Notice: the WebSM website has not been updated since the beginning of 2018.

Web Survey Bibliography

Title Mobile Device Use in Web Surveys Among College Students: Predictors and Consequences for Data Quality
Year 2016
Access date 04.06.2016
Increasing numbers of surveys are completed on mobile devices like Smartphones, and questions have been raised about the quality of mobile survey data versus data collected by laptop / PC. This paper examines predictors of mobile device useand the consequences of this choice for data quality in web surveys among undergraduates at the University of Pittsburgh (U.S.). The data come from an ongoing panel survey that has involved changes in the survey protocol over time and experimental manipulations related to device choice to explore data quality. This analysis incorporates combined data from 2013 (n=1,626), 2014 (n=1,431), 2015 (n=3,077), and 2016 (anticipated n~3,100; total n~ 9,200). Multivariate models with robust standard errors (for repeated individual observations) are tested for the following outcomes: choice of mobile vs. non-mobile device; and the following data quality indicators: (1) survey break-offs; (2) item missingdata; (3) survey completion time; (4) provision ofopen-ended text responses; (5) length of text responses; and (6) substantive survey outcomes (i.e., “mode effects”). Predictors include survey year, class (freshman, sophomore, junior), sex, race, SAT score, school, and survey history (# prior invitations / completions, prior devices used). The key predictor for the data quality models is use of mobile vs. non-mobile device. Two survey design factors that have varied over time will also be included as predictors: (1) survey format - mobile (no grids) vs. non-mobile (grids); (2) explicit mention in email invitation that the survey can be taken on mobile or non-mobile devices (vs. no mention). Main effects and interaction effects of predictor variables by type of device used will be tested. The goal of the paper is to use a large data set with multiple covariates to provide insights into the predictors and effects of mobile device use in web surveys.
Year of publication2016
Bibliographic typeConferences, workshops, tutorials, presentations

Web survey bibliography (4086)