Calendar Survey Methods: Verbal Behavior and Computer Applications

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

Surveys conducted in the social sciences depend on the quality of retrospective reports to provide valid estimates of social, economic, and health dynamics. Recall measures based on conventional standardized questionnaire list (Q-list) approaches have shown that response errors in retrospective reports become considerably more pronounced with events that have occurred in the more distant past. An Event History Calendar (EHC) questionnaire methodology has considerable potential toward improving the quality of retrospective reports in comparison to Q-list methods. Specifically, this project will 1) assess the impact of cognitive and communicative processes on the quality of retrospective reports that are obtained by Q-list and EHC interviewing methodologies. These processes will be examined through measures of interviewer and respondent verbal behaviors as coded from 500 audiotaped interviews of paper and pencil Q-list and EHC interviews, with particular attention focusing on the cueing strategies of interviewers and the retrieval strategies of respondents. 2) Partly based on the verbal behavior analyses, develop a 2-year reference period Computer Assisted Interviewing EHC (CAI-EHC) to measure key demographic, health, and economic variables that will be used in production interviews of the 2003 Panel Study of Income Dynamics (PSID). 3) Conduct usability and cognitive tests with CAI-EHC versions to assess and improve the interface between interviewers and the instrument, and the cognitive efficacy of the instrument. 4) Contribute to the cognitive science literature through additional discoveries concerning the structure of autobiographical memory. The dissemination of information is part of the educational goals of this project. Significant findings from the verbal behavior coding and usability tests will be presented in educational settings, at professional meetings, and through publication. The cognitive aspects of EHC interviewing will be introduced into the curricula of courses in survey methodology. Developments of the CAI-EHC interviewing methodology software will be publicly available for a reasonable charge. Ultimately, CAI-EHC interviewing may become the best state-of-the-art choice for the collection of retrospective reports in surveys, improving the ability of social scientists to model the life course and of governmental agencies to initiate sound social policies that are based on survey data.