This dataset contains records for each public elementary and secondary education agency in the 50 states, District of Columbia, and United States territories (American Samoa, Guam, Puerto Rico, the Virgin Islands, and the Marshall Islands), as reported to the National Center for Education Statistics by the state education agencies. Each record provides state and federal identification numbers, agency's name, address, and telephone number, county name and FIPS code, agency type code, student counts, graduates and other completers counts, and other codes for selected characteristics of the agency. Information on grade span and on the number of schools, classroom teachers, and staff is also included in most cases.

This dataset contains records for each public elementary and secondary school in the 50 states, the District of Columbia, and outlying areas (American Samoa, Guam, Puerto Rico, the Virgin Islands, and the Marshall Islands) in fall 1973. The data provide information on name, address, county, and district of the school, programs offered, and numbers of pupils and teachers by organizational level of government control. School-by-school data were obtained through various procedures chosen by the state education agencies from options established by the National Center for Education Statistics.

Abstract

The Service Delivery Indicators (SDI) are a set of health and education indicators that examine the effort and ability of staff and the availability of key inputs and resources that contribute to a functioning school or health facility. The indicators are standardized allowing comparison between and within countries over time.

The Education SDIs include teacher effort, teacher knowledge and ability, and the availability of key inputs (for example, textbooks, basic teaching equipment, and infrastructure such as blackboards and toilets). The indicators provide a snapshot of the learning environment and the key resources necessary for students to learn.

Kenya's Service Delivery Indicators Education Survey was implemented in May-July 2012 by the Economic Policy Research Center and Kimetrica, in close coordination with the World Bank SDI team. The data were collected from a stratified random sample of 239 public and 67 private schools to provide a representative snapshot of the learning environment in both public and private schools. The survey assessed the knowledge of 1,679 primary school teachers, surveyed 2,960 teachers for an absenteeism study, and observed 306 grade 4 lessons. In addition, learning outcomes were measured for almost 3,000 grade 4 students.

Geographic coverage

National

Analysis unit

Schools, teachers, students.

Universe

All primary schools

Kind of data

Sample survey data [ssd]

Sampling procedure

The sampling strategy for SDI surveys is designed towards attaining indicators that are accurate and representative at the national level, as this allows for proper cross-country (i.e. international benchmarking) and across time comparisons, when applicable. In addition, other levels of representativeness are sought to allow for further disaggregation (rural/urban areas, public/private facilities, subregions, etc.) during the analysis stage.

The sampling strategy for SDI surveys follows a multistage sampling approach. The main units of analysis are facilities (schools and health centers) and providers (health and education workers: teachers, doctors, nurses, facility managers, etc.). In the case of education, SDI surveys also aim to produce accurate information on grade four pupils’ performance through a student assessment. The multistage sampling approach makes sampling procedures more practical by dividing the selection of large populations of sampling units in a step-by-step fashion. After defining the sampling frame and categorizing it by stratum, a first stage selection of sampling units is carried out independently within each stratum. Often, the primary sampling units (PSU) for this stage are cluster locations (e.g. districts, communities, counties, neighborhoods, etc.) which are randomly drawn within each stratum with a probability proportional to the size (PPS) of the cluster (measured by the location’s number of facilities, providers or pupils). Once locations are selected, a second stage takes place by randomly selecting facilities within location (either with equal probability or with PPS) as secondary sampling units. At a third stage, a fixed number of health and education workers and pupils are randomly selected within facilities to provide information for the different questionnaire modules.

Detailed information about the specific sampling process conducted for the 2012 Kenya Education SDI is available in the SDI Country Report (“SDI-Report-Kenya”) included as part of the documentation that accompanies these datasets.

Mode of data collection

Face-to-face [f2f]

Research instrument

The SDI Education Survey Questionnaire consists of six modules:

Module 1: School Information - Administered to the head of the school to collect information on school type, facilities, school governance, pupil numbers, and school hours. It includes direct observations of school infrastructure by enumerators.

Module 2a: Teacher Absence and Information - Administered to the headteacher and individual teachers to obtain a list of all school teachers, to measure teacher absence, and to collect information on teacher characteristics.

Module 2b: Teacher Absence and Information - Unannounced visit to the school to assess the absence rate.

Module 3: School Finances - Administered to the headteacher to collect information on school finances (this data is unharmonized)

Module 4: Classroom Observation - An observation module to assess teaching activities and classroom conditions.

Module 5: Pupil Assessment - A test of pupils to have a measure of pupil learning outcomes in mathematics and language in grade four. The test is carried out orally and one-on-one with each student by the enumerator.

Module 6: Teacher Assessment - A test of teachers covering mathematics and language subject knowledge and teaching skills.

Cleaning operations

Data entry was done using CSPro; quality control was performed in Stata.

Sampling error estimates

At the national level, an anticipated standard error of 1.6 percentage points for absenteeism, and 4.4 percentage points for pupil literacy were calculated. At the county level, an anticipated standard error of 3.1 percent for absenteeism and 9.0 percent for literacy were estimated.

Abstract

The Impact evaluation tests the effectiveness of low-cost interventions aimed at increasing teaching quality. The primary objectives of the impact evaluation are to generate evidence on the following questions:

I. Does providing short, frequent teaching tips to teachers through SMS result in an improvement in teacher effort, resulting in better student performance?

ii. Does providing a detailed classroom feedback report to teachers lead to improvements in teacher performance through increasing their understanding of their students’ performance?

iii. Are there any positive complementarities between i. and ii.?

Analysis unit

The Uganda Private School Impact Evaluation describes: Individuals (Students, Teachers and Head Teachers)

Universe

All public secondary schools with more than 80 students per class in Uganda. From this list the following schools were dropped:

(i) schools already adopting Double Shift;

(ii) boarding schools;

(iii) hard to reach schools; and

(iv) schools in Karamoja where security issues were present.

Kind of data

Sample survey data [ssd]

Sampling procedure

Impact evaluation sample was constructed using Education Management Information System; headcount data for schools from 2010.

In collaboration with the MoES, these data were used to identify all public secondary schools with more than 80 students per class in Uganda.

There were 198 over-enrolled schools identified and selected to be part of the evaluation sample.

• 225 head teacher surveys

• 636 teachers tested

• 29,248 students tested

Mode of data collection

Computer Assisted Personal Interview [capi]

Research instrument

Questionnaires for this Impact Evaluation were structured to collect information about: Teachers, Head Teachers and Students.

Response rate

The team visited 231 schools with the following outcomes:

a.) Detailed information was obtained from 225 schools.

b.) 3 schools were no longer in operation

c.) 3 schools declined to participate in the survey.

Abstract

The 1993 Indonesia Family Life Survey (IFLS) provides data at the individual and family level on fertility, health, education, migration, and employment. Extensive community and facility data accompany the household data. The survey was a collaborative effort of Lembaga Demografi of the University of Indonesia and RAND, with support from the National Institute of Child Health and Human Development, USAID, Ford Foundation, and the World Health Organization. In Indonesia, the 1993 IFLS is also referred to as SAKERTI 93 (Survai Aspek Kehidupan Rumah Tangga Indonesia). The IFLS covers a sample of 7,224 households spread across 13 provinces on the islands of Java, Sumatra, Bali, West Nusa Tenggara, Kalimantan, and Sulawesi. Together these provinces encompass approximately 83 percent of the Indonesian population and much of its heterogeneity. The survey brings an interdisciplinary perspective to four broad topic areas:

• Fertility, family planning, and contraception • Infant and child health and survival • Education, migration and employment • The social, economic, and health status of adults, young and old

Additionally, extensive community and facility data accompany the household data. Village leaders and heads of the village women's group provided information in each of the 321 enumeration areas from which households were drawn, and data were collected from 6,385 schools and health facilities serving community residents.

Geographic coverage

National

Analysis unit

Households

Universe

Household Survey data were collected for household members through direct interviews (for adults) and proxy interviews (for children, infants and temporarily absent household members). The IFLS-1 conducted detailed interviews with the following household members:

• The household head and their spouse
• Two randomly selected children of the head and spouse aged 0 to 14 (interviewed by proxy)
• An individual age 50 and above and their spouse, randomly selected from remaining members
• For a randomly selected 25 percent of the households, an individual age 15 to 49 and their spouse, randomly selected from remaining members.

The Community and Facility Survey collected data from a variety of respondents including: the village leader and his staff and the leader of the village women's group; Ministry of Health clinics and subclinics; private practices of doctors, midwives, nurses, and paramedics; community-based health posts and contraceptive distribution centers; public, private, and religious elementary schools; public, private, and religious junior high schools; public, private, and religious senior high schools. Unlike many other surveys, the sample frame for the survey of facilities was drawn from the list of facilities used by household survey respondents in the area.

Kind of data

Sample survey data [ssd]

Sampling procedure

1. HOUSEHOLD SELECTION

(a) SAMPLING

The IFLS sampling scheme stratified on provinces, then randomly sampled within provinces. Provinces were selected to maximize representation of the population, capture the cultural and socioeconomic diversity of Indonesia, and be cost effective given the size and terrain of the country. The far eastern provinces of East Nusa Tenggara, East Timor, Maluku and Irian Jaya were readily excluded due to the high costs of preparing for and conducting fieldwork in these more remote provinces. Aceh, Sumatra's most northern province, was deleted out of concern for the area's political violence and the potential risk to interviewers. Finally, due to their relatively higher survey costs, we omitted three provinces on each of the major islands of Sumatra (Riau, Jambi, and Bengkulu), Kalimantan (West, Central, East), and Sulawesi (North, Central, Southeast). The resulting sample consists of 13 of Indonesia's 27 provinces: four on Sumatra (North Sumatra, West Sumatra, South Sumatra, and Lampung), all five of the Javanese provinces (DKI Jakarta, West Java, Central Java, DI Yogyakarta, and East Java), and four provinces covering the remaining major island groups (Bali, West Nusa Tenggara, South Kalimantan, and South Sulawesi). The resulting sample represents 83 percent of the Indonesian population. (see Figure 1.1 of the Overview and Field Report in External Documents). Table 2.1 of the same document shows the distribution of Indonesia's population across the 27 provinces, highlighting the 13 provinces included in the IFLS sample.

The IFLS randomly selected enumeration areas (EAs) within each of the 13 provinces. The EAs were chosen from a nationally representative sample frame used in the 1993 SUSENAS, a socioeconomic survey of about 60,000 households. The SUSENAS frame, designed by the Indonesian Central Bureau of Statistics (BPS), is based on the 1990 census. The IFLS was based on the SUSENAS sample because the BPS had recently listed and mapped each of the SUSENAS EAs (saving us time and money) and because supplementary EA-level information from the resulting 1993 SUSENAS sample could be matched to the IFLS-1 sample areas. Table 2.1 summarizes the distribution of the approximately 9,000 SUSENAS EAs included in the 13 provinces covered by the IFLS. The SUSENAS EAs each contain some 200 to 300 hundred households, although only a smaller area of about 60 to 70 households was listed by the BPS for purposes of the annual survey. Using the SUSENAS frame, the IFLS randomly selected 321 enumeration areas in the 13 provinces, over-sampling urban EAs and EAs in smaller provinces to facilitate urban rural and Javanese-non-Javanese comparisons. A straight proportional sample would likely be dominated by Javanese, who comprise more than 50 percent of the population. A total of 7,730 households were sampled to obtain a final sample size goal of 7,000 completed households. Table 2.1 shows the sampling rates that applied to each province and the resulting distribution of EAs in total, and separately by urban and rural status. Within a selected EA, households were randomly selected by field teams based upon the 1993 SUSENAS listings obtained from regional offices of the BPS. A household was defined as a group of people whose members reside in the same dwelling and share food from the same cooking pot (the standard BPS definition). Twenty households were selected from each urban EA, while thirty households were selected from each rural EA. This strategy minimizes expensive travel between rural EAs and reduces intra-cluster correlation across urban households, which tend to be more similar to one another than do rural households. Table 2.2 (Overview and Field Report) shows the resulting sample of IFLS households by province, separately by completion status.

(b) SELECTION OF RESPONDENTS WITHIN HOUSEHOLDS For each household selected, a representative member provided household-level demographic and economic information. In addition, several household members were randomly selected and asked to provide detailed individual information.

1. THE COMMUNITY SURVEY SAMPLING PROCEDURE

(a) SAMPLING

The goal of the CFS was to collect information about the communities of respondents to the household questionnaire. The information was solicited in two ways. First, the village leader of each community was interviewed about a variety of aspects of village life (the content of this questionnaire is described in the next section). Information from the village leader was supplemented by interviewing the head of the village women's group, who was asked questions regarding the availability of health facilities and schools in the area, as well as more general questions about family health in the community. In addition to the information on community characteristics provided by the two representatives of the village leadership, we visited a sample of schools and health facilities, in which we conducted detailed interviews regarding the institution's activities. A priori we wanted data on the major sources of outpatient health care, public and private, and on elementary, junior secondary, and senior secondary schools. We defined eight strata of facilities/institutions from which we wanted data. Different types of health providers make up five of the strata, while schools account for the other three. The five strata of health care providers are: government health centers and subcenters (puskesmas, puskesmas pembantu); private doctors and clinics (praktek umum/klinik); the private practices of midwives, nurses, and paramedics (perawats, bidans, paramedis, mantri); traditional practitioners (dukun, sinshe, tabib, orang pintar); and community health posts (posyandu, PPKBD).The three strata of schools are elementary, junior secondary, and senior secondary. Private, public, religious, vocational, and general schools are all eligible as long as they provide schooling at one of the three levels. Our protocol for selecting specific schools and health facilities for detailed interview reflects our desire that selected facilities represent the facilities available to members of the communities from which household survey respondents were drawn. For that reason, we were hesitant to select facilities based solely either on information from the village leader or on proximity to the village center. The option we selected instead was to sample schools and health care providers from lists provided by respondents to the household survey. For each enumeration area lists of facilities in each of the eight strata were constructed by compiling information provided by the household regarding the names and locations of facilities the household respondent either knew about or used. To generate lists of relevant health and family planning facilities, the CFS drew on two pieces of information from the household survey. The IFLS

Abstract

The Government of Kenya is interested in understanding how malaria prevention and treatment can improve the education of school children when it is combined with effective teaching. This project examines the impact of school-based malaria intermittent screening and treatment and enhanced literacy training and support for teachers on children's health and educational outcomes.

A cluster randomized trial was implemented with 5,233 children in 101 government primary schools on the south coast of Kenya in 2010-2012. The intervention was delivered to children randomly selected from classes 1 and 5 who were followed up for 24 months. Once a school term, children were screened by public health workers using malaria rapid diagnostic tests (RDTs), and children (with or without malaria symptoms) found to be RDT-positive were treated with a six dose regimen of artemether-lumefantrine (AL). Given the nature of the intervention, the trial was not blinded. The primary outcomes were anemia and sustained attention. Secondary outcomes were malaria parasitaemia and educational achievement. Data were analyzed on an intention to treat basis. The study is registered with ClinicalTrials.gov, NCT00878007.

The schools were randomly assigned to one of four experimental groups: some schools have been tested and treated for malaria; some schools have had extra support for teachers of English and Swahili; some schools have been both tested for malaria and received extra teacher support; and other schools have gotten neither of the two programs.

Following recruitment, baseline health and education surveys were undertaken in January-February 2010, which were followed by the first round of intermittent screening and treatment (IST) and the teacher training workshop. Classroom observations occurred in May 2010, followed by the second round of IST in June-July 2010. The third round of IST occurred in September 2010. The first follow-up education surveys were carried out in November 2010 and the first health surveys - in February and March 2011, followed by a round of IST as well as refresher teacher training for the literacy intervention. The final round of IST was conducted in September 2011 with the 24 months follow-up health and education survey in February-March 2012.

The data from the baseline and follow-up surveys is documented here.

Geographic coverage

Kwale and Msambweni districts

Analysis unit

Individuals, households, schools

Universe

The survey cover children in grades 1 and 5 in government schools in Kwale and Msambweni districts, their parents/guardians, head teaches of schools in above mentioned districts.

Kind of data

Sample survey data [ssd]

Sampling procedure

School selection was made from the 197 government primary schools in Kwale and Msambweni districts. In Kwale district, a separate study is evaluating the impact of an alternative literacy intervention in two of the four zones; therefore only 20 schools in this district were included in our study allowing the two interventions to proceed without leakage. In Msambweni district, 81 of 112 schools were selected; schools 70 km or further away from the project office, were excluded due to logistical constraints.

The randomization of the 101 schools into the four experimental groups was conducted in two stages, each involving public randomization ceremonies:

Stage 1 - Literacy intervention randomization a) Clusters of schools (groups of between 3-6 schools that meet and share information) were randomised either to receive the literacy intervention or to serve as a control schools. b) This randomization was stratified by (i) cluster size, to ensure equal numbers of schools in the experimental groups; and (ii) average primary school leaving exam scores across the cluster, to balance the two groups for school achievement. c) District officials and representatives from all 26 school clusters were invited to a meeting. Volunteers were asked to randomly draw envelopes each containing a cluster name from 10 pre-stratified ballot boxes and to sequentially place the envelopes in group A and group B.

Stage 2 - Health intervention randomization a) The health intervention was randomly allocated amongst the 51 schools assigned to the literacy intervention and the 50 schools allocated to serve as control schools during the first randomization. b) Schools were stratified by average primary school leaving exam scores into 5 quintiles and by literacy intervention group, producing 10 strata overall. c) Representatives from the 101 schools and local communities were invited to this randomization ceremony. Volunteers were asked to draw envelopes from the 10 pre-stratified ballot boxes and sequentially place the envelopes in group 1 and group 2.

During January and February 2010, schools were visited and a census of all children in classes 1 and 5 was conducted, including children absent on the day of visit. This census served as a basis for making a random selection of 25 children with consent from class 1 and 30 children with consent from class 5. Fewer children were selected from class 1 because of the extra educational assessments undertaken with these children and the practical feasibility of conducting the tests in a single day. Some of the classes were small, and this meant that in these classes all children with consent were recruited.

Of the 5,233 children enrolled initially, 4,446 (85.0%) were included in the 12 month follow-up health survey and 4201 (80.3%) were included in the 24 month health survey. Overall, 4,656 (89.0%) of children were included in the 9 month follow-up education survey and 4,106 (78.5%) in the 24 month follow-up survey.

Mode of data collection

Face-to-face [f2f]

Research instrument

The following questionnaires and forms are available:

1) School Questionnaire The school questionnaire is administered to the head teachers of each school during the initial school selection; if absent, the deputy head was interviewed. Information is collected on the characteristics of the school such as the number of boys and girls enrolled in each class, examination results in English, mathematics and Kiswahili, school features such as number of desks and teachers, facilities available such as latrines and the presence of school health activities and materials. Locations of each school were mapped using a handheld Global Positioning System (GPS) receiver, (eTrex Garmin Ltd., Olathe, KS).

2) Parent questionnaire for class 1 students The parent questionnaire for class 1 students assesses the educational and socio-economic environment of the children's households. This is administered to the parent or guardian at the time of consent. Questions relate to their own reading ability, schooling, and involvement in their children's school, as well as questions on family composition, household construction, asset ownership and mosquito net ownership and use.

3) Parent questionnaire for class 5 students The parent questionnaire for class 5 students assesses the educational and socio-economic environment of the children's households. This is administered to the parent or guardian at the time of consent. The section on education environment is reduced as the literacy intervention was focused on the class 1 children, so a less extensive knowledge of attitudes to education was required for parents of class 5 children. Questions relate to their schooling level achieved, as well as questions on family composition, household construction, asset ownership, and mosquito net ownership and use.

4) Nurse survey form for classes 1 and 5 The child ID, child name, and parent name of the randomly selected children are already entered on the form before arrival at the school. The nurse records the attendance of each child, completing the reasons using the codes at the bottom of the form. Height, weight and temperature of each child is recorded on the form. The child is also asked their age, which is recorded.

5) Health Technician survey form for classes 1 and 5 The child ID, child name, and parent name of the randomly selected children are already entered on the form before arrival at the school. The technician notes whether the child is present, and then records the hemoglobin reading, whether or not a blood slide has been taken, and the timing and result of the malaria rapid diagnostic test (RDT). This form is for assessment of children in the intervention schools where P falciparum infection is assessed.

Abstract

The GYTS is a school-based survey which uses a self-administered questionnaire to monitor tobacco use among youth and to guide the implementation and evaluation of tobacco prevention and control programmes.

Geographic coverage

National

Analysis unit

Individuals

Universe

School-going adolescents aged 13-15 years.

Kind of data

Sample survey data [ssd]

Sampling procedure

GYTS uses a global standardized methodology that includes a two-stage sample design with schools selected with a probability proportional to enrollment size. The classes within selected schools are chosen randomly and all students in selected classes are eligible to participate in the survey.

A total of 4,216 eligible students in grade 8 to 9 from elementary schools and grade 1 from secondary schools completed the survey, of which 3,896 were aged 13-15 years.

Mode of data collection

self-administered

Cleaning operations

All data processing (scanning, cleaning, editing, and weighting) was conducted at the US Centers for Disease Control.

Response rate

The overall response rate was 92.6%.

Abstract

The GYTS is a school-based survey which uses a self-administered questionnaire to monitor tobacco use among youth and to guide the implementation and evaluation of tobacco prevention and control programmes.

Geographic coverage

National coverage

Analysis unit

Individuals

Universe

School-going adolescents aged 13-15 years.

Kind of data

Sample survey data [ssd]

Sampling procedure

GYTS uses a global standardized methodology that includes a two-stage sample design with schools selected with a probability proportional to enrollment size. The classes within selected schools are chosen randomly and all students in selected classes are eligible to participate in the survey.

Eligible students in general and technical secondary schools (levels 1-4) completed the survey, of which 1,873 were aged 13-15 years.

Mode of data collection

self-administered

Cleaning operations

All data processing (scanning, cleaning, editing, and weighting) was conducted at the US Centers for Disease Control.

Response rate

The overall response rate was 88.3%.

Abstract

The Nigeria Teacher Development Programme (TDP) In-Service Training Component Impact Evaluation 2014 is a DFID-funded programme, managed by a consortium led by Mott Macdonald. Its aim is to increase the effectiveness of teachers and thus raise primary and junior secondary school (JSS) pupil learning levels, through support to the Federal and State institutions responsible for pre- and in-service training and development of basic education teachers. The programme has the following three components:

• In-service teacher training in the three core curriculum subjects of English, maths, and science & technology (S&T) for primary and JSS teachers;
• Pre-service teacher training; and
• Results & evidence-based research.

Schools eligible for TDP intervention are state (public) schools, excluding Integrated Quranic, Tsangaya Education (IQTE) schools, and special schools for children with disabilities.

The programme is being implemented in two phases (phase 1 covering Jigawa, Katsina and Zamfara (2013-2019) and phase 2 covering Kaduna and Kano (2016-2019)). For more information on the programme, please visit: www.tdpnigeria.org

The quantitative survey described in this documentation forms part of a DFID-funded phase 1 Impact Evaluation (IE) of the TDP in-service training component, which is the largest component of the programme (approximately 80% of total programme resources). The IE is being conducted by the Education Data, Research and Evaluation in Nigeria programme (EDOREN) which is a DFID-funded education research programme managed by Oxford Policy Management Ltd (OPM), and will focus on TDP activities in primary schools (since JSS activities do not start until the school year 2016/17).

The IE uses a theory-based, mixed-methods approach, and the results will be used to inform the implementation of the TDP for the remainder of phase 1 and the design and implementation in phase 2, as well as to assess whether the in-service training has contributed to improving teacher effectiveness and what could work otherwise and/or elsewhere in improving teacher effectiveness.

The three key objectives of the phase 1 IE quantitative baseline survey were to: - Establish baseline levels of teacher effectiveness and pupil learning before the start of TDP's in-service training, and to check that the IE's randomisation design yielded a valid counterfactual. - Inform and allow for potential adjustments to TDP design and implementation as deemed appropriate by the programme and DFID-Nigeria; and - Guide and interact with the qualitative baseline research by providing information on the situation prior to the start of the TDP as well as to allow for more in-depth examination of unanticipated quantitative findings.

Geographic coverage

The survey was carried out in the TDP phase 1 states (Jigawa, Katsina and Zamfara) but the results are NOT representative at the state-level, i.e. state-level estimates do not represent the average situation in a given state.

The data are not representative at any geographical level.

Analysis unit

The Primary Sampling Units (PSUs) of the survey are TDP-eligible state primary schools, at which level some analysis is performed (for example, characteristics of schools and head teachers).

However, the main units of analysis are:

· Teachers (selected prior to the PSUs and teaching grades 1-3 in any of the three subjects: English, maths, or science) · Pupils (in grade 3 at baseline, and taught English, math or science by at least one of the 'selected' teachers). · Lessons taught by the selected teachers (not sampled) · TDNAs administered to the selected teachers (not sampled)

Please refer to the 'Sampling Procedure' section for more details.

Universe

The target populations (the groups for which one would like to generalise the study findings) are the schools eligible for the TDP in treatment and control groups in the three states, and the eligible teachers and pupils within these schools.

Please refer to the 'Sampling Procedure' section for more details on the definition of eligibility.

Kind of data

Sample survey data [ssd]

Sampling procedure

(1) Aim of sampling design

The aim of the sampling design was to define a valid counterfactual 'control' group from which comparisons could be made with a 'treatment' group that participate in the TDP. The control group would not participate in the TDP in-service training but would have background characteristics which are, on average, similar to the treatment group that do participate in TDP in-service training.

The sampling design of the IE was based on a quasi-experimental 'constrained randomisation' approach. 'Constrained randomisation' means that certain parameters of the IE were already fixed - for example, the Local Government Areas (LGAs) where the programme operates. In addition, pre-determined groups of schools fulfilling certain criteria (described below) would constitute the sampling frame - this is in contrast to a fully randomised design approach where one might expect the random drawing of groups (or clusters) of schools from a list of all state primary schools in the region under study.

Randomisation was conducted only in allocating groups of schools to 'treatment' or 'control' status.

The sample design was determined to a large extent by practical programme considerations, and also by the available budget.

(2) Construction of sampling frame: Eligible primary schools

The sampling frame was constructed from scratch through the stages described below. The intended size of the frame was 1008 primary schools eligible for the TDP (504 'treatment' schools and 504 'control' schools) and would constitute the target population (or universe) of eligible schools, from which a sample of treatment and control schools would be drawn for the survey.

Stage 1: Selection of LGAs

In each state, 14 LGAs where the programme would operate had already been pre-determined by the TDP as per arrangements with the States.

· Jigawa: 14 out of 27 LGAs · Katsina: 14 out of 34 LGAs · Zamfara: 14 out of 14 LGAs

Stage 2: Selection of sets of primary schools

In each of the 14 LGAs in each state, 2 sets of 12 eligible primary schools each were to be selected;

To be eligible for the TDP: (1) each school should have one head teacher and at least another three teachers; (2) each school should have at least 8 grade-3 pupils.

Schools within each set were identified according to geographical proximity in order to facilitate any training and periodic meetings of teachers within each set, and to create a broader peer network within the locality.

It was the intention that the two sets of schools within each LGA would be selected to be broadly similar. State Education Boards (SUBEBs) were responsible for the selection and were provided with guidelines to assist them, such as taking into account the location of the schools (urban/rural), the size of the schools in terms of classrooms and pupils, presence of a School Based Management Committee (SBMC), and state of school infrastructure. In the case of Jigawa, nearly all schools would have had exposure to the also DFID-funded Education Sector Support Programme in Nigeria (ESSPIN). Therefore, care was taken to balance the level of exposure to ESSPIN across the pairs of sets in each LGA.

Stage 3: Selection of eligible teachers

Before the selection of schools which would participate in the TDP or not, the LGEA and head teacher from each school in every set was required to identify three other teachers who would potentially receive TDP support in addition to him/herself, based on the following criteria:

· Classroom teaching at early grade-level (grades 1-3); and · Classroom teaching in any of the three subjects: English, maths, or science.

Stage 4: Random allocation of treatment/control sets

After receiving lists of school sets and teachers from the TDP coordinators, the IE team randomly assigned one set of schools among every pair of sets to TDP 'treatment' status using a random number generator. The other set would therefore be assigned 'control' status.

This would result in 14 x 3 = 42 'treatment' sets of 12 schools each (504 'treatment' schools in total) and correspondingly 42 'control' sets of 12 schools each (504 'control' schools in total). In 'treatment' schools, all head teachers and identified teachers in the previous stage would receive TDP support.

(3) Drawing of the samples for the baseline survey

Selection of schools

This was performed in one stage, using implicit stratification by state, LGA and treatment/control status. In other words, each set of 12 schools described above was considered a stratum (42 'treatment' sets and 42 'control' sets).

4 schools were randomly selected from each set.

This yielded an intended sample size of 14 x 4 = 56 treatment schools in each state, and correspondingly 56 control schools in each state.

Thus the total intended sample size across all 3 states was 56 x 3 = 168 treatment schools and correspondingly 168 control schools = grand total of 336 schools.

Selection of teachers

At each sampled school, the head teacher and teachers identified during the construction of the sampling frame would be interviewed. Each teacher and head teachers who teach, would also be observed while they taught a lesson. Following the completion of the school survey, the teachers and head teachers (irrespective of whether they teach or not), would be administered a teacher development needs assessment (TDNA) at an examination centre.

Thus the intended numbers of interviews, lesson observations, and TDNAs were as follows:

· Head teacher interviews: 336 · Head teacher lesson observations: up to 336 · Teacher interviews: 336 x 3 = 1008 · Teacher lesson