Abstract
This paper reports two experiments in which the prominence of university sponsorship on Web surveys was systematically manipulated, and its effects on dropout and item non-response were observed. In Study 1, 498 participants were randomised to online surveys with either high or low university sponsorship. Overall, 13.9 percent of participants commenced, but did not complete the surveys, and there was no difference between the proportions of participants dropping out of each condition. However, counter to our predictions, participants in the high sponsorship condition displayed significantly higher item non-response. In Study 2 (N = 159), which addressed a rival explanation for the findings in Study 1, the overall dropout rate was 23.9 percent and sponsorship prominence had no effect on either outcome variable. Overall, these findings suggest that hosting information pages on university Web sites, placing university logos on survey pages, and including the name of the university in survey URLs do not reliably impact on dropout or item non-response. Although it may seem disappointing that enhancing sponsor visibility is not sufficient to reduce dropout and item non-response, researchers without ready access to university Web servers or branding will appreciate these findings, as they indicate that minimally visible sponsorship does not necessarily compromise data quality.
Highlights
Item non-response occurs when participants do not answer survey questions they have been exposed to, and are eligible to complete (Bosnjak and Tuten, 2001)
Because of the undesirable consequences associated with the forced-response feature, we need to consider other techniques and strategies that can be employed to reduce item non-response and dropout. One such potential strategy involves enhancing the prominence of survey sponsorship, which is the independent variable (IV) in the current research
When Boulianne, et al (2011) manipulated Web survey sponsor prominence, such that members of a university community were invited to complete a survey about transportation issues by either the university’s transportation department or its survey centre, it had no impact on response rate
Summary
Item non-response occurs when participants do not answer survey questions they have been exposed to, and are eligible to complete (Bosnjak and Tuten, 2001). When researchers have compared Web to mail-surveys in both experimental (Kwak and Radler, 2002; Bech and Kristensen, 2009; Messer, et al, 2012) and quasi-experimental designs (Haraldsen, et al, 2002; Denscombe, 2006; Lorenc, 2010; Israel and Lamm, 2012; Lesser, et al, 2012), they have tended to find less item non-response in the Web mode This finding is not unequivocal, with Millar and Dillman (2012) and Wolfe, et al (2009) both reporting no differences between modes.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
