The Adolescent Girls Initiative (AGI) in South Sudan was initiated in 2010 to support economic and social empowerment of young women (aged 15-24 years) in the country. Inspired by the success of similar interventions in Bangladesh, Uganda and Tanzania, where the program is known as Empowerment and Livelihoods of Adolescents (ELA), BRAC piloted AGI in four states of South Sudan. Before the commencement of the program, a baseline survey was carried out by BRAC Research and Evaluation Unit with the twin objectives of assessing the pre-program situation and impact evaluation. It is evident that households targeted by the program face severe poverty, food insecurity and poor quality of lives. Young women have additional vulnerabilities in their lives. Over 30 percent of adolescent girls are mothers and about 70 percent of them reported having unprotected sex. With limited education and livelihood skills, their engagements in earning activities are also very minimal. Less than 20 percent of the girls have ever received any skills training and only one-third are engaged in any form of earning activity. Despite these difficulties in their livelihoods, about a quarter of the girls have savings.There is significant scope of improvement in changing their knowledge and attitudes. Girls have also expressed their willingness to participate in the program. More importantly, those girls who are more likely to derive benefits from participating in such a program also have greater intention to participate. Therefore, the program has the potential of meeting their objectives, although excessively high expectations of the participant girls can discourage community ownership of the interventions. The baseline data will be used to compare relevant indicators to data from the endline data collection. Furthermore the data are an opportunity to assess the pre-program situation of households and adolescent girls.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Households and Individuals
This survey was designed to capture baseline indicators on the following:
-Household income and income generating activities
-Household assets, loans and expenditure
-General household conditions including housing, water and sanitation
-Education, financial literacy and income generating activities of adolescent girls
-Sexual and risky behavior/attitudes of adolescent girls
Research subjects were recruited from Juba, Bor, Yei and Torit. More details under “Sampling”.
This data set is the baseline survey dataset from a study of 4,075 households and adolescents in South Sudan who were targeted by the NGO BRAC in 2010-2013 for an adolescent girl’s initiative program with the aim of helping adolescent girls and young women make a successful transition from school to work. There were four program sites, in Juba, Bor, Yei and Torit.
Producers and sponsors
The randomization method used was over-selection of clusters (or villages). At the initial stage BRAC program management identified 10 branch offices from four states of South Sudan (Central Equatoria, Eastern Equatoria, Jongolei and Lakes) as intervention sites. The branches are Munuki, Kator, Hai Gwafa, Kanjoro, Inkas, Hai Police, Dukurut, Langbhar, Makuriric and Matangai. In each branch BRAC staff identified 20 clusters which could potentially have the intervention. A social mapping of the households living in each cluster was conducted.
Following the mapping a census of girls aged 15 to 24 was conducted in all 200 villages. The census contained basic identification and background information which was also used as a sampling framework for the baseline survey. Based on the information from the mapping and the census, 160 villages (16 villages per branch) were selected for project implementation.
In each branch 10 villages were randomly selected for the intervention, with the remaining six villages acting as the control group. Following the random assignation the baseline survey was conducted. In order to balance the sample between treatment and control groups, 6 intervention villages (from the 10 in each branch) were randomly selected for the survey. In total the survey was conducted in 120 villages, of which 60 belonged to the treatment and control groups.
From each cluster a random sample of 35 adolescent girls was drawn for the baseline survey from the census. One girl was interviewed in each household. The initial target of 4200 girls and their parents was not met, with the final sample size in the baseline survey of 4075. Interviews were conducted with the girls/young women and a separate instrument use to collect information from their parents, however in cases where the adolescent was found to be the household head, both instruments were administered to her.
There were resources available to implement the program in 100 villages. Therefore, in each branch 10 villages were randomly selected for the interventions, and the other 6 villages belong to the control group. Following this random assignation, the baseline survey was conducted. In order to balance the sample between treatment and control groups, 6 intervention villages (from 10 villages in each branch) were randomly selected to conduct the survey. In total, we conducted the survey in 120 villages, of which 60 villages belong to treatment and control groups each.
Dates of Data Collection
Data Collection Mode
Computer Assisted Personal Interview [capi]
Data Collection Notes
Much effort was put into the wording in order to adapt the survey to local and cultural context. A team of researchers from BRAC Research and Evaluation Unit and the World Bank worked together to finalize the questionnaire instrument for the two separate modules, which were then piloted prior to the survey.
Women with at least a secondary school leaving certificate and some previous field survey experience were hired as enumerators, and women or men with diploma and degree education level were employed as group supervisors. All field staff received training on data collection for 8 days. This consisted of instructive lectures and demonstrations followed by practice sessions at different households outside the sampled clusters, and finally special briefing about different field techniques. While the questionnaires were in English, this training equipped the enumerators with a good understanding of the questionnaires, which they could translate into Juba Arabic (and other local languages) in the field.
Groups of at least 10 enumerators, each led by a field supervisor, were deployed in the field for the survey outside sampled clusters to develop a rapport with the questionnaires. A team of researchers from BRAC closely monitored all the field activities through frequent visits for spot checking. It would also revisit random sub-sample of the respondents, called back the respondents for verification of information, and checked completed questionnaires for consistency. During their visits, assistance and guidance were provided to enumerators and their supervisors where needed. In addition, video and phone conferences were held regarding various field operations with the field team, the World Bank and BRAC Research and Evaluation Unit office to reach on working practical decisions.
You can find the detailed timeline in the study metadata file.
A - Household Module
0 - Identification and Consent
S1 - Household Members Characteristics
S2 - Household Members Education
S3 - Income Generating Activities of All HH members
S4 - Expectations for young (aged 5-25) HH members
S5 - Assets
S6 - Housing Conditions
S7 - Water and Sanitation
S8 - Loans Outstanding
S10 - Expenditure
Section 10 - Household Tracking Form
B - Adolescent Module
0 - Identification and Consent
S1 - Education
S2 - Income Generating Activities
S3 - Spare Time
S4 - Financial Literacy
S5. Loans and Savings
S7 - Expectations and Empowerment
S8 - Networks
S9 - Program Participation
S10 - Childhood
S11 - Risky Behaviours
S12 - Sexual Behaviours/AIDS awareness
Each completed questionnaire was scrutinized in the field and at the field office on the day of the interview by field supervisors. Further scrutiny took place during data editing. Consistency checks were done to yield cleaned datasets. A team of researchers worked on data analysis using STATA software.
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download.
Shubha Chakravarty and Niklas Buehren, World Bank. South Sudan - Empowering young women in South Sudan: The BRAC-World Bank Adolescent Girls Initiative (Baseline), Ref. SSD_2010_AGIIE-BL_v01_M_v01_A_PUF. Dataset downloaded from [url] on [date].
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.