Demographic and Health Survey, Special [hh/dhs-sp]
The 2004 NDES is the first education survey of its kind to be conducted in Nigeria.
The 2004 Nigeria DHS EdData Survey (NDES) was a nationally representative sample survey covering 4,268 households, 3,987 parent/guardians, 81 independent children age 13-16, and 9,695 children age 4-16. The primary objective of the 2004 NDES is to provide upto date household-based information on education among children of primary and secondary school age in order to inform the development, monitoring, and evaluation of education programmes in Nigeria. The survey focuses on the factors influencing household decisions about children’s school attendance. In addition, information is available on school attendance, costs of schooling (monetary and non-monetary) and parent/guardian attitudes about schooling. The 2004 NDES was the first education survey of its kind in Nigeria, and was linked to the 2003 Nigeria Demographic and Health Survey (DHS). The survey report (available under External Resources) presents information on adult educational attainment, children’s characteristics and rates of school attendance, absenteeism among primary school pupils and secondary school students, household expenditures on schooling and other contributions to schooling, and parent/guardian perceptions of schooling, among other topics.
The sample size for both the 2003 Nigeria DHS survey and the 2004 NDES was sufficiently large to provide estimates for indicators at the national level, by urban-rural residence, and at the regional level for most indicators. Twelve survey teams trained by the National Population Commission (NPC), in collaboration with the Federal Ministry of Education (FMOE), conducted the survey from February to July 2004.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
The scope of the 2004 NDES includes the following
- Background information on each parent/guardian respondent and on general education issues
- Parent/guardian's age, education, literacy, and religion
- Walking time and distance to the nearest primary and secondary schools
- Household support of and participation in school activities
- Parents' views on school quality
- Parents' views on the benefits and disadvantages of schooling
- Parents' views on reproductive health and HIV/AIDS education
- Primary school attended by the children for whom the parent/guardian responded, including the school type, location, and the reason for selection of that school.
Eligible Child :
- Schooling background and participation during the 2003-2004 school year (attended school during the 2003-2004 school year, dropped out of school, or never attended school)
- Frequency of and reasons for pupil absenteeism, household expenditures on schooling, and other costs of schooling (for children who attended school during the 2002-2003 school year)
- Reasons for dropping out of school (for children who had dropped out of school)
- Reasons for not attending school during the 2003-2004 school year (for children who had never attended school)
- Children's eating patterns
Producers and sponsors
National Population Commission
Federal Ministry of Education (FMOE)
United States Agency for International Development
Dr. Kristi Fair
Ms. Stephanie Gorin
Ms. Holly Newby
Mr. Glen Heller
Dr. Alfredo Alliaga
Department for International Development
United Nations Population Fund
Akintola Williams Deloitte and Touche
The sample for the 2004 NDES is based on the sampling frame for the 2003 Nigeria DHS survey, which was designed to provide estimates of health and demographic indicators for the country as a whole, urban and rural areas, and six geo-political zones (hereafter referred to as regions). This discussion will first address the sample design for the 2003 Nigeria DHS survey, then the subsequent design for the 2004 NDES.
The 2003 Nigeria DHS sample points (clusters) were systematically selected from a list of enumeration areas (EAs) defined in the 1991 Population Census. A total of 365 clusters was drawn from the census sample frame. After selecting the 365 clusters, the NPC trained teams to conduct the comprehensive listing of households and to update maps in the selected clusters. Following the listing operation, households to be included in the 2003 Nigeria DHS survey were selected, with the number of households selected per cluster being inversely proportional to the size of the cluster. In the 2003 Nigeria DHS sampling frame, the number of households by region was disproportional to population size, in order to have adequate numbers of cases for reporting by region. For both the 2003 Nigeria DHS survey and the 2004 NDES, the sample was constructed to allow for separate estimates for key indicators in each of the six geo-political regions in Nigeria (North Central, North East,
North West, South East, South South, and South West), with the result that the sample is not selfweighting at the national level.
Of the 365 clusters selected for the 2003 Nigeria DHS survey, 362 were successfully sampled. For the 2004 NDES, all of the 362 clusters completed for the 2003 Nigeria DHS survey were selected, and within those clusters, all households with children in the eligible child age range (4-16) were selected, comprising 4,563 households with one or more children age 4-16. Of these 362 clusters, 360 clusters were successfully completed for the 2004 NDES.
Deviations from the Sample Design
Of the 4,563 potential households selected, the 2004 NDES fieldwork teams successfully interviewed 4,268 households. The main reason that potential households were not interviewed was that the household had moved.
A total of 4,354 households were occupied, of which 4,268 were successfully interviewed, for an overall response rate of 98 percent. The household response rate was similar in urban and rural areas. In the interviewed households, 9,695 children were found and Eligible Child Questionnaires were completed for all of these children. In addition, 90 independent children were identified and interviews were completed with 81 of them, producing a response rate of 90 percent.
Since the 2003 Nigeria DHS sample is not selfweighting (non-proportional allocation by urban-rural residence and state), it requires sampling weights to provide estimates at every domain of study. In a given state, if c is the fixed number of households selected out of the total households (Li)—found in the 2003 listing process—for the i-th cluster, then the household probability in the selected i-th cluster can be expressed as
P2i = ( c / Li )
The final households overall probability in the i-th cluster could be calculated as
fi = P1i * P2i
and the sampling design weight for the i-th cluster is given as
1/ fi = 1 / ( P1i * P2i )
Dates of Data Collection
Data Collection Mode
Twelve interviewing teams carried out data collection for the 2004 NDES. Each team was composed of one supervisor, one driver, and three interviewers. Staff from NPC coordinated and supervised fieldwork activities, with the assistance of FMOE staff. ORC Macro staff also participated in field supervision.
Data Collection Notes
The pretest fieldwork was conducted over several days. Data were collected in 116 households, from 111 parent/guardian respondents, and on 227 eligible children age 4-16. Because none of the children age 13-16 in the households visited qualified as an independent child, the Independent Child Questionnaire was not used.
Based on the results of the pretest, changes in the pretest survey questionnaires and interviewers' manual were made before the main survey training was initiated and local language translations finalised. Among the changes made in survey instruments were the modification of answer categories for selected questions (including the classification system for religion), and the addition of new answer categories for some questions. For instance in terms of the benefits of primary schooling, a number of respondents mentioned improved hygiene, so this answer category was added to the relevant questions. Also,expenditures on furniture were mentioned frequently by respondents when they were asked about other expenditures on children's schooling, so a specific question on this expense was added to the questionnaire. One other change made to the questionnaire was in the sentences used to test literacy among children and adults, to ensure that across the various languages, the sentences included commonly recognized words and that the sentences were of comparable difficulty.
Twelve interviewing teams carried out data collection for the 2004 NDES. Each team was composed of one supervisor, one driver, and three interviewers. Staff from NPC coordinated and supervised fieldwork activities, with the assistance of FMOE staff. ORC Macro staff also participated in field supervision. Data were collected from February to July, 2004.
National Population Commission
Federal Ministry of Education
Four questionnaires were used for the 2004 NDES:
1. The Household Questionnaire
2. The Parent/Guardian Questionnaire
3. The Eligible Child Questionnaire
4. The Independent Child Questionnaire
These are all available under Appendix D of the Survey Report available under External Resources.
The Household questionnaire listed all of the people who were members of the household at the time the household was surveyed during the 2003 Nigeria DHS survey. The three purposes of the 2004 NDES Household Questionnaire were to:
- Confirm that the household was the same household surveyed by the 2003 Nigeria DHS survey;
- Identify which children were eligible (qualified) to be covered by the Eligible Child Questionnaire and those eligible to have anthropometric and literacy/numeracy data collected about them; and
- Identify a parent or guardian as the respondent for each eligible child.
Children who were age 4-16 at the time of the 2003 Nigeria DHS survey were eligible to be covered by the Eligible Child Questionnaire. Children age 4-9 at the time of the 2003 Nigeria DHS survey had their height and weight measured, and children age 4-12 were given a literacy/numeracy test.
The Parent/Guardian Questionnaire collected background information on each parent/guardian respondent and on general education issues. Information was collected on the parent/guardian’s age, education, literacy, and religion. Questions were asked about the walking time and distance to the nearest primary and secondary schools, as well as household support of and participation in school activities. Parent/guardians were also asked about their views on school quality, the benefits and disadvantages of schooling, and reproductive health and HIV/AIDS education. In addition, information was collected on each primary school attended by the children for whom the parent/guardian responded, including the school type, location, and the reason for selection of that school.
The Eligible Child Questionnaire collected different kinds of information about each eligible child age 4-16, depending on the child’s schooling status. While the subject of the Eligible Child Questionnaire was the individual child and his/her schooling, the respondent for the questionnaire was the child’s parent/guardian, as the purpose of the questionnaire was to collect information on issues from the parent/guardian’s perspective. Data were collected on the following topics, according to a child’s schooling status:
• Schooling background and participation during the 2003-2004 school year (attended school during the 2003-2004 school year, dropped out of school, or never attended school)
• Frequency of and reasons for pupil absenteeism, household expenditures on schooling, and other costs of schooling (for children who attended school during the 2002-2003 school year)
• Reasons for dropping out of school (for children who had dropped out of school)
• Reasons for not attending school during the 2003-2004 school year (for children who had never attended school)
• Children’s eating patterns
The Independent Child Questionnaire was used to interview directly a small percentage of the children age 13-16 in the selected households, rather than collecting information from a parent/guardian respondent. Independent children included those age 13-16 who were the head of the household, or the spouse of the head, or the son-in-law or daughter-in-law of the household head. Because these children did not have a parent/guardian who could answer questions about their schooling decisions, these children were interviewed directly. The same information was collected from these children themselves that otherwise would have been collected in the Eligible Child Questionnaire, and in terms of analysis, the data were grouped with data on other children in the eligible child age range.
The questionnaires were translated from English into three local languages—Hausa, Igbo, and Yoruba. Pretest training and fieldwork took place from 22 September to 4 October, 2003. For this exercise, six interviewers were trained (two per local language). The questionnaires were tested in Awka and Nibo (in Anambra State), Ibadan (in Oyo State), and Kano (in Kano State) in all languages, including English.
All questionnaires for the NDES were returned to the NPC headquarters in Abuja for data processing. Data processing consisted of office editing, the coding of open-ended questions, data entry, verification, and correcting of the computer-identified errors. A team of two data entry supervisors, a questionnaire administrator, three office editors, and ten data entry clerks processed the data. Data entry and editing started in late February, using the computer package CSPro (Census and Survey Processing System), which was specifically designed to process data from large-scale household surveys of this type. Data tables were produced using CSPro.
Estimates of Sampling Error
The estimates from a sample survey are affected by two types of errors: (1) nonsampling errors,and (2) sampling errors. Nonsampling errors are the results of mistakes made in implementing data collection and data processing, such as failure to locate and interview the correct household, misunderstanding of the questions on the part of either the interviewer or the respondent, and data entry errors. Although numerous efforts were made during the implementation of the 2003 Nigeria DHS and the 2004 NDES to minimize this type of error, nonsampling errors are impossible to avoid and difficult to evaluate statistically.
Sampling errors, on the other hand, can be evaluated statistically. The sample of respondents selected in the 2003 Nigeria DHS and the 2004 NDES is only one of many samples that could have been selected from the same population, using the same design and expected size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between all possible samples. Although the degree of variability is not known exactly, it can be estimated from the survey results.
A detailed account on estimatimation of sampling errors is available in Appendix B of the survey report available under External Resources.
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
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.