Designed for the device

Surveys are used across market research, political research, public policy, public health, and many other disciplines. Many survey forms are now mobile-responsive, but resizing correctly does not make an instrument ideal for mobile users. Established survey platforms still tend to preserve the structure of a paper questionnaire on a smaller screen.

If mobile devices are the primary collection channel, we can design around the full capability of the device: the screen, touch gestures, movement, audio, cameras, and immediate feedback. This prototype explores one small possibility. It turns a cumbersome pairwise-ranking task into a mobile-first interaction built around the familiar swipe actions used in other applications.

This experiment replaces a long grid with a choice between two options: a thumb flick across the full screen, with the selected option taking over the frame. Neither side is inherently better, and this is not a pass or fail task. A down-swipe option allows “too close to call”, so people are not forced into a false choice.

The prototype records the selected answer, decision time, and swipe path as separate signals. Timing and movement may eventually provide useful information about the interaction, particularly as automated responses become harder to identify (Westwood, 2025), but they are not validated measures of preference strength or proof that a respondent is human. The point is to collect the signals so those possibilities can be tested. Response sides and colours are randomised each round, and the full battery can be reversed to examine position effects.