Pile sorts as a means of improving the quality of survey data: malaria illness symptoms.

Much social research data on locally endemic diseases has been gathered in from surveys. In this form, the data may not yield insight into indigenous illness concepts and the relationships among the findings that is needed for a full understanding on how people classify and respond to illnesses. This paper reports on a trial effort to use the pile sort technique to strengthen quantitative survey data on malaria. Respondents sorted common malaria symptoms, previously reported on surveys conducted in southwestern Nigeria. Computer analysis of the 22 sortings with ANTHROPAC software produced a mapping of symptoms that corresponded with three locally perceived forms of malaria illness, as confirmed through discussions with key informants. This simple procedure improved the quality of survey data and enhanced its value for use for culturally appropriate communication in patient and community health education programmes.