Evaluating alternative techniques of questioning mentally retarded persons.

In an examination of methodological issues involved in interviewing retarded persons, alternatively worded or structured questions were embedded in interviews with three samples. Questioning techniques were evaluated according to the extent to which (a) interviewees could provide answers, (b) their answers agreed with parallel responses given by attendants or parents, and (c) answers were free of systematic response bias. Open-ended questions were found to be unanswerable by many persons, and supplementing them with clarifying examples and probes for additional information only exacerbated response bias. By comparison, yes-no checklists enhanced responsiveness but introduced serious acquiescence bias, whereas multiple choice questions, particularly with pictures, yielded valid answers from high proportions of interviewees. Implications for question design were discussed.