Teen sex from evidence to policy: The accuracy of the media's communication of quantitative methods

Health studies vary in appropriateness of their methodology to address a target policy question. The media may not report methodological issues accurately to the public, so studies with more appropriate statistical methods may not prevail in the policy arena. This paper compares the dissemination of the author's matched sampling study of sexual abstinence pledge efficacy with two regression-based papers of the same data. The study analyzes media reports obtained from Lexis-Nexis searches, coded for content, and evaluates which media accurately communicate methodological differences. We find that the media mentions statistical methods, allowing the comparison of statistical methods in the policy arena, and review communication strategies for authors to use in media interviews that may encourage media to report important methodological details and differentiate between studies.