Third Party Presence and the Political Salience of Ethnicity in Survey Data

In developing countries, in-person surveys are frequently conducted in the presence of respondents’ family, friends, or neighbors. What effect, if any, does their presence have on survey responses? We use data from an original survey in Karachi, Pakistan, to examine how such presence affects responses to questions related to ethnic identity and ethnic politics. We find that respondents are systematically more likely to express greater support for ethnic politics and greater feelings of perceived ethnic discrimination in the presence of known others. We present suggestive evidence that this finding is explained by social desirability bias due to a norm of in-group solidarity. Our findings have important implications for the study of ethnic politics and for survey researchers working in contexts where respondent privacy is rarely guaranteed.