Why “Fact Checking” Falls Short

Perhaps it’s my imagination, but “fact checking” this election season seems to have reached a fever pitch. This despite the historically weak tie between facts and politics in general, it would seem. Perhaps I’m overly sensitive to it, given the word-by-word scrutiny to which my own work and media interviews have been subject recently. (Not that the media would ever misquote someone…)

But after “lecturing” to a class of 12-year-olds yesterday on some themes in the book of Exodus, I am reminded again of the difference between moderns’ assumptions about detailed history–what we often mean by “the facts”–and historiography, the telling of history over time and from particular perspectives. Moreover, the former is not very easy to accomplish, and always, always misses material and meanings. It’s partial by definition. This came to mind when I briefly noted to the class that Exodus 1:6 simply states, “Then Joseph died, and all his brothers, and that whole generation.” The author (or authors/redactors) of that text clearly was not interested in conveying the details of the aging and death of the sons of Jacob–indeed, most of their lives and that of their children and their families–but rather with the rise of Moses and the Mosaic Law, from the perspective of those under it. So they paid some things no attention. So be it. (Then you have the lengthy lineages found in a variety of places in the Pentateuch, where we moderns feel like they paid too much attention to detail.)

People are often tempted to think that such a peculiar way of doing history is flawed, but in reality all accounts of “news” or “facts” are perspectival and partial. There is what actually happened–if it can be known–and then there’s the teller, who is a complex person (or organization) with interests, by default. There’s a philosophy of history embedded in all history writing, and indeed even in all news media. In a world increasingly short on attention span but long on bandwidth, this should only grow more familiar to us, not less. Take, for a recent example, the recent death of our Libyan ambassador. There are the facts, and frankly they may never be known with certainty, not simply because some people “won’t tell,” but because eyewitnesses saw different content and perceived different meanings, and have complex interests in relating “the facts” to those different sorts of persons (with quite different interests) who ask them. Add to that the untimely occurrence of this in an election season, and Senate and/or House hearings on the matter, and political sabre-rattling, and you can see how layers of interpretation are added.

And yet we still speak glowingly of “the facts.” In the Era of Science, we sense somehow that facts are always knowable. We presume someone is guardian of The Truth About Things.

This is normal behavior. What’s not normal, because it’s not really possibly in a strong way, is to have a very good grasp of “all the facts.” What’s relevant, after all? Even what counts (or is ignored) as evidence is constituted by particular perspectives. A recent critic of mine suspects I have been directly aiding the Romney campaign, but I’m not sure that the utter lack of evidence will convince them that I am not. To the critic, it’s simply evidence that my aid is more clandestine and thus I am even more suspect.

So it’s often an unrealistic challenge to learn all the facts about events that have already occurred, even recent ones, let alone those that have not. So “fact-checking” presidential candidates and their promises, budgets, plans, etc., is almost a joke. Almost. Moreover, to flippantly accuse one of them (but not the other) of lying–a ubiquitous occurrence of late–is to misunderstand all this.

It would behoove us all in this election season to understand that all politics involves some deceptions, and that human memory fails, and that people misspeak. We ought to remember that the public will always dislike “the facts” if they were all laid bare (and in today’s world, more are laid bare than ever before). This is true about most any of us, for that matter. Let’s be grateful that our thoughts–and for many, their words and actions–aren’t always an open book. We are flawed persons electing flawed candidates who will no doubt run flawed administrations. The two candidates for highest office have quite different philosophies on governance, rights, goods, economics, the role of the State, etc. (and probably most importantly, very different teams of trusted advisors and assistants). Vote on those. Not on some wistful idea of honesty and commitment to “the facts.” Politics has never dealt deeply in that.

Young Women and Porn Use: What Does the Data Say?

Recently I gave a talk at a Christian college about the contemporary mating market, and found myself in a conversation afterward with two women students and an administrator. I had mentioned during the talk that it was suboptimal for the mating market that a significant share of Christian men, frustrated by their own penchant for porn, had come to altogether problematize their sex drive and take themselves off of the mating market as being “damaged goods,” unsuitable for a woman. While the share of them that perceive themselves this way is unknown, and probably not large, it nevertheless cannot but cause further problems for the relationship prospects of women, since it means fewer men in the mating pool (at least for some time), thereby giving more power to the (minority of) remaining men to negotiate romantic relationships in ways they wish. In that sense, it should function like elevated male incarceration rates do in altering the relationship dynamics in the African-American community. Bottom line—when more and more men are considered less and less marriageable, this sex-ratio disparity tends to spell greater and greater problems for women in how they conduct their relationships.

Be that as it may, one of these young women declared to me that she knew more women who watch porn than men who did the same. While I know the data well enough to know that would not hold at a population level, it was true in her world and she was convinced that the sex drive of women thus exceeded that of men. Again, won’t hold at a population level.

But I know better than to shrug off her statement as simply incorrect. She was telling me something—that women her age were into porn at rates that people around her underestimated. I heard a similar story at another Christian college several months prior, where a counseling center employee made a comparable remark. So that prompted me to go to the data—to the NFSS data, that is, source of all things interesting and controversial. It’s nationally-representative, has a large sample, and can speak to what’s going on among young adults in America. What did it have to say about women and porn use? In a nutshell, the student is on to something. Just what it means, however, and how consequential it may be is more difficult to say.

I took the sample of unmarried 18-39-year-old women and split them into three age groups: 18-23, 24-32, and 33-39. When asked to report how often they tended to “view pornographic material (such as internet sites, magazines, or movies),” 21 percent of the youngest group of women reported doing so at rates exceeding more than once a month, up from 14 percent among 24-32-year-olds and well above the 8 percent among 33-39-year-old unmarried women. If we limited it to “every day or almost every day” porn usage, their numbers remain lower than among younger unmarried men (who clock in at 6.7 percent), but the age distinction still holds (3.9 percent among the youngest women, while around 0.5 percent among the older two groups). And the difference between 7 and 4 percent is not so stark. (Moreover, if social desirability bias is at work here—and it likely is—I would expect it to be more pronounced among women than men, thus lowering their likelihood of indicating elevated porn use rates.)

But if porn use is simply tied to sex drive, we should see greater use among older women here. Indeed, when asked whether they were “content with the amount of sex you are having,” the older two groups were far more likely to say no (43% and 41%, vs. 25% among 18-23-year-olds). And the correlation between actual reported frequency of sex and porn use is highest among the oldest group of women, reinforcing the standard explanation that—unlike men—women typically use porn-and-masturbation to augment sex, rather than replace it. So that standard explanation here fits the oldest group just fine, but not the youngest group. Actual sex and porn use are altogether unrelated among the youngest women.

What does this all mean? That’s a far more challenging thing to discern. Here are a few hunches, more speculations than anything else.

First, by their greater uptake of porn young women are acting more like men than women just 10-12 years older than them. But they don’t seem more or less permissive in their attitudes about uncommitted sex than women in their upper 20s and 30s. And their frequency of sexual behavior doesn’t seem strikingly different. Yet among the youngest group of women, the correlation between porn use and recent masturbation is twice as large as it is among the oldest group of women. (For men, the connection is, of course, very strong.) Again, the point is that the youngest women mirror men more than do women in their 30s.

Second, back to the basic question: why the greater uptake of porn among younger women? Here is where speculation is difficult to avoid. The easiest answer is that they’ve had more long-term exposure to porn’s availability, and perhaps less social desirability concerns about using it, then have the oldest women in the NFSS. In other words, it’s more normal to them than to others older than they are. But exposure need not mean attraction and uptake. Classically, the pursuit of porn was simply far more common among men than women, and that still remains true. So does this mean that women who watch it are somehow more sexually jaded—however defined—than men who do the same, simply because it’s more uncommon among women than men? Well, it is certainly worth inquiring (via interviews, perhaps) about their motivation for porn use. Heterosexual men are, on average, attracted to naked women. (File that in the “obvious” folder). Are women, on average, increasingly attracted to naked men? What sort of porn are these 21 percent consuming? (That, I should add, is a question seldom asked of anyone.)

Third, some speculate that women’s porn use is not the solitary thing that it often is for men, who typically use it as an aid in masturbation. Here too, the conventional wisdom is not absolute: whereas 92 percent of men who said they watched porn daily (or nearly daily) also said they had masturbated within the past day, only 40 percent of women who did so reported the same (well above the overall “yesterday or today” figure of 20 percent among women who ever have masturbated; for the record, 29 percent of women said they never have). So there is not nearly the linear association between porn and masturbation among women as among men. But a connection is no doubt present.

Here endeth the speculation. What can be known with confidence, however, is the basic message: porn use is notably higher among the youngest adult women than among women in their upper 20s or 30s. Duly noted. Consequential? Quite likely. How so? Not sure. But if porn use undermines classic ideas about marriage—such as sexual fidelity and relationship permanence—then it stands to reason that greater porn use among women should undermine those marital values. That much is certainly true: women who say they never watch porn are the least likely to report having cheated on a romantic partner and most likely to disagree with statements like “traditional marriage is outdated.” Makes sense.

NFSS data release

This isn’t much of a blog post, but I thought I’d note here that I’ve uploaded the New Family Structures Study data and documentation to ICPSR, the Interuniversity Consortium for Political and Social Research at the University of Michigan. For social scientists, it’s the most popular repository for data. After they process it, the NFSS should be publicly available there. I think it’s the right thing to do, and the right time, even though raw-data release is not required of privately-funded data collection projects in the social sciences like it is for federally-funded projects. I welcome other scholars to analyze the data, which should prove to be quite versatile. I hope its use spreads well beyond the narrow topical confines of parents who’ve had same-sex relationships and comparative young-adult outcomes. The NFSS fielded a comprehensive questionnaire to a random sample of Americans aged 18-39, includes lots of questions across numerous domains, and should attract secondary data analysts. I hope so.

The codebook and data design documentation remain available here.

A follow-up article in which I respond to critics will appear in the November issue of Social Science Research. It’s available here.


Amid all the rancor these days over who’s sexually attracted to whom—and what public policymakers ought to do about it, if anything—comes news of the emergence of another sexuality category, that of asexuality, or the state of not being sexually interested in men or women. It’s not really new news; acknowledgement of it has been around since Kinsey, if not before. It is thought to characterize around one percent of the population. I can imagine a good slogan about the one percent thing, should anyone wish to make something of it.

Some consider it a sexual orientation, while others think it’s the lack thereof. Some are pressing a case for the former because they’re concerned that asexuality might be considered a disorder, with all the politics pertaining thereunto. Thus the attempt to normalize asexuality. Except that it’s not really normal. It may not be a bad thing. In my mind, it’s neither here nor there. Why must we always assign normality a positive value and abnormality a negative one? I presume it’s because we’re hopelessly social creatures constituted in part by our perception of others’ impressions of us. (Ergo, we want to be normal, average.) If you are pursuing celibacy, then asexuality is probably pretty convenient. And it shouldn’t be mistaken for a low libido or diminishing testosterone during the aging process, but certainly could be confused with such. Apparently it’s even a legally “protected class” in Vermont and New York. Given that asexuality concerns an absence or invisibility, the protection of it is interesting.

A Canadian professor and supposed expert on the matter asserts that “asexuality has not been investigated enough.” He’s certainly right about that, and I am all for empirical investigation in this area, so long as the science isn’t beholden to pre-packaged answers. I know I have not investigated it much.

Alas, here’s to starting. In the NFSS, I asked 18-39-year-olds a question that should get at it pretty well. Respondents were asked to “choose the description that best fits how you think about yourself,” after which they were shown answer categories like “100% heterosexual (straight),” bisexual, “100% homosexual (gay), and a pair of categories in between those that I didn’t feel like typing out but you can look up in the survey instrument if you wish. The final category listed there, however, is “not sexually attracted to either males or females.”

So how many NFSS respondents selected that category? Well, wouldn’t you know it: one percent. Technically, six-tenths of one percent, but close enough.

What are they like? That is, who’s more or less likely to so identify? In the NFSS, gender does not appear to distinguish them (that is, men and women self-report comparable percentages of asexuality). Nor does race or age or religiosity or education, or experience with sexual abuse.

But as noted in my July Social Science Research article, 4.1 percent of women whose mothers have had a same-sex relationship reported asexuality, well above the 0.5 percent of women who came from stably-intact biological families. The figure is even higher among male children in such a situation: 7 percent of them report asexuality. Statistical coincidence? It’s possible; we’re dealing with a pretty small N of cases in the first place—only 23 people in the entire dataset indicated asexuality—so we shouldn’t read too much into this. But it’s one of the few variables I’ve seen elevate the level of asexuality among respondents to anything above 2 percent. As with lots of datasets, small numbers prevent us from making too much of this.

Nevertheless, the one percent figure is reassuring–the NFSS continues to display characteristics that suggest its validity and reliability vis-à-vis other nationally-representative datasets.

That’s all for now. It’s good to be back blogging a bit, following my self-imposed silence. (Only civil comments will be accepted…)