MULTIPLE-OUTPUT AGENCY INCENTIVES IN DATA PRODUCTION - EXPERIMENTAL-EVIDENCE

This paper interprets the production of sample surveys as a multiple o utput agency problem in which interviewers act as agents for the princ ipal investigator of the survey. The central aspect of data production that gives rise to the agency problem is that sampling error cannot b e separated out from interviewer performance. We argue that contract e ffects on data production are important because the quality-quantity i ncentives facing interviewers affect the survey, in terms of its bias and its sample size. Direct evidence pertaining to these quality-quant ity tradeoffs in the production of attitudinal surveys is examined thr ough consideration of an experiment in which piece-rate and fixed wage contracts were randomly assigned across interviewers. Using these ran domly assigned contracts, we fmd that observable quality dimensions do not respond to quantity incentives but unobservable quality dimension s do respond weakly in a negative way. (C) 1997 Elsevier Science B.V.