The bottom line is, you made a claim that you're unable to back up. That's not my fault.
What claim would that be?
The claim that there are peer-reviewed papers by feminist which use results from shelter samples to claim that "one in for women are beaten by their loved ones."
Can you cite a peer-reviewed paper which used results from surveying shelters to improperly claim that "one in four women are beaten by their loved ones"?
Not off the top of my head, but I know that they exist. For example, Johnath Archer (2002) wrote peer-reviewed responses to his critics outlining cases such as the ones I describe and you challenge...
I've now reviewed three different papers by Archer - the one you cited, his 2000 paper and his 2000 response to critics - and none of them support your claim.
Like I said, I haven't seen where Farrell has made any false claims about Koss' work, so until you show me something I can't comment on this.
I showed you two clear examples, neither of which you've rebutted in the slightest. I'll repeat what I wrote before:
In Myth of Male Power, Farrell wrote:
A Ms-sponsored study which the mass media widely quoted as saying that 25 percent of all women were raped by the time they were in college used this question to reach the 25 percent figure:
"Have you given in to sexual intercourse when you didn't want to because you were overwhelmed by a man's continual arguments and pressure?"
Two problems with this passage. First of all, the study in question found that 25% of college women have experienced rape or attempted rape at some time in their life; the number for completed rape is closer to 12%. One could argue that Farrell was just repeating how the study was reported in the mass media, but it's irresponsible to do so without also reporting the correct figure. Besides, Farrell clearly attributes the 25 percent figure to the study itself - it is the study, not the media, which (according to Farrell) "used this question to reach the 25 percent figure." But the study never claimed that 25 percent of women have been raped.
Second, and more important, problem: The study never used that question for calculating rape prevalence. (The study did contain that question, but used it only to report instances of "pressuring" - not rape.) Anyone could verify this by reading the study itself (The Journal of Consulting and Clinical Psychology v 55 (2) p. 162-170).
So Farrell, in "criticizing" the study, misrepresents both the study's results and its methodology. Although to be fair, this might be the result of extreme carelessness, rather than actual dishonesty.
Now, you say that you're not interested in Koss because you find her unreliable. But you also said that you found Farrell reliable; therefore, the fact that he badly misrepresented Koss' study should be of interest to you, since it shows that Farrell actually misrepresents what he talks about quite badly.
If what you say is true then indeed it shows that Farrell may have misrepresented the Koss work, however, I need to check it out in context. However, I can readily believe that he may indeed be criticizing the ridiculously common misuse by
feminists of Koss' work, i.e., the "one in four" number, which Koss seems clearly to have refuted in her work.
I'll have to look at
The Myth of Male Power again to verify your claims.
What do you mean, "maybe"? When three non-peer-reviewed anti-feminists agree that Koss' research is bad, you think that's strong evidence; but if other peer-reviewed research and reviews support Koss' conclusions, then that's not strong evidence? It seems to me that's a double-standard.
What peer-reviewed works support Koss' conclusions?
The main findings Dr. Koss made about rape (as opposed to about sexual coercion in general) are: One, that rape was much more common than the official data sources at the time indicated. Two, that a very large proportion of rapes are never reported to police. Third, that rape is usually committed by someone known to the victim, not by a stranger.
Not true. What Koss found was that college women (a very small subset of all women)
report rape 1) more commonly than initially believed; 2) that college women
allege rape in larger numbers on self-reported questionnaires than they do to police; and 3) that college women
report that rapes by intimates are more common than by strangers.
You're ignoring the extremely crucial point that Koss used 1) a very small subset of the general female population (making generalization to the population at large invalid); and 2) self-reported data which is notoriously unreliable unless it has been verfied by independent and objective methods such as physical exams, etc. In other words, we have no idea whether or not the rapes the women reported actually occured. This is a common issue when using self-reported data.
All three of these findings were widely suspected when Koss began her study, but had not been verified with social science research until Koss.
And indeed the Koss study did not verify them either - all Koss did was measure
reporting of rape by college women, not actual rapes. Which is why a baseline group for comparison, e.g., reporting by college men should have been measured as well.
These three findings have since been upheld by every nationwide survey designed to measure violence against women. That, in social science research, is usually the gold standard - if a finding can be repeated, then it should be taken seriously.
So which are these other studies?
* The NIJ/CDC "National Violence Against Women Study" found that 14.8% of American women experience a completed rape at some time in their lifetime. A typical rape-defining question was worded like this: "Has a man or boy ever made you have sex by using force or threatening to harm you or someone close to you? Just so there is no mistake, by sex we mean putting a penis in your vagina."
* The Department of Justice's Sexual Victimization of College Women study included a sub-study in which college women were asked about lifetime incidence of rape (most of the study asked about rape since the beginning of the school year, which isn't directly comparable to Koss). 10% of the women interviewed reported having been raped at some point in their lifetime. Rape was defined as "unwanted completed penetration by force or the threat of force."
* There's also The National Women's Study (NWS), a large-scale national study which found that 13% of American women have been raped in their lifetime. Unfortunately, this study doesn't seem to be available online, but this webpage (written by one of the study authors) includes a lot of info from the survey - scroll about halfway down the page, or search for the phrase "National Research on Rape." The NWS results were published in a number of peer-reviewed publications.
A bunch of other peer-reviewed studies have confirmed Koss' results using non-national samples, but those three are the biggest ones.
All of the studies you cite don't measure actual rapes, they measure reporting. Got any citations that use objective measures, e.g., physical exams, ER visits, etc.?
The researchers I cited in turn used and cited primary data from among others the U.S. govt., which I tend to trust more than feminist ideologues like Koss, et al. So sue me.
"Femnist idealogue" is an ad hom argument. If you had a decent case, you wouldn't rely on ad homs.
Political ideology is germane to the issue, as Archer et al. have shown in their meta-analyses. Investigator bias and conflict of interest are real issues, regardless of whether or not you have the integrity to admit it.
At the time Koss did her study, there were two government sources on rape prevalence (now there are more, which I cited above). One, the FBI's Uniform Crime Reports, counted only cases reported to police, which makes it useless for detecting the prevalence of an underreported crime like rape.
Not exactly - the FBI stats control for such things as false allegations, a common problem in rape crime statistics. We can't
really know that rape is "underreported" using unverified self-reported data. IMO the FBI data is the most reliable data set we have, and it's not even that good because all it measures is arrests, not convictions. Therefore, the FBI stats may actually
overestimate rape because it includes cases that end in acquittals, i.e., the man is proven to be innocent.
Unless of course you take the "women never lie" and "all accused men are guilty" approach to rape.
The other, the NCVS, at the time got their numbers on rape by asking "were you attacked in some other way?" at the end of a long survey about crime - a method guaranteed to produce underreporting. (As well as the obvious problem of not actually asking about forced sex, many scholars - including Archer, whom you cited - are convinced that asking about intimate violence in the context of a crime survey leads to significant underreporting).
There was no other source of primary nantional data at the time of Koss' study, apart from the data gathered by Koss. And although Koss' work wasn't perfect, it was much better designed for measuring rape prevalence than either the NCVS or the UCR.
Nope, Koss work did not measure rape, it measured reporting of rape by college women using her exapanded definitions. As far as I know, there was no validation of the data by independent, objective means to attempt to control for misclassification errors due to false reporting, investigator bias, etc.
(edited once for clarity)