Response Rates and Mode Preferences in Web-Mail Mixed-Mode Surveys: A Meta-Analysis

This meta-analysis examined (1) mode preference (i.e., response rate difference) between mail and Web survey modes in 43 mixed-mode surveys, and (2) the overall response rate in 52 Web-mail mixed-mode survey study results. Six study features were coded as potential factors that might contribute to the variation in the outcome variables (mode preference, and overall response rate) across the studies. Random-effect model analysis (forest plot) for mode preference indicates that, in general, mail survey mode was preferred over Web survey mode, but there was considerable variation in mode preference across these studies. All six study features were shown to contribute to the variation of mode preference across the studies. General linear model analysis for the outcome variable of overall response rate across these mixed-mode survey studies reveals that the six study features were statistically associated with the variation in the overall response rates across these mixed-mode survey studies. These findings are discussed, and some limitations of this study are noted.