent to respond with whether or
not they agreed with each statement. The noted trends are:
➢ The majority of respondents agree that customary leaders are able to prevent and/or resolve
disagreements in their community throughout the intervention area.
➢ The overwhelming majority of respondents agree that government authorities are able to
prevent and/or resolve disagreements in their community.
➢ The overwhelming majority of respondents agree that women/women's groups are able to
prevent and/or resolve disagreements in their community.
➢ The overwhelming majority of respondents agree that youth/youth groups are able to
prevent and/or resolve disagreements in their community.
Impact of the Crisis on the Income of Women and Men
Baseline Study IRTOUN III, February 2018
7
One of the main sources of income for women are IGAs. According to focus group participants, women
no longer practice Income Generating Activities (IGAs) like before because of lack of access or very
limited access to the market (especially intercommunity markets). The occurrence of robberies and rapes
are so frequent that women do not even leave to search for wood like they used to, out of fear of being
attacked and/or raped.
Baseline Study IRTOUN III, February 2018
8
Introduction
Since the beginning of the crisis in Mali in 2012, the conflict has had a profound impact on the population
of the northern region of the country, causing displacement, disrupting livelihoods and preventing
households from meeting their basic needs. Although a peace agreement was signed in June 2015,
implementation has been slow and the political commitment to long-term peace by all is in question.
Violence continues across the North, including attacks by Islamist groups, bandits and other armed actors
targeting the Malian army, MINUSMA, humanitarian actors and the civilian population.
Faced with this situation, a large part of the population, especially the most vulnerable households, are
unable to resume their livelihoods due to the loss of means of production, limited access to financial
resources to buy seeds and other inputs, and interrupted market systems. Mercy Corps assessments in
December 2016 showed that 63% of those polled in Timbuktu, Gourma-Rharous and Ansongo had
moderate or severe hunger. Meanwhile, food recalls in the last 24 hours showed extremely limited dietary
diversity, including only two different food groups that were consumed in the previous day. Women￾headed households are particularly vulnerable to food insecurity: 83% of these households experience
moderate hunger.
To help alleviate this difficult situation, Mercy Corps, with support from USAID/OFDA, has already
implemented two phases of the IRTOUN project (1 & 2). Based on the successes of these earlier phases,
Mercy Corps has sought and obtained the continued support of USAID/OFDA to implement Irtoun III in
the same areas of Gao and Timbuktu. For a period of 24 months, Irtoun III will respond to the immediate
needs of the most vulnerable households affected by conflict through direct cash or cash injections,
building local agricultural capacity and supporting the population in meeting their basic food security
needs and recovering their livelihoods. In the Ansongo and Gourma-Rharous circles, Mercy Corps will
help other marginalized households and communities severely affected by conflict, population
movements and drought, specifically targeting newly repatriated refugees, internally displaced persons
and their families, and host communities with immediate financial assistance and livelihood recovery
activities.
The aim of Irtoun III is for conflict-affected households to be able to meet their immediate needs, and
restore and diversify their livelihoods leading to enhanced food security, improved intra- and inter￾community conflict reduction, and increased resilience to shocks. Strategically, these households must
resume and diversify agricultural activities, and expand and diversify income-generating activities to meet
their needs, all the while being prepared to cope with shocks,
In line with its minimum program management standards and in line with its commitment with
USAID/OFDA, Mercy Corps is undertaking the Irtoun III baseline study internally. This baseline study is
the first phase of the pre-post evaluation cycle. The second phase will include the final evaluation, which
will be conducted in two years (after the end of the project). 
This baseline study has two components: (1) a representative household survey of the potential target
population to collect key effect/impact indicators; and (2) a qualitative study to collect additional data on
farmer/livestock contribution to household income, economic opportunities, conflict and livelihoods,
gender relations, and so on. The results of the baseline study will be used for the following purposes:
1. Establish benchmarks for key outcome and impact indicators;
2. Assist the project team in establishing target levels for improvement of these indicators during
the project cycle within two years of implementation; and
3. Inform the project team of the current status of dietary diversity, the place of agriculture in
household income, relevant opportunities, and current gender relations so that it can refine its
implementation strategy and improve its effectiveness.
This report begins with an overview of the methods of the household survey and the qualitative survey.
The results of the household survey are then presented for each indicator. The results of the qualitative
Baseline Study IRTOUN III, February 2018
9
study are incorporated into these results to provide more context and understanding. After the results, the
report ends with the main conclusions and recommendations.
Methodology
2.1. Household Survey Method
A. Study design and objectives
The main objective of the household survey, which constitutes the potential beneficiary population of the
project, is to assess the status of the main indicators before project implementation. The reference levels
will be used to calculate/assess the change in these indicators at the end of Irtoun III's two-year cycle,
when the same survey will be conducted again in the project areas. This pre-post model will measure the
change in indicators between the baseline and final evaluation.
B. Sampling
The size of the sample obtained using the moderate or severe hunger level indicator provides enough
households to measure target changes for all other indicators. The following criteria were used to
calculate the sample size:
n = sample size
deff = cluster effect (2)
z = confidence interval (95%)
p = proportion of households with moderate or severe hunger (50%)
e = margin of error (5%)
r = nonresponse rate (10%).
Based on these criteria, the maximum sample size is 825 households. A more detailed description of the
sampling methodology can be found in the SOW (attached in French).
The sampling frame of the project was built from all villages selected for intervention by the
implementation team. The team provided the final list of communities.
The sampling of 825 households was carried out in two stages: first, the sampling of geographical clusters
(villages), or 'first degree', and secondly the sampling of households in 'second degree' communities.
Baseline Study IRTOUN III, February 2018
10
The first-degree sample of 25 communities was selected using the systematic random sampling method
based on the list of 50 implementation communities provided by the team in a step of 2 (i.e. one out of
two villages). Table 1 below shows the total number of villages per district and commune as well as the
number of households sampled.
Selection of second-degree households was completed when field teams entered each community/village.
For each village, since the list of households was not available and the time was not enough for the teams
to draw up a nominative list, the teams observed the following steps:
• Ask the village chief, or any other local authority, the total number of households in the
community;
• Determine the appropriate sampling interval required to obtain 33 households, using the
number of households reported by the village chief;
• Place oneself in a landmark in the village and start surveying the households drawn using
the systematic random draw method.
Table 1: Village and Households Sampled by Commune
Region Circle Number of Villages
selected for
Intervention
Number of
Villages involved
in Baseline Study
Number of
Households
Surveyed
Gao Ansongo 6 3 99
10 5 165
12 6 198
Timbuktu Gourma Rharous 2 1 33
7 4 132
3 1 33
3 2 66
3 1 33
3 2 66
The systematic sampling method was used to select households. This method involved (1) randomly
choosing a starting point between 1 and n (the sampling interval), with household mentions 1, 2, ... n
starting at one end or a benchmark in the cluster/town; (2) conducting an interview in the first household
represented by the random starting point; and (3) choose an nth household afterwards for an interview
(where n is the sampling interval and is equal to the total number of households in the village, divided by
33), until all 33 households are covered. Investigators and their supervisors were trained to implement the
systematic sampling method before entering the field. Global Positioning System (GPS) units were used
to capture longitude and latitude for each interviewed household. However, unfortunately for some
households, these geographic coordinates could not be recorded due to the poor performance of some
tablets or mobile phones used to record the data.
For each household, a questionnaire was used to question the household head (or their representative in
case of absence or physical disability).
C. Questionnaire
The household questionnaire (see annexes) was developed through a series of consultations with the
country office team, the FANTA guides.
The questionnaire consists of separate modules covering the following topics:
• Module 0: Introduction and Informed Consent
• Module 1: Household Characteristics
• Module 2: Travel Information
• AMC1 Module: Participation in Previous Mercy Corps Activities
• Module 3: Number of Months of Food Self-Sufficiency
Baseline Study IRTOUN III, February 2018
11
• Module 4: Household Dietary Diversity Score
• Module 5: Household Hunger Index
• Module 6: Coping Strategy Index
• Module 7: Livelihoods and Sources of Income
• Module 8: Peaceful, Safe and Secure Communities
D. Field Steps
Investigator training and pre-test: For training purposes and field data collection, a brief checklist was
developed for investigators and supervisors.
The training session for the investigators and the pretest took place on January 11-13, 2018. This training
encompassed the roles and responsibilities of enumerators and supervisors, rules, behavioral principles,
and techniques for randomly choosing households and respondents. This also included the review of the
entire questionnaire, question by question.
Field data collection: The data collection team consists of two M&E assistants at the Gao and Timbuktu
regions and 15 investigators who were mobilized and/or recruited for the two-week collection period. The
survey was conducted in the local language according to the specificity of each area.
The data was collected using tablets via the ONA platform. This allowed the collectors to exercise the
maximum amount of control on the data collection starting with the conception of the models in order to
minimize errors.
E. Data Analysis
Calculation of household indicators and characteristics: The different modules listed above were used
to calculate the specific indicators and different characteristics of households and household heads.
Indicator definition and tabulation: Indicators were calculated according to the tabulation methods or
reference sheets developed by FANTA, some of which were created using the compendium of PAM
indicators.
2.2. Qualitative Survey Method
A. Design and Objectives
The overarching goal of the qualitative component of the baseline study is to elucidate and contextualize
the results of the household survey. More specifically, the qualitative component aims to uncover and
answer certain questions in the study through looking at factors such as the contribution of
agriculture/livestock to household income, the different economic opportunities available to women and
men, conflict and how it affects livelihoods, the socio-economic environment, gender relations, etc.
B. Sampling and Community Selection
For the qualitative study component, the sampling strategy is based on the object or purpose. This meant
that the M&E team targeted communities and individuals according to a number of criteria in order to
achieve the component's objective. Two main criteria are used in the process of choosing the circle and
the individuals.
All three circles were sampled, with two selected villages per circle. At the village level, two focus
groups, one for women and one for men, were conducted separately, i.e. 12 focus groups in total.
C. Participant Selection
The household survey focused on primary respondents who are heads of households (female or male) or
their representative. This same category was also the main focus of qualitative data collection for focus
Baseline Study IRTOUN III, February 2018
12
group participation. In this qualitative component, the collection teams worked with the following
categories of respondents or participants:
• Focus group men: head of household or another household member acting as a
representative.
• Focus group women: female head of household and/or women members of VSLA groups.
D. Focus Group Discussion Guide
Given the experiences of responses from group discussions, female/male household heads were able to
answer questions about different aspects explored in this qualitative component. For this purpose, a single
discussion guide was developed and used for discussions with women and men separately.
Aspects included in the discussion guide include:
• Agriculture/livestock in terms of contribution to household income;
• Economic opportunities;
• Conflict and livelihoods;
• Socio-cultural and economic environment;
• Gender relations.
E. Field Data Collection
Qualitative data was collected at the same time alongside the household questionnaire. In each village, the
teams agree to start first with the focus groups and then finish with household surveys or vice-versa
according to the realities of the village.
F. Data Preparation, Coding and Analysis
After returning from the field, the data collected on paper was entered in Excel by the M&E assistant and
then transmitted to the M&E officer to be codified, clustered, analyzed, and interpreted by crossing it with
the quantitative household data when possible.
2.3. Study Limitations and Encountered Difficulties.
The limitations and encountered difficulties during this basic study are summarized below:
Limited amount of time to conduct multiple studies at a time: a combination of expected delays in
conducting the baseline study and accidental coinciding with the start-up of other projects required that
several baseline studies were conducted at a time. This resulted in added pressure to design and validate
tools, conduct field collection, etc. in a relatively short time.
Security Constraints: The volatility of the security situation did not allow the teams to evolve as they
wish, given that they were not allowed to camp in the villages. This has resulted in an extension of
collection time without allowing investigators to live in communities for long enough to report certain
trends/information based on their informal observations and interactions with communities.
The insufficient number of tablets available to investigators at forced them to use their own or other staff
members' personal phones to fill this gap. This process of using tablets and phones of all kinds prevented
us from efficiently collecting the geographical coordinates of the households surveyed.
Timing for Analysis and Reporting: The tight timeline between the end of data collection and the final
report did not leave enough time for in-depth analysis and evaluation of the wealth of data collected from
the household questionnaires and qualitative study. Quantitative analysis focused on the development of
Baseline Study IRTOUN III, February 2018
13
indicators, with only a little analysis focused on qualitative data. However, this qualitative analysis may
continue even after the completion of this report if there is information that might still be useful in the
implementation of activities.
Results
In this section of the report, the results of the baseline study are presented according to the two categories
of content: (1) household characteristics, (2) indicators calculated at the household level. Additionally, the
relevant results of the qualitative study are included.
3.1. Household Characteristics
This section presents the demographic characteristics of households and household heads surveyed. 817
household interviews were conducted in the project intervention area, in two regions, including 451
interviews in the Ansongo district, 167 in Gourma-Rarhous and 199 in Timbuktu.
Table 2: Respondents' decision to participate in the survey.
# of respondents who decided to
participate in the survey
Total
Region
Gao
Gourma-rarhous
451
167
451
167
Timbuktu 199 199
Total 817 817
As the table opposite shows, all heads of households or their representatives agreed to be interviewed and
participate in this basic survey. If this 100% participation was completely motivated by sincere will
(without any ulterior motives) it demonstrates a good potential for collaboration with the households in
participating in project activities, which could contribute immensely to the durability of the program's
investments. On the other hand, this willingness to participate may also be due to a wait-and-see attitude,
or else related to local customs, which dictate that refusing an interview with a stranger is considered to
be an impolite rejection.
Table 3: Gender of Heads of Households
Women Men Total
Region
Gao Count 91 360 451
% within region 20.2% 79.8% 100.0%
Timbuktu Count 141 224 366
% within region 38.5% 61.2% 100.0%
Total Count 232 584 817
% within region 28.4% 71.5% 100.0%
Just over two-thirds of households are headed by men (72%), compared to 28% of households headed by
women. This proportion of female heads of household is still significant - in the context of the Sahel, the
percentage of women-headed households does not generally exceed 10%-15%. At the Timbuktu level (in
the project intervention area), 39% of households are headed by women compared to 20% in Gao. What
can explain why women carry such a burden to support their households in Timbuktu?
In general, and according to several findings in Africa, women heads of households suffer more in terms
of household management than men. The project can and must then deliberate targeting this vulnerable
demographic in its interventions.
Baseline Study IRTOUN III, February 2018
14
Table 4: Age of Household Heads
Statistics Global Women Men Ansongo
Gourma￾Rarhous 
Timbuktu
N Valid 816 232 584 451 166 199
Missing 1 0 0 0 1 0
Average 48.05 47.44 48.30 47.17 47.61 50.82
Standard Deviation 13.607 12.783 13.924 13.482 14.933 12.476
Minimum 15 16 15 15 18 20
Maximum 97 85 97 97 87 82
The average age of household heads in this Irtoun III intervention area is 48 years old. This age is roughly
the same regardless of gender (the difference is not statistically significant: t = -817 and p = 0.414), but
when looking at different regions things begin to vary. In Timbuktu, there heads of households tend to be
more elderly than in Ansongo or Gourma-Rarhous.
When it comes to extremity of age, both the youngest and the oldest heads of households were recorded
in Ansongo.
Table 5: Education Level of Household Heads
Level of Education Global Women Men Ansongo Rarhous Timbuktu
Freq % Freq % Freq % Freq % Freq % Freq %
Literate 64 7.8 29 12.5 35 6.0 16 3.5 24 14.4 24 12.1
No Education 320 39.2 112 48.3 208 35.6 170 37.7 67 40.1 83 41.7
Koranic School 261 31.9 60 25.9 201 34.4 129 28.6 51 30.5 81 40.7
Primary School 111 13.6 22 9.5 89 15.2 85 18.8 15 9.0 11 5.5
Secondary School 50 6.1 8 3.4 42 7.2 41 9.1 9 5.4 0 0.0
University 10 1.2 1 0.4 9 1.5 10 2.2 0 0.0 0 0.0
Total 817 100.0 232 100.0 584 100.0 451 100.0 167 100.0 199 100.0
There is a significant predominance of household heads with no level of education (39.2%) in the project
area. After this category, the second largest group of household heads have only attended Coranic School
(31.9%). Percentages of heads of households who have had a formal school education or adult literacy
courses is relatively low.
In this situation where heads of households have relatively low levels of education, women are more often
left behind, with 48.3% who have no level of education in comparison to 35.6% of men. The same trend
appears with Koranic School attendance, with 34.4% of men compared to 25.9% women.
From a geographical point of view, the proportion of heads of households without any level of education
is much higher in Timbuktu than in Ansongo or Gourma-Rarhous. The same is true for Koranic School,
with 36.1% of heads of households in Timbuktu having attended, in comparison to 28.6% in Ansongo.
However, when it comes to formal education, in Ansongo there are far more heads of household who
have attended Primary School (9.0%) than in Gourma-Rarhous and Timbuktu is a presence of heads of
primary level households far more than Gourma-Rarhous and Timbuktu (5.5%). This can perhaps be
Baseline Study IRTOUN III, February 2018
15
attributed to the higher number of youth heads of households in Ansongo in comparison to the two other
circles.
Table 6: Ethnicities of Heads of Households
Ethnicity Global Ansongo Gourma-Rarhous Timbuktu
Frequency % Frequency % Frequency % Frequency %
Arab 1 0.1 1 0.2 0 0.0 0 0.0
Other 11 1.3 3 0.7 2 1.2 6 3.0
Bambara 2 0.2 0 0.0 1 0.6 1 0.5
Dogon 5 0.6 5 1.1 0 0.0 0 0.0
Moorish 1 0.1 0 0.0 0 0.0 1 0.5
Peulh 28 3.4 21 4.7 2 1.2 5 2.5
Sonrhaï 684 83.7 394 87.4 145 86.8 145 72.9
Tuareg 84 10.3 27 6.0 16 9.6 41 20.6
Total 817 100.0 451 100.0 167 100.0 199 100.0
The intervention area of the project is almost completely Sonhraï with 83.7% of heads of households
belonging to this ethnic group. This ethnic group is much more dominant in Ansongo (87.4%) than the
other circles.
After the Sonhraï come the Tuareg, who represent 10% of the whole intervention area and who are more
present in Timbuktu and Gourma-Rarhous, where they respectively represent 20.6% and 9.6% of heads of
households.
In addition to these two dominant ethnic groups, there are other ethnic groups that are in the minority
such as the Peuhls, the Bozos, the Mossis and the Bellahs.
Table 7: Household Demographics
Total Household Members Total Number of Infants Total_mem_men_invalides
Statistics Global Ansongo Gourma￾Rarhous Timbuktu Gobal Ansongo Gourma￾Rarhous Timbuktu Global Ansongo Gourma￾Rarhous Timbuktu
N Valid 810 448 166 196 321 162 76 83 68 22 23 23
Missing 7 3 1 3 496 289 91 116 749 429 144 176
Average 8.24 8.03 8.50 8.48 3.0748 3.2593 3.0789 2.7108 2.5147 2.8636 2.6087 2.0870
Mode 7 7 7 7 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
Standard
Deviation
2.873 2.921 3.051 2.567 1.35116 1.59796 1.05531 0.94392 1.13942 1.69861 0.89133 0.28810
Minimum 2 2 2 3 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
Maximum 14 14 14 14 9.00 9.00 6.00 6.00 8.00 8.00 5.00 3.00
Sum 6671 3597 1411 1663 987.00 528.00 234.00 225.00 171.00 63.00 60.00 48.00
The table above shows the demographic characteristics of households. The average number of people per
household is 8.24 over the entire intervention area, with a slightly increasing trend in Gourma-Rarhous
(8.5%) and Timbuktu (8.49). The most common number of people per household in the intervention area
is seven (7). Overall, and at the level of each specific area, the least populated household was composed
of two (2) people and the most populated had up to fourteen (14) individuals.
Baseline Study IRTOUN III, February 2018
16
A focus was made on the presence of children under 5 in the project area. In total, 987 children under 5
years of age were counted in all the households surveyed in this study. According to the collected data,
children under 5 years old represents 15% of the population of these households. These children are
present in 39% of households, ie 321 out of 817 households. Among these 321 households, there are on
average 3 children under 5 years per household. In Ansongo, there tended to be more children under 5
years than in Gourma-Rarhous or Timbuktu.
We also noted the existence of some people with disabilities in certain households in the area. 8.3% of
households had a member with a disability. Thus, there are a total of 171 people with disabilities from 68
households, including 22 households containing 63 people with disabilities in Ansongo and 46
households with 108 disabled people in Gourma-Rarhous and Timbuktu. For the households containing
people with disabilities, there are an average of 2.5 people with disabilities per household.
Based on the above characteristics, it is recommended that Irtoun III and its targeting strategy pay
particular attention to households that are potentially more vulnerable (households with female heads and
households with people with disabilities) in order to better to achieve the intervention's desired impact.
3.2. Information on Household Displacement
This section provides information on the different categories of households in the intervention area, the existence of
IDPs within these households, and the most common explanations for their displacement.
Table 8: Household Categories
categorie_menage
Total Gao Timbuktu
Frequency % Frequency % Frequency %
Indigenous 776 95.0 427 94.7 349 95.4
Displaced Indigenous 2 0.2 2 0.4 0 0.0
Repatriated Indigenous 5 0.6 5 1.1 0 0.0
Repatriated and Displaced Indigenous 1 0.1 1 0.2 0 0.0
Returned Indigenous 7 0.9 7 1.6 0 0.0
Other 1 0.1 0 0.0 1 0.3
Displaced 7 0.9 2 0.4 5 1.4
Host 2 0.2 1 0.2 1 0.3
Repatriated 1 0.1 1 0.2 0 0.0
Returned 13 1.6 4 0.9 9 2.5
Total 817 100.0 451 100.0 366 100.0
Almost all households in the intervention zone are "indigenous" (95%). Households in the "returned"
category represent 1.6%. Apart from the few "displaced" and "returned indigenous" households, the
others are almost non-existent in the intervention zone. For the ten displaced households above, the
reasons for their displacement are mainly those relating to insecurity and/or conflict.
Almost all household heads are from their respective regions. 54% are from Gao and 45% from
Timbuktu. Only 1% are coming from Ménaka, with those coming from Bamako and Ségou making up
less than one percent of all heads of households.
3.2.1. Existence of IDPs Hosted in Households
Baseline Study IRTOUN III, February 2018
17
In 10% of households surveyed (83 households) there is at least one displaced person being hosted by
household. Of these households, 30% (25 households) are in Ansongo compared to 70% (58 households)
in Gourma-Rarhous and Timbuktu. The average number of hosted people per household is 2.5 persons,
with a minimum number of one person and a maximum of 20 people being hosted
These displaced people come from within each region itself, that is, Gao and Timbuktu.
The main reasons that these people and households migrate are 1) insecurity due to the conflict that is
raging in the zones (39%), and 2) the lack of cultivable land and/or available jobs (15%). Other reasons,
such as searching for open pasture and searching for trade opportunities, are each a driving motivation at
12% and 10% respectively.
Almost all (93%) of these IDP households confirm that they are planning to return home. This shows a
strong attachment to their origin despite the difficulties related to residual insecurity.
3.2.2. Socio-cultural Environment
Before the crisis, communities had very important cultural events. These events reinforce social cohesion,
fraternity, mutual aid and tolerance. These are mainly traditional dances that are organized during
ceremonies and celebrations (tam-tam, takamba, songs etc.) and inter and intra-community meetings.
For almost all focus group participants, the crisis had affected these very important cultural events. The
radical armed groups had banned these events and activities during ceremonies, so we are now witnessing
weddings or festivals with very few celebratory events/activities, fewer inter-community meetings, and
fewer demonstrations like dances because of the ban against congregating in groups.
The decrease or absence of these cultural events may negatively affect social cohesion, interethnic mixing
and the general culture of tolerance.
3.3. Household Indicators
In this section, quantitative and qualitative data is examined to calculate and describe the household
indicators in the Irtoun III intervention area. The section begins with the number of months of food self￾sufficiency, followed by discussion and other indicators such as dietary diversity score, household hunger
scale etc.
3.3.1. Increase in the number of months of self-sufficiency perceived by households.
The projected increase in the number of months of food self-sufficiency, due to the distribution of seed
and other improved agricultural inputs to households, is calculated for households engaged in at least one
agricultural production activity. A series of questions was asked of these households - namely the type of
agricultural production, the number of months usually covered by their production (without the project),
the precise months of the year during which these households experience a difficulty, and their perception
of the number of months that needed to be added to extend the typical food coverage.
3.3.1.1. Households that practice livestock breeding
Baseline Study IRTOUN III, February 2018
18
Many households in the intervention zone practice livestock breeding. As shown in Figure 2, opposite,
this activity is practiced by 71% of households.
Livestock activity is slightly more prevalent among households in Gao (72.3%) than in Timbuktu
(69.9%).
3.3.1.1.1. Number and type of animals owned by households
Table 9: Type of animals raised by households
Goats Poultry Sheep Cows Horses Camels Donkeys Other
Average 3.99 4.48 3.59 3.30 1.42 2.40 1.79 1.00
Mode 2 7 2 2 1 1 1 1
Standard Deviation 2.025 2.235 1.904 2.153 0.859 2.271 0.982 0.000
Minimum 1 1 1 1 1 1 1 1
Maximum 7 7 7 7 5 7 7 1
Sum/Total 1835 1652 1548 1068 71 24 343 3
Overall, small ruminant and poultry farming predominate in these targeted intervention areas. Goats are
the most common animals, with 71% of households them in this area. In total, 1,835 goats are raised
among the surveyed households, with an average of around 4 goats per household. The most common
number of goats per household is two. After goats comes poultry, with a population of at least 1652 heads
and an average of nearly 4.5 heads per household for poultry breeders. In this category of households, the
most common number of poultry per household is seven. After poultry comes sheep, with at least 1548
heads in the program area. Cattle and donkeys are ranked fourth and fifth with an average of 2 and 1 per
household, respectively.
The predominance of small ruminant and poultry farming is explained by the fact that they are easy to
breed without being too demanding in terms of space. Cattle and other large ruminants, while taking up
more space, are easy to breed without being too demanding in terms of food. They multiply with relative
speed, which offers an opportunity for households to sell one to two heads if necessary to meet their
needs or to cope with an economic shock. The sedentary nature of these animals also contributes to their
practicality.
29%
71%
Households that practice Livestock 
Rearing
No
Yes
Baseline Study IRTOUN III, February 2018
19
In Irtoun III's intervention area, herd ownership is a sign of wealth. The word 'animal' in the local
language, sonrhai, is almane (wealth). Culturally speaking, for the Sonhrai the imankoy (the rich) own
herds of cattle, while the poor own poultry and goats which can eventually lead to wealth (by multiplying
poultry and goats to purchase sheep, then later cattle).
3.3.1.1.2. Household access to pastures
A question was asked to households that breed animals about their access to pastures throughout last year
to meet their livestock's feeding requirements. Only 32% of households with livestock had adequate
access to pasture to meet the consumption needs of their animals during the twelve months preceding this
study. Over the same period, up to 68% did not have sufficient access to pasture for their animals' annual
food needs.
As can be seen in Figure 3 below, not having sufficient access to pastures is normal for households that
didn't have access in the past twelve months. Because of the repetitive or annual lean period, these
households claim to have never had access to sufficient food for their animals throughout an entire year.
This lean season was supported by the poor distribution of rainy seasons in the northern regions of Mali.
Insecurity is the second cause (26%) of livestock breeders' lack of sufficient access to sufficient pastures.
Armed groups are not only threatening security by stealing animals, but also by driving riverside
populations out of grazing areas and denying them access to pastures. Additionally, in some areas armed
groups are imposing taxes in exchange for access to pastures.
Cyclical drought, a serious environmental threat, has also been a major contributor to the 20% of
livestock breeders who cannot access pastures.
Although livestock in general, and small ruminants in particular, is a promising activity for building
resilience and livelihoods for households, the constraints mentioned above may prevent or discourage
many households from adopting or practicing this important activity.
3.3.1.1.3. Period of fodder need by households during the last year
26%
3%
5%
20%
46%
Reasons why grazing lands were insufficient last year
We cannot access lands due to
insecurity
Lands have been occupied by
Djihadists or armed groups
Available productive land has
diminished
Too much drought
We never have enough access
because of the lean season
Baseline Study IRTOUN III, February 2018
20
Since about 2/3 of the livestock-raising households did not have sufficient access to pasture for their
animals during the last year, those same households were in great need of pastures or feed in order to take
care of their animals.
Over a good period of about six months last year, these households had enormous needs for feeding their
animals. Starting in March 2017 (figure 4 below), this inability to feed their animals affected a large
number of livestock breeders. This need continued to grow until it reached its peak throughout the period
of May-June-July. During this period, the month of May was the most glaring in terms of level of need.
To meet these enormous needs for animal feed, households used several strategies. The main focus was
on the purchase of fodder (63%), followed by fodder collection and harvest (21%), a contribution or the
management of fodder by an NGO (Mercy Corps and/or other) (12%), or, finally, through trade (3%).
This feeding situation has become structural since the end of the 1980s because of the reduction of the
wintering period and the lengthening of the lean season. This is why wintering sinstead of June before)
stops already in mid-September (instead of October).
For these households that did not have sufficient access to pastures, 80% reported that the lean period of
the past year was normal. On the other hand, 11% reported that it was worse than normal, and 9%
reported it as better than normal.
An intervention aimed at providing livestock feed or setting up feed shops for the benefit of households
during this period of scarcity (April-July) could reinforce the practice of livestock breeding for these
households.
3.3.1.1.4. Animal
Health 
1% 1%
10%
15%
20% 19%
16%
9%
6%
3%
1% 0%
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Months when households need fodder 
Baseline Study IRTOUN III, February 2018
21
Households raising animals use a variety of strategies or means to ensure the health of their livestock.
These strategies range from no action to using state veterinary services. Through Figure 5 above, we note
that households most often use state veterinarians to ensure animal health, which indicates good
accessibility and quality of service.
Despite the use of agents, many households (32%) administer treatment themselves to ensure the health of
their animals. In 19% of cases, households call on community animal health workers (CAHWs).
In terms of access to animal products for animals, more than half of households raising animals have
access to these products (55%), compared to 45% who do not have this access. This phenomenon can be
partly explained by the difficulty of site accessibility by the veterinary services (material and logistic
means), the security situation, the lack of means of conservation of the products, etc.
In terms of geographical area, in Gao, these households access much more products with 79% versus only
23% in Timbuktu. The question that arises is why such a difference in access to pharmaceuticals for
animals? The isolation of Timbuktu in terms of the access road and the distance to a border and departure
of the veterinary representatives of the crisis are the reasons for this important difference. Another
explanation could lie in previous interventions that made efforts to strengthen the capacity of para￾veterinary agents in Gao. CAHWs come from the community and participate in the treatment of animals
in their community.
Overall, and in terms of animal health, there are still some practices that may be less interesting or even
risky in certain measures (32% treating their own animals and 7% doing nothing about animal health), for
lack of to know-how. To improve animal health, it would be interesting to conduct a barrier analysis, if
possible, to identify the causes and barriers that make some households resort to these somewhat
dangerous strategies in their livestock management. Once the barriers are known, they will be easier to
overcome in order to significantly improve animal health and at the same time also increase the demand
for animal health professionals in these areas.
3.3.1.2. Agricultural production practiced by households
State veterinarian 
42%
CAHW-para-vet
19%
We take care of our 
animals ourselves
32%
We don't do anything 
7%
Animal Care by Households 
Baseline Study IRTOUN III, February 2018
22
Households in the intervention zone practice even more agricultural production activities than livestock
raising. These same trends occur in each of the circles. In Ansongo , 83% of households practice
agriculture and 70% raise livestock. In Gourma-Rarhous and Timbuktu, households engaged in
agriculture account for 86% and with 70% who practice livestock farming.
There are several practices of agricultural production common in these communities:
The largest number of households (42%) who practice farming do rainfed and off-season rice production,
coupled with market gardening. After this dominant category of agricultural households, come those who
exclusively produce rainfed and off-season rice. They represent 29% of households. Specifically, at the
level of each region, the predominance of these same categories of agricultural households persists in
Ansongo / Gao where 53% of households practice rice production in the rainy season as well as market
gardening, where 28% of farmers in Timbuktu have the same mix of practices. In Timbuktu, more often
(34%) households grow only rain-fed and off-season rice.
The same trend of the predominance of cereal production (rice, millet, etc.) is confirmed by qualitative
data where this type of production and market gardening constitute the main agricultural activities of the
members of this zone. According to the qualitative data, the vast majority of the interviews highlight the
existence of a difference between men and women in agricultural production. There are many more
women involved in market gardening, and men are more present in the production of cereals such as rice
15%
85%
Households that practice agriculture
No Yes
Other
Gardening in rainy 
or dry season
Rainfed Rice + dry 
season gardening
Rainfed Rice + 
other
Rainfed or Dry 
Season Rice + 
gardening
Rainfed or Dry 
Season Rice + other
Agricultural Practices
Baseline Study IRTOUN III, February 2018
23
and millet. Nevertheless, some opinions have stated that there is no difference between men and women
in terms of production.
Why this dominance of men in grain production? Women are actively involved in the various agricultural
activities in the communities (sowing, plant maintenance, weeding, harvesting, transportation, storage,
etc.) However the men have control over grain production, as they are the landowners and have the
responsibility to supply the family breadbasket.
The typology of cultures produced in these areas is presented in Figure 8 below.
The typology of the crops produced through the above modes of production is summarized in rice
production, market gardening, millet, cowpea and sorghum.
In the entire project area, rice is by far the most produced crop by households. After rice, households also
engage in market gardening
(lettuce, cabbage, tomato, onion
etc.). In Ansongo, millet,
cowpea and sorghum are also
produced: this can be explained
by the shortening of the duration
of rains. Nowadays the rainfall
campaign rarely allows dry
crops to complete their cycle
(especially since for most
varieties the cycle is 70 to 90
days).
The project must take into
account these specificities in its seed support component and other modern agricultural inputs.
This situation is certainly due to the sedentary nature of these areas and their inhabitants, in addition to
the very important agricultural potential (along the river) that abounds in these geographical areas.
An intervention aimed at strengthening and further modernizing rice production (rainfed and off-season)
in particular and market gardening in general could have a greater impact, since these activities are the
most dominant in agricultural production habits of households in these project areas.
Agriculture and livestock are the main activities on which the livelihoods of households in these areas are
based. An intervention to strengthen these activities may have a much better chance of strengthening and
protecting the livelihoods of these majority Sonrai populations.
3.3.1.3. Household access to community water infrastructure
0% 10% 20% 30% 40% 50% 60% 70%
Millet
Sorghum
Cowpea
Vegetables
Rice
Type of crops produced 
Timbuktu Gourma Rarhous Ansongo Global
Baseline Study IRTOUN III, February 2018
24
Nearly two out of three households have
access to an infrastructure / water source
to support their farming or livestock
activities in this project area. Overall,
66% of households had access versus
34% who did not.
At the regional level, agricultural
households in the Timbuktu area (74%)
have access to water for their
agricultural production compared to
59% in Gao.
Figure 9 shows us that irrigation canals
are the most used infrastructure for these
households to undertake agricultural production. also common is the combined use of rainwater and deep
wells and those combining the use of deep wells with irrigation canals.The combination of irrigation
canals and shallow ponds is also practiced in the area.
The strong dominance of the use of irrigation canals as water infrastructure in the context of agricultural
production is certainly due to the geographical situation of several villages. These include villages located
along or nearby the river, where agricultural lands can be exploited along the river banks. This case is
more important in the area of Gourma Rharous and Timbuktu.
3.3.1.4. Reliability of access to a supply of water for agricultural activities
The reliability of the water supply for agricultural production was examined through our household
assessment, which shows that overall the water supply was less reliable or less regular during the
production season .This low reliability was felt much more by Gao's households than those of Timbuktu.
3.3.1.5. Environmental impact having a negative impact on agricultural production.
Less reliable 
63%
Same reliabiilty 
34%
More reliable
3%
SUPPLY OF WATER FOR AGRICULTURE 
.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0
captage_eau_pluie+puits_profond
captage_eau_pluie+puits_profond+…
captage_eau_pluie+puits_profond+…
captage_eau_pluie+canaux_irrigati…
puits_profond
puits_profond+canaux_irrigation+m…
canaux_irrigation+mare
Fig 9: Type of water access for agricultural activities 
Baseline Study IRTOUN III, February 2018
25
Table 10: Farming Households Experiencing an Environmental Incident
Environmental Incidents Total Gao Timbuktu
Frequency % Frequency % Frequency %
No 248 36% 37 10% 211 67%
Yes 440 64% 336 90% 104 33%
Total 688 100 % 373 100 % 315 100 %
Over the last twelve months preceding this
baseline study, 64% of households
experienced at least one environmental
incident having a negative impact on
agricultural activities.Households in
Ansongo were much more affected than
those in Timbuktu with 90% and 33%
respectively.
Zooming on the typology of these
environmental incidents that negatively affected households' agricultural activities in the past year, the
most common incidents are shown in Figure 11 above.
The most common incidents with the biggest impact on agricultural production included crops being
destroyed by birds, drought, and irregular rainfall.
3.3.1.6. Number of months of a household's food covered by own harvest
Table 11: Average number of months of household food supply
Statistics overall Female
HoH Male HoH Ansongo Gourma￾Rarhous Timbuktu
#
Valid 755 205 549 416 153 186
Missing 6 27 35 35 14 13
Average 3.58 3.31 3.68 4.06 2.92 3.03
Median 3.00 3 3
4.00 3.00 3.00
Mode 3 3 3
3 3 3
(gap-type) 1679 1.59 1703 1916 1.085 1,050
Minimum 0 1 0
0 1 1
Maximum 12 9 12 12 7 6
To assess this variable relating to the food self-sufficiency experienced by households thanks to their
agricultural production, for all the households in the intervention zone, their agricultural production alone
ensures them an average of 3.58 months of food self sufficiency.
Depending on the gender of household heads, male-headed households have nearly four months of food
supply from their production, while those headed by women have only three months of food supply
provided by their own harvest. This difference is statistically significant (t = -2.731 and p = 0.006).
Baseline Study IRTOUN III, February 2018
26
Geographically, Ansongo's agricultural production allows 4 months of supply on average as compared
with Gourma-Rarhous and Timbuktu, which only produce 3 months' supply of food on average. This may
be due to more diversity of agricultural production seen at Ansongo, where in addition to rice cultivation
and market gardening, households in this area also produce millet, sorghum and cowpea. This state of
affairs may support a hypothesis that the household that diversifies its agricultural production may be
more likely to experience food self-sufficiency.
3.3.1.7. Use made of household agricultural production
According to the focus group discussions on the main uses that households make of their own crop
production or harvest, five uses were cited including two most commonly. Household agricultural
production is used primarily for self-consumption by households to meet their food needs. After self￾consumption, a good part of the production is also sold to meet other needs that are either material or
ceremonial. One part is given to the parents / relative and another party used for the payment of either a
tax or to repay a debt to traders. Finally, a small percentage of harvest is used to purchase livestock.
3.3.1.8. Number of lean season months for households during the past year
Almost all the households in the intervention area experienced, during the last twelve months, periods
when they did not have enough food to meet the needs of their families. Up to 92% of households found
themselves in such a situation. This situation was more crucial in Ansongo and Gourma-Rarhous where
respectively 93% and 95% of households were in such a situation, compared with 85% in Timbuktu.
Figure 12 below shows the months during which households did not have enough food to meet the needs
of their families. Each respondent was asked to name the months during which their household did not
have enough food to cover their needs. It appears that the months of May, June, July and August were the
months during which access was very difficult.
3.3.1.9. Months of Adequate Household Food Supply
Food access depends on the ability of households to obtain food from their own production, food
reserves, purchases, collection or transfers of food from family, community members, government or
0%
5%
10%
15%
20%
25%
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Months when households don't have enough to eat
Global Gao Tombouctou
Baseline Study IRTOUN III, February 2018
27
donor agencies. Household access to food also depends on the resources available to family members and
the steps they must take to obtain these resources, especially the exchange of other goods and services.
This indicator in question and that of dietary diversity are indicators of the impact of household food
access.
The following table shows the reverential for this indicator for the IRTOUN III project
Table 12: Average Number of Months Adequate Food Supply for Households
Statistics Global Female
HoH Male HoH Ansongo Gourma￾Rarhous Timbuktu
#
817 232 584 451 167 199
0 0 0 0 0 0
Number of months of food supply
Median
5.99 5.02 6.38 7.05 4.63 4.73
Mode 9 4 9 9 4 6
(gap-type) 2726 2688 2645 2617 1.845 2547
Minimum 0 0 0 0 0 0
Maximum 10 10 10 10 10 10
In the last twelve months preceding this study, households in the project intervention area had, on
average, six months of adequate food supply. With all means of households (own production, reserve,
purchase, collection or transfer etc.).This confirms the trend shown in the figure above, or for about 6
months, these households did not have access to enough food to cover their needs.
Female-headed households have lower food coverage than male-headed households (p = 0.000) .And in
terms of geographical zoning, the area of Ansongo is much better than those of Gourma-Rarhous and
Timbuktu.
To this end, the project must aim to significantly increase the number of months of adequate food supply
for households. To do this, the project is proposed to target the average supply of the upper tercile (one
third), an average target of 8 months of adequate food supply for all beneficiary households.
For activities aimed at strengthening the capacity of households to meet their food needs, the project can
target female household heads and geographically more between Gourma-Rarhous and Timbuktu where
this indicator is the weakest.
3.3.1.10. Perceived increase in the number of months of household self-sufficiency by
households
To calculate this indicator, households were asked how they perceived the number of months that can be
added to the number of current months of their coverage / self-sufficiency, in case there is an improved
seed distribution, seed support, or other agricultural inputs.
Overall, 98% of the households responding to this question said that access and use of improved seeds
and other agricultural inputs allows for an increase in the number of months of food supply in their
household.
Table 13: Average number of months of increased coverage received by households
Statistics Global Female HoH Male HoH Ansongo Gourma￾Rarhous Timbuktu
#
Valid 675 171 503 364 138 173
Missing 142 61 81 87 29 26
Baseline Study IRTOUN III, February 2018
28
Average 3.88 4.12 3.80 3.96 4.04 3.59
Median 3.00 3.00 3.00 3.00 3.00 3.00
Mode 2 2 2 2 2 2
(gap-type) 2462 2637 2399 2724 2.313 1926
With support for improved seeds and other modern agricultural inputs, households foresee an average
increase of 3.88 months over the current four-month coverage of their agricultural production. In
summary, if their perception is true with this support, the households will know between 7 to 8 months of
food self-sufficiency thanks to the intervention of the project. However, since water is the driving force of
agricultural production, this period coincides with the withdrawal of water from the riverbed within km of
the parcels of market gardening. Perhaps it is necessary to think about the development alternative
solutions to continued water access for agriculture.
This average number of more months of food supply is the target that IRTOUN III should aim to achieve
(bring the average number of months of household food self-sufficiency to 8).
3.3.2. Household Dietary Diversity Score (HDDS)
The HDDS is based on the number of different food groups consumed by the head of household or by any
other household member in the last 24 hours. All 12 food groups are derived from the Food and
Agriculture Organization of the United Nations. HDDS varies from 0 to 12, with lower numbers
indicating less dietary diversity. Although the HDDS gives an indication of the food groups consumed in
the household, the HDDS should not be interpreted as a nutritional indicator reflecting the quality of the
diet, but rather as an indicator of access to food. Thus, it serves as a proxy for socio-economic status.
The results for the HDDS are presented in the table below. The overall average HDDS is 6.2 indicating a
dietary diversity that is relatively poor.
Table 14: Average household dietary diversity score (HDDS)
Statistics Global Female HoH Male HoH Ansongo Gourma￾Rarhous Timbuktu
Average 6.28 6.64 6.12 5.79 6.51 7.18
Mode 6 6 7 5 6 6
(gap-type) 1902 1671 1975 2.182 1230 1271
The dietary diversity of households shows that on average households have consumed 6 food groups
during the 24 hours preceding the present study. Female-headed households have a greater dietary
diversity than those headed by men. Statistically significant difference (p = 0.002).
Also, households in Timbuktu have a good food diversification compared to Gourma-Rarhous and
Ansongo.
Baseline Study IRTOUN III, February 2018
29
As shown in Figure 13 below, about 99% of households consume cereal-based foods such as wheat,
maize, rice, sorghum and / or millet. Sugar / honey or tea / coffee is the second most consumed food
group, followed by oil, milk / cheese and fish.
As a
target, it
is recommended that the project plan to raise this average HDDS to at least 8 by the end of the
implementation of IRTOUN III, by increasing the awareness of households to consume foods such as
beans, vegetables, etc. leaves that are generally more accessible than meat and can be / are produced by
households.
3.3.3. Household Hunger Scale (HHS)
Household hunger was measured using the HHS, a perceptual scale of food deprivation. The scale
includes three components that measure inadequate access to household food, each item being divided
into a question of occurrence (one episode occurring in the last four weeks or 30 days) and a question of
frequency of occurrence (how many times during the last four weeks). The answers to the questions are
coded and summed into a numerical score (with a possible minimum score of 0 and a possible maximum
score of 6) representing one of three levels of hunger: (1) Little or no hunger (HHS score = 0 at 1); (2)
moderate hunger (HHS score = 2 to 3); and (3) severe hunger (HHS score = 4 to 6).
Table 15: Household Hunger Scale
Indicator Global
Female
HoH Male HoH Ansongo
Gourma￾Rarhous Timbuktu
% household with little or no hunger 46% 32% 51 47% 37% 49%
% household with moderate hunger 47% 58% 42% 42% 54% 49%
% household with severe hunger 8% 10% 7% 10% 8% 2%
% household with moderate and / or severe
hunger 55% 68% 49% 52% 62% 51
The results for the HHS are shown in Table 15 above. Overall, 55% of households suffer from moderate
or severe hunger, with a higher prevalence at female- headed households (68%) and geographically at the
Gourma- Rarhous circle (62%). The HHS is based on perceptions of hunger over the past four weeks and
is therefore sensitive to the season in which the survey is conducted. In the case of this survey, data was
collected during the month of January 2018.
0%
20%
40%
60%
80%
100%
120%
% Households having consumed the following food 
groups
Global Ansongo Rarhous Timbuktu
Baseline Study IRTOUN III, February 2018
30
As this indicator has to change negatively or decrease, it would be difficult to predict a very realistic
target for a two-year intervention. But considering the effort that will be made under
IRTOUN III, we recommend targeting a reduction of 10 percentage points for this indicator at the end of
the project, ie an overall prevalence level of 45% of households with moderate or severe hunger levels.
3.3.4. Household Coping Strategy Index (CSI)
The Coping Strategy Index (CSI) is an indicator used to compare the difficulties faced by households by
measuring the frequency and severity of the behaviors they face in the event of a food shortage.
Strategies are divided into two types: strategies that affect food consumption and strategies that affect the
livelihoods of the concerned households.
3.3.4.1. Coping Strategy Index related to Food Consumption (CSIFC)
The consumption-based strategy index (CSIFC) measures behaviors [1] adopted by households when they
have difficulty meeting their food needs. This indicator assesses whether there has been a change in the
consumption patterns of a given household. It is calculated using strategies based on standard food
consumption and severity weighting. The advantage of the CSI is that it measures the same set of
behaviors and uses the same universal weights. By always measuring the same behaviors, the CSIFC has
greater applicability, so it can be used to compare the severity of crises and it helps with geographic
targeting.
For the calculation of this index at the level of each household, for each adaptation strategy, the frequency
score (0 to 7) is multiplied by the universal severity weight. The weighted frequency scores are summed
into a final score (CSIFC). The minimum possible CSIFC value is 0 and the maximum is 56. A higher
score indicates that more frequent and/or more severe coping strategies are being used by households. The
table below gives the average CSI score for Irtoun III's intervention areas for this period of the study
(January 2018).
Table 16: Average CSIFC Score
Statistics Global Women Men Ansongo Gourma￾Rarhous Timbuktu
#
Valid 817 232 584 451 167 199
Missing 0 0 0 0 0 0
CSIFC Average 14.67 16.72 13.86 18.00 11.65 9.67
Mode 3 3 7 7 17 3
Standard Deviation 12.711 13.371 12.369 14.942 7.529 7.269
Minimum 0 1 0 0 0 1
Maximum 56 56 56 56 43 37
As can be seen, in this period of the study, the frequency of using coping strategies related to food
consumption were relatively low in the project intervention area. The average score is 14.6 for the entire
area.
Although overall somewhat weak, this indicator still shows that during the period of this study, female￾headed households made more use of strategies more frequently than those led by men. The average score
Baseline Study IRTOUN III, February 2018
31
was 16.7 for female head of household, while the average score for men was 13.8 (statistically significant
difference with a value of p = 0.004).
This index (CSI) is sensitive to seasonality, and can change quickly over a short period of time. It is
therefore a sensitive indicator that can be included in regular monitoring activities - although the timing of
data collection needs to be taken into account due to possible seasonal biases (eg, lean period, period
harvesting, etc.). The frequency and timing of follow-up surveys depend on the context. However,
because of its sensitivity, it is important to ensure that the next measurement or collection (final
assessment) of this indicator coincides with the same season as this baseline study.
The following targets are recommended for the project:
Indicator Baseline Target
CSI Global Average 14.67 <=14
CSI Average for Women 16.72 <=14
CSI Average for Men 13.86 <=13
3.3.4.2. Coping Strategies Index related to Livelihoods (CSIL)
The Coping Strategies Index related to Livelihoods is an indicator that is used to better understand the
long-term resilience of households.
In general terms, the livelihoods and economic status of households are determined by income,
expenditures and assets. By understanding the behaviors that households have adopted to cope with recent
crises, such as the sale of productive assets, we can get a rough idea of how difficult their current situation
is and how likely they will be able to cope with challenges in the future. The higher this score, the more it
indicates a use of severe coping strategies.
As shown in Table 17 below, and like the CSI related to food consumption, this index is relatively low -
on average 12.8 for the entire intervention area. The smaller the number, the lower the indicated use of
severe coping strategies by households.
By looking at the sex of household heads, we note that the situation is slightly harder for female heads of
household (average of 16.7) in comparison to men (average of 11.2), with a significant statistical
difference (p = 0.000). In terms of geographical areas in Timbuktu, households use a little more severe
adaptation strategies than other areas
Table 17: Average Coping Strategies Index (CSI) related to Livelihoods Score
Statistics Global Women Men Ansongo Gourma￾Rarhous Timbuktu
#
Valid 817 232 584 451 167 199
Missing 0 0 0 0 0 0
CSIL Average 12.87 16.97 11.20 11.84 13.28 14.85
Mode 9 38 9 0 7 38
Standard Deviation 11.225 12.510 10.196 10.138 11.307 13.114
According to the recorded number of households organized by type of used strategies, we have:
Baseline Study IRTOUN III, February 2018
32
Prevalence of households using livelihood-related strategies
Neutral Stress Crisis Emergency
9% 33% 29% 29%
According to this classification of household prevalence by groups or categories of coping strategies, we
see that in the 30 days leading up to the present study, the greatest number of households (33%) used at
least one stress-based strategy. These strategies look like selling household goods/capital (radios,
furniture, TVs, jewlery, clothing etc.), buying food in credit or loans, spending savings, and borrowing
money). Crisis strategies (selling means of production, consuming seed stocks, removing children from
school, reducing health-related expenses) and emergency strategies (selling the house or land, begging,
selling female animals, migrating the whole household) were each used up to 29%. These percentages
indicate a high degree of household fragility in this area. To this end, to protect these households and their
livelihoods, we propose that the project adopt the following targets:
Baseline Target
% of households using neutral coping strategies 9% >30%
% of households using stress-related coping strategies 33% <33%
% of households using crisis coping strategies 29% <20%
% of households using emergency coping strategies 29% <17%
3.3.5. Livelihoods and sources of household income
To assess the households' livelihoods and sources of income, Mercy Corps posed a series of questions to
each participating household. Through these questions, each household cited sources of income in order
of important (from 'contributes the most' to 'contributes the least'). The following results were collected
during the survey.
Baseline Study IRTOUN III, February 2018
33
Each household numbered their sources of income from 1-7. Figure 14, above, shows that agriculture is
the main source of income for more than 55% of households in the area. After agriculture is, respectively,
trade, fishing, support from relatives, sending funds etc. The qualitative data collected showed the same
tendencies, where agriculture and livestock make up the most important shares of household income in
the intervention zone.
These main sources of income for households have not changed or are the same for 92% of households
(753 out of households 817). For the few households whose sources have changed, the reasons are
primarily the lack of opportunities to trade, in most cases due to insecurity, and the decline in agricultural
production caused by the drought.
Contrary to what the vast majority of households think in individual surveys, focus group discussions
have claimed there has been a negative impact on livelihoods in the last five years. Insecurity has
diminished the ability to freely trade because of fear of being attacked during a trip to a market or being
robbed and looted at home. For people who have remote fields, the lack of security reduces the amount of
time they're able to spend in their fields, which reduces their ability to produce crops. Cattle theft has also
affected several households.
For the vast majority of focus group participants, the above negative impacts on livelihoods are often
sources of conflict. How can these impacts become causes of conflict? Some people are ready to fight to
protect their assets from robbery or theft. There is also the feeling of intra- and inter-communal mistrust
created by the continuously diminishing livelihoods, which can also cause conflict.
As for the diversification of income sources, up to 45% of households say that they sometimes diversify
and/or change sources of income. The most important way in which these households have diversified is
the practice of petty trading. After trade, several households also became involved in the manufacture of
mats as a diversification activity, followed by fishing. Instead of relying exclusively on agriculture, which
is increasingly reducing in production, several households have undertaken small-scale businesses/IGRs
(such as selling wood, coal, etc.).
Migration : In the focus groups, migration is cited as the second most important means of income
diversification after small businesses. Primarily a strategy for dealing with food insecurity, labor
0 100 200 300 400 500 600 700 800
Agriculture
Transformation of Ag products
Trade
Fishing
Resource exploitation
Salaried work
Remittances
Work as migrant
Support from family
Other
Principle Income Sources of Households 
1st 2nd 3rd 4th 5th 6th 7th
Baseline Study IRTOUN III, February 2018
34
migration is a means of alleviating chronic poverty and increasing income opportunities. It is well known
that the decision to migrate may be related to several factors, including poor harvest, lack of food and
limited income to cover household expenses, insecurity, and so on.
3.3.5.1. Participation in Mercy Corps activities
More than half of the households interviewed said they had participated, through the head of household or
another family member, in at least one Mercy Corps-supported activity in their community prior to this
project. This demonstrates the wide impact of the program's previous phases. The activities that most
affected the beneficiaries in order of importance were: the distribution of vouchers for access to seeds, the
creation and implementation of Village Savings and Loan Associations (VSLAs), the vouchers for
livestock, sustained learning, Cash For Work (CFW) activities etc.
This is almost certainly due to the fact that the above-listed activities constitute the main activities that
have been implemented by previous phases of Irtoun I & II in these intervention areas. What is very
important to note at this point is that according to the perceptions of these beneficiaries, thanks to this
help they received through participation in the above activities, supported by Mercy Corps, they consider
themselves more apt to face the coming year or the year after, demonstrating a sense of resilience.
3.3.6. Economic opportunities
One of the key points to explore in this study was which economic opportunities are favorable to
communities, particularly when it comes to markets. This element was brought up during focus group
discussions.
Overall, according to participants, many economic opportunities within communities. Income generating
activities such as dyeing, sewing, selling items such as wood, charcoal, and condiments, etc., are the most
favorable economic opportunities according to the respondents. The next most favorable activities were
fish farming, livestock fattening and processing agricultural products.
From a gender perspective, the economic opportunities that are more favorable to women in order of
importance are: selling cosmetics (soap, ointments, etc.), handicrafts, processing agricultural products,
small businesses, livestock fattening, and dyeing.
As for men, the cattle trade and livestock fattening, selling fodder, and fishing or selling fish were the
most favorable.
In order to engage in economic opportunities, especially for women, there is a certain baseline level of
competence and knowledge that is required. This looks like professional training, acquiring adequate
materials, finding storage for products, etc.
3.3.7. Peaceful, safe and secure communities.
This indicator is measured using three separate questions on peace and security in the respondents'
communities. The community is defined as a group of people living close to each other in a
neighborhood, village or city depending on the target of the project.
Three questions were asked about the aspects of the indicator and the results of these questions were
aggregated to calculate the indicator.
Respondents were asked to comment on the guarantor of safety in their community. before beginning the
specific questions for calculating this indicator.
Baseline Study IRTOUN III, February 2018
35
3.3.7.1. Entities guaranteeing the security of the population
According to the opinion of the heads of households in the program's overall implementation area , the
primary entity for guaranteeing their safety remains the village chief. After the village chief comes armed
groups, then the gendarmerie, the marie, and finally 'others'. This tendency does not seem to be the same
for individual circles. In Ansongo, for example, respondents feel that the first guarantor of their security is
armed groups, who come before the village chief and the gendarmerie. In Gourma-Rarhous and
Timbuktu, the trend mirrors the global trend currently taking shape, except that the armed groups are
barely perceived as guaranteeing security in these zones.
As for 'other' entities that guarantee security, the military/FAMA are the first to be recognized by
households, followed by local youth, and finally the National Guard.
3.3.7.2. Levels of violence in communities
Respondents were asked to describe their community in relation to the level of violence experienced by
the community. In 68% of cases, they found their community quite peaceful, as shown in the table below.
The communities of Gourma-Rarhous and Timbuktu are perceived as being more peaceful than Ansongo,
where about 39% of respondents described their community as experiencing a significant level of
violence.
Table 18: Respondents' perception of the level of violence experienced in their communities
Households' Perceptions of the Community Global Ansongo Gourma￾Rarhous Timbuktu
% % % %
Somewhat Peaceful 67.8 47.0 97.6 89.9
Neither Violent, nor
Peaceful
5.4 5.1 1.2 9.5
Very Peaceful 1.1 1.3 1.2 0.5
Very Violent 4.0 7.3 0.0 0.0
Somewhat Violent 21.4 38.8 0.0 0.0
Total 100.0 100.0 100.0 100.0
As for feelings of insecurity and fear of crime in the last six months, more than 40% of respondents said
that in the last six months before this study, at least one member of their family felt insecure in their
neighborhood or community. More than 20% rarely felt insecure, while more than 15% always felt
insecure in the last six months.
For these same people, the main reasons they feel insecure are primarily because of jihadists (33%),
followed by the presence of bandits in the community or the surrounding area (40%), the absence of
police and security forces (27%), and intra- and inter-community tensions.
With regard to the frequency of fearing a crime taking place in the household, more than 50% reported
they have never feared this scenario, in contrast to the little more than 20% who have.
Baseline Study IRTOUN III, February 2018
36
Moving
beyond fears and perceptions, each respondent was asked to rate the frequency of crime that had occurred
in their neighborhood/community during the last 6 months. Overall, there are more respondents (32%)
who reported zero occurrences of crime in their area. But in 18% of cases, crimes occurred at least once a
month, 15% reported at least once in the last 6 months, 14% reported at least once a quarter, 11% at least
once per week, and finally 10% said a crime occurred in their community almost every day. According to
the different circles, the frequency of crimes is shown in Figure 20, below.
Ansongo experiences most the crimes, according to the different categories. The majority of respondents
reported that crime occurred either "at least once a month" or "at least once in the last 3 months", as
shown in the graph above. Just over 20% of respondents in this area reported that crimes occurred 'at least
once a month'.
3.3.7.3. % of respondents who believe that their community is peaceful, safe and
secure
15%
14%
18%
11%
32%
10%
Frequency of Crime
At least once in the last 6
months
At least once in the last 3
months
At least once a month
At least once a week
Never
Almost every day
0% 10% 20% 30% 40% 50% 60% 70% 80%
At least once in the last 6 months
At least once in the last 3 months
At least once a month
At least once a week
Never
Almost every day
Frequency of Crime By Circle 
Gourma-Rarhous Timbuktu Ansongo
Baseline Study IRTOUN III, February 2018
37
% of people (heads of households) who believe that their community is peaceful, safe and productive. 26.1%
At 26%, this indicator represents only a small percentage of household heads who believe that their
community or village is peaceful, safe and secure. This low percentage reflects the general level of
awareness that the local inhabitants have of how fragile the state of security can be in their area.
As this indicator is based on the perceptions and the experiences of the community, it will be difficult to
set a realistic target, especially without really knowing what the project will do to reinforce this feeling
among beneficiaries. Assuming that the project will raise awareness enough to reinforce the perception of
safety, the proposed target is 33%, which is 7 percentage points more than the current target of 26%.
3.3.8. Respondent perception of conflict management
The Likert scale was used to evaluate respondents' perceptions when it comes to conflict management.
Using the Likert technique, a series of four statements is posed to each respondent. The respondents then
respond with whether or not they agreed with each statement. These statements mainly concern four
categories of actors that are typically involved in the prevention and/or management of conflicts. These
include customary leaders, government authorities, women/women's groups, and youth/youth groups.
The results are as follows. In order to lay out all the information, there is a separate table for each of the
four statements.
Table 19: Respondent's Belief in Customary Leaders' Capacity to Resolve Conflicts
Customary leaders are able to prevent and/or resolve disagreements in my
community. Total
Agree Neither Agree Nor Disagree Disagree
Cercle
CERCLE (sig. 0.000)
Ansongo 390 4 55 449
Gourma-Rharous 166 0 1 167
Timbuktu 197 0 2 199
Total 753 4 58 815
Gender
GENDER (sig. 0.141)
Women 207 1 23 231
Men 545 3 35 583
Total 752 4 58 814
The majority of respondents agree that customary leaders are able to prevent and/or resolve disagreements
in their community throughout the intervention area. When disaggregated by location, however, the
differences in responses are statistically significant. On the other hand, the difference presented when
disaggregating the responses by gender was not significant.
Table 20: Respondent's Belief in Gov't Authorities' Capacity to Resolve Conflicts
Local government authorities are able to prevent and/or resolve disagreements in my
community. Total
Agree Neither Agree Nor Disagree Disagree
Cercle
CERCLE (sig. 0.000)
Ansongo 363 32 54 449
Gourma-Rharous 141 7 19 167
Timbuktu 156 1 42 199
Total 660 40 115 815
Gender
GENDER (sig. 0.005)
Women 175 9 47 231
Men 484 31 68 583
Total 659 40 115 814
Baseline Study IRTOUN III, February 2018
38
The overwhelming majority of respondents agree that government authorities are able to prevent and/or
resolve disagreements in their community. Depending on the cercle and the gender of the head of
household, the differences in the responses are statistically significant. In other words, this perception of
government authorities being trusted to resolve conflicts varies by who is giving the responses.
Table 21: Respondent's Belief in Women's Ability to Resolve Conflict
Women/women's groups are able to prevent and/or resolve disagreements in my
community
Total
Agree Neither Agree Nor Disagree Disagree
Cercle
CERCLE (sig. 0.000)
Ansongo 336 3 109 448
Gourma-Rharous 127 7 33 167
Timbuktu 134 0 65 199
Total 597 10 207 814
Gender
GENDER (sig. 0.000)
Women 145 7 79 231
Men 451 3 128 582
Total 596 10 207 813
The overwhelming majority of respondents agree that women/women's groups are able to prevent and/or
resolve disagreements in their community. This belief fluctuates depending on the geographic location
and the gender of the respondent.
Table 22: Respondents' Belief in Youth's Ability to Resolve Conflict
Youth/youth groups are able to prevent and/or resolve disagreements in my
community.
Total
Agree Neither Agree Nor Disagree Disagree
Cercle
CERCLE (sig. 0.000)
Ansongo 367 2 78 447
Gourma-Rharous 125 3 38 166
Timbuktu 133 0 66 199
Total 625 5 182 812
Gender
GENDER (sig. 0.000)
Women 149 4 78 231
Men 475 1 104 580
Total 624 5 182 811
The overwhelming majority of respondents agree that youth / youth groups are able to prevent and/or
resolve disagreements in their community. This belief is different depending on the geographic location
and the gender of the respondent.
Based on the responses to the statements above, the percentage of respondents who believe that the actors
responsible for conflict management in their area are able to prevent and/or manage conflict, is as
follows:
Baseline Study IRTOUN III, February 2018
39
Table 23: Perceptions of the actors responsible for conflict management
Variable Value
Total number of respondents (A) 815
Number of respondents who agree with D1 and D2 and D3 and D4 (B) 519
% of respondents who believe that the actors responsible for conflict management in their area are able to
prevent and/or resolve conflicts in their community. (B/A) 64%
Nearly two out of three respondents believe that customary leaders, government authorities,
women/women's groups and youth/youth groups are able to prevent or resolve conflict in their
community.
This perception is important and reflects the high level of confidence that household leaders in this area
have about the capacities of those responsible for conflict management. This perception is to be
reinforced through the sensitization of the population and at the same time the training of the relevant
actors on conflict management techniques. The project may target raising this proportion to 70% of
households by the end of the implementation period.
3.3.9. Gender dynamics
This section was only included in the qualitative part of this study. The section evaluated gender
dynamics in this crisis or post-crisis context. The investigated dynamics affect aspects such as the impact
of the crisis on the income of women and men, the way crisis affects demographics, and so on.
3.3.9.1. Impact of the crisis on the income of women and men
According to the participants in the focus groups, women no longer practice IGAs as much as before
because of lack of access or very limited access to the market (especially intercommunity markets).
Another limiting factor is the occurrence of robberies and rapes - most women avoid leaving to search for
wood as they used to do, out of fear of being attacked and/or raped.
As for men, they were more often victims of the robbery and animal kidnapping. Some were forced to
abandon their cattle or fields to flee their villages. All of these factors are in addition to the quotidian
limitation of moving between villages to participate in markets.
3.3.9.2. Demographic impact on households
This crisis has not been without impact on the demography of households in this area. Focus participants
thought that the crisis has increased the number of female-headed households, due to the death of male
heads of households. According to their estimate, in the villages affected in this study, there is an average
of 54 households (598 households estimated by 11 focus group) that are led by women due to the above￾stated crisis-related issue. This overwhelmingly increases the burden of maintaining households on
women.
In addition to the increase in female-headed households, the crisis has also led to an increase in
households headed by single men. There is an estimated average of 48 households headed by single men
per village.
Although defining which households are headed by children is difficult, the focus group participants still
think that the crisis has similarly increased the number of child-headed households. They estimated an
average of 35 of child-headed households per village.
In these villages, some grandparents also saw their household responsibilities and burdens increase due to
the conflict. For this type of household, due to the absence of the parents, the grandparents are forced to
Baseline Study IRTOUN III, February 2018
40
take care of the grandchildren. The participants estimated an average of 107 households in this situation
per village.
Finally, for the impact of the crisis on household demographics, some were forced to take in other
households that were displaced by the crisis. An average of 62 households that are hosting IDPs are
estimated per village.
The above demographic impacts of the crisis certainly play an important role in weakening the already
highly affected livelihoods of all the above household categories. If this trend continues, it will be very
difficult or even impossible for these households to get out of the cycle of poverty.
3.3.9.3. Change in certain factors for female heads of households
With regard to land tenure for women, there is a difference of opinion among the focus groups. For five
of the focus groups, it is seen as completely possible for female heads of households to own land without
necessarily having a husband. For three other groups, access to land is difficult for a female head of
household without a husband. What is the reason for this difficulty in access according to this group of
respondents? The investigators did not push to ask why land tenure was more or less accessible for single
women, thus limiting our knowledge in this report. However, there is definitely a possibility for future
exploration of this topic through focus group discussion in the same villages during activity
implementation.
As for access to income opportunities and social networks, when a woman is single, six groups said that
this had a positive affect. In other words, they have the same access as any individual when it comes to
different economic opportunities and social networks within their community. On the other hand, the
other six groups think that being single negatively affects women’s access to these things. On the basis of
these perceptions, it is easy to confirm that there is a general sentiment among some community members
that single women should not have the same access to economic opportunities as others. This could be an
important focus in conducting outreach activities, to potentially reverse this mindset.
3.3.9.4. How to deal with the current situation
There are several ways move forward with the situation, depending on whether it is a couple living with
their children, a single man with or without children, or a single woman with or without children.
The support shared between spouses is the most important way that couples ensure harmony in their
households. Husband and wife, everyone contributes the best to help each other and support the burden of
maintaining the household. As for single men, they depend mainly on their parents, as well as by relying
on their daily work which can act as a coping mechanism. For single women, their means for thriving are
fairly restricted to parental support or support from loved ones.
3.3.9.5. The most appropriate means of support.
According to the different focus groups, the most appropriate way to assist men and women in getting
back on their feet to be able to earn their living, is nearly identical for couples as well as single men and
women.
Provide vocational training in order to master a trades or form of manual work, support people with
materials necessary to conduct income-generating activities, and financial support are the three most
Baseline Study IRTOUN III, February 2018
41
useful tools for support in order of importance. Using these tactics could allow these people to earn their
living without being dependent on others.
3.3.10. Conclusion
Data from the Irtoun III baseline study was collected in January 2018 from approximately 817 households
in the Ansongo, Gourma-Rarhous and Timbuktu districts. The household survey collected data on dietary
diversity indicators, the household hunger index, household coping strategies, months of adequate
household food supply, perceptions of conflict, and so on. Qualitative surveys collected additional data
through group discussions with potential beneficiaries.
In line with the overall objective of the baseline study, the main findings and conclusions regarding the
project impact and outcome indicators are described below. The results are very descriptive and static
analyses too advanced that may require much more time, have not been made. Further analysis of the data
is possible, and the survey databases are available for in-depth analysis if necessary.
3.3.10.1. Household Dietary Diversity Score (HDDS)
The overall SDAM score of 6.2 indicates average dietary diversity, with six of the twelve food groups
consumed. Food diversity is lowest for male-headed households (HDDS = 6.1) compared to those headed
by women. From a geographical standpoint, the dietary diversity in Ansongo is lower than the others
(HDDS = 5.7).
Almost all households consume cereal-based foods such as rice, maize, sorghum and/or millet.
Sugar/honey or tea/coffee is the second most consumed food group, followed by oil, milk/cheese and fish.
3.3.10.2. Household Hunger Index (HHI)
Household survey results indicate that more than half of households (55%) suffer from moderate or
severe hunger, with a higher prevalence among female-headed households (68%). Geographically, this
prevalence is higher in the Gourma-Rarhous cercle (62%). Factors of household hunger are related to
access to and availability of food, which are influenced by seasonal conditions and sources of income.
3.3.10.3. Months of Household Food Self-Suffiency (MHFSS)
The results of the survey indicate that the average number of months of household food self-sufficiency is
5.9-6 months. But for women heads of households, it is 5 months. When divided geographically, food
self-sufficiency is weaker in Gourma-Rarhous and Timbuktu (about 4.6 months). Food access depends on
the ability of households to obtain food from their own production, food reserves, purchases, collection or
transfers of food from family, community members, government or donor agencies. Household access to
food also depends on the resources available to family members and the steps they must take to obtain
these resources, especially the exchange of other goods and services.
3.3.10.4. Perceived increase in the number of months of household food self-sufficiency
The results of the study indicate an average increase of 3.8 months of household food self-sufficiency.
According to household heads interviewed, with support for improved seeds and other modern
agricultural inputs, they see an average increase of 3.8 months over the current 4-month average coverage
period allowed by their agricultural production.
3.3.10.5. Coping Strategies Index related to Food Consumption (CSIFC)
Baseline Study IRTOUN III, February 2018
42
The results indicate an average index of coping strategies based on household food consumption of 14.6.
The minimum possible CSIFC value is 0 and the maximum is 56. A higher score indicates more frequent
and/or more severe coping strategies being used by households. This average of 14.6 shows that there is a
relatively low frequency of use of adaptation strategies related to food consumption by households in the
project intervention area.
Although overall somewhat weak, this indicator still shows that during the period of this study, female￾headed households made more use of strategies more frequently than those led by men. The average score
for female heads of households is 16.7, against 13.8 for men (statistically significant difference with a
value of p = 0.004).
3.3.10.6. Coping Strategies Index related to Livelihoods (CSIL)
The results indicate an average index of livelihoods-based coping strategies of 12.8 across the
intervention area. The smaller the number, the lower the indicated use of severe coping strategies by
households. According to the gender of household heads, we note that the situation is slightly harder for
female heads of household (average of 16.7) than male heads of households (average of 11.2), significant
difference (p = 0.000). In Timbuktu, households use a little more severe adaptation strategies than other
areas.
3.3.10.7. Peaceful, safe and secure communities
The results indicate that 26% of household heads surveyed think their community is peaceful, safe and
secure. This indicates that overall, only a small percentage of household heads believe that their
community or village is peaceful, safe and secure. This relatively low percentage indicates a general
awareness of the fragile state of each village.
3.3.10.8. Perception of participants on conflict management.
The results indicate that 64% of respondents believe that the actors responsible for conflict management
in their area are able to prevent and/or resolve conflicts in their community. This perception is important
and reflects the high level of confidence that household leaders in this area have about the capacities of
those responsible for conflict management.
Baseline Study IRTOUN III, February 2018
43
References
1. WFP Indicators, 2014-2017.
2. HDDS definition and guide, Indicator Guide, version2, 2006. USAID/FANTA ;
3. Household Hunger Index definition and Guide, 2011, USAID/FANTA ;
4. Month of food self sufficiency indicator definition and guide, june 2010,
USAID/FANTA.