Conclusion
In concluding remarks John Casterline outlined a framework for a future model survey instrument based on the workshop discussion. The framework consisted of a survey design similar to a DHS survey but with a refined core questionnaire. An essential feature would be a set of follow-up data collection activities, such as panel and longitudinal activities. The follow-up activities would gather detailed information from special populations or subgroups regarding specific topics or sensitive issues, such as STD/HIV. The timing of follow-up activities could vary between shortly after the initial survey to as much as 3 years later. A firm commitment of funding for targeting, design, and analysis of the data would be required. Proposed sample sizes for the initial survey would be larger than in previous surveys, and the frequency of the data collection of histories would not exceed 5 years, with possible intermediate surveys to assess contraceptive prevalence rates and other standard indicators as needed.
Further research would be needed to ensure that the components of the model adequately suffice for evaluating program outcomes and assessing program impact. Challenges of controlling for certain influential variables would need to be examined. The model framework would impose greater organizational challenges due to the complexities involved with conducting separate interviews with a subset of respondents. Casterline noted that although the organizational capacity to carry out these approaches exists at the international level, it may not within a particular country.
Several other themes were raised throughout the workshop:
- The demand for qualitative techniques has been rising as the need to collect information on behaviors and attitudes increases. Currently, there is limited agreement in the research community as to the best qualitative methods and techniques in a given situation. Further research was suggested, including a systematic assessment of the role and contribution of qualitative data collection and analysis. Participants proposed a separate meeting to specific address qualitative methods and issues.
- The value of data relating to the quality of services was acknowledged throughout the workshop. It was suggested that further research is needed to examine a combination of situation analysis and service availability modules with longitudinal designs at the areal level, if not at the individual household level.
- Many participants noted that indicators of program outcomes should reflect the needs of program managers in developing countries, including their interest in having data for local, provincial, or regional populations.
- Additional research was suggested on cost and expenditure data for monitoring and evaluating purposes. Increased emphasis on host countries to develop and maintain expenditure data was suggested.
- Suggestions were made for improved capacity building in developing countries. Improved training and fellowship programs were proposed, as well as greater use of existing expertise and facilities within countries.
Throughout the workshop there was much discussion about the invaluable contribution that household survey programs—the World Fertility Surveys, the Contraceptive Prevalence Surveys, and the Demographic and Health Surveys—have made in establishing parameters and baseline estimates for basic demographic information in developing countries. Household surveys have provided quality survey designs that have been a standard source for documenting key indicators. As programs strive to address the needs of their targeted populations, more information is desired about people's attitudes, perceptions, and behaviors, as well as the quality of services. Influential factors that motivate individuals to use or not use the services provided are intertwined in family and community norms and cultures and socioeconomic factors, as well as perceptions and realities of the availability, accessibility, and quality of services. Measurement of all these factors has become an ever-increasing challenge.