The Myth That Won’t Die

We were out to dinner at a pedestrian steakhouse last evening with a couple we’ve known for a very long time.  After running the course of the usual conversations, my pal Ralph mentioned that he was concerned for his niece who was a freshman at college because he hears there is a rape epidemic.

Oh crap.

He explained to me that 1 in 5 college women were raped.  I explained to him in return that the number was nonsense, a bit about the statistical failings until his eyes started to glaze over, and then about the Bureau of Justice Statistics number of .61%, not 20%. I had lost him.

For the handful of us who aren’t dedicated to perpetuating a lie that supports a political position, that the 1 in 5 is nonsense is old news.  And yet, the media continues to perpetuate the myth, fact-checkers be damned.  And it’s not yet done.

In the February 5th edition of the New York Times Review of Books, Zoë Heller writes about rape on campus.  And there it is, in her first line:

According to the most commonly cited estimate, 20 percent of women are sexually assaulted during their time at college and as few as 5 percent of these assaults are ever reported to the police.

The irony here is that Heller doesn’t write to demand castration of all incoming freshman males, but to move the discussion closer to center:

Few would disagree that the systems for preventing and prosecuting sexual assault on US campuses are in need of change. But the efficacy and fairness of recent reforms that focus on making college grievance procedures more favorable to complainants and on codifying strict new definitions of sexual consent remain highly questionable. Advocates of these reforms tend to dismiss their opponents as reactionaries and “rape apologists”—a characterization that is probably accurate in some cases—but feminists, too, have cause to view these measures and the protectionist principles on which they are based with alarm.

The gist of her argument is not that colleges aren’t rape incubators, where wild-eyed men are abusing women all day long, but that the evisceration of rights isn’t in feminists’ interests either.

For some, this is a salutary development, a necessary antidote to the unfair disadvantages that rape victims have traditionally suffered when seeking redress in college tribunals. According to Colby Bruno, senior counsel at the Victim Rights Law Center, the preponderance of evidence standard “helps counterbalance so much of the bias and rape culture that permeates these cases.” But the proper remedy for bias is surely not more bias in the opposite direction. And while there is certainly a long history of rape victims being demeaned and automatically disbelieved, not all of the difficulties associated with prosecuting rape are attributable to sexist prejudice. Rape cases, which often boil down to the relative credibility of two conflicting narratives, are inherently difficult to prove. No fair adjudication process can get around this fact by assuming a posture of reflexive credulity toward a victim’s testimony.

So?

The perils of this ends-justifies-the-means calculus (variants of which have been used in recent years to defend racial profiling, the mass government surveillance of US citizens, and the torture of terrorism suspects) ought to be self-evident. It is a moral and strategic error for feminism—or any movement that purports to care about social justice—to argue for undermining or suspending legitimate rights, even in the interests of combating egregious crimes. If the chance of an unfounded assault allegation is “only” eight in a hundred, that is reason enough to avoid basing standards of evidence on the assumed good faith of complainants.

And?

But to exempt women from the responsibility of stating their own sexual wishes without prompting—to insist that it is the man’s job to “figure out” those wishes—comes dangerously close to infantilizing women.

Therefore?

Laws that offer special protections to women based on their difference from men have a habit of redounding to women’s disadvantage. In the case of affirmative consent, the payback is readily apparent: women are deemed to have limited agency in their sexual relations with men, so men are designated as their sexual guardians—tenderly coaxing from them what it is they want or don’t want and occasionally overruling their stated wishes when they’ve had too much to drink. What a pity it will be if a campaign against sexual violence ends by undermining the very idea of female sexual autonomy that it seeks to defend.

Heller says many of the right things, many of the obvious conclusions that advocates prefer to ignore by screaming louder than anyone else.  But at the same time, she both perpetuates the myths and argues that the ultimate justification for not going down the path of infantilizing women by denying them responsibility for their choices is grounded in old-school feminism.

Or to put it differently, not because factual accuracy and due process for all is the right thing to do, but because it serves a more fundamental long-term feminist agenda.

But none of this matters to my pal Ralph.  All he hears is the 1 in 5 number, that his niece stands a pretty good chance of getting raped.  That his son, who will be going to college in a few years, stands a pretty good chance of being a rapist, because he’s a boy and that, according to the news, is what boys do these days.

Ralph asked me what I do with my son, how I dealt with keeping him from raping women and protecting him from being accused of rape.  As much as some would tell me that I should support articles like Heller’s because they’re better than the insanity generated by advocates for tossing all males out of college as rapists, it still makes it impossible to have a rational conversation about any of these issues with regular folks who aren’t on top of the statistics, the reports, the details that enlighten us as to the facts versus the advocacy.

It’s not good enough that Heller comes out on the right side for the wrong reason.  In her opening sentences, she perpetuates the myth.  As long as that continues, there will be no conversation with nice, smart, but not atop of the details folks, like Ralph, and they will continue to believe that a rape epidemic is going on at colleges.

And if there is such an epidemic, something must be done about it. Because we cannot have a rape epidemic.

50 thoughts on “The Myth That Won’t Die

  1. onewhoknows

    [Ed. Note: Apologies for posting this comment, but as it appeared and provides a teachable moment, the least I can do is use it for that purpose.]

    The Garner incident was an indictable offense, pure and simple. Take it from this former prosecutor. Not even submit manslaughter to the Grand Jury? Then there are rightful protests over this injustice, including the Mayor’s appropriate support for the incident to be reviewed by the Justice Department for a civil rights violation. That’s followed by the police and police union pushing back at the Mayor to show that they disagree with him, now don’t respect him and try to make it seem like he has a political problem? Really? Most cops don’t live in the city! ( I don’t count racist, segregated Staen Island). They don’t vote there. They make their living there with the taxpayer funds from the people that they serve and then they return to their ranches upstate in Monroe, NY or across the river in Putnam County and rag on the people they serve. Despicable. They should try getting decent wage paying jobs where they live and see what they come up with. While the homicide review is concluded on the Staten Island end, this incident serves to remind us about the attitude that many white police officers have towards the people they serve and the NEED TO PASS A RESIDENCY LAW to ensure that they live within the boundaries of NYC, which will foster respect for those residents.

    1. SHG Post author

      1. This comment is so bizarrely off-topic that I fear it’s the product of a bad acid trip. This post has nothing to with Garner, yet you post this comment here. Rational people don’t do such things. They may go slightly off-topic, but not whole-hog totally batshit crazy off-topic.
      2. You use the handle “onewhoknows” which puts you squarely in the category of flaming nutjob off the top. Only psychos use such handles. Are you trying to portray yourself as a psycho?
      3. You do not get to proclaim yourself a former prosecutor unless you use your real name and can be verified as a former prosecutor. Anyone can say they’re anything on the internets. The problem isn’t using a nym, but using a nym while trying to enjoy the credibility that comes with specific experience. You can’t have both.
      4. Use of all caps (“NEED TO PASS A RESIDENCY LAW”) is the hallmark of internet nutjobs, who have an inadequate command of the language such that they need to resort to caps (screaming) for emphasis. Educated, non-nutjob people don’t use all caps.
      5. When posting to a law blog, where the majority of readers are lawyers and judges, do not presume your puerile opinion to matter. Why should anyone give a damn what you think. You don’t exist.

      Perhaps you are a lawyer, even a former prosecutor. Perhaps you are educated and not a flaming nutjob. Perhaps you’re just new to the internets, unfamiliar with the etiquette, unaware of things that make you look crazy and ignorant, and just a generally unpersuasive individual. As someone who shares the concerns expressed in your absurdly off-topic comment, I offer this advice: this is not the way to accomplish anything beyond making yourself look truly bad.

  2. John

    Your link “that the 1 in 5 is nonsense is old news” appears to be going to the wrong place. The link leads to critiques of estimates of the rate of false rape reporting, not anything to do with estimates of rape rates on campuses.

  3. Not Jim Ardis

    “According to the most commonly cited estimate”

    Wait… So if I get enough people to claim that for every 5 women on a college campus, 7 of them are raped, I’ll get quoted by people who are otherwise serious individuals? Is that how it works? I don’t think that’s how it works.

    1. The Real Peterman

      Well, she has a point, that is the most commonly cited estimate. It’s like if someone went around saying “people often accuse Zoë Heller of being a lazy, dishonest hack.” People do say that, so why argue? I wonder if the New York Times will run a story about people who accuse Zoë Heller of being a lazy, dishonest hack. Maybe if enough people say it.

    2. Patrick Maupin

      The best estimate I have at hand is that 2 out of 5 males are raped in college. But this is significantly underreported, because no guy wants to admit that they’ve been a pussy.

      1. SHG Post author

        I note that you use a slang word for the female genitalia to describe the negative character flaw in males of being cowardly. Coincidence?

        1. Patrick Maupin

          Ironically, if he mans up and admits that he’s been used as a pussy, then he’s not a pussy, right?

  4. Barry Sheridan

    I often wonder why universities in America are not fully segregated, in fact given such thinking why stop there. The US is a large place, divide it into three, one part for all the females who loathe men, another for the men who loathe women and the rest of the space for those happy to get along with each other without spending all their lives inventing statistics aimed at inflicting harm on someone else. Be an interesting social experiment to see who does best.

    Apologises SHG, I could not resist making this point given that I find far too many modern females quite dispiriting. They have become far too confrontational to achieve any lasting good. What’s the gain??

    Oh by the way, next time you see Ralph ask him if he is a rapist. After all if he genuinely believes 20% of all college aged men are so then there is fair chance he might be for all you know. Doubtless he will indignantly reject the notion and get grumpy with you for even suggesting such a thing. You could then ask him how the guys he presumes are feel to be thought of in this way!

    1. SHG Post author

      You are laboring under a misapprehension. This isn’t about hating feminists or men’s rights, but about accuracy and due process for everyone. This is not your soap box to spew hatred toward feminists.

  5. Beth Clarkson

    “What a pity it will be if a campaign against sexual violence ends by undermining the very idea of female sexual autonomy that it seeks to defend.” I think this is a sentiment more important than the perpetuating the myth.

    Myths like that, with a basis in sketchy research and shifting definitions seem to go in and out of fashion over the decades. Remember the one about a single woman over 40 being more likely to be killed by a terrorist and find a husband?

    My condolences for the loss of your mother.

  6. Barry Deutsch

    “He explained to me that 1 in 5 college women were raped. I explained to him in return that the number was nonsense, a bit about the statistical failings until his eyes started to glaze over, and then about the Bureau of Justice Statistics number of .61%, not 20%. I had lost him.”

    I hope you explained to him that the .61% number is an annual figure, and so shouldn’t be directly compared to a figure for what happens over a 4 or 5 year college career. And that you also explained that there are many good reasons to suspect that the .61% significantly undercounts, and that its source – the NCVS – finds an extremely low rape prevalence rate compared to almost any other study of rape prevalence.

    Could you say what the “statistical failings” of the CSA are? The Yoffe article says that it surveys two colleges, not a representative sample, and that’s a fair point – but it doesn’t mean that the CSA is a “lie” or “nonsense” or a bad study, just that we shouldn’t assume that its findings are applicable to all colleges. (It’s also about a bunch of different kinds of sexual assaults, not only only rape – but the same is true of the NCVS statistic you favor.)

    I agree with you; your friend was wrong. However, I’d say his main mistake was not in thinking that rape prevalence figures are fairly high – the NCVS stat you prefer is an outlier, and is probably much lower than the reality – but in thinking that campuses are especially bad. As far as I know, there’s no reason to think that college-age women on campus are more likely to be raped than college-aged women off-campus are.

    1. SHG Post author

      The rampant statistical failings of the CSA has been dealt with here numerous times. No, just because you showed up today doesn’t mean we have to repeat everything that’s ever been written before for you. Yes, the .61 is annual. Multiply it be four (that would be 2.44%) and see how it compares to 20%?

      The reason the stats seem low is that it’s not biased by the political advocacy of their other studies and their absurd lack of definitions in what constitutes a sexual assault or rape, which have also been demonstrated over and over. Sorry. but the statistics don’t get higher just because you desperately wish them to do. The 1 in 5 is just flagrantly baseless, no matter how badly you wish more women were raped.

      1. Barry Deutsch

        So you’re saying that all studies that don’t show the outcome you want are politically biased? And you think I’m the one making desperate wishes?

        Just to be clear, I’m not arguing that “1 in 5 women are raped at college is correct.” I don’t think that’s correct, and I think the “real” number, as best as can be worked out, is significantly lower than that.

        I want rape and sexual assault prevalence rates to be as low as possible, and obviously you want the same thing. But there are many studies that have found prevalence rates much higher than the NCVS numbers, even if you limit the findings to nothing but completed rapes (which the NCVS doesn’t). And the NCVS has serious methodological problems.

        To pick out the study with the most-desired outcome and declare that all studies that don’t support the outcome you want are biased, as you seem to be doing, is indefensible.

        1. SHG Post author

          Don’t be silly. There are probably 25 posts here that address the statistics either directly or peripherally. My concern is that the 1 in 5 statistic is being use to drive new formulations of the offenses rape and sexual assault that are untethered from rational definition and undermine basic due process protections for those accused. I have no political interest in the numbers whatsoever, which you apparently can’t grasp as you impute political motive to me to justify your disagreement.

          The point is that this is hysteria driven by false statistics. In the name of false statistics, the burden of proof has been reduced, due process denied, definitions discarded and unworkable demanded imposed. Despite the 20% stat being universally rejected, it continues to be used to drive the hysteria, and hysteria is a dangerous and damaging driver of change.

          By the way, the NCVS isn’t a desired outcome, but the best statistics available by far. This has been discussed to death. Your rhetorical “serious methodological problems” just isn’t true, no matter how many time you try to say so. It may not be perfect, but it is a thousand times better than anything else out there, and it’s the only study not done by people or groups with an advocacy position in this issue. This has nothing to do with my political bias or desired outcome. It’s just statistics.

          1. Barry Deutsch

            If there’s a specific past post of yours you’d like me to read and take account of, please point me to it. But I don’t think I’m required to have read everything you’ve ever written to respond to a particular argument you publish, any more than Zoe Heller should require you to have read all of her works before commenting on her NYROB article.

            I certainly want due process protections for accused people, and don’t want people misusing rape statistics to remove due process. But that’s not relevant to the question of what the best current estimate of rape prevalence is. It’s possible to both want accurate numbers, and to oppose those numbers being used to remove due process, and I believe that describes both your position and mine.

            I’m quite ready to believe that you have no political interest in the numbers whatsoever. However, please do me the same courtesy, for example by not accusing me of wishing “more women were raped.” I’ll be civil if you will, but you shouldn’t demand treatment you’re not willing to offer it in return.

            Your rhetorical “serious methodological problems” just isn’t true, no matter how many time you try to say so. It may not be perfect, but it is a thousand times better than anything else out there, and it’s the only study not done by people or groups with an advocacy position in this issue.

            Dismissing studies because of the alleged beliefs of the people who ran them, without any actual evidence that they faked the study or used bad methodology, is an ad hominem argument. Legitimate arguments against studies, especially studies that are either from respected government agencies or peer reviewed, should argue based on the methodology, not based on what you imagine the ideology of the researchers to be.

            The NCVS is an excellent study of crime prevalence in most ways, but because it’s a decades-old annual survey, the questions and methodologies it uses don’t always reflect recent consensus on best practices. There are excellent reasons for the NCVS to be slow to change (the slower the questions change, the more legitimate comparisons between past and present NCVS results are). But it can also create problems, such as the fact that the NCVS has not followed most researchers in the use of behaviorally specific screening questions for measuring sexual violence.

            In 2000, the Bureau of Justice Statistics published a side-by-side comparison of NCVS screening questions and behaviorally specific screening questions. All other aspects of the study were identical; the only difference was the use of the NCVS screening questions (sample: “Incidents involving forced or
            unwanted sexual acts are often difficult to talk about. Have you been forced or coerced to engage in unwanted sexual activity by (a) someone you didn’t know before, (b) a casual acquaintance? OR (c) someone you know well?”) versus explicit, behaviorally specific screening questions (sample: “Has anyone made you have sexual intercourse by using force or threatening to harm you or someone close to you? Just so there is no mistake, by intercourse I mean putting a penis in your vagina.”). For the comparison, the NCVS definition of rape was used, and the NCVS two-stage model of interviewing was used.

            What the researchers found is that asking explicit, behaviorally specific screening questions led to 11 times more women reporting incidents that met the NCVS definition of rape, compared to the NCVS screening questions.

            So that’s the sort of thing I’d call a serious methodological problem. In empirical testing, NCVS screening questions have been shown to lead to a severe underestimate of rape prevalence. That’s not the only problem with the NCVS, but it’s certainly sufficient for reasonable people to be skeptical of their results.

            1. SHG Post author

              You can read or not read. To ask that I do your legwork for you, however, is just too blatantly narcissistic to bear, as if it’s my duty to prove something to you. If you prefer to wallow in ignorance, that’s your right. This is America, and every American has a right to be as stupid as they want to be. They just don’t have a right to make other people stupider. That’s my problem with Zoe Heller’s opening line.

              As for your attempt to critique the NCVS (and you might want to read this post to learn a bit about substantive analysis), it’s remarkable how you’ve misconstrued the nature of the questions to show exactly the opposite of what they show. By “begging” for responses to prove the point, it makes the study far less accurate and, as expected, proves the point desired by the study.

              Whether you are interested in understanding that fallaciousness of your arguments is up to you. I don’t write this blog to satisfy Barry Duetsch, and hard as this may be to imagine, your agreement is utterly irrelevant. The statistics are as good or bad as they are regardless of whether some random guy on the internet agrees with them or not.

              So put in the time to learn or not. If you prefer to wallow in ignorance, that’s entirely up to you. I didn’t come to you and demand that you agree with me. I neither care to prove it to your satisfaction nor spend my time schooling you, as if you are the center of my universe. There is an internet full of people who prefer to be clueless so they can believe whatever they want to believe. You’re just one of them.

            2. Sgt. Schultz

              Wow. First, narcissism rears its ugly head. Then you Gertrude SHG. And then you prove conclusively that you have no grasp of what makes survey questions valid or not.

              Very few people are capable of doing themselves so much damage in one comment. You win the internets today.

            3. Barry Deutsch

              Sgt:

              My comments about the relative advantages of behavioral specific questions are based on reading dozens of peer-reviewed studies written by experts. For one recent examples, see the 2014 study “Estimating the Incidence of Rape and Sexual Assault,” which spent quite a bit of time discussing how the unspecific and undefined questions used by the NVVS underestimates prevalence, and in conclusions recommended the use of behaviorally specific questions instead. This study was sponsored by the BJS and NIJ and conducted by the statistics committee of the National Research Council. I suggest that you address your criticism to the co-chair of the report, William D. Kalsbeek, Professor of Statistics, and former Director of UNC’s Survey Research Unit; he may be surprised to learn that he has “no grasp of what makes survey questions valid or not.”

              In contrast, I’ve never seen a published academic source documenting the problem of “begging” in sexual violence survey question design. SGH, if you can please provide a link or citation to a peer reviewed or a government published source documenting this concern, I’d be very interested in reading it, and open to changing my view.

            4. SHG Post author

              Now you’ve gone off the deep end. The Kruttschnitt/Kalsbeek paper was not a survey, but rather a “best practices” report on how to increase the rape prevalence numbers based upon the assumption that they must be doing it wrong because the numbers were too low.

              As for “begging,” this is basic stuff. Asking questions designed to evoke specific answers is run of the mill nonsense. I’ve already given you a link to Francis Walker’s post in my earlier comment which discusses this in depth, with links to studies, etc. But you can’t keep asking for me to spoon feed you. If you are open to changing your view, then read, but don’t keep coming back here asking for me to prove it to you. As I’ve already said, you are fully entitled to be as ignorant as you choose to be. Seriously, your narcissism is really insufferable.

            5. Myles

              I’ll be civil if you will, but you shouldn’t demand treatment you’re not willing to offer it in return.

              Since when do you come to someone else’s blog and dictate the terms of engagement? Nobody forces you to comment. Volenti non fit injuria. Sheesh.

            6. Andrew

              Just noting here, as I did on the other post, that Deutsch’s runs an extreme feminist blog, and his claim to be open minded is total bullshit. He’s a troll.

            7. EH

              Barry said:
              What the researchers found is that asking explicit, behaviorally specific screening questions led to 11 times more women reporting incidents that met the NCVS definition of rape, compared to the NCVS screening questions.

              That implies some combination of these:

              1) The CSA questions result in an undercount because they are bad questions.
              2) The behavior-specific questions overcount because they are bad questions.
              3) The combination of the NVCS definitions and the CSA questions undercount because of the interaction between the questions and definitions;
              4) The combination of the NVCS definitions and the behavioral questions overcount because of the interaction between the questions and definitions; and
              5) they’re all inaccurate and we don’t have a grasp on what is really happening.

              So that’s the sort of thing I’d call a serious methodological problem. In empirical testing, NCVS screening questions have been shown to lead to a severe underestimate of rape prevalence.
              The NCVS screening questions lead to a *lower* estimate of prevalence. It may be “under” or “over” or “accurate” but we don’t know. We know that two sources disagree, not which one is accurate.
              That’s not the only problem with the NCVS, but it’s certainly sufficient for reasonable people to be skeptical of their results.
              Yes, you can reasonably become skeptical of the NCVS because of the difference. No, you cannot use that same justification argue for reliance on the competing alternate.

            8. SHG Post author

              We know that two sources disagree, not which one is accurate.

              We do know that the CSA suffered from so many fundamental flaws that it is there is no statistical confidence. While the NCVS is imperfect, it cannot be compared to the CSA. Can someone still be reasonably skeptical of the NCVS? Of course. The nature of these studies will invariably present problems, but the choice is not the CSA or NCVS.

            9. David M.

              EH is right. I lean towards 2), because I’ve taken college statistics like everyone else and the sort of question Barry likes seems preposterously leading, but he’s right that we can’t discount the other possibilities.

            10. SHG Post author

              The point of statistics isn’t to find “an answer” so much as to gain a level of confidence that the results are accurate. In other words, while there is always the potential for other possibilities, the purpose is to distinguish which possibility bears the likelihood of accuracy on the whole. That’s why sample size, etc., matter.

            11. Barry Deutsch

              I hadn’t intended to comment here again, but EH made a really good comment, and I couldn’t resist responding.

              First of all, EH I think you made a slight mistake and sometimes typed “CSA” where you meant “NCVS.” But if I’m mistaken about that, please let me know.

              You’re right that if all we had was the two sets of screening questions and the responses, then we would be unable to distinguish between

              “1) The [NCVS] questions result in an undercount because they are bad questions.
              2) The behavior-specific questions overcount because they are bad questions.”

              And so on. Because we’d have no independent way of checking if someone who said “yes” to a particular screening question was actually referring to an incident of rape.

              However, this study did have a way of checking. The study, like the NCVS, uses a two-stage model of interviewing. In this model, the screening questions are used to find subjects who have had an “incident.” If someone says “yes” to one of the screening questions, that alone does not get them counted as a victim of sexual assault. Instead, they will be asked to fill out an incident report, describing what happened in detail.

              It is only if the subject fills out an incident report that clearly meets the NCVS definition of rape that the incident gets counted as a rape for purposes of the study.

              For this reason, the researchers were able to confirm that when subjects answered “yes” to the screening questions regarding rape, in nearly every case it was because the subject was referring to an incident which, when described, met the NCVS definition of rape.

              Does that alleviate some of your concerns?

            12. SHG Post author

              You’re back? How can we miss you if you won’t go away. I can’t speak for EH, but you’re still completely ignoring flagrant problems and trying to gloss over them.

              If someone says “yes” to one of the screening questions, that alone does not get them counted as a victim of sexual assault. Instead, they will be asked to fill out an incident report, describing what happened in detail.

              And then the people conducting the study decide to call it rape even though neither the respondent, nor anyone who doesn’t subscribe to feminist ideology, would call it a rape. They identified rapes and sexual assaults based upon their ideological interpretation of events and without resort to any legal definition. Sorry, Barry, but you can’t leave that part out as if there is anything neutral about the subjective interpretation of responses and it doesn’t matter.

              By the way, when you were busy telling me about how you were open to being proven wrong, you neglected to mention that you were a radical feminist ally. A minor omission? And you whine about it?

            13. David M.

              Barry, homeslice, I’m not as mean as Sgt. Schultz or anything, but that was the dumbest argument you’ve made yet and that’s saying something.

              You think that, if the CSA makes use of an expanded definition of rape, uses it to elicit positive responses (something you’d like to improve upon with leading questions, let’s not forget), then shoehorns those positive responses into the unexpanded NCVS definition and pretends they either A) match or B) arose organically, it’s evidence the CSA is reliable?

              Jesus Christ on a pogo stick, son! That is some hardcore stupid and no mistake.

            14. Barry Deutsch

              The study I’ve been referring to, that included direct comparison of NCVS screener questions and behaviorally specific screener questions, was the SVCW, published in 2000 by the DOJ. (I realize you probably already knew that, but I thought I’d mention it in case others reading didn’t know).

              And then the people conducting the study decided to call it rape even though neither the respondent, nor anyone who doesn’t subscribe to feminist ideology, would call it a rape. They identified rapes and sexual assaults based upon their ideological interpretation of events and without resort to any legal definition. Sorry, Barry, but you can’t leave that part out as if it doesn’t matter.

              Okay, let’s look at this. In the case of the SCVW, 46.5% of the women who were counted as having experienced a completed rape said that their experience was rape. 48.5% said it wasn’t rape, and the remainder said they didn’t know.

              There were also a few cases of women saying they’d been raped, which weren’t counted as rape because the incident described didn’t meet the study’s definition of rape).

              First of all, even if you decide to only count rapes in which the incident described meets the definition of rape and the woman agrees it was rape, then you’d still be left with the NCVS screening questions failing to correctly screen 80% of those women. So your current argument, even if accepted, won’t change the fact that the NCVS badly undercounts rape prevalence.

              Second of all, we know from both research and data that there are a large number of rape victims who won’t describe an event as rape even if it is a rape by virtually all legal definitions. Studies have show that a substantial number of women will describe being held down and forced to have sex when they didn’t want to, and when they had clearly said they didn’t want to have sex – and will still answer “no” to the “was it definitely rape?” question. Is it really just ideology to think that a woman in that case has been raped for the purposes of a rape prevalence study, even if she’s not willing to accept that label?

              Even in the most extreme cases, some women will say that what happened wasn’t rape. Here’s a real-life example:

              Victim’s male acquaintance breaks into her apartment and grabs her. He is in a rage because she had refused to go out with him. He roughs her up a bit, including belting her across the face and throttling her. He then forces her at gunpoint to drive him to his house, where he keeps her overnight. He specifically tells her that he will shoot her if she tries to escape. He is distraught and talks repeatedly about how much he loves her. He talks about wanting to live with her in Mexico. Her survival strategy was to pretend to go along with his plans. She wanted to gain his trust. When he had sex with her that night, she “went along with it” in order to survive.

              After finally escaping, she went to the police. She expected that he would prosecuted for kidnapping, assault and threatening. She was, however, shocked when I brought a rape charge against him. She didn’t feel that she had been raped because she had “gone along” with the sex.

              I should also point out that this standard – defining the crime by whether the incident meets the definition of the crime, not by whether or not the victim labels it a crime – is standard practice, and used by the NCVS and other crime studies when measuring prevalence of other kinds of crimes.

              And then the people conducting the study decided to call it rape even though neither the respondent, nor anyone who doesn’t subscribe to feminist ideology, would call it a rape. They identified rapes and sexual assaults based upon their ideological interpretation of events and without resort to any legal definition.

              As I already pointed out, the SCVW, for the comparison to NCVS questions, used the NCVS definition of rape. Elsewhere in the study, they used “Unwanted completed penetration by force or the threat of
              force” as their definition of completed rape.

              Do you have any evidence to support your accusation that they used some other definition?

            15. SHG Post author

              Aw, come on. You’re not going to make me go through every one of your comments to pull out your pimping the CSA, are you? Whatever, anyone who cares to wade through the million words you murdered can read what you wrote. This is just getting boring.

              You posted 8 comments, and about ten million words (just estimating, like it was rape), claimed neutrality despite being up to your eyeballs in feminist dogma, and lied and got caught. Enough for one day. Go watch the Super Bowl. You do watch the Super Bowl, right?

            16. EH

              I am skeptical of Barry’s approach. It seems that an ordinary woman should be capable of determining, for herself, whether or not she was actually raped.

              Most obviously, she is the single person other than the accused who is best suited to evaluate the knowledge, actions, and intent of the accused. And of course she is the single person best suited to evaluate her own actions, speech, and behavior.

              I question a methodology which removes her agency and which insists on making that determination for her.

              The comparison to other crimes is not apt. Other crimes are much less detail oriented and are not as dependent on minute and volatile interpersonal issues, so other crimes are less likely to be misinterpreted by study authors.

    2. Anon Prof

      Having read all the comments by Barry Deutsch, and in an effort to cut through the empty rhetoric, this is what I end up with:

      1. Everyone agrees the CSA’s “1 in 5” statistic is wrong.
      2. Deutsch contends that the .61% (or the 2.44% over four years) is better, but too low because it fails to conform to the CSA’s statistic, which everyone conceded is wrong.
      3. Therefore, the NCVS’s statistic of .61% (or 2.44% over four years) are too low and the prevalence must be higher.
      4. Therefore, we should assume some amorphous statistic lacking any foundation as a compromise between a bad study and a sound study that produced results that are unacceptable to feminists, where we have no clue what the accurate statistic might be, but we must assume rape and sexual assault to be an epidemic because that’s what ideology requires.
      5. And anyone who disagrees with this wrongly guided by political bias, dishonest or ignorant.

      What is special about Deutsch’s argument is how he makes such a preposterous position sound nearly credible, and makes anyone disagreeing with his position appear ignorant with his appeals to authority. Well done!

      1. SHG Post author

        Statistics isn’t a compromise game. They’re either accurate or they’re not. If not, then they must be rejected. While Deutsch claims to concede that the CSA stats are wrong, he really doesn’t, and continues to try to salvage the CSA if only to use it to show the NCVS must be wrong.

        We go with the best statistics available, as long as they reasonably sound, because we just don’t get to make them up to feed our ideology. Well, not Barry, but others.

        1. Barry Deutsch

          While Deutsch claims to concede that the CSA stats are wrong, he really doesn’t, and continues to try to salvage the CSA if only to use it to show the NCVS must be wrong.

          So many lies. First of all, I didn’t concede that the CSA stats are wrong; I said that those who misuse them are wrong. If someone says “1 out of 5 college women are raped at college” that is unambiguously wrong – but it’s also not what the CSA actually says, as I’m sure you know.

          Second of all, I haven’t been referring to the CSA at all in my criticisms of the NCVS.

          1. SHG Post author

            Yeah, I notice you keep moving the goal posts whenever the facts don’t fit the narrative. Got it. Not to be too big a prick about facts, but:

            “He explained to me that 1 in 5 college women were raped…

            Then:

            I agree with you; your friend was wrong.

            So yeah, so many lies. By the way, have you ever considered brevity?

  7. Jeff Gamso

    I gave it all day to see who would point it out, but since it seems nobody did, the job falls to the Cleveland Curmudgeon (not a bad name for a blogger should I decide to start an anonymous one) . Heller’s piece is in the New York Review of Books, not the New York Times Review of Books. My guess is that both the Times and NYRB would prefer that you get that right.

    1. Patrick Maupin

      The intertubes never forget, so anybody posting as the Cleveland Curmudgeon will henceforth be assumed to be one Jeff Gamso.

  8. Pingback: Statistics, With Rigor | Simple Justice

  9. Greg

    NY’s coming version of “yes means yes, until it doesn’t ” can be found buried in the Govs Budget Bill article VII bills, section on education, labor and family; part H at page 118

    Under threat of loss of Title IX funding, colleges will need to develop policies to implement a disciplinary procedure. It is gender neutral, and encourages the accused to speak to the investigors and at the prompt hearing, sans Miranda and counsel. It allows for police criminal investigation to proceed side by side with the colleges. The ‘hearing officer’ is given no burden or standard of proof to apply to the evidence.

    It requires affirmative consent and defines incapacity to consent to include “impairment” by drugs or alcohol. Borrowing the application of “impairment” from DWI litigation – for drugs, it means impaired “to any degree” – a couple of hits off a joint should just about do it; for alcohol, 0.05% BAC – about 2 drinks in an hour.

    All because somebody believed the ‘most often cited stat’.

    1. SHG Post author

      The NY Assembly and governor hate to be made to look like the caboose by California when it comes to cutting edge reforms like this. Whether the Senate will block it remains to be seen.

Comments are closed.