Summary of the sample design for PMA Kenya Phase 1:
The Performance Monitoring for Action (PMA) Kenya Phase 1 (KEP1) survey was conducted in 308 enumeration areas (EAs) from 11 select counties – Nairobi, Kilifi, Nandi, Nyamira, Kiambu, Bungoma, Siaya, Kericho, Kitui, Kakamega, and West Pokot. The EAs were selected from the master sample frame by the Kenya National Bureau of Statistics. After a complete listing of each EA, 35 households were randomly selected. All women age 15–49-years old in the selected households were eligible for the female survey. Public service delivery points (SDP) that served the EAs and up to 3 private facilities located within the EA were also selected. Any SDP with an average daily family planning client volume of three or more were selected for the client exit interview survey. At each selected SDP, family planning clients were interviewed over a two-day period. The household and female results are national and county-level representative. The final sample for the KEP1 includes completed interviews with 10,378 households, 9,477 de facto females, 945 SDPs and 3,930 family planning clients from selected SDPs. Data collection was conducted between November and December 2019.
Performance Monitoring for Action (PMA), a three-year longitudinal panel survey, builds on the previous success of PMA2020 surveys in Kenya by collecting routine data on key global indicators in family planning and reproductive health, while expanding content areas to address questions of contraceptive decision-making and autonomy, in order to better understand the determinants and consequences of unique contraceptive use and patterns of use. These indicators are measured through three-related data collection activities: household and female surveys, service delivery point (SDP) surveys, and SDP family planning client exit surveys. The household, female, and SDP panel baseline data are collected at Year 1 with follow-up data collection occurring in Year 2 and 3. By supplementing the sample and applying survey weights, cross-sectional results for households and females will be produced each year.
The PMA Kenya survey used a multi-stage cluster design with urban-rural and county as strata. The first stage of sampling was at the county level using probability proportional to size to select 11 out of 47 counties: Nairobi, Kilifi, Nandi, Nyamira, Kiambu, Bungoma, Siaya, Kericho, Kitui, Kakamega, and West Pokot. Within each county, enumeration areas (EAs) were selected proportional to size with urban-rural stratification. The national sample is composed of randomly selected EAs adjacent to the 115 PMA2020 EAs and 193 new EAs. The new EAs were drawn using the same stratified cluster design with urban-rural strata by the Kenya National Bureau of Statistics following the same sampling frame used for PMA2020. A total of 308 EAs from 11 counties are included in this survey.
A full household listing was done within each EA and field supervisors randomly selected 35 households for interview. Household listing will be updated annually and used to calculate the weights to generate cross-sectional estimates. All women age 15-49 years old in the selected households were eligible for the female survey. Public SDPs that served the EAs and up to 3 private facilities located within the EA were eligible for the SDP survey. For client exit interviews, a random convenience sample of female family planning clients were interviewed over two days from interviewed SDPs with daily average family planning client volume of three or more.
The final sample was designed to provide national and county level estimates of modern contraceptive prevalence rate (mCPR) among all women with a margin-of-error of 3% and 5% at the national and county level respectively.
Performance Monitoring for Action (PMA) uses standardized questionnaires to gather data about households, individual females, health facilities, and family planning clients that are comparable across program countries and consistent with existing national surveys. Prior to launching the survey in each country, local experts review and modify these questionnaires to ensure all questions are appropriate to each setting. Four questionnaires were used to collect data in the PMA Kenya Phase 1 (KEP1) survey: the household questionnaire, the female questionnaire, the service delivery point (SDP) questionnaire and the SDP family planning client exit interview.
The household questionnaire gathers basic information about the household, such as ownership of livestock and durable goods, as well as characteristics of the dwelling unit, including wall, floor and roof materials, water sources, and sanitation facilities. This information is used to construct a wealth index. The household questionnaire also collects information on age, sex, and marital status for all usual members of the household or visitors who slept in the household the night before.
The female questionnaire gathers specific information on education; fertility and fertility preferences; family planning access, choice and use; empowerment; quality of family planning services received; exposure to family planning messaging in the media; sexual activity; and a two-year contraceptive and reproductive calendar of events.
The service delivery point questionnaire gathers information on infrastructure and staffing, in addition to questions that focus on commodity availability and provider readiness. Facility questions cover the range of commodities and services provided in the MNH continuum, including, but not limited, to family planning services. The SDP family planning client exit interview includes questions on demographic information, family planning services received, and client satisfaction.
All PMA questionnaires are administered using Open Data Kit (ODK) software and Android smartphones. The PMA KEP1 questionnaires were translated from English into Kiswahili.
Training for the PMA Kenya Phase 1 (KEP1) survey was held in October 2018 and was led by ICRH with remote support from PMA staff at the Johns Hopkins Bloomberg School of Public Health. Resident enumerators (REs) were recruited from the selected enumeration areas included in the sample design. New REs to PMA were trained in general smartphone use, the PMA data collection platform called Open Data Kit (ODK) Collect, the art of asking questions, the ethics of survey research, and informed consent procedures. All REs were trained on the questionnaire content using a mix of lecture, paired practice, videos, and small group discussions. Throughout the training, REs and supervisors were evaluated based on their performance on several written and phone-based assessments as well as class participation. The RE trainings were conducted primarily in English, some small group sessions were conducted in Kiswahili.
Data Collection & Processing
PMA Kenya Phase 1 (KEP1) data collection was conducted between November and December 2019. PMA uses Open Data Kit (ODK) Collect, an open-source software application, to collect data using mobile phones. All the questionnaires were programmed using this software and installed onto project smartphones. The ODK questionnaires contain automatic skip-patterns and built-in response constraints to reduce data entry errors.
The ODK application enables resident enumerators (REs) and supervisors to collect and transfer survey data to a central ODK Aggregate cloud server. This instantaneous aggregation of data also allows for concurrent data processing and course corrections while still in the field. Throughout data collection, central staff at ICRH in Kenya and the data manager at the Johns Hopkins School of Public Health in the USA routinely monitored the incoming data and notified field staff of any potential errors, missing data or problems found with form submissions on the central server.
Once all data were on the server, data analysts cleaned and de-identified the data, applied survey weights, and prepared the final dataset for analysis using Stata® software.
In the PMA Kenya Phase 1 (KEP1) survey, a total of 10,780 households were selected for the survey and 10,581 households were found to be occupied at the time of the fieldwork. Of the occupied households, 10,378 (98.1%) consented to and completed a household-level interview. The response rate for the household interview was higher in the rural (98.6%) compared to the urban (97.2%) enumeration areas (EAs).
In the occupied households that completed the household interview, a total of 9,605 eligible women age 15 to 49 years old were identified. Overall, 9,477 (98.7%) of the eligible women were available, consented to and completed the interview. The female response rate was higher in the rural (98.8%) compared to the urban (98.4%) EAs. Only de facto females are included in the PMA analyses. There were 9,477 de facto females.
The final service delivery point (SDP) sample included 999 facilities, of which 945 completed the interview, for a response rate of 94.6%.
Weights were adjusted for non-response at the household and individual levels, and applied to all household and individual estimates in this report. SDP and family planning client exit interview estimates are not weighted.
International Centre for Reproductive Health Kenya (ICRHK) and The Bill & Melinda Gates Institute for Population and Reproductive Health at The Johns Hopkins Bloomberg School of Public Health. Performance Monitoring and Accountability (PMA) Survey Phase 1 Snapshot of Indicators, PMA/Kenya-P1 SOIs. 2019. Kenya and Baltimore, Maryland, USA.