Outsider, Locals, and Insiders: Results from a Methodological Experiment on Data Collection

Alexander Weinreb, University of Texas at Austin
Mariano Sana, Vanderbilt University
Guy Stecklov, Hebrew University of Jerusalem

It has been assumed since the early days of survey research that in order to collect unbiased and valid data, interviewers must have no prior social relationship with survey respondents. In an earlier paper, the theoretical underpinnings of this "stranger-interviewer norm" were found to be at odds with a substantial body of social theory, as were empirical data collected in rural areas of Kenya. In this paper we present results from a methodological experiment that we fielded in the Dominican Republic in 2010, the first-ever evaluation of the stranger-interviewer norm in which interviewers' degree of acquaintance with respondents was controlled by the researchers. Not surprisingly, results are mixed. Differences in the validity of the data by type of interviewer-respondent relationship appears to depend on the nature of the questions asked. We believe that the study’s findings will contribute to substantial improvements in data collection, especially in less developed countries.