Commissioned by the Dept. of Health and Social Welfare NW Province Odi Health and Welfare District (Winterveldt Survey -Findings)
Undertaken by: AMREF (Hosted by: Sapler Population Trust)
Open ended questions were used side by side with closed ended questions in order to capture both quantitative and qualitative information. To avoid unnecessary duplication, extensive use was made of secondary data and service records from various facilities in the study area.
Emphasis was put in making the approach as participatory as possible. To this extent deliberate efforts were made to involve members of the study team at every stage of study implementation process: conceptualization of the study including the refinement of the problem to be studied. development of the data collection instruments; data collection, compilation analysis and in report writing.
Simple size determination
A sample size of 240 households was determined by taking a 15% sample of all the stands in both 10 and 5 Morgan. In each of the stands, one household was selected randomly. In order to compensate for possible non-responses, a further, 40 households were included making a total of, a possible, 280 households.
The selection process
This involved selecting 15 stands using a systematic random sampling procedure. The sampling frame was the list of stands in both 10 and 5 Morgan areas of Winterveldt. Using, these stands as the reference points (centres), 16 households were selected randomly from the stands around. Only one household per stand.
This process effectively divided the area into 15 clusters, each with a representation of 16 households giving a total of 240.
Both primary and secondary data was collected. A questionnaire was used in the collection of primary data. The design of the questionnaire made provision for the collection of both qualitative and quantitative data. Secondary data collection consisted of review of past studies conducted in Winterveldt and a review of service records. more particularly from facilities from the area.
The survey team consisted of two consultants and interviewers. Interviewers were selected from Health and Welfare Departments, Non-governmental agencies operating in the area and volunteers from Winterveldt community.
Due to an emphasis on capacity building interviewers were trained over a period of 5 days. The training covered the following: Introduction to Participatory Action Research; objectives of the study; content and concepts; questionnaire design; interviewing techniques; practice interview, including actual pre-testing, and discussion on quality control.
Field work for the study was conducted in the 4th week of October, 1997. Data collection was carried out by 26 interviewers. Interviewers were divided into three groups. Each group selected a leader who was also expected to serve as a supervisor. For purposes of support and security, the interviewers paired up. At the beginning of the data collection the groups served as an effective support system more particularly in encouraging interviewers who initially were experiencing difficulties.
The Data collected was processed using the EPIINFO a data management and analysis package and analysed using SPSSPC, the statistical package for social science studies. Frequencies of various variable were generated and the relationships between various parameters assessed through cross tabulations. Graphic patterns describing the various situations and characteristics were generated using Word Perfect 6.1 for Windows.