I wonder how we can use digital trace data to validate the accuracy of self-reported measures of online activity in conventional survey data

Speaker(s):

Conor Gaughan, University of Manchester

Abstract:

Survey data is commonly considered to be the gold standard in social and political research. However, it is well-established that reliance on self-reported measures are fraught with measurement bias, where there is a mismatch between what a respondent reports about their attitudes and behaviours, and what they actually think and do. I wonder how we can use a respondent’s digital trace data such as their social media activity or web browsing history to assess and validate the accuracy of their self-reported measures of online behaviour in conventional survey data. Specifically, I would like to discuss the ways in which we could do this methodologically, the theoretical and ethical considerations that we would need to make, and the sorts of research questions that this could help answer for us as social scientists.