A meta-analysis looking at the correlation (or lack thereof) of self-report measures of media use and measures that track actual behavior. This would be useful to add to a discussion of types of measurement (contrasting the strengths and weaknesses of self-report and behavioral measures.
