Sunday, March 1, 2009

Back to the distinction between narrative and fact: on making it up to suit your politics in the blogosphere

Here are two post headlines referring to a new study of porn:

Classically Liberal: Born Again and Porn Again: Is There a Connection?

Delawareliberal: Porn in the USA: Conservatives are the biggest consumers

Both purport to chronicle, with considerable glee, the reporting of a new porn study by economist Benjamin Edelman, entitled Red Light States: Who Buys Online Adult Entertainment?.

Here's CL talking about the conclusions:

One state seems to subscribe to porn channels more often than the other states. I won’t leave you waiting—it was Utah. Yes, the Beehive state, the Deseret Kingdom of Mormonism has more porn subscribers than any other state. No doubt they are all fantasizing about their future multiple wives in the afterlife —or not. In second place is Mississippi, home of countless fundamentalists. The next tier of states in porn consumption are all Bible-belt states, but one: Oklahoma, Arkansas, Louisiana and Florida. The one exception is North Dakota, which at least has an excuse—it’s pretty lonesome up there....

The trend seems to imply that conservatives, with their moralistic agenda, are more likely to buy porn than those who lean to the left. Other aspects of the study tend to confirm that. So far 27 states have passed laws against gay marriage, an issue conservatives find very important these days. Those states have 11% more subscribers to porn than the states that haven’t passed these bans....

When porn consumption is compared to the percentage of adults expressing conservative religious viewpoints, porn subscriptions are also more prevalent in states with more "traditional" values.


Here's DL doing much the same:

According to Edelman’s paper, 8 of the top 10 states signing up for porn subscriptions voted for John McCain in last year’s presidential election, Florida and Hawaii were exceptions. Of the states with the lowest subscription rate, 6 out of the lowest 10 voted for Barack Obama.

So what does this tell me? That here is one other front of the Culture War that conservatives are losing badly — after all, they are (as was widely known) big consumers of the thing they claim is so harmful. This paper (while not exactly definitive, but good enough for its point) begins to document the real waste of energy it is to try to control what grown people view. The BushCo DOJ spent a great many resources targeting this industry, which always struck me as a particularly stupid policy. And their focus wasn’t entirely on child porn (which is beyond the pale and folks making that should be sent to the guillotine.) No doubt a corner of the family values set will continue to raise cain over grown up choices, but after this, you do know to ask them tips on the particularly hot sites.


Good writing in both cases, but unfortunately both are intellectually dishonest about (a) what Edelman's conclusions actually say, and (b) what they may mean.

This is a classic case of an interesting interpretive academic paper being usurped and twisted to build a particular political narrative.

Yes, Edelman does argue that certain types of porn, specifically AVN subscription porn, from which he has two years' worth of credit card receipts with zip codes attached, are slightly more prevalent in states known for socially conservative values and laws.

But not much more:

On the whole, these adult entertainment subscription patterns show a remarkable consistency: all but eleven states have between two and three subscribers to this service per thousand broadband households, and all but four have between 1.5 and 3.5. With interest in online adult entertainment relatively constant across regions, there’s little sign of a major divide.


Nor does Edelman find a significant correlation between porn use and voting, despite what our blog commenters would have us believe:

I found no significant relationship between subscriptions to this adult entertainment service and presidential voting in 2004, based on poll data by congressional district. However, using individual-level data from a Hitwise sample of ten million anonymized U.S. Internet users, Tancer (2008), finds that adult escort sites are more popular in “blue” states that voted for Gore in 2004, while visitors from the “red” states that voted for Bush in 2004 are more likely to visit wife-swapping sites, adult webcams, and sites about voyeurism.


He found strong correlations between marriage and divorce vis a vis porn viewing, a subject not noticed by our bloggers:

Both marriage and divorce are associated with reductions in subscription to this adult entertainment service. A 1 percent increase in marriage rates is associated with a 0.65 percent decrease in subscription rates, while a 1 percent increase in divorce rates is associated with a 0.28 percent decrease in subscription rates.


He found that youth, greater income, an urban setting, and bachelor's degrees correlated with higher porn use, while retirement, graduate degrees, and rural settings did not. Funny, none of that was takeaway for our bloggers.

Because Edelman is an academic, and is not willing to take his data further than the evidence supports, he never makes the claim that conservatives as individuals are more likely to purchase porn than anybody else. Instead, he deals in aggregate populations and statistical description.

Why? Because Edelman knows that there are other conclusions that this data set can support instead of the simplistic meme that hypocritical social and religious conservatives do more porn than anybody else.

For example: In a community in which there is a dominant conservative public value system and even social control legislation, people who are LGBT, people who are swingers, people who practice bondage, people who get into rape fantasies will find themselves driven further underground, and an AVN subscription service is far less risky that heading out to a club or trying to interact with real people. His data equally supports the thesis that conservative social repression sends moderates and liberals with different sexual orientations and interests underground.

To take another example: Edelman's data supports an interpretation that among large populations in the US there is little or no real statistical variation in porn consumption. But the problem is that in even the most conservative states there are thousands if not millions of liberals, and Edelman's data set cannot differentiate between them.

Every since Nathaniel Hawthorne committed an act of terrorism against modern high-school students by writing The Scarlet Letter, we have--as a nation--been obsessed with examining the perverse relationship between our religious beliefs, our public hypocrisies, and our sexuality.

The real finding in Edelman's study is not that the religious are any more likely to consume porn, but that they are not any less likely to do so.

That's an important finding (even if it is only tentative at this point), but one that's immediately lost in the rush to take good statistical detective work and bend it for partisan political purposes before the ink is even dry on the paper.

9 comments:

Anonymous said...

The survey says that the subscriptions in Utah were 5.47 per 1,000 homes. That is double the number in states like Connecticut, Delaware, New Jersey and Oregon. The report says:"Subscriptions are slightly more prevalent in states that have enacted conservative
legislation on sexuality" as well as "more prevalent in states where
surveys indicate conservative positions on religion, gender roles, and sexuality." He also said subscriptions are higher in states with strong religious views. (p 219)

Classically Liberal is not dishonest in calling those numbers a trend. That is all it said, and the rest reported on the numbers with several long quotes from the paper and links to the paper. Since Edelman said states that passed conservative legislation on sodomy were 11% more likely to subscribe to porn then how is is "intellectually dishonest" to use the term trend to describe that? How significant a trend may be debated but not that it is a trend.

Nothing I see in the quote from CL doesn't correspond with the report. The main assertions in what you quoted are:
1. Utah has more subscriptions -- it did by with rates far higher than the average state.
2. Conservatives seem more likely to buy porn -- as indicated by the higher rates in states that supported conservative views. The quotes from the report verify that is correct.

Miko said...

The real finding in Edelman's study is not that the religious are any more likely to consume porn, but that they are not any less likely to do so.

Even this isn't certain. At #1, Utah had only 0.547% of broadband residents subscribing to porn in the sample studied. So, as long as the percentage of religious people in the state is less that 99.453%, it could be that there's absolutely no overlap between the two groups. Unlikely, but possible. (Of course, as not every porn site was studied, the actual percentages would be somewhat higher.)

Steve Newton said...

Anon,
You'd be mostly right IF CL didn't draw the direct conclusion that individual conservative Mormons dreaming of multiple wives, or conservatives with their moralistic agendas had been shown to be greater users of porn. They weren't.

Moreover, CL made the jump to implying that there was a voting correlation to porn usage, which the study explicitly denies finding.

Nancy Willing said...

I really like the group Media Matters because they similarly seek out and debunk published news accounts that include distortions of fact. When a met a few of MM staff at a leftwing bloggers conference last summer, I asked if they addressed questions and distortions coming from the left. The woman looked a little sheepish and admitted that they were mostly only concentrating on rightwing material.

Shirley Vandever said...

I like to visit both conservative and liberal sites. It's amazing the different spins that can be put on the exact same story.

For example, in the Stanford Financial fraud story, left-wing sites showed only pictures of Allen Stanford with Republicans, while conservative sites showed only pictures of him with Democrats.

It's funny, really.

I think people really owe it to themselves to read both the left and the right; find ones that are intelligent and don't blast bullshit without any backup.

DailyKos is actually a great site. Though I may not always agree with their spin, the site provides alot of good information, especially if you want to know what is going on in Congress on a daily basis. My favorite conservative site is Ace of Spades HQ: the bloggers there do heavy-duty, intelligent and in-depth analysis (with a few snippets of war porn thrown in from time to time for good measure).

Trying to validate your own beliefs with a headline is intellectually dishonest (though quite tempting, is it not?). It is also lazy. I've probably succumbed to it from time to time, but I generally try to get as many angles on a story and as much information as I can before I open my big mouth.

Even then, I can be called out on an error. It happens.

Brian Shields said...

A friend of mine once said "All I do online is porn, everything else is recovery time. Online banking, news, You Tube.. all distractions from porn."

I have some strange friends. I also made a mental note not to check my email on his machine.

anca said...

Redtube cams

Nancy Willing said...

Cassandra really ought to change her headline.
She changed up her defenses in the posts' comments, even admitting that the article didn't come to the conclusion that conservatives are the biggest users of porn - because the data did not find a significant difference in use.
An honest headline would have said that the usage was equivalent along with the POV about hypocrisy and cultural denial.
She has yet to admit that she's wrong, or so said Steve Newton elsewhere on this blog.

Shirley Vandever said...

In other porn news, I read today (don't have the source, you'll have to trust me), that people who view porn commit less rapes.

I guess this means, following the logic of this study, that liberals commit rape more than conservatives.

I mean, it just naturally seems to follow, doesn't it?