Relevance & Research Question: Day of the week and time during the day for survey invitations are two of many possible factors that might affect participation rates in web surveys. Using two large scale experimental studies, we examine whether survey dispatch time is something survey practitioners need to take into consideration or not.

Methods & Data: In the first study, respondents were randomly assigned to one out of seven groups, one for each day of the week (n=11 200). In the second study, respondents were randomly assigned to one out of six different times of the day (n=47 279). Survey invitations were dispatched during the fall of 2014 to members of the Citizen Panel, a non-commercial web panel run by the Laboratory of Opinion Research at the University of Gothenburg.

Results: Findings indicate that initially Fridays and Saturdays, closely followed by Sundays, are the worst days to dispatch an online survey. On field day one, these days have significantly lower response rates than other days. However, after three to four days of data collection, the significant differences between dispatch days disappear, even without any follow-up reminder. Similarly, for the second study where time of the day is examined, results show that dispatch time only affects participation rates in the short run, within the first 24 hours. The results suggest that survey practitioners do not need to consider which day of week or what time to dispatch a survey, unless they really want quick answers and a very short period of field work.

Added Value: Further, and somewhat peculiar, respondents are more likely to say that they prefer to answer surveys on the day of the week when they actually received our invitation e-mail in the randomized experiment.