Privacy Attitudes toward Mouse-Tracking Paradata Collection

Abstract

Survey participants' mouse movements provide a rich, unobtrusive source of paradata, offering insight into the response process beyond the observed answers. However, the use of mouse tracking may require participants' explicit consent for their movements to be recorded and analyzed. Thus, the question arises of how its presence affects the willingness of participants to take part in a survey at all-if prospective respondents are reluctant to complete a survey if additional measures are recorded, collecting paradata may do more harm than good. Previous research has found that other paradata collection modes reduce the willingness to participate, and that this decrease may be influenced by the specific motivation provided to participants for collecting the data. However, the effects of mouse movement collection on survey consent and participation have not been addressed so far. In a vignette experiment, we show that reported willingness to participate in a survey decreased when mouse tracking was part of the overall consent. However, a larger proportion of the sample indicated willingness to both take part and provide mouse-tracking data when these decisions were combined, compared to an independent opt-in to paradata collection, separated from the decision to complete the study. This suggests that survey practitioners may face a trade-off between maximizing their overall participation rate and maximizing the number of participants who also provide mouse-tracking data. Explaining motivations for paradata collection did not have a positive effect and, in some cases, even reduced participants' reported willingness to take part in the survey.

Figure 1.

Conceptual overview of the study design. Participants were assigned to either the joint or separate consent conditions (top and bottom row), where they assessed one or two vignettes, respectively. Respondents reported their willingness to participate on an 11-point scale, which we analyze as a numeric value. Because, in a survey, consent is a binary choice, we dichotomize the ratings above the scale midpoint and assume that participants would progress if they responded above this value. Otherwise, we assumed they would drop out of the study entirely (red) or consent to the survey but not paradata collection (yellow).

Figure 2.

Conceptual overview of our hypotheses and analyses. We compare reported willingness to participate between a survey-only consent and a joint consent in which paradata collection is included, assuming that paradata collection would reduce willingness to participate (H1). We then compare different reasons given for paradata collection, assuming that the presence of a reason, and additional specificity, will increase willingness to participate (H2). Finally, we estimate overall participation rates in the survey alone or in combination with paradata collection, by dichotomizing our continuous measure of willingness to participate.

Figure 3.

Mean willingness to participate in the survey (as a continuous scale, 0–10) separately for participants who were informed about mouse-tracking data (conditions 1–5, solid orange line) collection or not (conditions 6–10, dashed grey line), depending on the paradata collection purpose. Error bars indicate 95% confidence intervals. Please note that the conditions 6–10 had not yet been exposed to the different stated purposes for mouse tracking; they may serve as an indicator of the variability between groups under identical conditions.

Figure 4.

Overall willingness to participate in the survey, in proportion of participants indicating a willingness to participate above the scale midpoint. The different lines compare the willingness to participate in the survey in general, drawn from the separate consent condition, where mouse tracking was not included in the initial consent (light gray dashed line, as in figure 3) and consent rates to mouse-tracking data collection specifically (solid lines) for joint (orange) and separate (blue) conditions split by paradata collection purpose (X-axis). Error bars indicate 95 percent confidence intervals.