With respondents increasingly completing web surveys on tablet computers and smartphones, several studies have examined the potential effects of the switch from PCs to mobile devices. The studies have looked at a range of outcomes, including completion rates, breakoffs, and item nonresponse. We carried out a field experiment that compared responses obtained by smartphones, tablets, and laptop computers, focusing on the potential effects of the different devices on measurement errors. We examined whether the differences across devices in screen size (and the related need to scroll to see the entire question or the full set of response options) might moderate the effects of response order, affect the strategy respondents used to decide which of two options was preferable, change the effect of question context, or influence the use of definitions. Our experiments were based on the principle of visual prominence—the idea that respondents are more likely to notice and consider information that is easy to see. The experiments were deliberately designed to maximize the impact of screen size on the results, since the screen size would affect the visual prominence of key information. However, like many of the prior studies examining mobile devices, although response order, context, and evaluation strategy affected the answers respondents gave, few device effects emerged.