National Agricultural Sample Census 1993-1994
Nigeria, 1993 - 1994
Identification
NGA_1993_NASC_v01_M
National Agricultural Sample Census 1993-1994
| Name | Country code |
|---|---|
| Nigeria | NGA |
Agricultural Census [ag/census]
The Federal Office of Statistics (now National Bureau of Statistics) first took part in the World Census Agriculture in 1950 and 1960. This was in response to a FAO request for all countries to undertake decennial censuses. However, these cenuses were not carried out simultaneously in all regions. Since then FOS has carried out two Censuses of Agriculture - 1974/75 and 1984/85. These censuses are generally referred to as The National Agricultural Sample Census (NASC) since a large sample is used to collect information rather than a 100% enumeration of holdings.
The two series of NASC were successful on date collection and a preliminary report on farm practice had been published.
The 1993/94 National Agricultural Sample Census was undertaken by the Federal Office of Statistics (FOS) in collaboration with the Federal Ministry of Agriculture, Water Resources and Rural Development. It had technical inputs from FAO.
The operation involved the complete listing of household units and households within units. Farming households which for this purpose includes households engaged in crop farming, livestock and fishing were identified from the listing forms. One out of every four farming households was selected for study. A Holding questionnaire which dealth with farm practice and other agricultural structural issues was administered to all selected households. Also the basic questionnaire, that is the General Household Survey questionnaire, dealing with socio-economic activities of the household (health, education, detailed demographic information, housing status, employment, etc) was applied to all selected households.
The three principal objectives of the census were:
a) To provide structural data on Agriculture in Nigeria mostly on those aspects that do not change frequently. In the context of this census, agriculture has been defined to include crop production (temporary and permanent), livestock rearing, keeping of poultry and fishing and forestry.
b) To obtain the socio-economic activities, health and educational status, detailed demographic and housing status formation on households, household heads and household members. This would provide Local Government Areas with baseline statistics.
c) To obtain production figures at the state level. The Census was in two phases: the first was to meet its first two objectives while the second phase was to meet its third objective. The attached report only deals with the phase 1 of the census.
Scope
The type of information collected include :
Identification,
General characteristics
Demographic characteristics
Land tenure
Area planted
Crops grown and yield
Farm prices
Farm inputs
Livestock
Fisheries
Other insular activities
These conform to the categories of items of FAO statistical development series for the program of World Census of Agriculture.
Coverage
The NASC Phase 1 covered 36 EAs in each LGA. At that time, 540 LGAs had been gazetted by the National Population Commission. Of these 18 LGAs on basis of their relative sizes compared to other LGAs in their respective states were split into sub-LGAs each. Therefore, there were 526 strata each with 36 sample EAs giving a total national sample of 20 232 EAs. Out of each EA, 12 households were selected for study giving a total national sample of 242 794 households.
The 540 LGAs gazetted did not reflect the last exercise of Government to creat new LGAs. Therefore some LGAs on the gazette were in fact two or more LGA8 on the ground. Since for such LGAs it was not possible at HQ to sort out the frame of EAs into their respective LGAs, it became necessary to select multiples of sample of EAs in the gazetted LGA. The selection of additional EAs was a condition- exercise and a total of about 59 additional samples (ach of 26 EA's) were added to the 242,784 indicated.
The pretest was in two phases in line with the design anticipated for the census. The phase one operation was carried out in five pretest states, namely, Anambra, Bauchi, Kano, Osun and Ondo.
Producers and sponsors
| Name |
|---|
| Federal Office of Statistics |
| Federal Ministry of Agriculture, Water Resources and Rural Development |
| Name | Role |
|---|---|
| Food and Agriculture Organization | Technical Input |
| Name | Role |
|---|---|
| Nigerian Federal Government | |
| Food and Agriculture Organization of the United Nations | Financial and Technical Assistance |
Sampling
Sample Design
The Sampling Scheme adopted was a two phase stage sampling selection: Phase One involved three levels of stratification.
The basic objective of Phase I was to provide some baseline data on every local Government. Area (LGA) in Nigeria. The LGA thus became the primary of first level of stratification. The EAs in each LGA were stratified into urban or rural, which thus the second level of stratification, Thus, in listing the EAs within each LGA the urban EAs were listed first, followed by the rural EAs. Systematic sampling from the EA list was to ensure that. the sample was distributed between urban EAs in the same proportion as for the whole population, without the need for calculating urban and rural sampling rates separately.
The third level of stratification, again implicitly, reflected general agro-ecological variation. Thus within the rural sector, the listing of EA in each LGA prior to selection was in a serpentine order on the map. 36 EAs were to be selected in each LGA using systematic selection with probability proportional to site. 12 households were selected per EA for study, the household being the primary sampling unit.
Sample Selection and the Associated Problems
The methodology of sample selection for NASC was as contained in the survey design by Chris Scott, FAO consultant. The preferred design which had several levels of stratification as state and Local Government had the Local Government further stratified into urban and rural sector,with additional level ofstratification, this time implicitly imposed on the rural sector to stratify it by cropping pattern. This design was believed to have the twin advantage of marrying most of what was good in the previous sample while at the same time remaining simple in application with regard to methodology of sample selection and estimation procedure. Over all it was believed that the resulting sample will provide us with better estimates than before.
The following steps were taken in the selection process.
(i) Stratification or grouping of EAs in each Local Government Areas (LGA) into urban and rural
(ii) The grouping of area within the rural EAs that produce similar crops together in a systematic manner until all the EAs within the rural sector of each LGA was strung together.
(iii) The selection of 36 EAs systematically in a continious manner from each Local Government Area. By this implicit stratification, the urban EAs will appear in proportion to their weight. Rural EAs with different cropping will also appear according to their presence or weight.
To facilitate the work a two week training of the staff for the sample selection was put in place. During the period, effort was also intensified to get the EAs frame from the National Population Commission (NPC).It was however discovered that the format in which the frame was compiled by NPC did not include areas by locality. This made both the distinction between urban and rural EAs blurred and affected rural stratification by crop. At this junction the methodology for sample selection was reviewed. The above method was then replaced by a simple straight forward systematic selection of EAs via the cumulation and selection of households which are contained in the frame as supplied by NPC. Under this method 36 EAs were in most cases systematically selected from each LGA. However, due to the marked difference or variance in the sizes of EAs, it was decided that some criteria was needed to separate urge LGAs from the average ones to avoid some LGA
being relatively over sampled or under sampled, with these 36 EAs were selected in each LGA while 72,108 EAs were selected in large EAs.
Soon after the rule guiding the sample selection for this revised method was established, the proper selection started. Once the initiall part of the frame came out of the computer, the work of selecting sample was done simultaneouslv with computer production of the frame. As the sample list of EAs per state were compiled arrangement was made to collect the corresponding sketch maps from NPC.
There were various problems in the course of compiling the frame for NASC. These were.
(i) Repeated requests and visits to NPC before the frame from which the sample list of EAs was selected.
(ii) The frame obtained was somewhat defective and incomplete. It was about 95% complete and listing of EAs did not contain listing of localities.
(iii) Because of the incompleteness of the frame a few LGAs in a few of the states were missing and so sample list for each LGA could not be obtained. Also the non-listing of EAs by localities in the frame presented some sampling problems leading to the review of the methodology of
sample selection .
(iv) Difficulties arising from further state creation was also encountered but it was easier to resolved since in nearly all cases it was a matter of reallocation of LGAs within the affected, state, except where they were subdivided and boundaries were not clearly defined.
(v) Where LGAs were split there was the need to draw additional samples.
Survey instrument
Listing Questionnaire: This was used to list households in the selected EA and to obtain data on crops grown, livestock/poultry kept and fishing activities.
General Household Questionnaire: This was used for sample household in Phase I and contained data on socioeconomic characteristics of each member of the household as well as housing conditions for the household.
Holding Questionnaire: This was for every holding identified as being operated by a member of the sample households in Phase 1 EA. Data was collected in respect of general farm practice, area of holding, tenure, use of inputs, farm implements, kinds of livestock/ poultry kept, access to
credit and marketing channels. Most responses on the questionnaire were precoded using international standard classifications.
Data collection
| Start | End |
|---|---|
| 1993 | 1994 |
| Name |
|---|
| Federal Office of Statistics |
Supervisors were based at strategic centres each overseeing the work of the three or four enumerators. Assistant field officers (AFO) formed the next level in the supervisory cadre, three supervisors were supervised by an AFO within each state. There were two or three field officers at the next level to oversee the work of the assistant field officers. The officer was in charge of the work in the state. Each state fell into one of four zones into which the country is divided. There was a zonal controller in each zone.
Data collection was undertaken by enumerators from FOS. FOS maintains a team of permanent field staff to collect data. The level of education and general development of the country demands this. Enumerators are posted to enumeration areas and live there throughout the period such areas are studied.
For the purpose of the census temporary enumerators were recruited to supplement the existing staff. About 4,602 enumerators and supervisors were deployed for the census and data collection took the form of personal interviews.An enumerator was assigned one EA and a supervisor coordinated a team of 6 enumerators while an editor worked with the supervisor in carrying out field edit on records.
Training
NASC was executed in collaboration with the Federal Ministry of Agriculture. Since some of the enumerators were temporary and the content of the questionnaires and manuals were detailed, there was thus the need to ensure that enough time was devoted to training of all categories of staff. There were four levels of training in preparation for the start of NASC in 1993/94.
1st Level: The first level training took place at the Centre for Management Development, Shangisha on the 2nd and 3rd of April 1993. The trainers were the officers of the Headquarters selected from the Agriculture and Household Statistics as well as the Field and Evaluation departments. The trainers were the Zonal Controller, Zonal Coordinators and State Officers of the pre-test state and some other officers from the Headquarters.
2nd Level: The second level training took place at the Zonal Headquarters: Kaduna for NW zone, Jos for NE zone, Enugu for SE zone and Ibadan for SW zone, from the 12th to 14th of April, 1993. The trainers were the zonal controllers with eight officers from headquarters - complementing the trainers as well as acting as moderators. The trainers were the state officers, scrutiny officers and the field officers.
3rd Level: The 3rd level training took place from the 19th to the 22nd of April 1993 in all the headquarters of state offices. All the supervisors for the Census were trained by those trained during the 2nd level training.
4th Level: The 4th training took place from the 26th to 29th April, 1993. The training was held in at least 3 training centres within the states. All the interviews for the census were trained here by the sate trainers at the 3rd level and a few selected supervisors.
State Statistical Agencies: Training was also conducted for data collection for staff of the statistical agencies from the 2nd to 5th of June in four zonal offices of the FOS.
Editors and Coders : Training for coders and editors took place at the Agriculture Survey Division, Lagos from 14th to 25th June, followed by field practicals for one week. Senior officers from the Federal Ministry of Agriculture were trained during the first level training for the pretest and they were trainers during second level training. They however were unable to take part during the training for the main census.
The pretest was in two phases in line with the design anticipated for the census. The phase one operation was carried out in five pretest states, namely, Anambra, Bauchi, Kano, Osun and Ondo.The experience from the pretest and the result of the pretest data processing were used extensively to effect changes in the training for census and drafting of the final instruments for the census.
Data processing
Data Processing: Questionnaires were retrieved from the field for processing at Headquarters. The retrieved questionnaires were first edited and coded manually by trained statistical clerks before being sent to the data entry clerks for computerisation. After data entry had been completed and checked by the programmers, the data diskettes were sent to the Statistician for computer editing and tabulation. The programme for Data Entry was written by FOS programmers, while editing and tabulation programmes were written by an FAO Consultant who worked with FOS for about six months. The FAO Consultant did a lot in building computer capability among the staff of the Division. All aspects Of Data Processing were carried out by our statisticians and programmers.
Data appraisal
Spot/Quality Checks:
Right from the planning stage of the NASC need for control of the quality of the various aspects of the programme was realised. Particular attention was paid to field enumeration, this was one major operation whose activities at every stage were evaluated. Several measures
were out in place
(i) adequate training of the scrutiny staff
(ii) development of appropriate instructions manuals
(iii) close supervision and monitoring of file activities
(iv development of editing specification for each data collection instruments
(v) well documented programme of spot checks and re-interviews
(vi) validating checks through the use of data from Family Planning Survey (FPS), Labour Force Survey (LFS) and NASC Phase II listing.
The measures were helped to assess fully the quality of the data collected during the field operations.
Consistency Checks: The office has constituted a group headed by a very senior officer from the Methodology Division to evaluate the degree of consistency of the current census data. This evaluation covered the Listing, Holding and General Household Questionnaires. Agricultural data for the past years were examined to see the extent of compatibility with the current census data.
National Technical Committee of the Census: A technical committee of the NASC was set up. It was made up of officers from FOS, Federal Ministry of Agriculture and some other relevant agencies. The committee was responsible for the design for the pilot test and some aspects of the draft of the main census questionnaire The committee could not meet during the actual census exercise due to logistics problems but the
Federal Ministry of Agriculture was always informed of the progress of the census.
Census Monitoring Committee: This was constituted to review and monitor the progress of the census. Members of this committee were senior officers from the Agricultural Statistics Division, Household Statistics Divsion, Field Operations and Evaluation Department, Planning Division and
Data Processing Department. This committee met weekly. The committee was very useful and helped to identify areas requiring intervention at every stage of the exerse.
Data Access
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
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.
Metadata production
DDI_NGA_1993_NASC_v01_M_WB
| Name | Affiliation | Role |
|---|---|---|
| Development Economics Data Group | The World Bank | Documentation of the DDI |
2013-07-09
Metadata version
Version 01 (July 2013)