This should be free if you haven’t used up your quota of free articles for the month:
Here’s a link to the critical article, whose principal author is Kate Carey, a researcher at Brown. The NYT piece mentions it in connection with discussing the Canadian study: http://i2.cdn.turner.com/cnn/2015/images/05/20/carey_jah_proof.pdf
The critical paragraph, somewhat edited by me to make the alphabet soup intelligible, and to incorporate numbers from the table, appears below. Note that “rape” as used in this paragraph always means the sum of attempted and completed rapes:
The study involved over 460 women at a university that seems to resemble Brown.
Here’s [the Krebs study.](https://www.ncjrs.gov/pdffiles1/nij/grants/221153.pdf) As I read the study, they looked at each of the factors (alcohol use, prior rape, going to fraternity parties, race, etc.) in isolation, and did not control for the others when computing the predictions for each one.
The study says that a woman who (says she) had been raped by alcohol/drug incapacitation in high school is more than 6 (six) times as likely to be (say she was) raped by alcohol/drug incapacitation in college. That’s a big, big increase. They don’t seem to have even gathered data about whether the study participants drink to incapacitation. That is, looking at the Krebs study I can’t tell whether a propensity for drinking to incapacitation fully explains why the students raped by incapacitation in high school were much more likely to be raped by incapacitation in college.
A few comments on the study linked in my prior post:
- The overall incidence of experiencing attempted or completed rape reported in the study is astounding: over 37% by the end of the first college year. (And I should clarify that, on rereading the paper, I fear I was somewhat misleading above. The figures include incidents during both freshman year of college and the ensuing summer.) In every case, attempted rapes significantly outnumber completed rapes. That is especially true for women reporting their pre-college experience, both for incapacitated rapes and forcible rapes. For incidents reported during college, the rates are much closer.
However the percentage of women reporting one or the other is always far less than the sum of the percentages of women reporting either attempts or completions, and generally is only a little higher than the percentage of women reporting attempts in each period. That suggests to me that women who report that they are victims of a completed rape are also very often reporting that they are victims of an attempted rate, in each category, and for each period examined (which during the study is essentially a semester).
- Similarly, but not as strongly, the data in the study indicates that a meaningful percentage of women -- about a quarter of those reporting some experience with rape -- reported being confronted with both incapacitated rape and forcible rape over their lifetimes. The ratio of both to either-or-both is only about 20% for pre-college and first-semester reports, but then goes to about 25% for experiences with multiple rape types within a single semester-long period. It isn't clear from the paper, however, whether the same incident might be characterized as both an incapacitated rape and a forcible one. I can imagine lots of circumstances that could fit both categories.
- Incapacitated rape incidents outnumbered forcible rape incidents by a considerable margin: about 5-1 in the study.
Pretty much by definition, only women who drink (or use drugs) to a point affecting capacity can be victims of attempted or completed incapacitated rape (putting to one side, I suppose, women whose non-alcoholic drinks may have been “roofied,” or spiked with a drug like rohypnol, and who are aware of that). It would have been interesting to know the comparative incidence of rape and revictimization as between drinkers and non-drinkers.
I imagine that women who drank to incapacity in high school may well be somewhat more likely to drink to incapacity their first year of college, compared to women who didn’t. That alone has to put them more at risk for incapacitated rape than other women.
- The survey on which the paper is based does not ask women to characterize what happened to them as attempted or completed "rape." Instead, as part of a larger health survey the participating women were asked about a range of various "seduction" techniques, from verbal argument to physical force to threats to engaging in sexual activity while the woman is drunk, and then to report on the outcome of that over a range of sexual practices from touching or kissing to intercourse and beyond. The paper is fairly clear what it means by "completed" rape -- oral, vaginal or anal penetration by something -- but it is completely unclear how an "attempt" is identified (other than by one response option that seems to have been something like "tried to have intercourse with me but it did not happen"). And as far as I can tell, it ignores the element of consent in the case of incapacitated rape.
So there’s a real possibility that this paper is coding “My boyfriend and I got drunk and had sex” as a completed incapacitated rape, and " We got drunk and wanted to have sex, but my boyfriend fell asleep before we got anywhere" as attempted incapacitated rape. That seems less useful than one might hope. It certainly helps explain the high incidence of women experiencing rape, and the high rate of multiple experiences. It also allays my concern that the study was to some extent measuring women’s willingness to label a marginal situation as rape. But it undercuts, at least for me, the meaningfulness of the data on prior experiences predicting future ones.
That’s a pretty broad definition of seduction. There’s a difference between “you’re so beautiful,” “come on, it would feel so good,” “you know you want it,” and “you’re not leaving this room.”
More recent studies use what I think is a better set of questions, the [SES (Sexual Experiences Survey)]( http://www.midss.org/content/sexual-experiences-survey-short-form-victimization-ses-sfv). The SES asks participants whether they’ve experienced different sexual specific situations, not characterizing them as assault or rape or anything like that. Here’s an example question:
For each of (a)-(e) in the questions, the survey participant answers how many times it happened in the last year (0, 1, 2, 3+), and how many times it happened in their lifetime. This survey thus distinguishes, for example, between the case where the assailant forcibly removes the person’s clothes, but then does not go on to penetrate them, and the case where the assailant forcibly removes their clothes and then rapes them.
It would seem to me that c, d, and e are what a college and/or police should concern themselves with.
The Sexual Experiences Survey as revised in 2010 (or thereabouts) is what the study used. And, yes, only actions associated with c, d, or e were considered “rape,” whether attempted or completed.
Sorry, JHS, I though you were talking about the Krebs study in #403. I now understand you’re talking about the study by Carey et. al., that came out last year. When we have talked about that study before, we’ve called it the Syracuse study because many of us thought the “large private university” it was done at was Syracuse, the affiliation of several of the authors.
But then I don’t quite understand this:
When the questionnaire asks about attempted rape, it clearly indicates that it’s asking about failed attempts only, not successful attempts. “Even though it didn’t happen, a man tried to put his ***** in my vagina…” (CC censors, but the survey used the real word for the male organ)
It may have been Syracuse. Two of the authors were at Brown, and one at an institute in Syracuse, but it occurs to me that Carey came to Brown from Syracuse. So it could well have been Syracuse where the actual study was done.
I can’t imagine why any difference between Brown and Syracuse would matter much to anyone looking at this study, in any event.
We on CC had this long conversation about the Syracuse study before, and I wasn’t realizing we were talking about the same thing now, that’s all.
I looked at the study you linked to. There are a few multivariate regressions reported in Exhibit 5-6 (see commentary pages 5-7 to 5-14). These allow you to look at the influence of a factor while trying to control for the other factors.The analysis divides victims of completed sexual assaults into three categories - those for whom the sexual assault(s) were enabled by alcohol and drugs, those for whom they weren’t, and those who were victims of both types.
Even after controlling for factors like drug and alcohol use, number of male sexual partners, demographics, etc., being a victim of forcible sexual assault before college increases the odds of being a victim of forcible sexual assault (only) in college by a factor of 6.58; being a pre-college victim of a sexual assault while incapacitated increases the odds of being a victim of an alcohol/drug enabled sexual assault in college by a factor of 3.58, etc. This partially addresses your question, but it doesn’t distinguish between using alcohol and drugs and using them to incapacitation (see also commentary on pages 2-5 and 2-6, which summarizes the findings of other studies that have addressed similar issues).
From Exhibit 5-6, being threatened/humiliated/physically hurt by dating partner is a very strong predictor of or correlator with sexual assaults too. I wonder if a lot of these sexual assaults are being perpetrated by the women’s “boyfriends”?
Exhibit 5-1 states that 11.3% (weighted) of women in their sample reported a completed sexual assault before college. 13.7% reported a completed sexual assault while in college. It looks like the women had been in college for a little less than 2 years on average when they answered the survey, but on the other hand I (hopefully?) wouldn’t expect girls younger than 16 to be going to many parties where there’s a lot of drinking either. It’s hard to tell, but the college environment or being away from home isn’t as big a factor as I might have thought … I guess it’s a risk that just comes from being a 16-21 year old woman.
OK, I see now that I misunderstood Table 5-6. It’s a result of multivariate logistic regression.
OTOH, the girls under 16 who go to parties with a lot of drinking probably are at a huge risk of sexual assault. It could well be that the risk of sexual assault per year for girls age 14-17 does not rise with age. Or maybe it does, but I don’t think that’s a safe guess.
Maybe this has been addressed already…in the gazzilion pages of posts. But wouldn’t you tell your daughter the same thing as your son?!
I would not necessarily have the same conversations about consent with a daughter as with a son. In the world we live in, the issue for men is more usually getting consent, not getting consent, being accused of not getting consent when they did. The issue for women is giving consent, not giving consent, being said to give consent when they didn’t. Rape victims and false accusers are overwhelmingly women. Rapists and people who are falsely accused of sexual assault are overwhelmingly men.
“Don’t have sex unless you’re in a committed relationship” and “don’t have sex if either person appears to be under the influence of alcohol or drugs” seems to cover most of it.
CCardinal Fang, I didn’t realize there had been a big discussion of that paper before. I haven’t seen the actual survey, obviously you have.
My impression from the paper itself is that things like oral sex and digital penetration, as well as good ol’ fashioned intercourse, would all be coded as completed rape if associated with force, threat, or incapacitation. What was completely unclear to me is how “attempt” was identified, if it was associated with incapacitation. Does the survey ask “attempt” questions for each of the different kinds of sex act?
Also, it seemed to me that the same event could be counted as many events in the survey, e.g., a session involving both oral and anal sex. Or a session involving oral sex and attempted (but not completed) vaginal sex. Do you know how that is handled? (Just speculating here, but based on popular literature and a dim memory of my own past, up to and including a few weeks ago, sexual encounters may not always be just one thing.)
Finally, I am somewhat puzzled by the relationship of “[t]aking advantage of me when I was too drunk or out of it
to stop what was happening” and attempt. I understand what it means if someone says, “He tried to force me to have intercourse with him without my consent, but I resisted and he gave up.” I’m not sure what it means if someone says, “He tried to have intercourse with me while I was too drunk to stop what was happening, but he gave up.” Did he give up because, in fact, it turned out I wasn’t so drunk or out of it that I couldn’t say “no”? Did he give up because I passed out and he realized it was wrong? How do I know what happened when I passed out, anyway?
Generally in criminal law, I think, “attempt” and “abandonment” are mutually exclusive. If I am trying to commit a crime, but think better of it and stop, I am not guilty of attempt to commit that crime. If, however, I try to commit a crime and am prevented from doing so by external forces, then I am guilty of attempt. (Obviously, there are many circumstances where it may be hard to tell one from the other, so this is always a contentious area of the law at the level of detail.) If Person A tries to force himself on Person B, and Person B yells and resists, and Person A then stops after struggling for awhile and leaves, that’s clearly an attempted rape. If Person A pins Person B down and thinks about forcing himself on Person B, but thinks better of it and lets Person B go, that’s not an attempted rape, although it may be a completed battery.
Anyway, the survey is producing a whole lot of attempted incapacitated rapes, and I’m not certain what that means in the real world if the victim-designate is really too drunk to stop what is happening. If the other person is stopping by his own volition, it seems misleading to characterize that as an attempted rape. If the victim is stopping it, that seems inconsistent with the premise of the question. And if someone else, or something else is stopping it . . . well, I’m not sure what that would be – a Neighborhood Watch foot patrol, maybe? – but it sure seems to be happening alot to the women who took the survey.
For those of you who know the survey, is there any real possibility that a woman who would describe her experience as shown would reflect it on the survey in a way that would be scored as a completed or attempted rape?:
“We met at my sister’s party. It was lust at first sight for both of us. We got completely drunk, came back to my room, and had sex.”
“We got completely drunk. He was trying to get me to have sex, but I said no, I don’t feel so good, and he said OK and stopped.”
“We got completely drunk. He started trying to have sex, but I didn’t feel good, so we didn’t. But I went down on him so he wouldn’t be disappointed.”
“I was so drunk that first night, I couldn’t even remember having sex with him. I had been chasing him for weeks, and when I finally caught him I almost blew it! I’m glad he stuck around long enough so we could do it again for real the next morning.”
Here is the [link to the actual SES survey.](http://www.midss.org/content/sexual-experiences-survey-short-form-victimization-ses-sfv) You’ve got to download the pdf to look at it. The survey is longish. It asks about a laundry list of various sexual situations (someone tried to remove my clothes without my consent but didn’t attempt penetration, someone attempted anal penetration without consent but didn’t succeed, someone had oral sex or made me have oral sex without my consent, someone inserted his ***** in my vagina without my consent, etc.). For each of the items on the list, the respondent has to say how many times someone did it by
So for example if I take the survey I have to answer specifically how many times in the last twelve months someone has tried to put his **** in my vagina by threatening to physically harm me or someone close to me, and so forth for the cross product of the sexual situations, the five methods and last twelve months/lifetime.
For the four scenarios JHS presents, the relevant words in the survey are going to be “without my consent.” For the second one, where the guy is trying to get her to have sex while they are both drunk, but takes no for an answer, there is no way that the survey participant is going to answer yes to the survey question. His verbal pressure did not continue after she said no.
For the other three scenarios, where the two people had apparently consensual sex in a situation where both of them might have been deemed to have been too drunk to consent, we have to consider whether a respondent would say that the guy “took advantage” of her when she was too drunk to stop what was happening. I don’t think she’d say any of those guys took advantage of her. It was lust at first sight before they got drunk, she gave him the blow job, she had been chasing him for weeks. Took advantage? I don’t think women would answer that way.
But it’s not always easy to know how survey questions get answered.
The survey will get arguably questionable positives in situations where while sober she would never consent to sex with him, but she drank a lot and agreed to the sex. She might say that was without her consent, whereas we would say she consented. There’s no agreement on how drunk is too drunk to consent.
287. "I'm gonna go with option C. We should focus resources on training men to not rape."
Not necessarily, unless you are primarily interested in apolitical rather than practical outcome.
I’d suggest that we should devote resources on only those things that are proven by the data to be the most EFFECTIVE.
If that means training guys, so be it. If that means training girls, so be it. If that means training both, so be it. If that means spinning up a big Dear Colleague/title ix campaign, so be it.
But if the data says that any of those things are ineffective, then we should stop doing them.
My hunch is that the data would say that training girls is the most effective. And that the title ix effort will prove least effective. But the data (rather than hunches) should determine.