SOLVING THE UNINTENTIONAL MOBILE CHALLENGE

Web surveys are increasingly completed on small-format mobile devices (e.g., smartphones) regardless of the intention of the researchers or the design of the instruments. In this experiment, we learn whether "mobile friendly" designs improve completion rates, survey length, self-reported user experience and respondent engagement for survey takers who choose to do a survey on smartphones, while still generating the same response distributions of surveys completed on PCs and tablets. By "mobile friendly," we mean designs that use larger fonts and larger interface elements, and still size to the available screen real estate on a smartphone. We randomly assigned respondents to complete the same survey using one of eight survey presentation alternatives—six smartphone alternatives and two PC alternatives. We found that mobile friendly designs improve completion rates, survey length and user experience, but only one mobile friendly design consistently matched the response distributions from the PC surveys. We found very few differences in measure of engagement such as straightlining, speeding and satisficing in any of the alternatives.