Multiple Indicator Cluster Survey 2013-2014, Bungoma County
Household Survey [hh]
The Bungoma County Multiple Indicator Cluster Survey (MICS5) was conducted in collaboration with the Population Studies and Research Institute (PSRI) of the University of Nairobi, the Kenya National Bureau of Statistics (KNBS) and the United Nations Children's Fund (UNICEF).The Kenya National Bureau of Statistics implemented (MICS5) in 2013-2014 in the three counties of Bungoma, Kakamega and Turkana as part of Global MICS round five.
The global MICS program was developed by UNICEF in the 1990s as an international household survey program to support countries in the collection of internationally comparable data on a wide range of indicators on the situation of children and women. MICS surveys measure key indicators that allow countries to generate data for use in policies and programs and to monitor progress towards the Millennium Development Goals (MDGs) and other internationally agreed upon commitments. Technical and financial support were provided by the United Nations Children's Fund.
The results of this survey provided requisite baseline information that can be used to facilitate evidence-based planning, budgeting and programming by policymakers and stakeholders at the county levels. The survey will go a long way in encouraging increased demand for use of statistics by policy makers at devolved levels and will ensure that resources at both county and national levels are used most effectively through well-planned projects/programs that will benefit especially the women and children of the three counties. The MICS5 results were critical in gauging milestones achieved in the field of education, nutrition, child development, health for women and children in the three counties and in evaluating the various health based policies that the government has formulated over the years towards achieving the national welfare objectives.
The 2013-14 MICS5 data was critical in informing the future planning for the three counties, especially in view of the new constitutional dispensation and Vision 2030. It was anticipated that MICS5 would supplement the data collected during the 2014 Kenya Demographic and Health Survey (KDHS). In addition the information collected would inform strategic communication for social and behavior change interventions by government and partners including UNICEF. Furthermore the data contributed to the improvement of data and monitoring systems in the three counties. The primary objectives of the Bungoma County survey are:
1. To provide up-to-date information for assessing the situation of children and women in Bungoma County.
2. To generate data for the critical assessment of the progress made in various areas, and to put additional efforts in those areas that require more attention.
3. To furnish data needed for monitoring progress toward goals established in the Millennium Declaration, and other internationally agreed upon goals, as a basis for future action.
4. To collect disaggregated data for the identification of disparities, to allow for evidence based policy-making aimed at social inclusion of the most vulnerable.
5. To contribute to the generation of baseline data for the post-2015 agenda.
6. To validate data from other sources and the results of focused interventions.
7. To contribute to the improvement of data and monitoring systems in Kenya and to strengthen technical expertise in the design, implementation, and analysis of such systems.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Households and individuals
v1: Edited, anonymised dataset for public distribution.
The scope of the Bungoma County Multiple Indicator Cluster Survey includes:
- List of household members
- Child labor
- Child discipline
- Household characteristics
- Insecticide treated nets
- Indoor residual spraying
- Water and sanitation
- Salt iodization
INDIVIDUAL WOMEN QUESTIONNAIRE
- Woman's background
- Access to mass media and use of information/communication technology
- Fertility/birth history
- Desire for last birth
- Maternal and newborn health
- Post-natal health checks
- Illness symptoms
- Unmet need
- Female genital mutilation/cutting
- Attitudes toward domestic violence
- Sexual behavior
- Tobacco and alcohol use
- Life satisfaction
CHILDREN UNDER 5 QUESTIONNAIRE
- Birth registration
- Early childhood development
- Breastfeeding and dietary intake
- Care of illness
The survey covered Bungoma County in Kenya
The survey covered all de jure household members (usual residents), all women aged between 15-49 years and all children under 5 living in the household.
Producers and sponsors
Kenya National Bureau of Statistics
Government of Kenya
Population Studies and Research Institute
University of Nairobi
United Nations Children’s Fund
Financial and technical support
The primary objective of the sample design for the Bungoma County MICS was to produce statistically reliable estimates of indicators, at county level. The urban and rural areas in Bungoma County were the sampling strata. A multi-stage, stratified cluster sampling approach was used for the selection of the survey sample.
MICS5 utilized the recently created fifth National Sample Survey and Evaluation Program (NASSEP V) frame which is a household based master sampling frame developed and maintained by KNBS. The frame was implemented using a multi-tiered structure, in which a set of 4 sub-samples (C1, C2, C3, C4) were developed. It is based on the list of enumeration areas (EAs) from the 2009 Kenya Population and Housing Census. The frame is stratified according to County and further into rural and urban. Each of the sub-samples is representative at county level and at national (i.e. Urban/rural) level and contains 1,340 clusters.
The Primary Sampling Units (PSUs) for the survey were clusters drawn from the NASSEP V sampling frame, so the first component of the probabilities and weights are based on that master sample. Within each stratum the PSUs for the MICS were selected independently from one of the subsamples of the master sample using Equal Probability Selection Method (EPSEM). A total of 50 clusters were selected from the master sample in this way.
Out of the 50 sample clusters selected for Bungoma County, it was established that 30 had been listed more than six months prior to the start of the survey. These listing for these clusters was updated prior to selection of households. For this purpose, listing teams visited each cluster, and listed all occupied households. For the remaining 20 sample clusters a more recent listing was available, so it was used for selecting the sample households.
Information was collected from a total of 1,246 households representing 95 percent response rate. The composition of these households was 5,983 household members comprising 2,797 males and 3,186 females. The mean household size was 4.8 persons. About 48 percent of the sampled households' population is below 15 years, 48 percent are between age 15-64 years and four percent are age 65 years and above.
Due to data quality issues, data relating to mortality and anthropometric measures were not analyzed and reported. Anthropometric data suffered digit preference for both weight and height, while for mortality, deaths especially among children under-five years were under reported. KDHS 2014 had similar shortcomings.
The MICS5 sample was not self-weighting and thus a weighting process was required to provide estimates representative of the target population. Two main sampling weights were calculated: household weights and individual (women and children) weights. The base weights incorporated the probabilities of selection of the clusters from the census EAs database into the NASSEP V sample frame, the probabilities of selection of the MICS clusters from NASSEP V frame and the probabilities of selection of the households from each of the NASSEP V frame clusters.
Base weights were then adjusted for cluster and household non-response by multiplying them by the inverse of the clusters and households response rates. The individual weight of a woman or child was calculated as the household weight multiplied by the inverse of the individual response rate. Given that the MICS5 sample was a two-stage stratified cluster sample, sampling probabilities were calculated separately for each sampling stage.
Dates of Data Collection
Data Collection Mode
Kenya National Bureau of Statistics
Government of Kenya
Population Studies and Research Institute
University of Nairobi
A set of three questionnaires was used in the survey:
1. A household questionnaire which was administered to the household head or any other responsible member of the household.
2. A questionnaire for individual women administered in each household to all women age 15-49 years.
3. An under-5 questionnaire administered to mothers (or caretakers) for all children under-5 years living in the household.
Data were entered into the computers using the Census and Surveys Processing System (CSPro) software package, Version 5.0. Data entry was done by a trained team of 14 data entry operators, one archivist/system administrator and one data entry supervisor. For quality assurance purposes, all questionnaires were double-entered and internal consistency checks performed. Procedures and standard programs developed under the global MICS program and adapted to the Bungoma County MICS questionnaire were used throughout. Data processing began simultaneously with data collection in November 2013 and was completed in February 2014. Data were analyzed using the Statistical Package for Social Sciences (SPSS) software, Version 21. Model syntax and tabulation plans developed by UNICEF were customized and used for this purpose.
Estimates of Sampling Error
The sample of respondents selected in the Bungoma County MICS is only one of the samples that could have been selected from the same population, using the same design and 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 the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.
The following sampling error measures are presented in this appendix for each of the selected indicators:
- Standard error (se): Standard error is the square root of the variance of the estimate. For survey indicators those are means, proportions or ratios, the Taylor series linearization method is used for the estimation of standard errors. For more complex statistics, such as fertility and mortality rates, the Jackknife repeated replication method is used for standard error estimation.
- Coefficient of variation (se/r) is the ratio of the standard error to the value (r) of the indicator, and is a measure of the relative sampling error.
- Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling based on the same sample size. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design of the survey is as efficient as a simple random sample for a particular indicator, while a deft value above 1.0 indicates an increase in the standard error due to the use of a more complex sample design.
- Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2.se or r - 2.se) of the statistic in 95 percent of all possible samples of identical size and design.
For the calculation of sampling errors from the MICS data, programs developed in CSPro Version 5.0, SPSS Version 21 Complex Samples module and CMRJack116 have been used. The results are shown in the tables that follow. In addition to the sampling error measures described above, the tables also include weighted and unweighted counts of denominators for each indicator. Given the use of normalized weights, by comparing the weighted and unweighted counts it is possible to determine whether a particular domain has been under-sampled or over-sampled compared to the average sampling rate. If the weighted count is smaller than the unweighted count, this means that the particular domain had been over-sampled. As explained later in the footnote of Table SE.1, there is an exception in the case of indicators 4.1 and 4.3, for which the unweighted count represents the number of sample households, and the weighted counts reflect the total population.
Sampling errors are calculated for indicators of primary interest, at the county level, and for urban and rural areas within Bungoma County. Three of the selected indicators are based on household's members, eight are based on women, and two are based on children under 5. Table SE.1 shows the list of indicators for which sampling errors are calculated, including the base population (denominator) for each indicator. Tables SE.2 to SE.4 show the calculated sampling errors for selected domains.