The politics of local research production: surveying in a context of ethnic competition

Using observations from rural Malawi in 2010, this paper describes and studies two significant challenges when surveying in a context of ethnic competition. First, I provide an empirical account of the local reality of ethnoregional competition as demonstrated in local research assistant selection. Negative local perceptions about how many research assistants came from the field site generated some study refusals. Second, I use subsequent survey data to demonstrate nuanced findings on ethnicity-of-interviewer effects: overall item non-response patterns are not associated with interviewer–respondent ethnic matching; refusal to respond to questions about sexual behavior was higher in co-ethnic interview settings; and respondents inconsistent about reporting their ethnic identity often reported their ethnicity to match that of the interviewer. The findings are critical for individual-level research conducted in divided societies, especially but not exclusively in situations where scholars are studying the causes and effects of social division.