Web-based research: Methodological variables' effects on dropout and sample characteristicsWeb-based research: Methodological variables' effects on dropout and sample characteristics

Authors

Kevin M. O'Neil, John Jay College of Criminal Justice, New York, New York
Steven D. Penrod, John Jay College of Criminal Justice, New York, New York
Brian H. Bornstein, University of Nebraska-LincolnFollow

Document Type

Article

Date of this Version

January 2003

Comments

Published in Behavior Research Methods, Instruments, & Computers 2003, 35 (2), 217-226. Copyright 2003 Psychonomic Society, Inc. Used by permission.

Abstract

Previous research has made a beginning in addressing the importance of methodological differences in Web-based research. The present paper presents four studies investigating whether sample type, financial incentives, time when personal information is requested, table design, and method of obtaining informed consent influence dropout and sample characteristics (both demographics and measured attitudes). Undergraduates were less likely to drop out than nonstudents, and nonstudents offered a financial incentive were less likely to drop out than those offered no incentive. Complex tables, tables that were too wide, requests for personal information on the first page, and the imposing of additional informed consent procedures each provoked early dropout. As was expected, nonstudents and those presented with complex tables showed more measurement error and attitude differences. Asking for personal information and imposing additional consent procedures affected the demographic makeup, raising challenges to external validity.