Teacher, Child, and Caretaker Surveys from the Modern Daaras Impact Evaluation 2016, Baseline Survey
Senegal, 2016
Get MicrodataIdentification
SEN_2016_MDIE-BL_v01_M
Teacher, Child, and Caretaker Surveys from the Modern Daaras Impact Evaluation 2016
Baseline Survey
| Name | Country code |
|---|---|
| Senegal | SEN |
A sub-component of this wide initiative is the objective to reach children who do not typically access formal education and are enrolled in religious education in Koranic schools known as Daaras. With Muslims comprising around 95% of the Senegalese population, a vast majority of Senegalese males would have attended Daaras at one time or another, and it is estimated that between 800,000 and one million children and youth attend Daaras (D’Aoust, 2013; as cited in Goensch, 2016). In Senegal, a “Traditional Daara” is dedicated only to memorization of the Koran and advanced studies (Islamic law, etc.) and do not offer any additional instruction in science, math, French or other core courses under the official curriculum. “Modern Daaras”, on the other hand, train students not only in religious education like memorization of Koran but also in Math and French as per the official curriculum.
This subcomponent of the PAQEEB project aims to upgrade and improve Traditional Daaras to have language and math curricula like the Modern Daaras. This is an innovative intervention that provides pedagogical support through disbursement of “grants for results.” In return for this funding, project schools commit to perform the following activities: (1) to implement the specific “Modern Daaras” math and French curriculum; and (2) to ensure that students achieve learning results as reported through indicators measuring their levels of proficiency in reading and mathematics. The project stakeholders selected 100 Daaras in 20 counties based on the lowest gross enrollment ratios out of the 46 counties in Senegal (effectively, counties with gross enrollment ratios between 29 percent and 69 percent) to pilot the Daara modernization efforts (Bureau des Statistiques Scolaires et Universitaires, 2007).
This survey was used to evaluate this sub-component of the larger PAQEEB project that provides “grants for results” to selected Daaras. The survey consists of three distinct instruments that collected relevant data from teachers, caretakers of students, and students.
Sample survey data [ssd]
Individuals, households, and schools.
Version
Please refer to the PII removal document that has been provided in our submission packet for a description of the removed and edited variables.
Coverage
Rural and peri-urban areas only.
Primary school and Daara teachers, students, and caretakers of students.
Producers and sponsors
| Name |
|---|
| IMPAQ International, LLC |
| Name | Role |
|---|---|
| Jean Paul Petraud, Ph.D. | Questionnaire design, sampling methodology, data collection |
| Daniel Elliot, M.P.P. | Data collection, data processing & data analysis |
| Lucy Cutting, M.S. | Data processing & data analysis |
| Name |
|---|
| World Bank Strategic Impact Evaluation Fund |
Sampling
Identifying Eligible Treatment and Control Schools
The first step in the sampling implementation process was to identify a list of eligible treatment and control schools for the sample. To this effect, The World Bank and the Inspections des Daara committee provided IMPAQ with lists of Daaras that participated in the selection process for the Daara sub-component of the PAQEEB program within each of the 20 included districts. These lists included details on the ranking assigned to each candidate Daara and which Daaras were selected into the program (treatment Daaras) based on those rankings. Using this data on school rankings and characteristics, as well as information gathered during an initial visit to candidate Daaras, IMPAQ began the sample selection process by disqualifying schools from the sample that have previously been deemed ineligible for program allocation based on PAQEEB guidelines.
Selecting from Eligible Treatment and Comparison Schools
In order to decrease spillover effects between individuals in treatment school communities and those in comparison school communities, IMPAQ used GPS data to apply a set of minimum distance criteria to all eligible comparison schools and remove any that were too close to treatment schools. More specifically, comparison Daaras were removed from the sample if they were less than 2 kilometers from a treatment school. This decision was based on Theunynck (2009), who shows that distance to school is inversely related to the probability of being enrolled in school in Senegal. Additionally, Theunynck explains that evidence from multiple countries in Africa shows that enrollment and retention decline significantly when students must walk more that 1 to 2 kilometers to get to school. This trend is particularly strong among younger children. Thus, at a distance of two kilometers, we should see minimal interference between treatment and comparison Daaras.
Additionally, in order to be able to distinguish the communities around comparison schools, comparison Daaras were removed from the sample if they were less than ½ kilometer away from other comparison school. In these cases, one school out of the two was randomly chosen to remain as eligible for selection. The radius around comparison schools is smaller because there is no concern of spillover effects between these Daaras. Rather, this radius ensured the research team that they were not measuring the outcomes of two comparison Daaras within the same community. The concern that children from comparison communities may enroll in other nearby comparison Daaras is not considered a major source of bias in the ITT estimate, as the comparison Daaras are generally considered to be of similar quality, making it less likely for a child in a comparison community to commute to a Daara in a different comparison community.
Remaining eligible Daaras were selected for inclusion in the sample based on their ranking in the PAQEEB program selection process. Specifically, Daaras included in the PAQEEB program that were ranked closest to (just above) the program selection threshold were identified as treatment Daaras. Daaras not included in the PAQEEB program that were ranked closest to (just below) the program selection threshold were identified as comparison Daaras. In this way, IMPAQ ensured that treatment and comparison Daaras were as similar as possible concerning the key criteria used for program selection. In the event that multiple comparison schools received equivalent rankings, a random number generator was used to select among them for inclusion into the sample. If an appropriate comparison school could not be identified within a given IEF, all schools from that IEF were dropped from the sample. In most IEFs, IMPAQ selected 3 treatment Daaras and 3 comparison Daaras into the sample.
Selecting Eligible Secondary Comparison Schools
In addition to the comparison Daaras, IMPAQ included a second comparison group consisting of formal government schools. These schools were selected based on proximity to treatment Daaras, while still meeting the minimum distance criteria outlined above for comparison Daaras (i.e. 2-kilometer distance).
Household and Child Selection
IMPAQ performed a house-listing census of all households with children under the age 16 within a 1-kilometer perimeter (school catchment areas) of each Daara and formal school selected into the sample. For details on this house listing please see section 6.3.3 below. Once all households within the established perimeter of a selected school that had at least one child aged 7-10 were identified, IMPAQ randomly selected 15 households with at least one girl aged 7-10 and 15 households with at least one boy aged 7-10 for inclusion in the study. Only one child of each gender was selected from a given household in order to minimize the influence of larger households on the study outcome. Lastly, if a selected household had more than one child aged 7-10 of a single gender, IMPAQ randomly selected which of those children would be included in the sample, in order to prevent any bias in the selection of children within households.
Survey instrument
All instruments were originally developed in French, but have been translated to English as well.
Instruments
The baseline survey consisted of three unique instruments: A caretaker survey, a child survey and academic assessment, and a teacher survey.
Caretakers’ instrument (Enquête sur les personnes qui s’occupent des enfants)
The caretaker survey was designed to learn about the decisions and opinions within each household in the sample. A caretaker was defined as “the person who takes care of the child and makes decisions about what he/she eats and how he/she spends his/her time.” The survey instrument was divided into a schooling section and a household information section. Within the schooling section, caretakers were asked about schools and Daaras in their community, last year’s schooling choices, this year’s schooling choices, their opinions about education, and the child’s school participation/attendance. The household information section briefly captured some basic household characteristics, such as household size, number of children, education levels, and household assets.
Children’s instrument (Enquête sur les enfants)
The children’s survey begins with a few questions for the child’s caretaker, which are used to confirm the child’s name, age, and the school he or she attends. The rest of the survey is addressed to the child. First, the enumerator spent 3 to 5 minutes speaking with the child and setting him/her at ease. Next, the child answers questions about the school/Daara they attend. There are different sets of questions depending on whether he/she attends a formal school or a Daara. In general, both sets of questions tease out what the child does on a typical school day, including chores, playing, learning math or French, reading the Koran, reciting lessons, etc. Next, the child is asked questions regarding opinions about school. Questions cover whether or not he/she likes studying, what his/her favorite subject is, whether or not he/she likes the type of school currently attended, and what is important about education.
Children’s ASER Math and French instruments
After completing the survey, each child participated in an academic assessment based on the ASER instrument. ASER is a household survey that measures reading and arithmetic. ASER (meaning “impact” in Hindi and Urdu), began in India in 2005 as an evaluation tool for Pratham’s literacy enrichment programs in India. Since then, the process and tools have been adapted for use in other countries, including Tanzania, Kenya, Uganda, Mali, Senegal, and Mexico. The tests conducted in Kenya, Tanzania, and Uganda are known as the Uwezo surveys. The instruments function as follows:
On the reading portion, children would start with three simple sentences. They choose one to read out loud. If the child successfully reads the sentence, they move on to reading one of three complex sentences. Continued correct answers will lead to reading a short paragraph, answering comprehension questions about the paragraph, reading a short story, and finally, answering comprehension questions about the story. If these final comprehension questions are answered correctly, the child earns a “level J,” the highest score on the test. If they cannot read a simple sentence, they move on to read individual multisyllabic words. If the child cannot read these words, they move to monosyllabic words. If they cannot read the monosyllabic words, they are given complex sounds to read, then simple sounds, and finally letters. Letter recognition is “level A,” or the lowest level on the test.
The numeracy portion has a similar structure to the reading evaluation. It starts with two digit by one digit division, and moving up, they complete 2 digit by 2 digit multiplication, then multiplication with decimals, then three digit by two digit division, and finally a word problem. Moving down, they do 2 digit by 1 digit multiplication, double digit subtraction, double digit addition, single digit subtraction, and number recognition 10-99. Levels range from “A” to “J”.
Teachers’ instrument (Enquête Ecole)
The teacher survey included an introduction and four sections. First, basic information was collected about the person responding to the survey. Following this brief introduction, section 1 collected information about the school, including number of pupils, number of classrooms, subjects taught, etc. Section 2 covered school scheduling and activities. Section 3 elicited information about pedagogical methods within the school, including teachers’ classroom methods and activities. Finally, section 4 collected information about teachers’ qualifications and certifications.
Data collection
| Start | End |
|---|---|
| 2016-04 | 2016-06 |
| Name |
|---|
| Centre de Recherche Pour Le Développement Économique et Social |
Oversight for the data collection effort was applied on multiple levels through the evaluation staff, the CRDES headquarters management team, and a series of CRDES field managers. The evaluation team worked closely with CRDES throughout the preparatory activities leading up to baseline data collection, including the final week of training. Additionally, IMPAQ staff remained in country to provide technical oversight for the implementation of the survey pilot and first two weeks of data collection. During the baseline data collection, each team of six enumerators included a field manager to oversee daily operations of the team and answer any immediate enumerator questions while in the field. Questions were escalated to the CRDES management team and the evaluation staff, as necessary.
Data integrity was a primary focus of the evaluation team. In advance of piloting, the programmed CAPI instruments were reviewed in thorough detail for consistency in universes, skip patterns, and other logical arrangements. Survey piloting data was analyzed to confirm whether the CAPI instruments were working as expected in the field. Once in the field, data was uploaded to CRDES’s servers on a nightly basis using mobile hotspots. Data was then analyzed daily by the primary data contact at CRDES and a weekly basis by the IMPAQ team to ensure data integrity and oversee progress for the data collection effort.
Enumerator Training and Survey Piloting
Enumerator Training and survey piloting were conducted over the course of a week in late March of 2016. The CRDES data collection team, the evaluation PI, and 30 potential enumerators participated. Two full days led by the CRDES management team and the evaluation PI were devoted to discussion of the study and a detailed question by question review of the baseline survey instruments. A third day was dedicated entirely to familiarizing potential enumerators with the instruments on the CAPI device used for baseline data collection. At the end of the first three days, 24 enumerators were selected to be part of the data collection team based on their participation and engagement during the training process. The remaining days were devoted to survey piloting, within which enumerators were sent to a rural area outside of Dakar closely resembling the study area. Baseline survey instruments were used to collect pilot data, which was then analyzed for any remaining issues to be addressed before launch of baseline data collection.
Baseline Data Collection
Baseline data collection launched in the beginning of April and spanned two months. Four teams of six enumerators were used to collect baseline data from caretakers and children selected into the evaluation. Baseline surveys were administered in Wolof wherever possible, except in cases where the respondents spoke only French, as it is the predominant spoken language in the study area. Within each household, enumerators first administered the caretaker survey, then asked to speak with the child in order to conduct the children’s survey. Both surveys were conducted in an appropriately semi-private setting wherever possible, in order to minimize response bias due to the presence of surrounding family members. After completion of the Children’s survey, the ASER instruments were administered to the child to determine their level of proficiency in math and French. Children were given a small unannounced treat at the end of the ASER instruments in order to leave them with a positive experience of the interview. After completion of the baseline caretaker, children’s, and ASER surveys in all selected school neighborhoods, the Teachers survey was quickly conducted in all selected schools at the beginning of June.
Technical and Administrative Oversight
Oversight for the data collection effort was applied on multiple levels through the evaluation staff, the CRDES headquarters management team, and a series of CRDES field managers. The evaluation team worked closely with CRDES throughout the preparatory activities leading up to baseline data collection, including the final week of training. Additionally, IMPAQ staff remained in country to provide technical oversight for the implementation of the survey pilot and first two weeks of data collection. During the baseline data collection, each team of six enumerators included a field manager to oversee daily operations of the team and answer any immediate enumerator questions while in the field. Questions were escalated to the CRDES management team and the evaluation staff, as necessary.
Data integrity was a primary focus of the evaluation team. In advance of piloting, the programmed CAPI instruments were reviewed in thorough detail for consistency in universes, skip patterns, and other logical arrangements. Survey piloting data was analyzed to confirm whether the CAPI instruments were working as expected in the field. Once in the field, data was uploaded to CRDES’s servers on a nightly basis using mobile hotspots. Data was then analyzed daily by the primary data contact at CRDES and a weekly basis by the IMPAQ team to ensure data integrity and oversee progress for the data collection effort.
Data Access
Public use files, accessible to all after registration.
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
Example:
IMPAQ International, LLC. Teacher, Child, and Caretaker Surveys from the Modern Daaras Impact Evaluation 2016- Baseline Survey. Ref. SEN_2016_MDIE-BL_v01_M. 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.
Contacts
| Name | URL |
|---|---|
| IMPAQ International | https://www.impaqint.com/forms/contact-us |
Metadata production
DDI_SEN_2016_MDIE-BL_v01_M_WB
| Name | Affiliation | Role |
|---|---|---|
| Development Economics Data Group | The World Bank | Documentation of the DDI |
2018-03-26
Metadata version
Version 01 (March 2018)