UWSC staff are presenting a paper on straightlining in a mail-web mixed-mode survey at the annual meeting of the Midwest Association for Public Opinion Research (MAPOR) in Chicago, Illinois, November 22-23: “Straightlining: Overview of Measurement, Comparison of Indicators, and Effects in a Web-Mail Mixed-Mode Survey” by Yujin Kim, Jennifer Dykema, John Stevenson, Penny Black, and Paul Moberg (staff names appear in bold).
According to the abstract: “Straightlining occurs when survey respondents give identical (or nearly identical) answers to items in a battery of questions using the same response scale, which may lead to reduced data quality. Despite its potential importance, the small body of research examining straightlining does not use a standard measurement technique. Further, while mixed-mode studies increase in prevalence, few examine straightlining in mail versus web surveys. Our paper has the following goals: (1) compare methods for detecting straightlining; and (2) examine effects of mode, education, and incentives on straightlining in a sequential web-mail mixed-mode survey. Data are from a 2010 survey of alcohol beliefs and consumption in which 7,200 young adults were sampled from driver’s license records in Wisconsin and randomly assigned to web-mail or mail-web treatments. Respondents initially received either a postal letter with URL (web-mail) or a mailed survey (mail-web) along with a pre-incentive ($1 versus $2). For those not responding to the initial mode, mode was switched and respondents received contacts in opposing mode. We identify five distinct straightlining measures: 1) proportion of respondents using the same category in a battery (Herzog & Bachman, 1981); 2) standard deviation or variance around a battery (Krosnick & Alwin, 1988); 3) rho (Krosnick & Alwin, 1988); 4) mean of the root of absolute differences between pairs of items in a battery (Mulligan et al., 2001); and 5) maximum number of identical categories used (Holbrook et al., 2003). While the first two measures provide useful descriptive summaries, we argue they are not as refined as other measures, which are highly correlated. Preliminary results indicate that education is positively related to straightlining. However, while we find no significant effects of mode of assignment or incentives on straightlining, some evidence indicates that the mode in which the respondent completed the survey is associated with straightlining.”
UWSC is proud to support MAPOR by contributing as a Bronze Level Sponsor of the Conference, and Associate Director, John Stevenson, and Survey Methodologist, Jen Dykema, serve on MAPOR’s Executive Council.
To visit the Midwest Association for Public Opinion Research (MAPOR) click here.