It is actually the first variable he lists in his results. Blame ABC.

a) The statistics are fine. First, a cutoff of p=.1 is normal in the social sciences. Second, the article DOES use multiple regression (check out the first paragraph on p.216). Third, the suspect AVN figures are fluff in the intro, he doesn’t use them in his analysis. He gets his real numbers from respectable sources.

b) The author’s claim is fairly clear at the end (the sentence bridging the last two pages)–and it’s not what ABC says. He finds that online porn subscriptions “show remarkable consistency” across state and demographic boundaries. His map (table 2) shows this: Minnesota and Texas are the same, New York is higher than North Carolina, Washington State is quite high–there is no red/blue divide on that map.

c) The author does not make any casual claim tying conservatism to porn buying. From what I read, the argument (stated in other comments) that people in conservative places buy it online because they can’t get it in stores fits perfectly with his findings, and he offers no reason to think he’d disagree.

d) The issue of whether or not to control for broadband use is an important one which ABC neglects. In the article, he discusses it, and shows two pictures–one controlling for it and one not. By ignoring this, ABC twists the article. For example, the ABC article says that 8 of the top 10 states voted for McCain. That’s true, controlling for broadband use. In a straight per capita list (which the author provides) 7 of the 10 vote for Obama. The author remains equivocal about which list is best, since he cannot get data on what percent of broadband users get broadband in order to watch porn.

In other words, the article is a solid piece of research that makes no claims it can’t back up with empirical proof. It does not put a partisan spin (or even provide data that could be honestly spun). The problem here is that the ABC reporters (who probably don’t understand the statistical issues) cherry picked fragments of the article and strung them together into something new.

West Virginia’s Median HH income: $36,088
Utah: $53,324 (which is above the U.S. average of $50,007)

But that’s boring and doesn’t push the liberal agenda.

1. What possible use is this study to anyone outside the porn industry?
2. How much money went into this study?
3. How much does Harvard cost to attend a year?
4. How much do studies like this add to the tuition costs?
5. How much government money goes into Harvard every year, in the forms of PEL grants and research grants?
6. How much money will Harvard receive from the “stimulus” package?

I have to tell you, the idea that I spent even a second working in order to pay the government so that Harvard can do studies like this, really brightens up my day.

This was in regards to consumption of pornographic materials online. In fact, it IS entirely possible that social conservatism has something to do with it, but not in the way ABC made it out to sound.

First off, try looking for adult bookstores in particularly socially conservative areas. They tend not to exist. In a state like Utah, I think it would be a relatively safe assumption that the number of adult bookstores per capita is probably significantly lower than the national average, as the general population doesn’t really want them there.

Secondly, there’s the factor of the expectations of others. The main appeal to online porn is anonymity; you’re not having to walk out of an adult boutique carrying a brown paper bag, for instance. In a state like Utah, where the general population is very socially conservative, porn consumers would be drawn online in greater numbers, mainly to avoid having to answer questions from their neighbors.

In short, I think the author’s general premise may be right after all, but for the wrong reasons, and ABC’s own little editorializing is more than likely completely off base.

If you’re going to look a hundreds of variabless and set overly lenient significance levels, you’re going to find lots of p-values that are significant, if only from the laws of chance. If anything, the author should have set his significance level at .01.

Since many of these variable are co-correlated (i.e. not independent), the only way to sort out what are the genuinely significant correlates is to do a multiple regression analysis to identify and rank the variables.

The failure of the authors to do this represents a failure of peer-reviewer - and also enables the author to cherry pick and expound all sorts of innuendo without a basis in reality.

Which I guess is the new paradigm for our nation in this Orwellian era of Obama - until reality explosively intrudes, as history teaches us it always does in the end.

It’s common knowledge that people don’t read articles anymore; they only skim the headlines.

Controlling for broadband access, states show remarkable similarity in their subscription quantities, as shown in Table 2.

Even after holding constant income, age, and education, adult entertainment
subscriptions are most prevalent in urban areas.

The fourth column reports that in regions where more people report regularly attending religious services (per National Election Studies 2004), overall subscription rates are not statistically significantly different from subscriptions elsewhere (p = 0.848).

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.

And this bit didn’t make in to ABC for some strange reason

The fifth column shows that subscriptions are more prevalent in regions with higher measures of social capital. Following the Social Capital Index developed in Putnam (2000), I obtain DDB Life Style Survey data from 1991 to 1998 on consumers in 340 metropolitan areas, and I form a social capital index based on prevalence of people donating blood, engaging in volunteer activities, or participating in community projects. In a region where 1 percent more people participate in these activities, subscriptions to the adult entertainment service are 0.09 percent more widespread.

Wish I had time for more thorough debunking. But one point, I’ve never before seen a p value of .1 accepted as statisically significant in a scientific paper. Table 3 was totally incomprehensible, which is where most of the “findings” were drawn from.

Worse, the study principal author in the ABC inerview went far beyond the study findings in his comments, clearly showing that he had an political agenda of “hypocrisy” that he wanted to get out despite the non-cooperation of the data.

Since religiousity informs conservative belief even for those who aren’t very religious, the Left has to attack religion and the religious. The easy way has proven to be the sexualization of the Country then it is easy to attack those who criticize that sexuality.

Just more mindless tripe from the center of American stupidity, our MSM

Also, I’d be curious to look at the sales figures — is this just magazines? Videos? Sex hotlines?

Where are the sales numbers coming from?

Maybe this could be answered by reading the study — but I don’t have the time, so I’m tossing this out to see if these are legitimate complaints or if they’re answered.

It sounds like the study doesn’t mean much.

Non-thinking public will construe this as “knowledge” without even analyzing the veracity and logic of the arguments.

ABC, through it’s headline, clearly misinterpreted the Harvard report in question.

Let’s say that the main end-users of the porn materials are in the Red states. Did Harvard examine the political affiliations of the porn buyers in those states?

One possible hypothesis could be this one:

“Since dems and libs don’t have enough places or opportunities to express their sexual urges in Red States, they resort to porn materials for cheap releases.” [sorry...that's not my intention here]

I am not saying that the above statement is true. The same should undergo a standard social analysis through sampling before a formal conclusion can be made.

But since the report did not provide elucidation to the above proposition, I can fairly say that there’s something missing in the research design of the said Harvard study…. if it’s real intention aligns with the ABC’s drivel.

However, the diary’s hints indicate that ABC’s story is inconsistent with the objective of the Harvard study. Thus, it’s another classic example of MSM ’s disinformation campaign against the conservative movement.

Another example: Yesterday, Saturday, I was listening to the 1:00 PM news on my AM radio. A study that ABC News reported on suggested that the very rich were responsible for our financial crisis because they buy large homes. The poor people that build those homes must than take out risky loans to purchase their homes…the result is the wealthy are to blame for the poorer people’s bad decisions.

ABC reports this as a fact and ,unfortunately, the general public believes it to be true.

Talk about truth in lending.

Barnee

Since the conservatives are the biggest readers of the Bible… so it’s just natural that Harvard will identify the conservatives as biggest porn consumers.