Self-reporting measuring how socially popular respondents think an action is vs. their actual actions is likely a more significant issue (we know this from, e.g., comparisons of actual election results to pre- and post-retirement polls; immediately pre-existing polls tend to be very close to the actual results, post election polls of how people say they voted consistently show a much better result for the declared winner than the actual election results.)

Now, we don't know how big of an effect perceived popularity has on other self-reports of past actions, but we have plenty of reason to believe it plays a significant role.