Lies, Damned Lies, and Statistics.
Mar. 9th, 2009 09:46 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
elfsCertainly, there's been a lot of brouhaha over a report that appeared in New Scientis article stating that, in the US, porn's biggest consumers are conservatives. Henry over at Crooked Timber ran the numbers and concluded that the report doesn't really say what the New Scientist says it does. At best, New Scientist oversold the conclusions (which is hardly new; New Scientist is the shock magazine of the science journal set). The consumption data is gathered from anonymized credit card records, but the voting and religiousity patterns are all derived from regional statistics, so making a one-to-one correlation is impossible. The summary of the article itself points out that so-called "blue" voting districts tend to use escort service websites more often, while "wife-swapping sites, adult webcams, and sites about voyeurism" were more popular with "red" ones. All in all, the actual consumption patterns are mostly a wash, and as one friend of mine pointed out, the outlier pattern of Utah buying the most porn over the Internet is consistent with porn being difficult to acquire any other way in Utah.
But reading the American Family Network's take has its own entertainments. Their reporter points out, if Utah has a 1.69 subscribers per thousand and there are 2.7 million people there, that means that Utah has 5,000 people buying porn. "That means Utah is hardly full of 'moralistic deviants.'"
What reporter Taranto fails to note is that the number is only small because the population is small; what matters is the extent. If 1.69 permille (yes, that's a real word) is a dismissible number for 2.7 million people, it is a dismissable number for 270 million.
I can only conclude from the AFA's article that they would not think it noteworthy if porn consumption across the rest of the nation went up to match Utah's numbers.
Somehow, I doubt that's the conclusion they wanted their readers to reach.
But reading the American Family Network's take has its own entertainments. Their reporter points out, if Utah has a 1.69 subscribers per thousand and there are 2.7 million people there, that means that Utah has 5,000 people buying porn. "That means Utah is hardly full of 'moralistic deviants.'"
What reporter Taranto fails to note is that the number is only small because the population is small; what matters is the extent. If 1.69 permille (yes, that's a real word) is a dismissible number for 2.7 million people, it is a dismissable number for 270 million.
I can only conclude from the AFA's article that they would not think it noteworthy if porn consumption across the rest of the nation went up to match Utah's numbers.
Somehow, I doubt that's the conclusion they wanted their readers to reach.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2009-03-09 05:09 pm (UTC)no subject
Date: 2009-03-09 07:18 pm (UTC)Anonymous Blog Reader #127
no subject
Date: 2009-03-09 07:32 pm (UTC)One of the biggest sources of CGI T&A models for the Poser software is in Utah. Two other major sources are in Tennessee. Apparently, you can get away with more if you're paying their taxes than if you're in California.