Forcing respondents of online surveys to slow down: An effective strategy?

Abstract

The goal of this paper is to investigate the effects of forcing the respondents to slow down while answering online surveys. This is done by setting up an experiment with respondents from the Netquest panel in Mexico in which respondents are randomly assigned to a control group (free to go as fast as their Internet connection allows them) or to different treatment groups. The treatment groups are forced to slow down by using a timing control (minimum time per page necessary before they can go on), by showing the different blocks on the page progressively (first the question, later the scale, later the “next” button), or by fading-in the text of the question character by character. The minimum time per page is fixed by us in half of the groups and determined by tuning in the other half. The results show that there are some positive effects on the quality, mainly from the treatment which shows progressively the different blocks (question-scale). There are also some negative effects in all treatments (more complaints) but the abandon rates are higher only in the timing control condition. Overall, forcing respondents to slow down seems to be a strategy with more positive than negative effects if it is done by presenting progressively the different elements. We believe it should be further considered for future research.