More on Porn and Rape

Commenter Ballgame asked me why I say the porn-rape study is flawed. Here’s why:

1. The control variables leave out a lot to be desired. In particular, Kendall controls for poverty and income, even though the strongest correlate of at least murder is inequality. Further, he doesn’t control for urbanization, only for population density; some largely empty states, like Colorado, can nonetheless have high urbanization rates.

2. Kendall uses reported rape as a proxy for rape. But the reporting rate changes over time and over place, and at any rate the Uniform Crime Report has a number of rapes far higher than the reporting rate times the number of survey rapes.

3. The National Crime Victimization Survey’s rape rate has a margin of error of about 25% – that is, there are 110,000 (+/-) 30,000 rapes every year in the US. A reduction of 2% or 4% or even 7% in rape is statistically insignificant.

4. The US rape rate started to crash in 2000, after stagnating in the late 1990s. If pornography reduced rape, we’d see a sudden upsurge in American access to pornography around 1999 or 2000; however, we don’t.

4 Responses to More on Porn and Rape

  1. ballgame says:

    Alon: Thanks for your reply. Some commenter* posted this link to Falstaff who seems to make a pretty cogent point about multicollinearity and the reversed correlations between internet access and rape, and PC ownership and rape, that one of Kendall’s charts suggests. You’re likely much better equipped to quickly assess Falstaff’s mathematical reasoning than I.

    As to your points:

    1. This might be less than optimal but it doesn’t seem particularly damaging unless you can fill in the blank of how exactly the urbanization/population density dichotomy bears on the issue being studied.

    2. Does the change in incidence in reported rate not strongly correlate to the change in incidence of other measures of rape?

    3. I’m not quite clear on how the sampling limitations of NCVS bear on the statistics of the FBI Uniform Crime Reports which Kendall used, which I presume is a compilation and not a survey. I guess if you were to say that there’s only a weak correlation between reported crime and actual crime, but I don’t think I would buy that without an actual argument being presented. (Kendall himself makes it clear that his findings bear primarily on the ‘stranger rape’ phenomenon and not the frequently unreported ‘acquaintance rape’ crimes.)

    4. I’m totally confused by this comment. Kendall’s whole point was that, indeed, access to the internet (and therefore access to pornography for 15-19 year old men) DID dramatically increase from 1998 to 2003.

    *I thought I picked up this link from a comment on one of the Pandagon or Feministe threads but now I can’t find it.

  2. SLC says:

    Anybody who has perused the internet knows that it has replaced dirty book stores as the main source of porn of all types. This has occurred since the early 1990s as access to broadband internet has exploded. Therefore I don’t know how Mr. Levy is able to deny that there has been a massive increase in the access to porn during this time period; one doesnt have to visit the local dirty book store wearing a raincoat, wig and false mustache anymore. This also includes areas in which dirty book stores were prohibited (like Cincinnati where Larry Flynt was indicted for violating laws which declared Hustler magazine illegal porn. In fact, Hustler, which was pretty bad at the time, is now positively mild compared with the filth now available for downloading.

  3. double-soup tuesday says:

    To SLC —

    Despite increased access to teh internets, my personal internet porn consumption is way down relative to the 90s.

    That’s anecdotal, but I’d have to attribute a part of that to the proliferation of pay for porn sites and dangerous sites (including malware) sometimes associated with current porn site offerings. I’d never give these folk my CC numbers; but in the mid-90s, the internet porn was more readily available and freely downloadeable — mainly the domain of sickos sharing their deviance vs. capitalist entrepreneurs so more of it was downright free although of lesser quality.

    Still, free porn is free porn. And like Ernest Cline says,

    “Like a preacher needs pain, like a needle needs a vein,”
    Guys need porn.

  4. Alon Levy says:

    Ballgame:

    1. Inequality is the main problem. It’s a stronger correlate of homicide than all other major socioeconomic factors, so it’s strange that the study controls for income and poverty but not inequality.

    2. Actually, the change in reporting doesn’t necessarily correlate with a change in the real rate. The long-term trend is for reporting to increase, so a look at the Uniform Crime Report will make it look as if rape increased in the 70s and early 80s, peaked in the late 80s and early 90s, and then decreased slowly, whereas in fact it crashed in the early 80s, then started crashing again along with all other violent crimes in the 1990s, and began crashing again in 2000 after a pause.

    3. The UCR is not based on a sample, but it’s not necessarily even an accurate reflection of stranger rape. The NCVS shows that the stranger-to-acquaintance rape ratio has been relatively constant in the last 10 years, even though rape rates got cut in half. However, the UCR doesn’t show a similar decrease in reported rates; the decrease it documents is very mild.

    4. The thing is, the US rape rate didn’t decrease steadily from 1998 to 2003. It was steady between 1998 and 2000, and then crashed from 2000 to 2004.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: