Using online surveys for health behavior research with college students: How does recruitment method impact study results?

Online surveys are a popular way to collect data from college students for health behavior research. However, there remain gaps in understanding the usefulness, challenges, and validity threats of different online survey recruitment methods. The purpose of this paper is to compare the experiences and sample characteristics of two cross-sectional samples of undergraduate college students who took the same online survey during a global pandemic but were recruited using different methods: an online panel and online community recruitment. Students took the survey ‘COVID-19 and Digital Health Literacy in University Students’ during the summer of 2020. Sample characteristics were compared using chi-square tests of independence, z-tests of proportions, and t-tests of independent sample means. There were distinct drawbacks and benefits to each recruitment modality. Findings suggest online community recruitment is less costly, but more time intensive, and could produce a more homogenous sample. The panel-based sample was higher cost, but less time intensive, and produced a much more diverse sample of college students. It is important to consider how recruitment modalities for online surveys can impact required resources, influence study limitations, and impact the results and conclusions of health behavior research.