A further thought on all these rape statistics

[This is a reprint of a post from 2002, following up on the previous post, which is also a reprint from 2002.]

In the comments for my earlier post discussing rape, Rob Lyman wrote:

There’s been strikingly little discussion about the fact that this survey was limited to college students, who are NOT a representative population for the country at large. Sure the students were selected randomly, but that just makes the conclusions good for college students, not for all Americans.

We can speculate about relative income and educational levels, but without more data, we can’t generalize from “people who attend college” to “all Americans” meaningfully.

I did say, at the start of that post, that I was discussing Mary Koss’ survey of college men. But in retrospect, I should have emphasized that there are problems generalizing from a college population to a general population.

However, I’m leery that “we don’t know for certain, because we lack data” sometimes becomes an excuse to ignore the data we do have. (This is an “in general” comment, not a criticism of Rob). To my knowledge, only three U.S. studies have used behaviorally-specific questions to ask how many men commit rape, and only one – the Koss study I cited, which surveyed college students – had a national sample. (I haven’t read the other two yet, but I’m told they found similar results). In other words, the Koss study is the best information we currently have on this subject. And considering how under-studied this area is, it may be the best information we ever get.

From an academic point of view, that’s not good enough. One cannot say in an academic journal, “from the data we have, the number of rapists among college men is pretty damn terrifying, which makes it seem plausible that the number among the general male population may be terrifying as well.”

That’s a reasonable point of view – for academic journals. Outside the academic world, however, it’s sometimes necessary to draw the best inferences we can from imperfect data. It is not reasonable or possible to postpone drawing conclusions and addressing problems until perfect data exists, because it is likely that perfect data will never exist.

So I think Rob was right. I have to admit, it’s theoretically possible that men who go to college are enormously more likely to have committed rape since turning 14, than those who don’t go to college. But – speaking as a non-academic – it doesn’t seem likely.

Also in the comments, Ardinger asks “if 4.5% of men are rapists, what percentage of the women you meet are rape victims?”

According to Dr. Koss’ study (which was conducted in the early to mid eighties), about 12% of college women have been victims of completed rape at some point since age 14.

This result of Koss’ study has been frequently criticized by anti-feminists. But at least three other nationwide studies of lifetime rape prevalence came to similar conclusions; the National Women’s Study found 13% (not available online, sorry), the Centers for Disease Control study found 14.8%, and the Bureau of Justice Statistics study of college women found 10%.

I don’t think there will ever be a single number that people can point to and say “this is the right answer.” Unreported rape is inherently difficult to measure; there will never be a study that someone can’t find reason to doubt, and every new study will raise new questions. But from the best studies currently available, somewhere between 10% and 15% of American women have been raped in their lifetimes.

My conclusion: Maybe the “real” number is that 2% of men commit rape sometime in their life, and 8% of women are raped. Maybe it’s more like 8% and 20%. We’ll never know for sure. But from the data that’s currently available, we can say this: Rape is a scary, serious, widespread national problem. It is not something committed by a freakishly small minority of men (unlike, say, serial killing); it is not something that happens to a small number of women.

Feminists want a society in which rape is rare (or nonexistent), and rapists are freakishly unusual deviants. But the first step in building that society is realizing we’re not there yet; we’re not even close. We’ll never change if we can’t even admit the scope of the problem.

This entry posted in Mary Koss controversy, Rape, intimate violence, & related issues. Bookmark the permalink. 

10 Responses to A further thought on all these rape statistics

  1. 1
    Kerim Friedman says:

    I generally agree with this post. But I do think it is important to ask about how College students might differ from the rest of the population. For instance, there are specific sub-cultures related to varsity sports and fraternities which seem to encourage a culture of sexual violence. I think we could learn a lot by looking at what portion of college rapes occur within these subcultures. That would tell us if this is simply a pervasive pattern of male behavior in our society, or if it is something that only thrives within specific cultural settings. It would not tell us what cultural settings it might thrive in outside of College campuses, but it would be a start. Perhaps I am an optimist, but I don’t personally believe that rape is “natural.” I believe that these subcultures of violence have a tremendous facilitating influence upon the percentage of rapes. I call this view optimistic because it means that the problem can be managed by policies which seek to change the behavior of such subcultures, or else eliminate them altogether.

  2. good pair of posts. tiny quibble – It is not something committed by a freakishly small minority of men (unlike, say, serial killing)
    unless you are using serial killing in some technical sense, i don’t think it’s so freakishly small, -if- we include those whose killing is state sanctioned (soldiers etc.) The Milgram experiments showed that people will try to hurt other people, when told to by an authority figure, who would never do so on their own.
    another quibble – there’s nothing too reliable about the 4.5% figure. college students have taken enough tests to know a question with “ever” signals a yes. “Have you ever been to mars?” 4.5% will answer yes. But there are other factors that suggest the 4.5 is too low, or about right.
    The exact definitions of what it is to rape or be raped will affect the percentages. A narrower definition gives a smaller number, a broader def gives a higher number. For axample, i would say that i have been sexually assaulted, but not raped. But stuff has happened to me that would meet some people definitions of rape. And i’ve done some stuff that might look like rape without the right follow up questions – I held him down while i boinked him, after he’d suggested we do it that way. Easy example to talk about openly because there was clearly negotiated consent. In my experience, the numbers look like 1 in 3 women, 1 in 5 men, get raped at some point (north america, no stats on global). I don’t have any estimate of how many men rape, partly because i’m not sure how i would define the term. I can’t say that 4.5% is wrong.

  3. 3
    Morphienne says:

    Um, so 4/5% of people said they had been to Mars? I must have missed that IMAX… damn.

    Amp, please catch me if I misinterpret the parameters of Koss’ study, but I thought that the only things the study counted as raped were penetrative intercourse WITH AN UNWILLING PARTNER attained by physical force or violence or the threat of physical force or violence. Did I not read it correctly? Because that would seem to “count” as rape to me, and I feel that that’s a VERY narrow definition of rape, which means, by aardvark’s reasoning, that it underrepresents the number of rapes and/or sexual assaults committed.

  4. 4
    Random says:

    Examining rape in the context of college life is equivalent to examing rape in prisons or the military. All are examples of institutions, and may have only passing similarities to life on “the outside.”

    As a former college student at both resident and commuter colleges, I suspect that the culture of a residential college in particular promotes a variety of borderline behavioural traits. Kids (young adults?) are on their own in an environment that encourages experimentation and minimizes potential repercussions. Mix with a smidgen of rebelliousness (or more than just a smidgen…) and you have a heady cocktail that can send the wrong message to a person trying to discern the rules and mores that bind a society.

    I therefore suspect that the incidence of rape at colleges is not indicative of this aberrant behaviour in the populations at large. Now if you were to examine this sort of behaviour in the context of a military organization… Yep, like the youngsters currently serving in Iraq; there you might find a different percentage of aberrant (abhorrent) behaviour.



  5. 5
    Echidne says:

    Amp, do you know of any studies done elsewhere in the world? It would be interesting to compare U.S. data to that from other countries. All I recall is seeing statistics that indicate that reported rape figures are higher here than in most European countries, but that may be older data. I think awareness in Europe still lags somewhat behind the U.S., though mostly in the Southern and Eastern European countries.

  6. 6
    Ampersand says:

    Random: Keep in mind that Koss (if it’s Koss your talking about) did a study of lifetime incidence of rape among college students, not a study of rape at college. That is, although the people surveyed were college students, they were asked about all incidents of rape (and other things) up to that point in their lives, not just things that had happened to them since they began attending college.

    Also, the CDC rape study I’ve linked to a couple of times this week has a representative sample of the entire USA, not just college students.


    I do have some info on that, but I’m too burned out right now to look it up and type it in. I’ll try to remember to do something on it later, but feel free to remind me if I forget.

  7. 7
    Sheelzebub says:

    Just a note about stats, I’d be a little skeptical of “officially reported via law-enforcement” stats.

    When I lived in Japan, I was told that the rape rate was low. What was low was the actual reportage of rape; women who come forward are put through the grinder. They tend to keep quiet about it because they are blamed.

  8. 8
    Spicy says:

    ‘Amp, do you know of any studies done elsewhere in the world? It would be interesting to compare U.S. data to that from other countries. ‘

    Here’s a link to one in the UK showing an incidence of 4.9% for rape and 9.7% for sexual victimisation for women since the age of 16:


  9. 9
    curiousgirl says:

    one thing to point out. College students are SIGNIFICANTLY YOUNGER than the average poplution. All other things being equal, this would create underreporting of lifetime rapes committed by the average man, i think significantly.

  10. 10
    AhClem says:

    Last night, contemplating after a friend related to me that her friend had been raped the night before that, it dawned on me that rape statistics rarely focus on the number, or percentage, of men who commit rape. Instead, we are given the number or percentage of victims of rape. How can we solve the problem if we only focus on the symptom?