Files in this item



application/pdf9624475.pdf (8MB)Restricted to U of Illinois
(no description provided)PDF


Title:Wording formats and biases in interview responses by adults with mental retardation
Author(s):Rubin, Stephen Scott
Doctoral Committee Chair(s):Heal, Laird W.
Department / Program:Education
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Education, Tests and Measurements
Sociology, Theory and Methods
Speech Communication
Education, Special
Abstract:The interview has been a common method for gathering information. Despite its ease of administration, the interview can produce considerable measurement error. As artifacts of the interview, specific biases are exhibited when particular question formats are used. If the interview is to be used as a scientific technique, interviewers should be aware of the types of biases produced. Previous survey research has suggested that individuals with mental retardation are particularly susceptible to giving inaccurate responses depending on the type of question asked. The purpose of the present investigation was to examine further the types of questions which produce inaccurate responding, and suggest ways to circumvent any bias. In order to study response biases systematically, an experiment was completed that disentangled question content from question format in a factorial experimental design. Three types of question formats were studied, (a) yes-no, (b) either-or, and (c) multiple-choice. Each question type was presented with or without "happy" or "sad" faces. Each format was designed to detect an acquiescence, wording order, or position bias, and was repeated across three positively regarded contents (home, specifically named friends, friends in general), and three negatively regarded contents (prison, specifically named physical pain, and physical pain in general). Based on the interview responses of 115 midwestern adults with mental retardation, four major findings emerged. First, question formats made a huge difference in the responses given by adults with mental retardation. The respondents' claims of satisfaction were determined by the way the question was worded, as well as the topic addressed. Second, only 15% of the cohort made no self-contradictions across the three couplets of yes-no questions designed to detect acquiescence and nay-saying. Third, an echolalia effect was found in 50% of the sample when they responded to either-or and multiple-choice question types. Fourth, the multiple-choice, five-faces picture format yielded the most accurate responding with the highest response rate; and the either-or format, "usually satisfied or usually dissatisfied," elicited no recency effect, suggesting that these types of questioning may circumvent wording order bias. These findings have implications for human service providers who make policy decisions based on the interview responses given by their clients. Understanding the importance of wording order and the types of biases produced from closed-ended question formats will promote better question structuring in order to maximize the credibility of information gathered from interviews.
Issue Date:1995
Rights Information:Copyright 1995 Rubin, Stephen Scott
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9624475
OCLC Identifier:(UMI)AAI9624475

This item appears in the following Collection(s)

Item Statistics