In a previous post, I summarized a half-dozen recent studies showing that all over the world—in the U.S., Denmark, Japan, China, Hong Kong, and the Czech Republic—as pornography became more easily available because of legislative changes or the arrival of the Internet, rates of sexual assault dropped consistently and significantly. As a result, I concluded: More porn, less rape.

My conclusion appalled one reader whose critique included citations of studies she claimed demonstrated a strong connection between exposure to porn and sexual assault. Here are summaries of those studies' findings:

• Canadian researchers interviewed 118 imprisoned rapists to identify factors that had caused their crimes. Three stood out: their own childhood sexual victimization, recurring fantasies of committing sexual assault, and consumption of pornography.

• At a crisis center, Philadelphia sociologists interviewed 100 survivors of domestic violence that often included sexual assault. Many said their violent husband/boyfriend regularly viewed porn.

• Researchers in Singapore interviewed 62 convicted rapists. Most had consumed pornography within the six months before their crimes.

• UCLA researchers reviewed the literature on sexual violence and pornography and found a “reliable association” between the two.

My critic’s studies show: More porn, more rape. Meanwhile, the studies I cited in my previous post show the opposite: More porn, less rape.

How can we make sense of such contradictory findings? By looking more deeply into the studies, specifically into their methodologies.

Prospective vs. Retrospective

Research is expensive, and scientists always struggle with inadequate funding. But let’s assume money were no object, and researchers wanted to design the best study imaginable to discover the link, if any, between porn and sexual assault. What would it look like?

It would track all men on Earth continually from birth to death, determine how much pornography they consumed, track their every move, including any sexual violence they committed, and see if the two were related. Of course, such a study would be impossible. Follow billions of men for every moment of their entire lives? Expensive beyond belief and not even the most intrusive police state could do this.

Fortunately, we don’t need to follow billions of men for life. Researchers can approximate what happens in real life by using smaller—but carefully chosen—population samples.

Perhaps you’ve heard of the Framingham Heart Study. In 1948, researchers recruited 5,000 residents of Framingham, Massachusetts, into what quickly became a landmark study of heart disease, ongoing now for almost 70 years. The study began with a survey of participants’ diets, lifestyles, medications, etc. Participants have been re-surveyed regularly ever since. This approach, a “prospective trial” that moves forward in time, comes close to how people really live.

Compared with studying everyone on earth, prospective trials are much less costly. Nonetheless, they still cost a fortune. It’s very expensive to track thousands of people for many years. As a result, not many prospective trials get funded.

The alternative is to look back in time with “retrospective” studies. The researchers begin with select individuals, in our case, rapists, and explore their backgrounds to see if they share anything in common, for example, exposure to pornography.

Retrospective studies start with known rapists, so the researchers need not wait years to pinpoint the small proportion of men who commit sexual assault. As a result, retrospective trials are much cheaper and produce results much faster. Most social science research is retrospective.

Unfortunately, retrospective trials have major—and unavoidable—flaws:

• They usually involve only small numbers of subjects, nowhere near whole populations.

• They depend on memory, and over time memory fades. Memory also plays tricks, especially when researchers ask about stigmatized activities like consuming porn. People generally underestimate their use of things like cigarettes, alcohol, and porn. Statisticians have mathematical ways to minimize recall errors. But in retrospective trials, recall errors cannot be eliminated.

• Finally, retrospective trials explore only the factors the researchers think might play a role in the phenomenon they’re studying. However, it’s possible that the researchers ignored other important factors.

Retrospective trials are certainly valuable, but scientists don’t consider them definitive. For real answers, researchers rely on prospective trials.

Sometimes, retrospective trials point one way—and then prospective research points another. During the 1980s a dozen retrospective studies unanimously agreed that post-menopausal estrogen—hormone replacement therapy or HRT—reduced women’s risk of heart disease without increasing their risk of breast cancer. But questions remained, so in 1991, the National Institutes of Health enrolled 160,000 post-menopausal women in a prospective study, the Women’s Health Initiative (WHI). Fifteen years later, the WHI showed that, contrary to all the retrospective trials, HRT actually increased risk of both heart disease and breast cancer. The prospective WHI trumped all the retrospective trials, and after its results were announced, HRT prescriptions plummeted.  

Why did so many retrospective studies show HRT beneficial when, in fact, it was harmful? The women in the retrospective trials shared an attribute the researchers didn’t account for: They were healthier than average to begin with. Their rates of heart disease and breast cancer were disproportionately low. HRT boosted their risk of heart disease and cancer, but because they started out so healthy, their rates of those conditions remained below average, leading to the spurious conclusion that HRT was a boon and not the bane it actually was.

Bottom line: Be cautious about drawing conclusions from retrospective trials.

Which Studies Are Most Credible?

All the studies my critic cited are retrospective. Now, when dozens of such studies show a significant association between exposure to porn and violence against women, it’s tempting to conclude that the former predisposes men to the latter. But retrospective studies are small. They’re prone to recall errors. And in this case, they’re plagued by a key researcher error, the assumption that viewing porn is unusual, that most men don’t do it, only bad guys. In fact, virtually ALL men have been exposed to pornography but only a tiny fraction become rapists.

Many women find it hard to believe that all men have viewed porn. They insist: Not my grandfather-father-uncle-brother-boyfriend-husband-son. Sorry, ladies, but yes, it’s almost certain that all the men in your life have viewed porn.

• Google “porn.” The search produces hundreds of millions of web pages, more than one for every man, woman, and child in the U.S.

• Google Alexa.com, the site that monitors web traffic. After Google, Facebook, YouTube, Amazon, and a few others, guess what follows close behind. Very few women visit porn sites, yet with only around half the potential web audience—pretty much just men—XXX-rated sites still rank among the world’s most visited.

• Finally, in 2009, for a study on porn and violence against women, University of Montreal researchers advertised to find men who had never been exposed to X-rated media. They couldn’t find any. They were surprised and tried harder, advertising more extensively. They still couldn’t find any.

So, yes, rapists have viewed porn. But every other man has, too. If 1,000 men view porn, and 999 immediately commit rape, then we can say with confidence that the one spurs the other. But if 1,000 men view porn, and 10 commit rape, what can we say? Not much.

Meanwhile, the studies I cited in my previous post were all prospective, and therefore, inherently more credible. They used huge natural experiments—rape rates before and after porn became easily available—in several countries with male populations in the hundreds of millions. Those enormous numbers add to the studies’ credibility. So does their consistency. In every prospective trial, as the availability of porn increased, rates of sexual assault declined.

Prospective trials are more believable than retrospective studies. All the prospective trials show that as porn became more easily available, rates of sexual assault declined. I stand by my original conclusion: More porn, less rape.

References:

The retrospective studies my critic cited:

Beauregard, E., et al. “An Exploration of Developmental Factors Related to Deviant Sexual Preferences Among Adult Rapists,” Sexual Abuse: Journal of Research and Treatment (2004). 16:151.

Bergen, R.K. and K.A. Bogle. “Exploring the Connection Between Pornography and Sexual Violence,” Violence and Victims (2000) 15:227.

Gwee, K.P. et al. “The Sexual Profile of Rapists in Singapore,” Medicine, Science and Law (2002) 42:51.

Malamuth, N.M. et al. “Pornography and Sexual Aggression: Are There Reliable Effects and Can We Understand Them?” Annual Review of Sex Research (2000) 11:26.

The prospective studies I cited:

Diamond, M. et al. “Pornography and Sex Crimes in the Czech Republic,” Archives of Sexual Behavior (2011) 40:1037  

Diamond, M. “The Effects of Pornography: An International Perspective,” in Pornography 101: Eroticism, Sexuality, and the First Amendment, edited by J. Elias et al. Prometheus Press, Amherst, NY, 1999.

Diamond, M. and A. Uchiyama. “Pornography, Rape, and Sex Crimes in Japan,” International Journal of Law and Psychiatry (1999) 22:1.

Goldstein, M. et al. “Experience with Pornography: Rapists, Pedophiles, Homosexuals, Transsexuals, and Controls,” Archives of Sexual Behavior (1991) 1:1.

Kutchinsky, B. Pornography and Rape: Theory and Practice? Evidence from crime Data in Four Countries, Where Pornography is Easily Available,” International Journal of Law and Psychiatry (1991) 14:47.

Kutchinsky, B. “The Effect of Easy Availability of Pornography on the Incidence of Sex Crimes: The Danish Experience,” Journal of Social Issues (1973) 29:163.

Poipovic, M. “Pornography Use and Closeness with Others in Men,” Archives of Sexual Behavior (2011) 40:449

The University of Montreal study:

http://www.sciencedaily.com/releases/2009/12/091201111202.htm

You are reading

All About Sex

How to Discuss Sex With Your Children

Child and teen sex education resources are deeply flawed.

Are Sexual Lubricants Safe?

New findings: Lubricants may damage cells and increase sexual infection risk

The New Erotic Frontier: Sex in Nursing Homes

Until recently, nursing homes frowned on sex. Now, increasingly, they smile.