Inference from Biased Polls

Poll respondents often attempt to present a positive image by overstating virtuous behaviors. We examine whether people account for this “socially desirable responding” (SDR) when drawing inferences from poll data. In an experiment, we incentivize “predictors” to guess others' choice behaviors across eight actions with varying social desirability. To aid guessing, predictors observe random subsamples of (i) incentivized choices or (ii) hypothetical claims from polls. Predictors show reasonable skepticism towards hypothetical claims, which exhibit predictable SDR. However, their skepticism is not tailored to the direction or magnitude of SDR. This under-correction occurs even though subjects' explicit responses can predict SDR