APPLIED COGNITIVE PSYCHOLOGY Appl. Cognit. Psychol. 18: 169–188 (2004) Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/acp.955 Misunderstanding Standardized Language in Research Interviews MICHAEL F. SCHOBER 1 *, FREDERICK G. CONRAD 2 and SCOTT S. FRICKER 3 1 New School for Social Research, USA 2 University of Michigan, USA 3 Bureau of Labor Statistics, USA SUMMARY Leaving the interpretation of words up to participants in standardized survey interviews, aptitude tests, and experiment instructions can lead to unintended interpretation; more collaborative interviewing methods can promote uniform understanding. In two laboratory studies (a factorial experiment and a more naturalistic investigation), respondents interpreted ordinary survey concepts like ‘household furniture’ and ‘living in a house’ quite differently than intended in strictly standardized interviews, when the interpretation was left entirely up to them. Comprehension was more accurate when interviewers responded to requests for clarification with non-standardized paraphrased definitions, and most accurate when interviewers also provided clarification whenever they suspected respondents needed it. Copyright # 2004 John Wiley & Sons, Ltd. Most people probably assume that competent listeners or readers exposed to the same words or phrases will all end up with more or less the same interpretations. How else would people reach common understandings of, for example, newspaper articles, adver- tisements, or the items in a drop-down menu on a computer? Psychologists implicitly subscribe to this view when they assess attitudes and opinions in questionnaires, when they administer aptitude tests and clinical assessments, and when they instruct experimental participants. In order to reduce bias and promote comparability of responses, they standardize the materials: they expose all respondents and subjects to exactly the same words. They also require participants to interpret those words for themselves, so that researchers will not influence responses, as in these instructions for administering the Information subtest of the WAIS-III intelligence test: Read each question aloud exactly as it is written. If the response to a question is incomplete or unclear ... it is permissible to say: ‘Explain what you mean’ or ‘Tell me more about it’. However, do not ask leading questions or spell the words. Do not alter the wording of any question. (Wechsler, 1997, p. 136). Copyright # 2004 John Wiley & Sons, Ltd.  Correspondence to: Michael F. Schober, Department of Psychology F330, New School University, 65 Fifth Avenue, New York, NY 10003, USA. E-mail: schober@newschool.edu Contract/grant sponsor: National Science Foundation; contract/grant numbers: SBR-9730140 and ISS-0081550. Contract/grant sponsor: Bureau of Labor Statistics.