Don't Blame Pop Culture For Teen Misbehavior

A barrage of studies now proclaims that a single source causes nearly all bad behaviors among young people: "the media." Researchers announce that teenagers' smoking, drinking, pregnancy, obesity, poor academic achievement, violence, and other troubles overwhelmingly derive from television, movies, video games, music, and internet use.

Studies announcing that new forms of media damage kids now claim previously feared influences are unimportant. A 2008 study declaring violent Internet sites the most important media factor promoting teen violence concluded that television, music, movies, and video games really aren't big deals. A new report branding sexy TV shows a major cause of teenage pregnancy also finds, contrary to previous beliefs, that viewing other TV shows predicts less pregnancy.

Old notions that cigarette ads incite teen smoking have yielded to newer studies blaming television viewing, or simply having a TV in the bedroom. Now, the latest study fingers smoking in movies as the pivotal instigator of teen smoking, downplaying television and other previously suspected influences.

Researchers disagree about what forms of media most corrupt youth, but they agree that "media are increasingly pervasive in the lives of children and adolescents" and "newer forms of media seem to be especially concerning." Indeed, just 15 years ago, the types of media researchers now find so devastating didn't exist or were less graphic. Explicit rap music, ultra-violent first-person video games, Internet sites, increased smoking in movies, sexual explicitness in prime time, advertisements employing sophisticated suggestion, and brutally realistic movie mayhem intensified in the last decade and a half.

And this leads to a bizarre contradiction: The more objectionable media available to youth, the bigger the declines in their rates of violence, pregnancy, and other risky behaviors.

The FBI reports that from 1990 through 2007, rates of serious violent and property crime among youths under age 18 plunged by 49%, including unprecedented declines in murder (down 66%), rape (down 52%), robbery (down 32%), and serious assault (down 28%). The National Crime Victimization Survey finds even larger declines in teens' violent victimizations. The Centers for Disease Control reports massive declines in teenagers' rates of giving birth (down 30% since 1990), pregnancies (down 40%), gun deaths (down 55%), suicide (down 30%), and violent deaths (down 37%). Large-scale surveys such as Monitoring the Future and The American Freshman find students today reporting higher levels of happiness, optimism, leadership interest, and volunteerism and lower rates of smoking, drinking, depression, dropout, and materialism. Youngest teens show the biggest improvements.

How do we reconcile surveys claiming young people increasingly suffer from media-inflicted damage with solid statistics showing massive improvements in young people's real-life attitudes and behaviors? Decades of research warn that surveys and experiments are easily biased tools vulnerable to unreliable results. Subjects can be powerfully influenced to confirm the researchers' beliefs even when researchers conduct studies ethically. That's why researchers usually produce results consistent with what they and their funders already believe. In a notable example, respected media-violence scientists claimed Sesame Street and Mr. Rogers' Neighborhood provoked aggressive behaviors in children.

Recently, researchers reported that most published studies find the media are "crucial" contributors to "negative health results for children." The study, not incidentally, was funded by Common Sense Media, a lobbying group whose mission statement declares that "media and entertainment profoundly impact the social, emotional, and physical development of our nation's children." Its researchers produced favorable results by downplaying a key fact: journal editors and reviewers notoriously favor studies that find significant results and rarely publish studies that find no effects.