At the last ESRA conference, we presented the results of a vignette-based study in the Netherlands exploring how best to inform respondents about the capture of paradata in Web surveys. We found that any disclosure of paradata capture reduced willingness to participate.

In an attempt to improve willingness to allow the use of paradata, we carried out two further experiments in the U.S., one in a probability-based panel (n=5,550) and one in an opt-in panel (n=9,206). We varied a number of factors across the two studies, including the descriptions of paradata capture (e.g., reasons given, amount of information provided), the time of the consent request (at the beginning or end of the survey), the sensitivity of the topic, and (in one condition) directly asked for permission to use the paradata from the current survey (as opposed to using hypothetical vignettes).

Although there is considerable variation in results between the two studies, in general they reinforce the earlier finding that informing respondents of paradata capture is likely to reduce their willingness to participate in Web surveys, especially when they are asked for their consent directly. In the condition where we asked respondents at the end of the survey for permission to use their paradata, 31% declined. We will describe the results of the three studies, consider possible alternatives to the methods of informing respondents we have been using, and discuss implications for practice.