The Botswana Family Health Survey II 1988 is the second DHS survey carried out in Botswana (the first one was done in 1984).
The BFHS-II was a national sample survey designed to provide information on fertility, family planning, and health in Botswana. The BFHS-II involved interviewing a randomly selected group of women between 15 and 49 years of age. These women were asked questions about their background, the children they had given births to, their knowledge and use of family planning methods, some health matters and other information which will be helpful to policy-makers and administrators in the health and family planning areas.
The objectives of the BFHS-II are to provide information on family planning awareness, approval and use, basic indicators of maternal and child health, and other topics related to family health. In addition, the BFHS-II complements the data collected in the BDS, by obtaining information needed to explore trends in fertility and mortality, and to examine the factors that influence these basic demographic indicators, particularly, the proximate determinants of fertility.
Specific objectives are:
- To collect information on fertility and family planning;
- To find out what type of women are likely to have more or fewer children or to use or not use family planning;
- To collect information on certain health-related matters such as antenatal checkups, supervised deliveries, postnatal care, brcastfeeding, immunisation, and diarrhoea treatment;
- To develop skills in conducting periodic surveys designed to monitor changes in demographic rates, health status, and the use of family planning; and
- To provide internationally comparable data which can be used by researchers investigating topics related to fertility, mortality and maternal-child health.
- The BFHS-II found that current fertility levels in Botswana remain high; however, the results show a decline in fertility in recent years.
- The BFHS-II found that traditional practices of breastfeeding and post-partum abstinence continue to be important factors in protecting women from a subsequent pregnancy. However, there is evidence that the duration of these practices is being curtailed among urban women.
- Knowledge of family planning methods and of places to obtain them is critical in the decision whether to use family planning and which method to use. The BFHS-II found that the MCH/FP programme has been quite successful in educating women about family planning.
- Use of contraception is the most important measurement of success in a family planning programme. The BFHS-II found that more than half of Batswana women have used a modern method of family planning at sometime and three out of ten women are currently using a contraceptive method to delay or avoid a birth.
- The BFHS-II found that women who are not currently using family planning, but do not wish to become pregnant soon, report a number of barriers to using family planning. A significant number of these women do intend to use family planning in the future.
- The BFHS-II also looked at the issue of women's perceptions about their partner's attitudes toward family planning. Within couples, male approval is much lower than that of females, although the reported level of partner's approval has increased since 1984.
- The BFHS-II results suggest that Batswana women have a growing interest in spacing births but continue to have a preference for many children.
- The BFHS-II documents that many women had a birth sooner that they would have liked although only a minority of women had another birth when they preferred not to have any more children.
- Women can be considered in need of family planning if they are not currently using a method of contraception and either want no more births or want to postpone the next birth for two or more years. The BFHS-II found that 45 percent of women in union are in need of family planning.
- Since teenage pregnancy places the health and welfare of teenagers and their births at risk, the Government of Botswana encourages women to wait until age 20 before their first pregnancy. The BFHS-II found that nearly one-quarter of teenagers had at least one birth and an additional 5 percent were pregnant with their first child at the time of the survey.
- The BFHS-II also documents that the MCH/FP programme has made a successful contribution to the reduction of infant and child mortality, though children of mothers with no education and children born soon after a previous birth have higher mortality rates.
- The BFHS-II documents that maternal and child health services are widely used by women in Botswana and the programme has expanded significantly since 1984.
- The BFHS-II found that a significant proportion of ill children received appropriate treatment. Although a large proportion of children who suffered from diarrhoea were treated with oral rehydration therapy, of concern is the significant proportion for whom fluid and food intake was cut down during the diarrhoeal episode.
- The BFHS-II included questions on knowledge of AIDS, the ways the disease is transmitted, who is at highest risk, and behaviors that will help someone avoid the disease. In the absence of either a vaccine or a cure for AIDS, education about prevention is the main strategy for combatting the epidemic. Nearly all women interviewed in the BFHS-II had heard of AIDS. However, many women lack correct information or have misconceptions about the disease.
The results of the 1984 BFHS showed that the Botswana MCH/FP programme has made considerable progress in providing health and contraceptive services to women of childbearing age. The 1988 BFHS (BFHS-II) confirms this and documents the further progress made between 1984 and 1988. The results of the BFHS-II indicate that utilization of MCH services has increased, along with knowledge and use of family planning. However, the 1988 findings also point to areas of the MCH/FP programme that need improvement.
I. An area where additional effort is needed is in Information, Education, and Communication (also recommended in 1984):
- Counselling services should be strengthened so that they are better able to disseminate information about family planning and dispel misconceptions women have regarding the use of contraception. The strengthening of the services should be targeted not only towards clients but also health workers.
- Information, education, and communication (IEC) activities at the district level need strengthening by training or designating officers specifically to carry out these services.
II. Outstanding recommendations from the 1984 BFHS should continue to receive emphasis:
- Further efforts should be directed toward educating and counseling teenagers (both boys and girls) about responsible sexual behavior.
- Additional attention should be placed on informing men about the health and other benefits of family planning. Emphasis should be placed on the importance of couple communication in this area and on the fact that childbearing is the joint responsibility of the couple and not the choice of the man or woman alone.
- Stress should continue to be placed on the health benefits of traditional practices such as breastfeeding and post-partum abstinence.
IEC materials targeting special population subgroups, e.g., illiterate women, should be developed.
- Emphasis should be placed on identifying women in need of family planning services, particularly those concerned about limiting their family size. Counseling about family planning during the provision of antenatal and post-partum services is a key mechanism in reaching these women.
- Potential acceptors should be counseled about the most appropriate methods for their age, life situation and fertility intentions.
- Research should be undertaken to further investigate the determinants and consequences of adolescent childbearing.
- Acceptors should be informed about possible side effects associated with the method they adopt, and follow-up of acceptors should be emphasized to reduce the levels of discontinuation due to side effects.
Kind of Data
Sample survey data
Unit of Analysis
- Women age 15-49
The Botswana Family Health Survey II 1988 covers the following topics:
- Family planning knowledge
- Mortality and health
- Post partum insuscecptibility
- Knowledge of AIDS
- Reproductive Behavior
- Teenage Pregnancy
The population covered by the 1994 ZDHS is defined as the universe of all eligible women, defined as those age 15-49 years who spent the night prior to the household interview in the selected household, irrespective of whether they were usual members of the household.
Producers and sponsors
Family Health Division
Ministry of Health
Central Statistics Office
Ministry of Finance and Development Planning
Institute for Resource Development/Macro Systems, Inc
United States Agency for International Development
Finance section of MFDP
Ministry of Finance and Development Planning
Administration of Funding
The Botswana Institute of Administration and Commerce (BIAC)
Statistics Department of the University of Botswana
Comparison of the age distribution of the women in the BFHS-II sample with the distribution of women 15-49 in the 1981 census and 1984 Botswana Family Health Survey (BFHS). The BFHS-II sample has a greater concentration of women at the ages 20-34 than the other two data sets. There is apparent under-sampling of teenagers in both the 1984 BFHS and the 1988 BFHS-II. An examination of the distribution of household members by age and sex enumerated in the BFHS-II household listing indicates a greater than expected number of women in the 10- 14 age group for females and a dearth in the 15-19 age group. Some interviewers may have recorded women in the 15-19 year age group as having a younger age in the household listing in order to make them ineligible for the individual interview and thus lighten their work load. Similarly, it was also found that females in the 45-49 age group was under-enumerated relative to the 50-54 age group.
The greater concentration of women in the prime reproductive ages in the BFHS-II may also result from the fact that interviewers were more successful in interviewing women in selected households in the urban areas, where more young women are found. One consequence of the greater concentration of younger women is that estimates of contraceptive prevalence may be higher, and fertility lower, than if more older women had been interviewed.
The distribution of women by marital status in the BFHS-II is similar to that found in the 1981 census, whereas the 1984 BFHS classified a much greater proportion of women as currently in union. The 1984 BFHS included two additional probes to determine how many women reporting their marital status as separated, divorced, widowed, or single were actually living with a partner at the time of the interview. In response to these probes, almost half of the women who initially did not report themselves as married or in a consensual union said that they were currently living with a partner, resulting in a much higher estimate of the proportion currently in union.
The sample shows a rapid increase in the proportion of the Botswana population living in urban areas. The proportion of respondents residing in urban areas increased from 21 percent in 1981 to 24 percent in 1984, and rose to 30 percent by 1988. However, the BFHS-II may include a slightly greater proportion of urban women than is found in the population. There has also been a increase in the education of women in the 1980s. Only 30 percent of women 15-49 at the time of the 1981 census reported that they had completed primary school or higher, compared with more than 50 percent of women in the BFHS-II. In 1981, 35 percent of women of reproductive age had not attended any school; by 1988, only 24 percent had received no education.
Distribution of the surveyed women by education and according to age, urban-rural residence, and religion. Education is a major factor which determines the level of participation of women in the various sectors of the modcrn economy. Generally, women in Botswana play an active and significant role in the educational system both as students and as teachers. For the last ten years, female students have dominated the primary and junior secondary school system. However, this situation changes at senior secondary and higher levels of education.
The percent of women by education according to age cohort shows the increasing level of education among Batswana women. The percent of women with no education drops dramatically with decreasing age and, conversely, the proportion with at least completed primary schooling rises. As expected, urban women are better educated than their rural counterparts. The data also show variations in education by religion. Women who belong to the Spiritual-African Church, or profess to have no religion, have substantially less education than Catholic or Protestant women.
Deviations from the Sample Design
The large difference in the proportion of eligible households between urban and rural areas is because many rural residents have more than one house, which they occupy at different times of the year. Households which were occupied for only part of the time were included in the household listing used for selection, but some proportion of them would necessarily be empty at the time of the survey. Among eligible households, the same proportion of households were successfully interviewed in urban and rural areas.
Number of households and women selected and successfully interviewed by urban/rural residence: It indicates that 4620 households, or 80 percent, of the 5776 selected households were eligible to be interviewed, Thirteen percent of households were ineligible because no member of the household had slept in the house the night before the interview and another 4 percent of the selected households were vacant or not dwellings. Of the 4620 eligible households, 4473 households or 97 percent, were successfully interviewed. In the urban and rural areas, 90 and 72 percent, respectively, of the households were eligible for interview.
The household questionnaire identified 4648 eligible women, of which 95 percent were successfully interviewed. This rate did not vary between urban and rural areas. The overall response rate, the product of the household response rate and the individual response rate, was 92 percent.
The weighted and unweighted number of women. Weighting of data is necessary to compensate for differences in the selection probabilities and response rates. The weights are determined in such a way that the total number of weighted cases equals the total number of women interviewed. Therefore, for most of the sample, the weighted number of cases can serve as a rough guide for the actual number of cases. The main exceptions are when results are tabulated by the criteria used to define the sampling domains, in this case urban or rural residence, or any characteristics strongly associated with urban-rural residence. All results presented in this report are weighted.
Dates of Data Collection
Data Collection Mode
The supervisor was responsible for the overall management of the team, including work assignments, locating selected households, and enlisting the cooperation of the community in the selected areas, as well as control of the quality of data collection. The latter was done through field editing all questionnaires, observation of interviews and re-interviewing women when necessary. Supervisors were in frequent contact with CSO by telephone. Additionally, central survey staff from CSO and DHS participated in fieldwork observation. The objective of these visits was to monitor the progress of fieldwork, to help solve problems, and to enhance the morale of the fieldworkers.
Data Collection Notes
The BFHS-II questionnaires were pretested in April and May, 1988. Eight female interviewers, two female supervisors and five male interviewers/supervisors, all of whom had participated in the 1987 Botswana Demographic Survey, were trained for 12 days and conducted 166 interviews during the pretest.
Immediately following the pretest, a listing of dwellings from the selected EAs was carried out by 12 male CHIPS interviewers (the five male interview/supervisors included). The exercise began in late May and was concluded mid-October, 1988. Due to the experience accumulated through previous surveys, the 12 listers were trained for a period of less than a week. For listing in towns, plot maps with plot numbers and street names were used; while in the rural areas, villages and district maps were used to locate households.
Training for the main fieldwork was held in July, 1988 and lasted three weeks. CSO and DHS staff were primarily responsible for training. In addition, staff from the Family Health Division, Ministry of Health, conducted several sessions on human reproduction, contraceptive methods, and maternal-child health. A separate training course was held for supervisors (9 of 10 who had participated in the pretest as supervisors or interviewers).
Fieldwork started on 4th August, 1988 and was completed on 13th December, 1988. In all, 25 female interviewers, 9 supervisors (6 female and 3 male), and 9 drivers participated in the fieldwork. Fieldwork was conducted by nine teams composed of 2 or 3 interviewers and a supervisor. Each team was assigned a vehicle and a driver.
Central Statistics Office
Ministry of Finance and Development Planning
Two questionnaires were used for the BFHS-II: a household and an individual questionnaire. The questionnaires were adapted from the DHS Model "B" Questionnaire, intended for use in countries with low contraceptive prevalence, with the addition of a modified version of the family planning section from the DHS Model "A" Questionnaire for high prevalence countries. The household and individual questionnaires were administered in either Setswana or English.
Information on the age and sex of all usual members and visitors in the selected households was recorded in the household questionnaire. This information was used to identify women eligible for the individual interview. Data on fostering for children age 0-14 were also collected in the household questionnaire.
The individual questionnaire was used to collect data for all eligible women, defined as those age 15-49 years who spent the night prior to the household interview in the selected household, irrespective of whether they were usual members of the household. The individual questionnaire was used to collect information on the following topics:
4.Knowledge and Use of Family Planning
5.Maternal and Child Health and Breastfeeding
7.Knowledge of AIDS
9.Husband's Background, Women's Work, and Child Support
Completed questionnaires were delivered to CSO regularly. Coding, data entry and machine editing went on concurrently at the CSO as the fieldwork progressed. All data processing was performed on microcomputers using the Integrated System for Survey Analysis (ISSA) software developed by IRD. Both coding and data entry, which were started in mid-September, were completed by mid-December, 1988. Subsequently, approximately 20 percent of the questionnaires were re-entered to verify the accuracy of the initial data entry. Before tabulation, the data were edited for consistency and inconsistencies were resolved, when possible, following the rules developed for the Demographic and Health Survey programme. Senior survey staff from CSO were responsible for supervising data entry and for resolving inconsistencies in questionnaires detected during secondary machine editing. The tabulations for the preliminary report were produced in Botswana in the week fieldwork was completed. Tabulations for this report were initially run at IRD and sent to CSO and FHD for review. An initial draft of this report was prepared by CSO, FHD, and DHS staff in Gaborone. Subsequently, one analyst from CSO and one from FHD spent two weeks in Columbia, Maryland to finalize the report.
Estimates of Sampling Error
The results from sample surveys are affected by two types of errors: nonsampling error and sampling error. The former is due to mistakes in implementing the field activities, such as failing to locate and interview the correct household, errors in asking questions, data entry errors, etc. While numerous steps were taken to minimize this sort of error in the BFHS-II, nonsampling errors are impossible to avoid entirely, and are difficult to evaluate statistically.
Sampling errors, on the other hand, can be evaluated statistically. The sample of women selected in the BFHS-II is only one of many samples of the same size that could have been drawn from the population using the same design. Each sample would have yielded slightly different results from the sample actually selected. The variability observed among all possible samples constitutes sampling error, which can be estimated from survey results (though not measured exactly).
Sampling error is usually measured in terms of the "standard error" (SE) of a particular statistic (mean, percentage, etc.) which is the square root of the variance of the statistic across all possible samples of equal size and design. The standard error can be used to calculate confidence intervals within which one can be reasonably sure 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 of identical size and design will fall within a range of plus or minus two times the standard error of that statistic.
If simple random sampling had been used to select women for the BFHS-II, it would have been possible to use straightforward formulas for calculating sampling errors. However, the BFHS-II sample design used two stages and clusters of households, and it was necessary to use more complex formulas. Therefore, the computer package CLUSTERS, developed for the World Fertility Survey, was used to compute sampling errors.
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 indicates that the sample design is as efficient as a simple random sample; a value greater than 1 indicates that the increase in the sampling error is due to the use of a more complex and less statistically efficient design.
Sampling errors are presented for selected variables and sub-populations of women in Tables B.1-B.7. In addition to the standard error and value of DEFT for each variable, the tables include the weighted number of cases on which the statistic is based, the relative error (the standard error divided by the value of the statistic) and the 95 percent confidence limits. The confidence limits may be interpreted by using the following example: the overall estimate of the mean number of children ever born (CEB) is 2.580 and its standard error is .050. To obtain the 95 confidence interval, twice the standard error is added to and subtracted from the estimate of CEB, 2.580 + 2 * 0.050. Thus, there is a 95 percent probability that the true value of CEB lies between 2.480 and 2.681.
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
Botswana Central Statistics Office, Family Health Division, Ministry of Health [Bostswana], and Institute for Resource Development/ Macro Systems, Inc., Columbia, Maryland USA. Botswana Family Health Survey II (BFHS) 1988. BWA_1988_DHS_v01_M. Datase 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.