The Kenya Demographic and Health Survey (KDHS) was conducted between December 1988 and May 1989 to collect data regarding fertility, family planning and maternal and child health. The survey covered 7,150 women aged 15-49 and a subsample of 1,116 husbands of these women, selected from a sample covering 95 percent of the population. The purpose of the survey was to provide planners and policymakers with data useful in making informed programme decisions.
On March 1, 1988, 'on behalf of the Government of Kenya, the National Council for Population and Development (NCPD) signed an agreement with the Institute for Resource Development (IRD) to carry out the Kenya Demographic and Health Survey (KDHS).
The KDHS is intended to serve as a source of population and health data for policymakers and for the research community. In general, the objectives of the KDHS are to:
assess the overall demographic situation in Kenya, assist in the evaluation of the population and health programmes in Kenya, advance survey methodology, and assist the NCPD strengthen and improve its technical skills to conduct demographic and health surveys.
The KDHS was specifically designed to:
- provide data on the family planning and fertility behaviour of the Kcnyan population to enable the NCPD to evaluate and enhance the National Family Planning Programme,
- measure changes in fertility and contraceptive prevalence and at the same time study the factors which affect these changes, such as marriage patterns, urban/rural residence, availability of contraception, breastfeeding habits and other socioeconomic factors, and
- examine the basic indicators of maternal and child health in Kenya.
SUMMARY OF FINDINGS
The survey data can also be used to evaluate Kenya's efforts to reduce fertility and the picture that emerges shows significant strides have been made toward this goal. KDHS data provide the first evidence of a major decline in fertility. If young women continue to have children at current rates, they will have an average of 6.7 births in their lifetime. This is down considerably from the average of 7.5 births for women now at the end of their childbearing years. The fertility rate in 1984 was estimated at 7.7 births per woman.
A major cause of the decline in fertility is increased use of family pIanning. Twenty-seven percent of married women in Kenya are currcntly using a contraceptive method, compared to 17 percent in 1984. Although periodic abstinence continues to he the most common method (8 percent), of interest to programme planners is the fact that two-thirds of marricd women using contraception have chosen a modern method--either the pill (5 percent) or female sterilisation (5 percent). Contraccptive use varies by province, with those closest to Nairobi having the highest levels. Further evidence of the success in promoting family planning is the fact that more than 90 percent of married women know at least one modern method of contraception (and where to obtain it), and 45 percent have used a contraceptive method at some time in their life.
The survey indicates a high level of knowledge, use and approval of family planning by husbands of interviewed women. Ninety-three percent of husbands know a modern method of family planning. Sixty-five percent of husbands have used a method at some time and almost 49 percent are currently using a method, half of which are modern methods. Husbands in Kenya are strongly supportive of family planning. Ninety-one percent of those surveyed approve of family planning use by couples, compared to 88 percent of married women.
If couples are able to realise their childbearing preferences, fertility may continue to decline in the future. One half of married women say that they want no more children; another 26 percent want to wait at least two years before having another child. Husbands report similar views on limiting births--one-half say they want no more children. The desire to limit childbearing appears to be greater in Kenya than in other subSaharan countries. In Botswana and Zimbabwe, for example, only 33 percent of married women want no more children. Another indicator of possible future decline in fertility in Kenya is the decrease in ideal family size. According to the KDHS, the mean ideal family size declined from 5.8 in 1984 to 4.4 in 1989.
The KDHS indicates that in the area of health, government programmes have been effective in providing health services for womcn and children. Eight in ten births benefit from ante-natal care from a doctor, nurse, or midwife and one-half of births are assisted at delivery by a doctor, nurse, or midwife. At least 44 percent of children 12-23 months of age are fully immunised against the major childhood diseases, Almost all children benefit from an extended period of breastfeeding. The average duration of breastfeeding is 19 months and the practice does not appear to be waning among either younger women or urban women. Another encouraging piece of information is the high level of ORT (oral rehydration therapy) use for treating childhood diarrhoea. Among children under five reported to have had an episode of diarrhoea in the two weeks before the survey, half were treated with a homemade solution and almost one-quarter were given a solution prepared from commercially prepared packets.
The survey indicates several areas where there is room for improvement. Although young women are marrying later, many are still having births at young ages. More than 20 percent of teen-age girls have had at least one child and 7 percent were pregnant at the time of the survey. There is also evidence of an unmet need for family planning services. Of the births occurring in the 12 months before the survey, over half were either mistimed or unwanted; one fifth occurred less than 24 months after a previous birth.
Kind of data
Sample survey data
The 1989 KDHS sample is national in scope, with the exclusion of all three districts in North Eastern Province and four other northern districts (Samburu and Turkana in Rift Valley Province and Isiolo and 4 Marsabit in Eastern Province). Together the excluded areas account for less than 4 percent of Kenya's population.
Unit of analysis
- Women age 15-49
- Men age not specified
The population covered by the 1989 KDHS is defined as the universe of all women age 15-49 in Kenya and all husband living in the household.
Producers and sponsors
National Council for Population Development (NCPD)
Institute for Resource Development/Macro Systems, Inc.
Central Bureau of Statistics
U.S. Agency for International Development
The sample for the KDHS is based on the National Sample Survey and Ewduation Programme (NASSEP) master sample maintained by the CBS. The KDHS sample is national in coverage, with the exclusion of North Eastern Province and four northern districts which together account for only about five percent of Kenya's population. The KDHS sample was designed to produce completed interviews with 7,500 women aged 15-49 and with a subsample of 1,000 husbands of these women.
The NASSEP master sample is a two-stage design, stratified by urban-rural residence, and within the rural stratum, by individual district. In the first stage, 1979 census enumeration areas (EAs) were selected with probability proportional to size. The selected EAs were segmented into the expected number of standard-sized clusters, one of which was selected at random to form the NASSEP cluster. The selected clusters were then mapped and listed by CBS field staff. In rural areas, household listings made betwecn 1984 and 1985 were used to select the KDHS households, while KDHS pretest staff were used to relist households in the selected urban clusters.
Despite the emphasis on obtaining district-level data for phoning purposes, it was decided that reliable estimates could not be produced from the KDHS for all 32 districts in NASSEP, unless the sample were expanded to an unmanageable size. However, it was felt that reliable estimates of certain variables could be produced lbr the rural areas in the 13 districts that have been initially targeted by the NCPD: Kilifi, Machakos, Meru, Nyeri, Murang'a, Kirinyaga, Kericho, Uasin Gishu, South Nyanza, Kisii, Siaya, Kakamega, and Bungoma. Thus, all 24 rural clusters in the NASSEP were selected for inclusion in the KDHS sample in these 13 districts. About 450 rural households were selected in each of these districts, just over 1000 rural households in other districts, and about 3000 households in urban areas, for a total of almost 10,000 households. Sample weights were used to compensate for the unequal probability of selection between strata, and weighted figures are used throughout the remainder of this report.
A total of 9,836 households were selected in the Kenya Demographic and Health Survey. Of these, 8,343 were identified as occupied households during the fieldwork and 8,173 were successfully interviewed. Respondents for the individual interview were women aged 15-49 who had spent the night before the interview in the selected household. In the interviewed households, 7,424 eligible women were identified and 7,150 were successfully interviewed. In general, few problems were encountered during the interviewing and the response rate was high--98 percent for households and 96 percent for individual female respondents. In addition, 1,116 husbands were interviewed out of a total of 1,397 eligible, for a response rate of 81 percent. Eligible husbands were defined as those who spent the night before the interview in the selected households and whose wives were successfully interviewed. Every other household was considered eligible for the husband interview.
The distribution of all women by province indicates only minor differences among the three sources of data. For purposes of comparison, Respondents are classified into 4 educational categories, according to the highest grade attained at each level. These categories are: no education, 1-4 years, 5-8 years, and 9 or more years. 1 The data show a strong increase in the educational attainment of women over time. The proportion of women with no education declined from 44 percent in 1977/78 to 25 percent in 1989. The proportion of women who have 5 to 8 years of education is higher in 1989 (43 percent) than in 1984 (32 percent) and 1977/78 (27 percent).
Dates of collection
Mode of data collection
The KDHS utilised three questionnaires: a household questionnaire, a woman's questionnaire, and a husband's questionnaire. The first two were based on the DHS Programme's Model "B" Questionnaire that was designed for low contraceptive prevalence countries, while the husband's questionnaire was based on similar questionnaires used in the DHS surveys in Ghana and Burundi. A two-day seminar was held in Nyeri in November 1987 to develop the questionnaire design. Participants included representatives from the Central Bureau of Statistics (CBS), the Population Studies Research Institute at the University of Nairobi, the Community Health Department of Kenyatta Hospital, and USAID. The decision to include a survey of husbands was based on the recommendation of the seminar participants. The questionnaires were subsequently translated into eight local languages (Kalenjin, Kamba, Kikuyu, Kisii, Luhya, Luo, Meru and Mijikenda), in addition to Kiswahili.
Data processing staff for the KDHS consisted of five data entry clerks, two data entry supervisors and a control clerk who logged in questionnaires when they arrived at the office. The staff was supervised by two NCPD officers with periodic assistance from IRD staff. All the data processing staff completed the interviewer training course in November 1988 and received further instruction in data processing from the IRD staff.
Three IBM-compatible desktop microcomputers were installed in a temporary office on the Kenyatta National Hospital compound and wcre used to process the data. The Integrated System for Survey Analysis (ISSA) program was used for data entry, editing and tabulations. The supervisors and the NCPD officers were responsible for supervising data entry, and for resolving inconsistencies in questionnaires detected during secondary machine editing.
Data processing started in February 1989, once a sufficient number of questionnaires had been returned to Nairobi. Data entry was completed in early June and tabulations for the preliminary report were run in mid-June, two wceks after the last interview took place. The preliminary report was printed in July, tabulations for the final report were also produced in July, and this report was drafted in August and September.
The sample of women selected in the KDHS is only one of many samples that could have been selected from the same population, using the same design and expected size. Each one would have yielded results that differed somewhat from the actual sample selected. The sampling error is a measure of the variability between all possible samples; although it is not known exactly, it can be estimated from the survey results. Sampling error is usually measured in terms of the "standard error" of a particular statistic (mean, percentage, etc.), which is the square root of the variance. The standard error can be used to calculate confidence intervals within which one can be reasonably assured that, apart from non-sampling errors, the true value of the variable for the whole population falls. For example, for any given statistic calculated from a sample survey, the value of that same statistic as measured in 95 percent of all possible samples with the same design (and expected size) will fall within a range of plus or minus two times the standard error of that statistic.
If the sample of women had been selected as a simple random sample, it would have been possible to use strightforward formulas for calculating sampling errors. However, the KDHS sample design depended on stratification, stages, and clusters; consequently, it was necessary to utilize more complex formulas. The computer package CLUSTERS was used to assist in computing the sampling errors with the proper statistical methodology.
In addition to the standard errors, CLUSTERS computes the design effect (DEFT) for each estimate, which is defined as the ratio between the standard error using the given sample design and the standard error that would result if a simple random sample had been used. A DEFT value of 1.0 indicates that the sample design is as efficient as a simple random sample; a value greater than 1.0 indicates the increase in the sampling error due to the use of a more complex and less statistically efficient design.
Sampling errors are presented in Table B.2 through B.4 in the Appendix to the Final Report for 45 variables considered to be of major interest. Results are presented for the whole country and for urban and rural areas. In Tables B.5 through B,11, results are presented by province for 30 variables. Finally, Table B.12 contains sampling errors for current contraceptive use for the 13 targctted districts. For each variable, the type of statistic (mean, proportion) and the base population are given in Table B.1. For each variable, Tables B.2 through B.12 present the value of the statistic, its standard error, the number of unwcighted and weighted cases, the design effect, the relative standard error, and the 95 percent confidence limits.
The confidence interval has the following interpretation. For current use of family planning (CURUSE), the overall proportion of married women using is 0.269 or 26.9 percent and its standard error is 0.010. Therefore, to obtain the 95 percent confidence limits, one adds and subtracts twice the standard error to the sample estimate, i.e., 0.269 + or -(2 x 0.010), which means that there is a high probability (95 percent) that the true contraceptive prevalence rate falls within the interval of 0.250 to 0.288 (25 to 29 percent).
The relative standard error for most estimates for the country as a whole is not large, except for estimates of very small proportions. The magnitude of the error increases as estimates for subpopulations such as particular provinces or districts are considered. For contraceptive prevalence, for example, the relative standard error (as a percentage of the cstimated proportion) for the whole country, urban areas, Nairobi and Kilifi District is, respectively, 3.6 percent, 6.2 percent, 7.6 percent, and 23.3 percent. By district, this means that the prevalence rate of 31.3 for Murang'a District cannot be said with certainty to differ from the rate of 20.2 for Kisii District, since the confidence intervals overlap. Similarly, the difference between the rates for Kirinyaga (52.2 percent) and Machakos Districts (40.4 percent) might be explained by sampling error.
Other forms of data appraisal
Nonsampling error is due to mistakes made in carrying out field activities, such as failure to locate and interview the correct household, errors in the way questions are asked, misunderstanding of the questions on the part of either the interviewer or the respondent, data entry errors, etc. Although efforts were made during the design and implementation of the KDHS to minimize this type of error, nonsampling errors are impossible to avoid and difficult to evaluate analytically.
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.