Social Assessments and Program Evaluation with Limited Formal Data : Thinking Quantitatively, Acting Qualitatively

This note revisits the long-standing tension between qualitative and quantitative approaches to poverty analysis, with reference to social assessments and program evaluation. It presents a summary of recent work in St. Lucia and Colombia, where innovative efforts were made to integrate the guiding principles of quantitative approaches with the practice of qualitative approaches. While neither case should be seen as ideal or a substitute for a more comprehensive analysis, they nonetheless present a series of strategies for generating some meaningful and useful results in environments where, for any number of reasons, formal data is weak or absent. Such environments, of course, are all too common in low-income countries. The first case, a social assessment of poverty, comes from St. Lucia. The task manager had funds sufficient to cover key informant and focus group interviews in sixteen communities around the island. Given this small number, he elected not to work with a "random sample" as such but rather to maximize coverage on as many key variables as possible (rural/urban, access to clean water, distance to main road, level of poverty, etc). Our St Lucia-based colleagues happened to have access to a 1990 census, but it did not contain data on the full set of variables that would have enabled us to generate a final sample meeting all our criteria.