How Interviewers' Conversational Flexibility Affects the Accuracy of Survey Data

摘要

Standardized survey interviewing is widely advocated in order to reduce interviewer-related error (e.g. Fowler & Mangione, 1990). But Suchman and Jordan (1990, 1991) argue that standardized wording may decrease response accuracy because it prevents the conversational flexibility that respondents need in order to understand what the survey designers have intended. We evaluated these competing positions - standardized versus flexible interviewing approaches -- in a laboratory experiment. Professional interviewers asked the respondents questions from three large, government surveys, using either standardized or flexible interviewing techniques. The respondents provided answers on the basis of fictional descriptions enabling the experimenters to measure response accuracy. Accuracy and certain characteristics of the interaction were compared under standardized and flexible interviewing conditions. The two interviewing techniques led to virtually perfect accuracy when the concepts in the questions clearly mapped onto the fictional situations. When the mapping was less clear, flexible interviewing increased accuracy substantially, from 27% to 87%. We discuss theoretical and practical implications and propose which circumstances justify the use of either interviewing technique.