ERIC Number: ED271437
Record Type: RIE
Publication Date: 1986-Feb
Reference Count: 0
Reducing Interviewer Effects on Health Survey Data. Executive Summary.
Fowler, Floyd J., Jr.; Mangione, Thomas W.
This large-scale field experiment examined the potential of various training and supervision programs to affect the performance of health survey interviewers and the quality of data they collect. It was found that interviewers who received less than one day of basic training generally displayed inadequate interviewing skills. A program of tape recording as part of the supervision of household interviewers was associated with more precise and less biased data if interviewers were more than minimally trained. Training and supervision were found not to be compensatory but, rather, to interact so that if either was inadequate the data were adversely affected. The results also point to the value of designing questions to minimize the need for probing, a significant source of interviewer effects, and the value of procedures to communicate the importance of accuracy to respondents. Overall, attention to a variety of aspects of interviewer management--their training and supervision, the design of questions, the procedures they are to use, and the size of their assignments--are cost-effective ways to improve the quality of survey-based estimates. (Author)
Publication Type: Reports - Descriptive
Education Level: N/A
Sponsor: National Center for Health Services Research and Development (DHEW/PHS), Rockville, MD.
Authoring Institution: National Center for Health Services Research and Health Care Technology (DHHS/PHS), Rockville, MD.