Senegal Yaajeende Baseline Study of the Kolda Region: A Formative Evaluation
January 25, 2017 This publication was produced at the request of the United States Agency for International Development. It was prepared independently by International Development Group LLC (IDG).
Kolda Baseline Study
Learning, Evaluation and Analysis Project-II (LEAP-II)
Senegal Yaajeende Baseline Study of the Kolda Region: A Formative Evaluation
Final Report
Contract Number: AID-OAA-I-12-00042/AID-OAA-TO-14-00046
DISCLAIMER The contents of this report are the sole responsibility of IDG and do not necessarily reflect the views of USAID or the United States Government.
Kolda Baseline Study
TABLE OF CONTENTS
Executive Summary ...... 4 1. Project Background ...... 11 1.1 Overview of the Yaajeende Program ...... 11 1.2 Yaajeende Activities ...... 12 1.3 Yaajeende Activities in Kolda ...... 17 1.4 Population of Kolda ...... 20 2. Purpose of the Kolda Baseline StudY ...... 22 2.1 Indicators ...... 23 3. Methodology ...... 27 3.1 Household Survey ...... 27 3.2 Qualitative Research ...... 38 4. Limitations of the Baseline Study ...... 43 5. Findings of the survey ...... 45 5.1 Profile of the Households in Kolda ...... 45 5.2 Basic Indicators ...... 49 5.3 Study Question 1: Nutrition ...... 49 5.4 Study Question 2: Healthy Household Practices ...... 55 5.5 Further Discussion on Hygiene and Sanitation Practices ...... 61 5.6 Study Question 3: Agricultural Practices and Production ...... 62 5.7 Further Discussions on Land Preparation and Access, Agricultural Inputs and Practices ...... 63 6. General Observations on Yaajeende Implementation in the Kolda Region ...... 75 7. Conclusion ...... 77 7.1 Conclusion for Study Question 1: Nutrition ...... 77 7.2 Conclusion for Study Question 2: Healthy Household Practices ...... 78 7.3 Conclusion for Study Question 3: Agricultural Practices and Production ...... 79 8. Recommendations ...... 80 Annex I: Abridged Results Framework ...... 81 Annex II: Correspondence of Indicators with PMP ...... 82 Annex III: USAID/Yaajeende Program – Kolda Baseline Scope of Work ...... 84 Annex IV: Kolda Baseline Survey Questionnaire ...... 88
Kolda Baseline Study
ACRONYMS
BCC Behavior Change Communication CAPI Computer- based Personal Interviewing CBSP Community-based Service Provider (Agents de Prestation de Service or APS) CNV Community Nutrition Volunteers (Volontaire de Nutrition Communautaire or VNC) DEFF Design Effect FGD Focus Group Discussion FTF Feed the Future GOS Government of Senegal GTC Citizen Work Groups (Groupes de Travail Citoyen or GTC) HH Households IEC Information, Education, and Communication KII Key Informant Interview LSC Local Steering Committee (Comités Locaux de Pilotage or CLP) MDES Minimum Detectable Effect Size MIE Midterm Impact Evaluation NEP Nutrition Enhancement Program NLA Nutrition Led Agriculture PG Producer Groups PG Producer Organization PMP Performance Monitoring Plan PNDL Programme national de développement local POs Producer Organizations PPS Probability Proportional to Size PSU Primary Sampling Unit WASH Water, Sanitation and Hygiene ZOI Zone of Influence
Kolda Baseline Study
EXECUTIVE SUMMARY
Background. Food insecurity continues to be a serious problem throughout Senegal and the West African region, leaving large segments of the population vulnerable to famine and hunger. USAID/Senegal is working with the Government of Senegal to promote greater agricultural productivity and improve regional food security. The goal of the Yaajeende Agricultural Development Program, commonly known as Yaajeende, is to accelerate the participation of the very poor in rural economic growth and to improve their nutritional status. This Program falls under the Feed the Future Initiative (FTF) which was designed based on the four Intermediate Results (IR) of the USAID Economic Growth Results Framework (see Annex 1)1: IR 1 Inclusive Agriculture Sector Growth; IR 2 Increased Trade; IR 3 Improved Nutritional Status, especially of women and children; and IR 4 Improved Management of Natural Resources.
Purpose. This baseline study in Kolda was conducted in preparation for an impact evaluation of the USAID/Yaajeende in Senegal. The program has been implemented in the regions of Matam, Bakel, and Kedougou since 2011, and started in Kolda in April 2014. Because this baseline study was conducted fifteen months after the start of USAID/Yaajeende implementation in Kolda (data were collected November 2015 through January 2016), the study’s objectives are to: 1. Establish baseline values of 16 key target indicators defined in the USAID/Yaajeende performance monitoring plan (PMP), 2. Provide supporting information on dynamics around household revenues, income, and adoption of key behaviors thought to impact nutritional status, 3. Provide a description of what is happening in the field among the intervention villages where USAID/Yaajeende has started its activities, 4. Identify if USAID/Yaajeende communities are already witnessing change and how they perceive this change happening; and 5. Make recommendations to enhance the project’s implementation and performance in Kolda.
Methodology. The Kolda baseline study uses a mixed-method approach involving a non- experimental quantitative strategy and qualitative techniques. To collect quantitative data, the Kolda baseline study utilizes a household survey to collect information from a randomly selected sample of households in comparison and intervention areas defined by Yaajeende. The survey is specifically designed to establish the values of the 16 basic indicators listed in the Scope of Work for the assignment. To collect qualitative data, the Kolda baseline study utilizes key informant interviews (KIIs) and focus group discussions (FGDs) with stakeholders in intervention areas only.
Sixteen basic indicators have been calculated using the survey data. They are reported for the intervention and control areas and by several demographic variables to provide a profile of the
1 USAID/Yaajeende Performance Monitoring Plan (PMP) Revised: December 6, 2013
Page 4
Kolda Baseline Study
population of Kolda. These demographic variables are disaggregated by gender whenever possible. Information on farming activities and practices was also analyzed to describe the agricultural activities in Kolda. The results of the qualitative research are triangulated with the results of the quantitative survey.
Summary of Key Findings. The baseline key findings are related to the following areas: (1) Population Characteristics, (2) Study Question 1: Nutrition, (3) Study Question 2: Healthy Household Practices, and (4) Study Question 3: Agricultural Practices and Production.
1) Population Characteristics
Summary: The household survey shows that the average household size in the Kolda region is 13.8 among intervention groups and 13 in control groups. The average number of women aged 15 to 49 was 3 in intervention areas and 2.7 in control areas. Overall, respondents indicated having about 3 children less than 5 years old in their household (2.7 in intervention households and 2.5 in control). Most households in both intervention and control areas are headed by men (about 95 percent).
2) Study Question 1: Nutrition
Table 1: Nutrition Indicators Nutrition P- Indicator at Baseline 2015 Intervention Control Value 1.1 Proportion of children under 5 that are wasted 8.00% 7.81% 0.90 1.2 Proportion of children under 5 that are stunted 28.24% 31.96% 0.34 1.3 Proportion of children under 5 that are underweight 18.98% 21.46% 0.41 1.4 Proportion of underweight women 22.37% 27.26% 0.09 1.5 Proportion of children 6-23 with MAD 20.00% 0.00% NA 1.6 Number of months of reduced food intake 2.28 2.26 0.85 1.7 Proportion of HHs consuming < 2 meals per day (%) 1.83% 2.56% 0.64
Summary: To assess the nutritional status of children under 5 in the Kolda region, three anthropometric indicators were reported: wasting, stunting, and underweight (see Table 1). Wasting is a measure of acute malnutrition and is reported for children under the age of 5. The prevalence of wasting among children under 5 is higher in intervention areas (8 percent) than in control areas (7.81 percent).
Stunting is an indicator that reflects chronic undernutrition. The prevalence among children under 5 is higher among control areas (31.96 percent) than intervention areas (28.24 percent). Underweight is a weight-for-age measurement and reflects acute and/or chronic undernutrition. About 21.5 percent of children under 5 in control areas are underweight. The proportion of underweight children is slightly lower among children in intervention areas – 18.98 percent.
Page 5
Kolda Baseline Study
According to the household survey data, 22.37 percent of women in intervention areas are underweight. The prevalence of underweight is higher among women in control areas at a statistically significant level, with 27.26 percent measuring as underweight.
The minimal acceptable diet (MAD indicator) is calculated using data from the FTF population- based survey (PBS) conducted concurrently to the Kolda Baseline Study. The proportion of children between 6 and 23 months of age with MAD is 20 percent for the intervention group and 0 percent for the control group While the prevalence of children with MAD is higher among the intervention group than the control, these results are not statistically reliable due to the small sample size since the data from PBS is limited to the intervention and control villages that overlap with the Kolda Baseline Study.
In rural Senegal, it is common for households to reduce food intake for a period of time each year, also known as soudure, typically during the rainy season. The intervention group and the control group experience an average 2.28 and 2.26 of reduced food intake respectively. The results are not statistically significant between the two groups. Additionally, the survey data indicated that 1.83 percent of intervention households and 2.56 percent of control households have less than 2 meals per day.
3) Study Question 2: Healthy Household Practices
Kitchen Hygiene (Indicator 2.1), Safe Food Storage (Indicator 2.2), Water Treatment (Indicator 2.3), and Food Conservation Techniques (Indicator 2.4)
Table 2A: Healthy Household Practices Healthy Household Practices P- Indicator at Baseline 2015 Intervention Control Value 2.1 HHs practicing at least one Yaajeende kitchen hygiene 91.18% 92.97% 0.55 behavior 2.2 HHs practicing safe food storage 68.30% 70.63% 0.69 2.3 HHs treating drinking water 22.78% 13.72% 0.01 2.4 HHs practicing at least one food conservation technique 75.05% 57.61% 0.04
Summary: The household survey data in Table 2A show that a high percentage of households in both intervention and control areas (Indicator 2.1) practice at least one hygienic practice in the kitchen (92 percent and 93 percent respectively). Participants in the qualitative research demonstrated widespread knowledge of proper handwashing techniques and several indicated they have learned these techniques from training held by Yaajeende staff. During the focus groups discussions, many respondents highlighted their households have limited access to water. This constraint may hinder further improvements in kitchen hygiene practices.
Households in control areas (70.63 percent) are more likely than those in intervention areas (68.30 percent) to practice safe food storage. In contrast, a higher percentage of households in intervention areas practiced at least one conservation technique. According to the survey data,
Page 6
Kolda Baseline Study
75.05 percent of intervention households and 57.61 percent of control households report using at least one food conservation technique. The difference is statistically significant at the 0.01 percent level. While the difference cannot be attributed to Yaajeende activities, particularly at baseline, participants in the qualitative interviews have mentioned learning about food enrichment techniques from Yaajeende staff.
Household water treatment practices are more prevalent in intervention areas. The household survey data show that there is a 9.06 percentage points difference between the two groups. The difference is statistically significant at the 0.01 percent level. Most FGDs respondents reported knowledge and awareness of water treatment practices.
Salt Iodation and Storage (Indicator 2.5), Exclusive Maternal Breastfeeding (Indicator 2.6), and Food Diversity (Indicator 2.7)
Table 2B: Healthy Household Practices Healthy Household Practices Indicator at Baseline 2015 Intervention Control P-Value 2.5 HHs using and storing iodized salts properly (%) 9.28% 10.04% 0.79 2.6 Proportion of children 6-months that were 60.08% 66.67% 0.44 exclusively breastfed 2.7 Average Household Dietary Diversity Score 5.53 4.23 0.00 (HHDS)
Summary: The proportion of households using and storing iodized salt is low in both intervention and control groups (see Table 2B). The household quantitative survey reveals that only 9.28 percent of intervention households and 10.04 percent of control households use iodized salt and store it adequately. The qualitative research corroborated the results of the household survey. Several FGD participants reported not knowing about the benefits of iodized salt or only learning about its benefits from recent Yaajeende training.
The rate of exclusive breastfeeding among intervention households is lower than the rate among control households (60 percent versus 66 percent). These results are not statistically reliable due to the small sample size since the indicator is calculated using PBS data and data was limited to the intervention and control villages that overlap with the Kolda Baseline Study. FGDs, however, showed that most participants in the qualitative study are aware of the importance of exclusively breastfeeding during the first 6 months of life.
The average household dietary diversity score (HHDS) of 5.53 indicates that on average households in intervention areas consume between five to six of the 12 food groups. Households in control areas have lower dietary diversity (4.23), consuming on average between four to five food groups. The difference between the two groups is statistically significant at the 0.01% level.
Page 7
Kolda Baseline Study
Handwashing Station in Common Use (Indicator 2.8), Drinking Water from an Improved Source (Indicator 2.9), and Cooking Water from an Improved Source (Indicator 2.10)
Table 2C: Healthy Household Practices Healthy Household Practices Indicator at Baseline 2015 Intervention Control P-Value 2.8 Proportion of HHs with soap and water at a 0.6% 1.13% 0.39 handwashing station 2.9 Proportion of HHs using an improved water 41.02% 14.14% 0.01 source for drinking 2.10 Proportion of HHs using improved water 37.54% 20.75% 0.05 source for cooking
Summary: Access to clean water sources continues to be one of the biggest challenges faced by households in both control and intervention areas. The proportion of households with soap and water at a handwashing station is very low, with only 0.6 percent of households in intervention areas and 1.13 percent in control areas reporting having these available in their house. The low rates are most likely due to limited access to clean water.
About 41.02 percent of households in intervention areas report using an improved water source. The percentage among control households is significantly lower with only 14.14 percent of households having access to an improved water source for drinking. Similarly, the percentage of households using an improved water source for cooking (37.64 percent) in intervention areas was higher than in control areas (20.74 percent). The difference is statistically significant at the 0.05 percent level. However, it is important to note that, despite improvements in using improved water sources, most participants in the focus groups discussions mentioned they had limited access to clean water sources.
4) Study Question 3: Agricultural Practices and Production
The majority of respondents in the household survey in both intervention and control areas indicate they practiced farming, with 74.32 percent of all households breeding livestock. According to Table 3, households with livestock production is higher in intervention areas (78.35 percent) than in control areas (67.36 percent). Additionally, household income from the top five sources of income is much higher in intervention than control areas. The average income for the intervention group is 424,027 CFA (USD 703), whereas for the control group the average income is 247,963 CFA (USD 411). Both differences are significant at the 0.10 percent level.
Table 3: Agricultural Practices and Production Agricultural practices and production Indicator at Baseline 2015 Intervention Control P-Value 3.6 Households with Livestock Production 78.35% 67.36% 0.08 3.7 Percentage of Households with Livestock 7.32% 11.7% 0.05 Income 3.7 Average Income from The Top Five Sources 424,027 CFA 247,963 CFA 0.01 of Revenue in CFA
Page 8
Kolda Baseline Study
Farming plays an important role in the Kolda region as it helps households produce food not only for immediate consumption but also to generate income. The percentage of households with livestock income in Kolda, however, is currently low. The share of livestock income is higher among control groups – 11.7 percent – than intervention groups – 7.32 percent. Therefore, USAID/Yaajeende interventions that support livestock breeders such as Animal Placements and Passing on the Gift are particularly relevant in this region. The qualitative study revealed that farmers in Kolda are very receptive to the program. This confirms farmers and livestock producers in Kolda are willing to adopt new farming technologies – a key assumption of Yaajeende programming
While the qualitative study has found that agricultural training is generally well-received in Kolda, water supply and irrigation of farmland are significant challenges for the success and sustainability of the program in this region. The survey data show that rain-fed cereal is the main farming system in Kolda and is practiced by 90 percent of households in both intervention and control groups; all other systems of production are almost nonexistent in this region with around two percent of the population practicing irrigated cereal farming. However, qualitative research shows that, in recent times, rainfall in the rainy season has been irregular, unpredictable, and unevenly distributed creating either droughts or heavy floods. During the rainy season there can be dry spells that last a couple of days, which can have a damaging effect on the crop yields. Therefore, relying on rain-fed farming is risky and has a direct impact on food security, animal farming, and farm income, since households do not practice farming outside the rainy season.
Recommendations
1. Account for the timing of Kolda Baseline during the final impact evaluation. A number of nutrition-related indicators are sensitive to the timing of data collection. For instance, prevalence of wasting (Indicator 1.1) and underweight (Indicator 1.3) may sharply increase during an acute shortage of food. Similarly, the MAD indicator (Indicator 1.5) is partly based on meal frequency and the rate of children satisfying the MAD may decrease during the season of relative food insecurity. Household meal frequency (Indicator 1.7) may also sharply decline during the lean season because the indicator only accounts for the number of meals prepared in the last 24 hours. For Kolda Baseline Study, data were collected between November 2015 and January 2016, while the data collection for Yaajeende MIE and the baseline for the other regions occurred in May and June. If the final impact evaluation is conducted around May or June to align with the MIE and the baseline, Kolda’s season-sensitive nutritional baseline data may not be comparable. Therefore, in theory, data collection for the final impact evaluation should align with the timing of the Kolda baseline and occur between November and January. If this is not viable in practice, the final impact evaluation of Yaajeende should take into account that Yaajeende’s effect on nutrition may be underestimated for Kolda.
2. Track improvements in women’s dietary diversity. Yaajeende monitors the prevalence of underweight women in its intervention villages. However, this indicator is insufficient to accurately measure Yaajeende’s activities on nutrition of women. A number of women noted during the focus group discussions that they learned about the importance of nutrient-rich
Page 9
Kolda Baseline Study
diet for pregnant and breastfeeding women through Yaajeende. Therefore, it is recommended that Yaajeende track the number of women achieving minimum dietary diversity.2
3. Add Household Hunger Scale (HHS) to the final impact evaluation. HHS is a light- touch way of calculating the prevalence of households experiencing moderate or severe hunger. By adding three additional questions to the final impact evaluation questionnaire, the hunger level of Yaajeende’s intervention areas can be compared to other regions in Senegal using the Senegal PBS data and that of other countries.
2 This recommendation is based on the project performance indicators provided to the team in June 2015. Yaajeende’s revised M&E plan now includes an indicator measuring women’s dietary diversity as “percentage of direct female beneficiaries of nutrition-sensitive agricultural activities consuming a diet of minimum diversity”.
Page 10
Kolda Baseline Study
1. PROJECT BACKGROUND
1.1 Overview of the Yaajeende Program
Food insecurity continues to be a serious problem throughout Senegal and the West African region, leaving large segments of the population vulnerable to famine and hunger. USAID/Senegal is working with the Government of Senegal (GOS) to promote greater agricultural productivity and improve regional food security. The USAID Yaajeende Agriculture and Nutrition Development Program, commonly known as Yaajeende, is a five-year program that received a two-year extension (November 2010 to September 2017), implemented by NCBA/CLUSA, Counterpart International, Heifer International, and Sheladia Associates in four regions of Senegal.
Taking a structural approach to the question of food security, Yaajeende’s goal is to accelerate the participation of the very poor in rural economic growth and to improve the population’s nutritional status. The program falls under the Feed the Future Initiative (FTF) which was designed based on four Intermediate Result (IR) of the USAID Economic Growth Results Framework (See Annex 1)3: IR 1 Inclusive Agriculture Sector Growth; IR 2 Increased Trade; IR 3 Improved Nutritional Status, especially of women and children; and IR 4 Improved Management of Natural Resources.
IR 2, Increased Trade, is not directly targeted by Yaajeende since the program focuses on poor, food-deficit areas. Additional production in these areas is expected to be auto-consumed or marketed within Senegal.
At the core of the Yaajeende program is the Nutrition Led Agriculture (NLA) approach, which promotes improved production, trade, and local consumption of high quality, nutritious foods, including foods that resolve priority nutritional deficiencies. The NLA theory of change is that mutually supporting programs of nutrition and agriculture will be more efficacious in improving nutritional status than either of the mutually reinforcing components on their own. The NLA approach supports the development of a set of skills and techniques that are thought to have direct influence on local food security, defined as availability, access, utilization, and governance of food resources, in the communities where the project operates. Components of the approach include improving access to, and understanding of, relevant technologies and techniques, promoting structural changes, and developing the capacity of local institutions to create local markets for high quality nutritious foods.
Yaajeende program activities span the four FTF pillars of food security:
3 Yaajeende Performance Monitoring Plan (PMP) Revised: December 6, 2013
Page 11
Kolda Baseline Study
1. Availability. Interventions related to farm production – this bundle of activities is aimed at introducing and increasing production of key crops such as Vitamin A-rich orange flesh sweet potato, millet bio-fortified in iron and zinc, and micro-nutrient rich fruits and vegetables; 2. Access. Interventions related to farm productivity – these activities are aimed at increasing farmers’ access to inputs and agriculture services that permit enhanced production of nutritious food crops via a network of private sector, Community-Based Service Providers (CBSPs or Agents de Prestation de Service (APS)); 3. Utilization. Interventions aimed at creating demand for nutritious foods and potable water – these activities educate the public about the need for a diverse diet that includes fruits and vegetables and increases their ability to prepare these foods in ways that preserve and maximize the food’s nutritional content; and 4. Sustainable governance. Interventions aimed at strengthening local government and civil society – These activities strengthen local actors’ ability to engage in creative and dynamic partnerships to guide food production and water-related activities, and administer related resources, in an equitable and inclusive fashion that prioritizes the needs of the most nutritionally vulnerable populations.
Distinct from traditional development programs, Yaajeende trains and relies on local agents of change to effect behavior change. Agents include CBSPs, Community Nutrition Volunteers (CNVs or Volontaire de Nutrition Communautaire (VNC)), relays, and auxiliaries who work closely with Yaajeende staff. CBSPs are trained on components of Yaajeende programs that they can then sell as services in their local communities. They are organized into regional networks with regional steering committees, which help them to easily purchase products and resell them locally, thereby addressing the challenge of limited access to goods. Set up by the World Bank’s Nutrition Enhancement Program (NEP), Yaajeende utilizes the existing network of CNVs and train them on nutrition topics and give starter kits, allowing them to hold Mother to Mother (MtM) groups, provide local training, and conduct Q/A sessions (causeries in French) to ensure participants have absorbed the material. CNVs are also trained on animal health and support Yaajeende’s livestock program.4 Relays benefit from, and assist with, Yaajeende agriculture or livestock training, serving as a local resource for communities. Auxiliaries are government extension agents that similarly serve as local resources and ensure animal health based on Yaajeende husbandry training.
1.2 Yaajeende Activities
Within the NLA framework, Yaajeende’s nutrition program activities cover a wide range of behaviors concerning food choice, food preparation, food conservation, water, and sanitation, while agriculture program activities promote techniques for commercial and subsistence agriculture, horticulture, arboriculture, and livestock. Agriculture and livestock activities focus on local food production over international supply chains. Nutrition programs complement locally available foods, rather than advocating for the adoption of exotic foods and techniques. Yaajeende provides minimal subsidies to key activities and slowly transfers project ownership to local
4 CNVs can sell services to their local communities and thus become Nutrition CBSPs.
Page 12
Kolda Baseline Study
populations in activity areas.
The following sections provide further details on activities implemented in each of the four FTF pillar categories: Pillar 1: Availability
Conservation Agriculture and Soil Health Program. In this intervention, a cadre of CBSPs offer a package of farming practices to local producers. This package, called the Zero Risk Package, consists of skills including composting and intercropping along with tractors and rippers, improved short-cycle seeds, organic fertilizer, microdose chemical fertilizers, and crop insurance. Women’s groups are eligible to participate in a bio-reclamation of degraded lands (BDL) program that teaches women techniques to transform unused, biodegraded land into productive land. Men are also trained on these techniques in order to assist their wives with intense labor. Yaajeende assists some of these women’s groups to procure deeds to lands for at least 25 years. It is otherwise difficult for women’s groups and individual women to own land, as fields are typically passed down from father to son.
Agroforestry and Arboriculture. Private sector nurseries maintained by Arboriculture CBSPs are the primary method of extending agroforestry to participating communities. Such nurseries sell trees like Acacia Melifera, Mango, Moringa, Bauhinia, Cashew, Madd and Henna. They install live fencing on community projects including schools, land reclamation sites, commercial gardens, and community gardens. They also plant fruit trees on land adjacent to houses, schools, and health huts. Yaajeende also provides training on grafting the Sahel apple, rich in Vitamin C, onto jujube trees.
Rain-fed Agriculture. Rain-fed agriculture is the most common form of agriculture in Yaajeende intervention zones and its successful use during the rainy season is crucial to farmers. Yaajeende recognizes this by identifying appropriate seasonal strategies. The program focuses its efforts on a group of leading producers (grands producteurs in French) that adopt a package of techniques, including intercropping, use of improved seed varieties, and use of fertilizer. CBSPs offer a number of key agricultural services: tillage, organic fertilizer, improved seeds, enhanced crop varieties, cereal processing, and cereal storage. Yaajeende promotes deep urea placement for rice cultivation to avoid nutrient losses caused by surface broadcasting.
Irrigated and Flood Recession Agriculture. In partnership with the Government of Senegal and other projects, Yaajeende promotes irrigation and flood recession agriculture in appropriate regions so that farmers can be productive outside of the rainy season. CBSPs provide seed, fertilizer, seedlings, and training to local populations. They also lease and sell irrigation pumps to communities with optional service agreements for repair. Large producers that purchase irrigation pumps are trained in accounting and financial planning. CBSPs also offer crop insurance with improved seed and tillage.
Commercial Horticulture. Rainy and counter-season commercial horticulture aim to increase the sale and consumption of fruits and vegetables in the project zones. Large-scale commercial gardens are supported by CBSPs through a broad range of services: financing, accounting,
Page 13
Kolda Baseline Study
marketing, irrigation, seed, fertilizer, and various technical training to increase farmers’ skills in different seasons. Women’s groups farm community gardens with Yaajeende support. Crops include tomato, okra, bissap, hot pepper, eggplant, bitter eggplant, cabbage, lettuce, and onion.
Bio-fortified Crops Program. Both CBSPs and producer organizations promote the adoption of nutritionally enhanced hybrid varieties of maize, rice, millet, sweet potato, and beans. MtM groups and large producers participated in trials of bio-fortified orange flesh sweet potatoes that they have since adopted.
Seed Production Program. In partnership with public and private sector organizations, Yaajeende promotes seed multiplication activities. Specific partners include ISRA, Tropicasem, Hortis, Agroseed, Regional Rural Development Agency (DRDR), and the Association of Producers of Corn and Sorghum in the Senegal River Valley.
Livestock Enterprise Program. MtM and Citizen Work Groups (CWG or GTC in French) are involved in many phases of the animal husbandry program, including targeting of direct animal subsidies of chicken, goat, and sheep. CBSPs conduct training on animal care, breeding, marketing, and dairy products. Animal insurance was introduced using microfinance institutions in 2014 in certain locations. Recipients of direct animal subsidies have begun growing forage cowpeas and dolich for animal feed across the project zones. Emerging livestock breeders are supported with training from Yaajeende.
Livestock Health Program. CBSPs, relays, and auxiliaries trained in animal husbandry provide veterinary care and track the health of animals issued from the Passing on the Gift (POG) program, having identified animal health as an area in need of significant support in their areas of intervention. Yaajeende supports these businesses with marketing in cooperation with the Government of Senegal. The program has assisted the first private veterinarian to open a practice in Kédougou and involves veterinarians in the other regions in the livestock program. CBSPs also fabricate and sell mineral licks to breeders to increase consumption of important nutrients.
Passing on the Gift, or passage du don, is a livestock program with roots in traditional West African lending practices. POG is a direct livestock subsidiary to beneficiaries. Beneficiaries receive lots of livestock, either ten fowl or three small ruminants. Each head of livestock is intended to be repaid to the community’s pool of animals for subsequent subsidy. Beneficiaries are educated on animal care and dairy practices. Beneficiaries can sell or consume the dairy products from livestock, and learn the nutritional and financial benefits of livestock. Livestock recipients are also obliged to purchase insurance and veterinary services. In POG villages, the project also encourages local markets for pasture services, meaning herders are available for hire to care for animals in the pasture on behalf of beneficiaries. Targeting of the POG program is a participatory community process. Where POG has long been active, some villages describe that the pool of beneficiaries has been completely saturated. Pillar 2: Access
Financial and Insurance Program. Yaajeende has trained CBSPs and Coach LRPs on issues related to credit, including Decentralized Finance Systems (DFS), financial education, and how to
Page 14
Kolda Baseline Study
submit credit applications. Coach LRPs help CBSPs in managing credit records. Credit has been obtained for agriculture, livestock, horticulture, CBSP activities, staple crops, and processing foods. Agricultural insurance, for livestock mortality and crops, is another component of this program.
CBSP Mechanization, Postharvest and Marketing Program. This program focuses on reinforcing CBSPs to address challenges in the agriculture program. CBSPs receive training to professionalize and reinforce the capacity of their network. This increases linkages between private sector firms and CBSPs in order to distribute more inputs. This program includes investigating postharvest options such as cold storage for seeds and postharvest equipment facilitation. Marketing of surplus agricultural production by CBSPs is also part of this program.
Nutrition-based Enterprises. This Year 4 program aims to promote the emergence of enterprises that transform and process nutritious foods that can be marketed through the CBSP networks, by supporting CNVs to begin providing services as Nutrition CBSP. Pillar 3: Utilization
Educational and Nutritional Gardening. Yaajeende supports community gardens and trains CNVs on micro-gardening itineraries and the nutritional importance of vegetables. CNVs, in turn, train MtM groups on these topics and work with MtM groups in their community gardens. MtM participants keep micro-gardens in their homes and consume the vegetables they grow within the home. MtM groups are taught to use compost in their gardening activities and are given inputs at the beginning of the program to support these activities. CNVs track the amount of produce grown, sold, and consumed from community gardens. School gardens are set up to educate students on the growth and consumption of vegetables and to improve the students’ diets through the school cafeteria. A recipe book incorporating wild and nutritious foods is currently being assembled for distribution amongst program beneficiaries.
Potable Water, Sanitation and Hygiene Program. CNVs train MtM groups on Water, Sanitation, and Hygiene (WASH) issues and about Community Led Total Sanitation (CLTS). This training aims to develop participants’ skills in areas such as latrine management, trash collection, and hand-washing to reduce diarrheal diseases, especially in children. They have taught people to create holes filled with rocks and charcoal to receive used water from latrines to prevent this water from being absorbed into the water table. CNVs also teach MtM groups to make simple handwashing stations called TippyTaps and drying racks covered with mosquito nets to protect utensils, plates, etc. from animals and insects when drying after being washed. CBSPs sell soap, bleach, and filtered water as part of this program.
Food Fortification and Transformation Program. Wild foods activities include promotion and distribution of recipes with micronutrients (iron, zinc, vitamin A, iodine). Household fortification of flour includes incorporating cowpeas, peanuts, and/or corn for later use in locally produced enriched flours that mothers are taught to incorporate into their families’ diets for improved nutrition. Participants are trained to process milk into cheese and yogurt. In cooperation with the NGO ACCRA, Yaajeende participants have been trained to use a solar dryer to dry grains, okra, and beans in the Matam region.
Page 15
Kolda Baseline Study
Behavior Change Communication (BCC) Program. BCC activities include community meals, MtM meetings, WASH activities, and awareness caravans. This program includes implementing activities around the Essential Nutrition Actions (ENA or AEN in French) and teaching members of the MtM groups the principles of the ENA. CNVs also teach MtMs the importance of consuming iodized salt and proper storage techniques so the salt retains its nutrients. This also includes behavior change activities targeting grandmothers and men.
Social Marketing Program. Vitamin A is a major concern in all Yaajeende intervention areas. To build demand to support commercial production of Vitamin A-rich produce, facilitate understanding around the contribution of Vitamin A to good health, and encourage orange foods consumption, the project developed a major social marketing campaign called “Eat Orange.” This campaign focuses on getting people to consume orange flesh sweet potato, mangoes, carrots, papaya, and squash. Campaigns have also been run on conservation agriculture, biofortified crops, seed breeding, and livestock vaccination. Pillar 4: Sustainable Governance
Local Governance and Civil Society Organization (CSO) Capacity Building Program. CWGs for Food Security engage in the leadership of food security issues within localities in collaboration with LRPs, which include CBSPs, CNVs, and Producer Organization Agents (POAs). CWGs apply for land grants for women to help them gain formal access to bio-degraded lands. Yaajeende develops the capacity of CWGs by having Governance Coaches teach them elements necessary to run an organization.
Local Partner Capacity Building Program. Yaajeende signs contracts with federations and other partners on targeted techniques for the implementation and follow-up of food security activities. This program strengthens local partner producer organizations on identified weaknesses according to capacity building plans. Cross-cutting Activities
In addition to activities falling under the four FTF Pillars, Yaajeende implements activities in a number of cross-cutting areas. These include:
Gender. The majority of participants in the counter-season commercial horticulture program are women. The gender dimension is widely present in project interventions designed to improve maternal, infant and child health; in the rehabilitation of biodegraded lands; in the promotion of hygienic cooking practices; etc. Women’s unique contribution to health and nutrition is recognized in, e.g., the putting in place of MtM networks.
Partnerships. Yaajeende has benefited from many partnerships since 2011, such as with USAID/PSSCII (Community Health Program), Child Fund’s Community Health Program, USAID/PCE (Economic Growth Project), ARD (Regional Development Agency), PRN (Nutritional Reinforcement Program), and Teranga Gold Operations. Yaajeende collaborates with Africare and World Vision and does research with University of Cheikh Anta Diop and
Page 16
Kolda Baseline Study
ICRISAT (International Crop Research Institute for the Semi-Arid Tropics). The biggest collaboration has been with the Institut du Sénégal pour la Recherche Agricole (ISRA) for the introduction and trial of improved and bio-fortified seeds, fertilizer, and orange fleshed sweet potato cuttings.
Climate Change Adaptation. The “Zero Risk” package promotes conservation agriculture which is best suited to climate change adaptation. Training
Nutrition training for CBSPs and CNVs covers the following topics: ENAs, enriched flour, transformation of fruits and vegetables, sweet potato marmalade, handwashing, purification of water, preparation of enriched flour, diet of children aged 6-24 months, importance of the three food groups, transformation of sweet potato, hibiscus syrup, transformation of onion, transformation and conservation of milk, transformation of jujube into jujube galette, WASH, training on the 1000 days from pregnancy to 24 months, fabrication of mango jam and transformation of fruits and vegetables, use and maintenance of latrines, nutrition education, importance of Vitamin A, micronutrients, and domestic water treatment. Agriculture training covers: agricultural techniques, bio-restoration of degraded lands, market gardening, rice growing, and husbandry. Training for school children covers WASH and nutrition education.
1.3 Yaajeende Activities in Kolda
The implementation of Yaajeende started in 2011 in the Matam, Bakel and Kédougou regions and expanded to Kolda region in April 2014. Yaajeende interventions in the Kolda region of Senegal differ from the three other regions. Because the Kolda implementation team consists primarily of former staff of Yaajeende in Kedougou, the team has had the advantage of capitalizing on lessons learned from previous implementation. As a result, the Kolda Regional Office has omitted a number of activities implemented in the other regions, and, in turn, has developed new activities tailored for Kolda.
Table 1.1 below compares the activities of Yaajeende in Kolda to the interventions in the Matam, Bakel and Kedougou regions by beneficiary population, and Table 1.2 compares the activities by stakeholder type. Several new initiatives have been implemented in Kolda, such as “Child Granaries (Le Grenier des Enfants in French)”. Farmers are asked to donate part of their harvest of cereals for the children of the village. The donations are stored and managed by the CNVs who have been provided with a recipe to make enriched flour. In turn, the CNVs teach women how to make the flour and distribute the donated cereals to the households with children during the time of food scarcity. This recipe of enriched flour is nutritious and pregnant women and children are encouraged to consume it as well as the entire family whenever possible.
Page 17
Kolda Baseline Study
Table 1.1 Activities and Interventions of Yaajeende by Beneficiary Population5
INTERVENTIONS IN MATAM, BENEFICIARIES INTERVENTIONS IN KOLDA BAKEL AND KÉDOUGOU Nutrition Eat Orange Enriched flour Enriched flour Non-timber food Non-timber food Child Granaries (New for Bio-fortified foods (mill, yellow Kolda) sweet potato) Diversified diet Diversified diet Gardening Commercial and nutritional Micro nutritional gardening in market gardening the households Access to Land MOTHER TO Restoration of degraded land Restoration of degraded land MOTHER Access to land Access to land 6 (MTM) Animal Production Passing on the Gift (Passage du Breeding of quails Don) (New for Kolda) Vaccination Direct subsidies Water and Sanitation Awareness and Communication Awareness and Communication Campaign for Behaviour Change Campaign for Behaviour Change for: for: Sanitation Sanitation Hand washing Hand washing Involvement of men in activities Involvement of men in activities promoting nutrition and hygiene promoting nutrition and hygiene Agriculture Conservation farming Conservation farming HEAD OF (motorised ripping, offsetting, (motorised ripping, offsetting, HOUSEHOLDS access to credit, utilization of access to credit, utilization of AND FARMERS compost, assurance, etc.) compost, assurance, etc.) Seed production Access to quality farming input
5 The table lists Yaajeende interventions and activities implemented as of December 2015, during the time of the data collection. This may not reflect Yaajeende’s activities for FY16 of FY17. 6 The Passing on the Gift (POG or Passage du Don in French) started in FY16 for Kolda.
Page 18
Kolda Baseline Study
INTERVENTIONS IN MATAM, BENEFICIARIES INTERVENTIONS IN KOLDA BAKEL AND KÉDOUGOU Irrigation, flood recession Involvement of men in activities agriculture promoting nutrition and “Child Involvement of men in activities Granaries” promoting nutrition Hygiene in the households and Hygiene in the households villages
Table 1.31.2 Activities and Interventions of Yaajeende by Stakeholder Type7
STAKEHOLD INTERVENTIONS IN MATAM, INTERVENTIONS IN KOLDA ERS BAKEL AND KÉDOUGOU Service Delivery Agriculture: plowing, motorized Agriculture: plowing, motorized ripping, offsetting, sale of ripping, offsetting, sale of farming farming input and pesticides input and pesticides CBSP Insurance brokerage et farm Insurance brokerage et farm credit credit Gardening and arboriculture Gardening and arboriculture sale sale of farm input, nursery and of farm input, nursery and plants plants Marketing of nutritional Marketing of nutritional products products Food processing of cereals, fruits Food processing of cereals, and vegetables NUTRITION fruits and vegetables CBSP Marketing of hygiene products Marketing of hygiene products Sensitization and messaging for Sensitization and messaging for behavior change behavior change Sensitization and Messaging for Behavior Change Nutrition (enriched flour, Nutrition (enriched flour, diversified diet, iodized salt diversified diet, iodized salt) CNV etc.) Hand washing with Tippy Tap Hand washing with Tippy Tap Set-up and management of community garden VETERINARY Livestock vaccination Not implemented in Kolda
7 The table lists Yaajeende interventions and activities implemented as of December 2015. This may not reflect Yaajeende’s activities for FY16 or FY17.
Page 19
Kolda Baseline Study
STAKEHOLD INTERVENTIONS IN MATAM, INTERVENTIONS IN KOLDA ERS BAKEL AND KÉDOUGOU Animal health (medical supplies) Field management of Yaajeende Field management of Yaajeende activities activities Engagement in the design of Engagement in the design of food food security and nutrition security and nutrition planning planning Engagement of local elected Engagement of local elected representatives in the activities CWG AND representatives in the activities Advocacy for woman access to LSC Advocacy for woman access to land land Promotion of access to high Promotion of access to high quality farm input for producers quality farm input for Advocacy for water and producers sanitation; drinking water Advocacy for water and Breeding of quails managed by sanitation; drinking water LSC EMERGING Livestock companies (poultry Not implemented in Kolda BREEDERS rearing, animal husbandry)
Yaajeende is encouraging men to play a greater role in nutrition and WASH in their communities. As food producers, they are encouraged to donate part of their cereal harvest to the “Child Granaries”. Yaajeende has also promoted the involvement of men in its hygiene and sanitation activities in Kolda, including a campaign and community clean-up of public areas in the village.
Whereas animal farming was encouraged in the three other regions, it is not currently implemented in Kolda. The program has instead focused on the rearing of quails. Quail eggs have been introduced to households as a very nutritious addition to their diet. The LSC manage quail rearing and distribute the eggs to the households. Yaajeende plans to introduce quail rearing and Quail Passing of Gift in the households. The approach by Yaajeende in Kolda has been to first introduce and test a new activity in the community, then establish it at the household level, and finally to scale up the activity in other targeted regions.
Gardening at the household level and among women is encouraged in Kolda, while commercial and nutritional market gardening is promoted in the three other regions.
1.4 Population of Kolda
Senegal is divided in 14 regions, which are further sub-divided into departments and communes. The region of Kolda is made up of three departments: Kolda,
Page 20
Kolda Baseline Study
Medina Yoro Foulah, and Velingara.
In total, there are 40 rural communes in Kolda distributed among the three departments as shown in Table 1.3 (the distinction between intervention and control villages will be elaborated in the Sampling section of this report). According to the 2013 census data published by the National Agency of Statistics and Demography (ANSD), Kolda has a population of over 662,000. At the time of this study, there were 13 communes in Kolda that had received interventions from the Yaajeende program8.
The Zone of Influence (ZOI) of the Yaajeende program has been defined as the communes where the interventions have been implemented. The non-intervention or control areas for this study are the communes where there are no interventions as of September 2015.
Table 1.3 Population of Kolda by Intervention and Control Groups
Total Intervention Control Department Commune Persons Commune Persons Commune Persons Kolda 15 245,990 6 83,501 9 162,489 Medina Yoro Foulah 11 138,084 3 61,163 8 76,921 Velingara 14 278,382 4 68,981 10 209,401 Total 40 662,456 13 213,645 27 448,811
8 Yaajeende is implementing its program on a rolling basis. The list of villages used for this study was established in September 2015.
Page 21
Kolda Baseline Study
2. PURPOSE OF THE KOLDA BASELINE STUDY
USAID is committed to measuring and documenting the progress of its projects to gain an understanding of return on investment and to maximize program impact. It recognizes the importance of utilizing rigorous evaluations to obtain systematic and meaningful feedback on its development initiatives. Evaluations, contextualized to beneficiary countries, permit evidence- based decision-making and are fundamental to the Agency's success9.
The purpose of the Kolda Baseline Study, therefore, is to establish baseline measures in preparation for a final impact evaluation of Yaajeende activities in the Kolda region of Senegal. The specific objectives of the impact evaluation are to:
1. Establish baseline values of 16 key target indicators defined in the Yaajeende Performance Monitoring Plan (see indicators below), and 2. Provide supporting information on dynamics around household revenues, income, and adoption of key behaviors thought to impact nutritional status.
Although villages in Kolda were surveyed in a baseline assessment that took place in 2011, the data are over three years old and may not reflect the current status of indicators in the region. Therefore, this study is intended to produce up-to-date indicator values for Kolda. It is important to note that because the Baseline Study is being conducted fifteen months after the program activities launched in Kolda, the final indicator values may deviate slightly from indicator measures prior to the arrival of Yaajeende. Because the Baseline Study is being conducted after the start of program activities, it has the added an objective of:
3. Providing a description of what is happening in the field among the intervention villages where Yaajeende has started its activities.
It is expected that the Kolda sample will eventually be included and analyzed with the overall final impact evaluation of the Yaajeende program. For this reason, this Baseline Study follows the same methodology as the 2011 Baseline and the 2015 Mid-term Impact Evaluation (MIE) of Yaajeende in the Matam, Bakel, and Kedougou regions of Senegal.
The final impact evaluation for the four regions of Senegal will attempt to answer the following questions when Yaajeende finalizes its activities in these areas:
1. Nutrition: Did households and individuals living in villages located in project treatment intervention areas see greater improvement in nutritional status indicators than those residing in non-project areas? 2. Healthy household practices: Did households living in villages located in project treatment areas see greater adoption of healthy nutritional and WASH practices than those residing in non-project areas?
9 USAID Evaluation Policy January 2011
Page 22
Kolda Baseline Study
3. Agricultural practices and production: Did households living in villages located in project intervention areas see greater use of improved agriculture and livestock practices and technologies than households living in non-project areas? Did those practices and technologies lead to greater agriculture production and greater productivity? 4. Nutrition-led agriculture: Did individuals and households who benefitted from both nutrition and agricultural project components experience greater improvement than those who benefitted from neither or from only one? The Yaajeende theory of change holds that nutrition-led agriculture will be more effective in improving nutritional status than the sum of the individual effects of nutrition and agriculture interventions alone.
The results of the household survey and qualitative research conducted to establish a baseline for these study questions are presented in this report, along with recommendations for Yaajeende based on the results.
2.1 Indicators
Each of the study questions above is informed by a series of specific, measurable, and concrete key indicators. The Baseline key indicators correspond closely with the Performance Monitoring Plan (PMP) key indicators and in many cases are taken directly from the PMP definitions. Where indicators differ from PMP definitions, they have been developed in dialogue with project and USAID personnel. Annex II presents the correspondence of indicators with the PMP indicators.
Most of the key indicators are binary variables that take the values zero or one – e.g., a child either meets the criteria for wasting or he does not; a household either consumed fewer than two meals yesterday, or it did not. However, for convenience and clarity, indicators are generally described in terms of prevalence among all the individuals or all the households in the survey (i.e., proportion of individuals or proportion of households, a number between zero and one). The key indicators associated with the Study Questions above are as follows:
Study Question 1 – Nutrition 1.1 Wasting among children aged 6-59 months.10 Defined according to the child’s z- score on a weight-for-length curve using World Health Organization (WHO) reference data. 1.2 Stunting among children aged 6-59 months. Defined according to the child’s z-score on a length-for-age curve using WHO reference data. 1.3 Underweight among children aged 6-59 months. Defined according to the child’s z- score on a weight-for-age curve using WHO reference data. 1.4 Underweight among women aged 15-49 years. Defined as a body mass index (BMI) below 18.5. 1.5 Minimum acceptable diet (MAD) among children aged 6-23 months.
10 Indicators 1.1, 1.2, 1.3, 1.4, and 1.5 are binary variables observed at the individual level. Trends, treatment effects, and related PMP targets are estimated as a change in prevalence of the key indicator among beneficiary populations.
Page 23
Kolda Baseline Study
1.6 Average duration of reduced food intake in the household (months). 1.7 Fewer than two meals in the previous 24 hours prepared in the household.11
Study Question 2 – Healthy household practices 2.1 Household practices at least one hygienic kitchen behavior, including handwashing and hair covering.12 2.2 Households practices improved food storage practices, including cold storage and covered storage. 2.3 Household treats drinking water using at least one of the following: bleach, filters, and silver filters (binary). 2.4 Household practices at least one food conservation technique, including fermentation, germination, torrefaction, drying, or fortification (mélange). 2.5 Household uses and properly stores iodized salt. 2.6 Exclusive maternal breastfeeding of infants under 6 months of age. 2.7 Household food diversity score (1.12). 2.8 Household has a handwashing station in common use. 2.9 Household drinks water from an improved source, meaning from a covered well, faucet, or deep well. 2.10 Household cooks with water from an improved source, meaning from a covered well, faucet, or deep well.
Study Question 3 – Agricultural practices and production 3.1 Poverty, estimated as a propensity to fall below the World Bank’s USD $1.25 Purchasing Power Parity (PPP) daily income line.13 3.2 Surface area devoted to agriculture (ha). 3.3 Surface area devoted to horticulture (ha). 3.4 Surface area devoted to irrigated agriculture (ha). 3.5 Surface area devoted to decrue, or flood-plain agriculture (ha). 3.6 Total agricultural production (kg). 3.7 Total revenue from agriculture (FCFA). 3.8 Household purchases seed.14
11 Indicator 1.7 is a binary indicator observed at the household level. Trends, treatment effects, and related PMP targets are estimated as changes in prevalence among households. 12 Indicators 2.1, 2.2, 2.3, 2.4, 2.5, 2.8, 2.9, and 2.10 are binary variables observed at the household level. Trends, treatment effects, and PMP targets are estimated as changes in prevalence among households. 13 Poverty is estimated using the scorecard approach in Marc Schreiner (2009) “A simple poverty scorecard for Senegal,” available at http://www.microfinance.com/#Senegal. The household’s likelihood of poverty is estimated as a two-digit percentage, based on responses to ten simple, multiple-choice questions. The scorecard can be calibrated to any of a menu of poverty lines, including USAID extreme poverty, the national poverty lines, and multiples of the preceding. By multiples, we mean double or triple the income of that poverty line; for example, the scorecard can be calibrated not only to the World Bank’s $1.25 PPP daily income line, but also to $2.50 PPP and to $3.75 PPP daily income. 14 Indicators 3.8, 3.9, 3.11, 3.12, and 3.13 are binary variables observed at the household level. Trends, treatment effects, and PMP targets are estimated as a change in prevalence among households.
Page 24
Kolda Baseline Study
3.9 Household purchases fertilizer. 3.10 Index of agriculture technology adoption.15 3.11 Household uses an improved seed source: government technical service, specialized vendor, non-governmental organization (NGO) or CBSP. 3.12 Household purchases goods or services from a Yaajeende CBSP. 3.13 Household uses an improved fertilizer source: government technical service, specialized vendor, NGO or CBSP. 3.14 Head count of individuals in household that have attended agricultural training in last 12 months.
The indicators related to Study Question 3 are designed to highlight changes to the practice of agriculture and livestock rather than the value of direct subsidies received. Study Question 3 implicitly also asks whether project participation affects overall poverty rates, so we include a poverty assessment scorecard based on simple observable questions such as house construction materials, educational attainment, and ownership of consumer durables.
Study Question 4 asks whether nutrition-led agriculture is more effective than nutrition and agriculture programs independently for each of the indicators under Study Question 1, 2, and 3. Therefore, no additional indicators are listed under Study Question 4.
Table 2.1 below lists the PMP indicators that are also studied in the Baseline Study. Targets for change at the endline are as listed in the PMP as of December 2013. “Track only” in the table refers to case where no quantitative target was set.
15 Indicator 3.10 is a count of specific Yaajeende agriculture techniques adopted by the household. See Annex II for a complete list of the techniques scored in the index. It takes natural number values of 0 or greater, and the maximum value in the sample is 17.
Page 25
Kolda Baseline Study
Table 2.1 PMP Targets Related to Future Impact Evaluation16
Cumulative Ind. Target Type No. Project Performance Indicator Change % of Households that have increased dietary diversity score Outcome 1 80% by at least 8%. Number of HHs adopting improved practices or behavior Outcome 6 Track only after training by USAID/Yaajeende (new) Outcome 8 Number of Households with improved livestock production Track only Number of Households with Increased livestock related Outcome 9 Track only income Total number of months of the previous 12 months a Outcome 27 household was unable to meet its food needs (annual -30% reduction compared to the baseline) Prevalence of stunted children under five years of age Impact 28 -20% (Reduction) Prevalence of underweight children under five years of age Impact 29 -25% (Reduction) Impact 30 Prevalence of wasted children under five years of age (new) Track only Impact 31 Prevalence of underweight women (new) Track only Reduction in % of households that consume fewer than 2 Outcome 32 -50% meals per day Prevalence of children 6-23 months receiving a Minimum Outcome 33 Track only Acceptable Diet MAD) Percentage increase over baseline of households using Outcome 35 30% iodized salt and storing it properly Number of HH adopting food processing, food safety or Outcome 36 25000 nutrition practices due to USAID/Yaajeende (new) Number of HH adopting improved water, sanitation and Outcome 40 9,500 hygiene practices due to USAID/Yaajeende Percent of households with soap and water at a Outcome 41 30% handwashing station commonly used by family members Outcome 42 Percent of households using a drinking water source 50%
16 Table 4 is as listed in the Performance Monitoring Plan (PMP) as of December 2013. Since then a revised PMP has been released for FY16 and FY17.
Page 26
Kolda Baseline Study
3. METHODOLOGY
The Kolda Baseline Study used a mixed-method approach involving a non-experimental quantitative strategy and qualitative techniques. To collect quantitative data, the Kolda Baseline Study utilizes a household survey to collect information from a randomly selected sample of households in comparison and intervention areas as defined by the Yaajeende program. The survey is specifically designed to establish the values of the 16 basic indicators listed in the Scope to Work for the assignment (Annex III).
The initial values of the basic indicators are intended to provide a benchmark for measuring changes that can be reasonably attributed to the Yaajeende intervention using a difference-in- difference (DD) model. This model assumes that changes within a control group over time that has not received assistance from the program represents what would have happened without the intervention and provides a comparative value with which to assess the actual impact of the program. While Yaajeende does not expect to observe tangible changes in the short-term among slow-moving indicators, there are changes in practices and behaviors that can be measured soon after interventions have been implemented (e.g. the practice of washing hands or drinking treated water, which can result in a reduction in illness and, in turn, improve health).
To collect qualitative data, the Kolda Baseline Study utilizes key informant interviews (KIIs) and focus group discussions (FGDs) in intervention areas only. These methods are intended to investigate how Yaajeende activities are being implemented in Kolda and how they are being received by beneficiaries and collaborators. The qualitative component of the study has the added benefit of providing preliminary information to stakeholders that can be used to modify programming to overcome any unexpected outcomes, as needed.
The results of the qualitative research are triangulated with the results of the quantitative survey, and the outcomes of this analysis are presented in the Findings section of this report.
3.1 Household Survey
An electronic household survey was conducted on handheld tablets in all communes of Kolda by a local survey firm, selected through a competitive procurement process. The data collection period ran for approximately six weeks from 8 November to 13 December 2015. In total, 971 households were visited in 75 selected villages and 894 were found eligible for the survey and interviewed. Data were collected for 12,645 persons, of whom 2,784 were women aged 15 to 49 with children less than five years old and a total of 2,360 children. The target sample size of 900 households was not reached as one of the sampled villages had fewer than 12 households (see Sample Design/Size/Allocation below for more information).
The baseline survey questionnaire contained two components, each targeted to a different respondent:
1. Household, Revenue, and Agriculture Questionnaire: the respondent was the Head of Household or a person assigned by the Head of Household; and
Page 27
Kolda Baseline Study
2. Nutrition and Child Health Questionnaire: the respondent was a woman age 15 to 49 with children under the age of five in charge of the household.
The survey consisted of nine modules: 1) list of persons; 2) household goods and furnishings; 3) revenue sources; 4) surface area cultivated; 5) agriculture and livestock; 6) debts and financial services; 7) food consumption; 8) nutrition and health; and 9) anthropometry of women and children. Administering the survey to a household took, on average, 3-4 hours. At every point in the survey, from enumerator training to data cleaning, a comprehensive data quality assurance (DQA) program was implemented, described below (see Data Quality Assurance). Anthropometric measurements of women and children between the ages of 6 months and 5 years were collected by anthropometry experts with the assistance of team supervisors. The survey questionnaire is in Annex IV. Sample Design
The Kolda Baseline household survey followed the sample design of the Yaajeende Mid-term Impact Evaluation, a two-stage stratified sample design. The target population was all households with children under 5 years of age in the Kolda region. The population was stratified by department, the ZOI and non-intervention areas. The ZOI includes the communes where the Yaajeende program has been implemented. Non-intervention areas are the communes where there are currently no interventions. For this survey, villages were the primary sample units (PSUs) and were selected from the communes. In total, 75 villages were randomly selected from intervention and control communes. Households were the secondary sampling units (SSUs). Twelve households were randomly selected from each of the 75 selected villages. The villages were selected in the following manner:
1. Intervention villages. The Yaajeende program has been implemented at varying levels of intensity (high, medium, or low)17 in each intervention village. The list of intervention villages as of September 2015 was provided by Yaajeende and contained a total of 142 villages. This list included the number of persons living in the village as per the Programme National de Développement local (PNDL)18 repository as well as the program intensity level. Using the Probability Proportional to Size (PPS) method, 45 intervention villages were randomly selected from among all program intensity stratum. 2. Control villages: These are villages in the communes where no intervention is taking place and which are geographically distant from the villages in the ZOI. The latter specification was
17 Based on the interventions that the project provided to each village, the village also received an intensity score. The intensity score could be low, medium or high, based on how many of the packages the village received. Details of the project intensity can be found in Table 3.4 Sample Allocation. 18 Yaajeende uses this repository when selecting villages to implement interventions. This repository is used to design and select the sample of villages for the Kolda Baseline Survey. The aim of the PNDL is to contribute to poverty reduction through the combined action of government departments, local authorities, and the private sector. The PNDL keeps an up-to-date repository of the villages in each department in Senegal. PNDL is an element of the Government of Senegal’s efforts to achieve its Millennium Development Goals (MDGs).
Page 28
Kolda Baseline Study
used to account for the fact that Yaajeende activities in Kolda include mobilization events to sensitize the population on many issues that reach populations beyond the direct intervention villages. Hence, there are expected spillover effects in neighboring villages and bordering communities. The 30 control villages were randomly selected using the PNDL repository and were cross-checked with 2013 Census Data and maps for quality assurance. Sample Size
The sample size for the Baseline Study was calculated to provide precise point-estimate of its key indicators. Following the FTF Guidance, the sample size was calculated to detect precise point- estimates of the key indicators of the Baseline Study.19 As stunting and wasting indicators tend to require the largest sample sizes, the sample size needs to ensure detecting precise estimates for these two key indicators. The sample size was also adjusted to account for the probability of encountering a household with at least one child under five, given the number of children per household. The following formula was used to determine the sample size: