“…3 Abstracts were inspected, and reports were excluded if the research (a) was not written in English or German; (b) contained no empirical study or no reporting of original data (e.g., interviews, theoretical considerations on risk taking or reanalysis of previous data; Aitken, Eadie, Leathar, McNeill, & Scott, 1988;Garfield, Chung, & Rathouz, 2003;Kunkel, Cope, & Biely, 1999;Roberts & Foehr, 2008); (d) had no relevant measure of exposure to risk-glorifying media content (e.g., Buchanan & Lev, 1989;Eaton et al, 2006;Unger, Johnson, & Rohrbach, 1995); (e) was conducted with a clinical participant sample (such as people who had previously attempted suicide; Berman, 1988); (f) did not have risk-glorifying media content as its stimulus (e.g., antitobacco advertisements; Lee & Ferguson, 2002;Loukas, Murphy, & Gottlieb, 2008;Moyer-Guse, 2008;Tyler & Cook, 1984); (g) did not principally focus on risk-glorifying media content (such as those on media violence and aggression); (h) did not include riskglorifying pictures, films, music, video games, advertisements, or TV programs (e.g., Griffiths, 2005;Lasorsa & Shoemaker, 1988); (i) did not focus on one of the classic risk-taking domains, such as risky driving, smoking, drug consumption, alcohol consumption, gambling, delinquency, or sexual risk taking (thus excluding those on aggression, suicide, eating behavior, body perception, or skipping school, e.g., Ata, Ludden, & Lally, 2007;Robinson, 1999;Scheel & Westefeld, 1999;Vandewater, Shim, & Caplovitz;; (k) included neuropsychological methods (Brewer-Smyth, 2006;Brewer-Smyth, Burgess, & Shults, 2004); (l) did not provide sufficient statistical data to compute an effect size (or where we received no response from the researchers who were contacted for additional statistical data); (m) did not use risk-glorifying but rather risk-negative media depictions (e.g., antitobacco advertisement); and (n) merely assessed the relationship between media exposure and risk taking with path analytical procedures, because no clearly comparable effect size indicators were provided (e.g., Tickle, Hull, Sargent, Dalton, & Heatherton, 2006;Wyllie, Zhang, & Casswell, 1998).…”