the hunting ground and the university as organization

The Hunting Ground is a new documentary about the epidemic of sexual assaults on college campuses. It’s also a story of the birth of a new social movement targeting universities using innovative legal tactics alongside traditional organizing and protests. And in a gut-wrenching way, it’s a story about universities as organizations, and the dark side of the organizational imperative for self-protection and survival. In addition to being important for faculty to see in their roles as advisors, teachers, and participants in university governance, I think the film will also make an excellent teaching tool in a range of sociology classes on anything from sex and gender, to social movements, sociology of law, crime and deviance, and especially organizations. In theaters now!

And please use the comments as a place to discuss your thoughts on the film.

Edit: Comments are now closed.

Author: Dan Hirschman

I am a sociologist interested in the use of numbers in organizations, markets, and policy. For more info, see here.

33 thoughts on “the hunting ground and the university as organization”

  1. I’m not sure that any documentary is a good teaching tool in isolation. I think that it’s worthwhile to present students with critiques of resources, especially resources that take a strong point of view. It thus might be a good idea to have students read this critique of the film by Emily Yoffe at Slate. I’d be interested in hearing thoughts on whether her criticism is incorrect.


    1. Thanks so much for the link! I’d been looking for critical commentary and hadn’t hit on much yet.

      I should also clarify that I meant teaching it as part of a unit looking in more detail at one of those themes (how social movements target organizations, how actors find novel uses for existing laws, how universities respond to challenges and their resource dependence on different kinds of status, etc.). As with any primary source, it requires contextualizing (both theoretically and empirically). I’ll respond to the Slate piece separately.

      Liked by 1 person

    2. So, in re: Yoffe’s criticism, in no particular order…

      1) “The film traffics in alarmist statistics and terrifying assertions, but fails to acknowledge both the recent changes in the way the government and universities approach sexual assault charges and the critiques that those changes go too far.” The statistics that the movie cites are, to my knowledge, reasonable and accurate. You could maybe quibble with the use of stats that compare the number of sexual assault claims to the number of students expelled.. but only because we know the number of assaults reported is a dramatic undercount, so that ratio actually makes the schools look *more* proactive (e.g. if they say University X had 100 reported assaults in the last 10 years and expelled only 5 students, you get an impression of 5%.. but actually the number is lower because universities make it hard to report and survivors have incentives not to report in the first place). But Yoffe’s ‘alarmist statistics’ critique is, as far as I know, wrong.

      Elizabeth Armstrong and Jamie Budnick (both of whom I watched the documentary with) have a brief coming out soon at CCF that summarizes these findings and while they do fine a lot of debate about the 1-in-5 number, most sources agree that it’s reasonable. The percentage falls to something like 7-10% if you only count ‘forcible rape’, but for the broader definition of sexual assault (as used by the Bureau of Justice Statistics), the numbers range from 14-26% or so. Yoffe bases her critique on studies citing the National Crime Victimization Survey, which have been determined to radically undercount sexual assault for a host of methodological reasons, to the point where an entire volume was put out by the National Research Council calling for new measurements. It’s Yoffe, in other words, who is behind the times.

      2) In terms of not responding to the most recent changes – I don’t know what she’s talking about. If she means something that happened in the past year, it’s a bit unfair – movies take time to make. If she’s talking about some major development before that – beyond simply the big expansion in Title IX investigations by the Department of Education, which is a major subject of the film – I don’t know what it is.

      3) Yoffe cites statistics from a higher ed insurance company about the percentage of sexual assault claims that end in a punishment of expulsion or suspension. Here she seems to be cherry picking: the claims that end up at the insurance company are surely much more egregious than average, and so they are not comparable to broad claims at a school level. In other words, it’s totally consistent that 305 claims to the insurance company, you would have ~35% expulsions, and yet for the population of assault complaints as a whole you’d have a rate much, much lower. In fact, it would almost have to be that way.

      4) The lack of mention of the “Dear Colleague” letter and the circa 2011 push by Obama administration to take action on sexual assault at colleges is a fair critique, and one that was made by the experts I watched the film with.

      5) “The Chronicle of Higher Education notes that “a shadow justice system” is being set up on campuses, with schools hiring new staff—often lawyers—to investigate and adjudicate sexual assault cases and that law firms are offering their services as freelance finders of fact, at a fee of $5,000 to $20,000 per investigation… But The Hunting Ground does not address the changes OCR has wrought in the ways campuses handle sexual assault accusations; instead it declares there are virtually no resources for victims.”

      Part of the point of the movie – and one of the reasons why I think it makes a useful organizational case study – is to show how all of the bureaucratic systems put in place are not designed to provide resources for victims, but rather to protect the institution. Universities responding to new pressure by the federal government and to activists (who Yoffe gives no credit to in explaining why universities are taking this seriously, or why the federal government is continuing to pursue the issue) by hiring new staff to investigate complaints makes sense in this context. The calculus has shifted: it’s not as easy to bury cases as it once was, and the costs of doing so are higher. But hiring lawyers to investigate claims is not the same thing as providing resources for victims, in terms of emotional, academic, and legal support. Note that the staff mentioned are investigators and adjudicators, not victims’ advocates.

      6) “The Hunting Ground never even makes a feint at acknowledging that dozens of young men like John Doe have filed similar lawsuits, saying they were deprived of an education over dubious charges.”

      I think the movie does tackle the issue of false reports in a reasonable fashion (acknowledging that they occur but contextualizing their size both relative to the overall magnitude of the problem, and other crimes). But asking a documentary tackling a large issue to focus on a small number of exceptions is just a form of sealioning.

      7) A critique Yoffe sort of alludes to, but fails to really make, is that the film is weak on solutions. It’s mostly a call to action and awareness rather than an analysis of what policies might succeed. This leads to Yoffe’s slippery slop critique:

      “This definition means that if a male Columbia University student persuades a Barnard student to make out with him, and she later feels she was pressured, she could file a complaint that leads to his removal from school.”

      I don’t think the film argued that all individuals found responsible for any form of sexual misconduct should be expelled – the example Yoffe uses is extreme, and the film tends to focus on the most violent assaults, or those where the victim is drugged or otherwise incapacitated where expulsion is much more clearly warranted. Given the lack of engagement with an array of possible solutions, we’re not given a strong sense of how the filmmakers want universities to handle a broader array of cases. But I can see how Yoffe – already so inclined to distrust the film based on her misreading of the data – would take issue here.

      In sum, I think Yoffe’s critique is sloppy if not disingenous, but touches on a few places where the film (rightly or wrongly) is a bit narrower than it otherwise could have been.

      Liked by 2 people

      1. Hi Dan,

        Thank you for the detailed comment about Yoffe’s post.

        From what I can tell, the volume that you linked to discusses only *potential* sources of error in the National Crime Victimization Survey. What is the basis for the claim that the NCVS radically undercounts sexual assault, other than the fact that different surveys give higher rates?

        Part of the critique of the NCVS in the volume that you linked to is that “it does not measure the low incidence events of rape and sexual assault with the precision needed for policy and research purposes” (p. 5); that criticism makes sense if the NCVS estimate is in the right ballpark, but the criticism does not make much sense if the true rate of rape and sexual assault on college campuses is 20%. The Digest of Education Statistics reported that there were roughly 12 million women enrolled in degree-granting post-secondary institutions in 2012, so that would make 2.4 million victims of rape and sexual assault, just among women in college.

        For comparison, here is a link to crime statistics for crimes reported to the University of Pittsburgh campus police, university officials, or the local police. Pitt has roughly 18,000 undergraduates and 10,000 graduate students, and there were 19 reports of on-campus rape from 2011 to 2013. Those statistics seem closer to what we’d expect based on the NCVS than on the 1-in-5 statistic. (See this link for more data on Pittsburgh universities.)

        The NCSV estimate for the rape and sexual assault rate was lower than I expected, so I have no trouble believing that the NCVS is an underestimate. But the research design for the NCVS seemed of higher quality than the other reports that I have read about, so I don’t have another estimate to replace it with.


      2. Here I’m going to have to leave it to the experts for the full explanation – I’ll link to the Armstrong and Budnick piece when it’s posted, as it deals with this exact topic. My quick summary is that the NCVS doesn’t define rape or sexual assault, and many women do not apply those labels to their experiences and thus respond negatively (no, I have not experience rape/sexual assault), but when presented with survey questions that offer definitions answer yes. Also, the NCVS surveys whole households at once, and so there’s not a lot of protection for respondent confidentiality. In other words, the similar forces that keep victims from reporting to authorities keep them from reporting their experiences in the survey. Put differently, the NCVS respondent environment is more like the reporting-to-authorities context and so it’s not surprising that the two rates would be similar: both require victims to self-classify as victims of rape/sexual assault and be willing to share that experience publicly. But that’s a smaller subset than we are interested in.

        Again, though, I should really punt to the experts. Will report back.

        Liked by 1 person

      3. Ok, last bit because I’m poking around myself now. Here’s a 2012 paper from the National Academies conference: Measuring Rape and Sexual Assault: Successive Approximations to Consensus.

        There’s a lot of detail in there about question wording and reporting time periods and how they can affect estimates. There’s also an issue in that a lot of the surveys report victimization rates for six-month or 1-year periods, not lifetime prevalence, so some of the stats are apples-and-oranges (the 1-in-5 number is not a yearly stat, but usually framed as something like by the time they leave college). p. 17 of the linked pdf reports some of the higher estimates, from surveys that asked more detailed questions about a broader range of assaults (including ones that take place when the victim is unconscious/passed out – a common theme in the film, but not included in the NCVS).

        One survey experiment actually compared the question wording differences and found a huge difference:

        “behavior-specific wording produced estimates over 10 times larger than the survey relying on NCVS question wording.” (p. 21)

        That’s the debate, in effect. Read the whole paper for more details on why the NCVS measures are pretty restrictive.

        Liked by 2 people

      4. Thanks for the link, Dan. What a mess of statistics. Perhaps it’s representative of this research area that one of the studies extrapolated 24 unweighted cases into an estimate of 876,064 rapes (p. 12-13); not 900,000 or 880,000 or 876,060, but 876,064.

        For what it’s worth, the survey experiment appears to be a quasi-experiment, so there are other factors uncontrolled for. But it’s believable that specific items provide higher and presumably more accurate estimates.

        Liked by 1 person

  2. UNC hasn’t had the best track record over the last five years responding to scandals, but our administration has sent out two mass emails encouraging everyone to see the Hunting Ground. According to our Chancellor, the film “shows how the activism and courage of students across the nation, including two of Carolina’s former students, have helped drive change in the way colleges and universities respond to reports of sexual violence.”

    Liked by 1 person

  3. Has anyone replicated David Lisak’s survey? He conducted it on a commuter campus and got a gigantic age range of men to self-report their rapes. But two things are striking.

    (1) Men have no problem attesting their rapes. Lisak explains that these men are proud of or feel justified in their actions. Collin’s book on violence and now Fiske’s book on violence corroborates the idea that perpetrators of violence feel like victims who are settling scores and restoring justice.

    (2) The distribution he generated was exponential. That is, a tiny minority of the men he surveyed committed multiple rapes, and accounted for (I think) just over half of all the rapes he counted.

    Lisak’s survey should be replicated for a cross section of institutions, and victims and their rapists can be matched with one another in aggregate. This will tell us if rape is a product of a widespread culture that confuses young men about consent, or whether a small minority of men feel like it’s their right to get theirs after their own (warped) boundaries have been violated by women who refuse sex.

    Overall, the campus rape issue is a moral panic. It emerged, not coincidentally, alongside the fervor over campus hookup culture, which does not exist. The distribution of numbers of sexual partners that college kids take has not changed substantially in 40 years. Age of first intercourse has *gone up* and it remains that the vast majority of people have a handful of partners over the life course.

    What has changed is the dating ritual. The sex comes first and the romantic courting comes later — and that seems to be what leads parents and feminists to fear that women are newly exposed to the risk of sexual violence. But they aren’t.

    In fact, given that traditional courting rituals have dropped, where parents, churches, friend groups, and other community interests no longer rubber stamp matches on the front end, young women and men are more liberated to match across traits they value without logrolling their own preferences with those of their communities. This is evidence of the liberation of women, expansion of individuality and freedom of choice, and degeneration of ascriptive kin ties and social control over balkanized communities.

    The UVA Rolling Stone debacle is the best evidence that this issue has gotten completely out of hand. I fail to see how a documentary with ominous music will improve discussion or raise standards of evidence in the classroom, journalism, or published research.

    Liked by 1 person

  4. Dan,

    This is an important point: ” all of the bureaucratic systems put in place are not designed to provide resources for victims, but rather to protect the institution.”

    Administrative bloat might be going in the direction opposite to the one we would like it to in this case. Providing more deanlets and staffers may in fact lead to a more coordinated effort to protect the institution rather than victims.

    But you also say, “hiring lawyers to investigate claims is not the same thing as providing resources for victims, in terms of emotional, academic, and legal support. Note that the staff mentioned are investigators and adjudicators, not victims’ advocates.”

    As I understand it, advocacy, mental health, and disability resources have expanded, along with the capture and redirection of rising tuitions, by administrators, to salaries, centers, and amenities.

    I think an ideal solution is to divert funds away from kangaroo courts and into mental health and academic support, who can competently and professionally address trauma on a personal level for victims (instead of professors and student groups moonlighting in these roles), while campuses otherwise refer the criminal matter and decision of guilt to the courts.


    1. A couple quick responses.

      1. “As I understand it, advocacy, mental health, and disability resources have expanded, along with the capture and redirection of rising tuitions, by administrators, to salaries, centers, and amenities.” I hope this is true. I’d like to see data. The things Yoffe pointed to were about investigators, not mental health or advocates.

      2. “while campuses otherwise refer the criminal matter and decision of guilt to the courts.” As the documentary persuasively argues, this isn’t an either/or. The standard of proof for kicking someone out of school (or suspending them, etc.) is lower than for criminal prosecution – and that’s probably a good thing. Schools can and should take action alongside the police and prosecutors, and in fact have a (Title IX) obligation to do so to the extent that failing to do so creates a hostile environment for (mostly women) survivors of assaults on campus.

      We don’t tell people not to sue in civil court just because there is also a criminal proceeding, or because a criminal prosecution lacks sufficient evidence. I think something similar applies to university disciplinary hearings. [Which is not quite the same as saying whether some universities have shifted the burden too far – a possibility, but one that I have not yet seen a lot of compelling evidence for.]


  5. Epidemic: “is the rapid spread of infectious disease to a large number of people in a given population within a short period of time, usually two weeks or less.”

    There is no campus “rape epidemic”. What we have is an epidemic of “moral panic”. A sociologist worth his salt would investigate the social causes of this particular episode of moral panic instead of fueling it.

    Liked by 1 person

  6. When I showed up on campus in 1991 events like Take Back the Night were contentious, but already institutionalized. In other words, the fight against sexual assault on campus has a pretty long history, and I suspect is strongly tied to the growth of organized feminism on campus during the 1970s. In addition to other organizational and institutional factors mentioned above, one major difference this time is that campus activists have a bureaucratic ally in the Department of Education that they have been able to successful leverage. I’m pretty sure standard org/movement/political theory explains things quite well, and certainly much better than largely descriptive theories that state that sexual assault an exaggerated threat to the social order.

    Liked by 1 person

    1. BJS numbers on sexual violence may suffer extraordinary measurement bias, but that bias washes out as long as the methodology stays constant over time. The time trend on sexual violence shows that it has declined secularly for a couple decades.

      If that has happened while a discourse that self-consciously warns against the large and growing risk of sexual violence enjoys rapid diffusion and adoption, I think that’s pretty good evidence of moral panic. Also, if that discourse better correlates to a cascade in discourse about campus hookup culture than it does, say, the Department of Education stepping up enforcement, I think that’s also pretty good evidence of a moral panic.

      And those claims are testable with text analysis by computational movements scholars.


      1. “but that bias washes out as long as the methodology stays constant over time…” No, it doesn’t (necessarily). Suppose there are two types of events, with different characteristics, one of which is mostly reported to the BJS and one which is mostly not. Changes in the BJS data tell us *nothing* about trends in the second type of event. Trends in the BJS data are only meaningful if we assume that they are biased only in magnitude, but are otherwise capturing a random sample of events that would be captured by a broader measure. But that’s exactly what we think isn’t happening – the BJS data are not just undercounting, but are excluding a large class of events.


    2. Agreed. It’s a classic social movements thing. The problem (grievance) existed for a long time (at least since the 1970s?). Awareness among set of actors existed as well. So explaining the public success of the movement involves looking for changes in the resources, political opportunities, etc. Hence a useful teaching case – one that’s contemporary and close to students, but which actually follows patterns we’d largely expect.

      And I think there’s a similar story to be told about the use of Title IX filings as a kind of endogenous legal change, which kind of mashes up a law & orgs story with an orgs & social movements story.

      Liked by 1 person

      1. That’s a nice point, Dan. But excluding “unwanted kissing” from measures of campus sexual violence isn’t excluding a particularly meaningful class of events.


      2. I think you’re absolutely wrong about that – and also you are jumping to overly dichotomize the outcomes, as so often happens in this discourse. But even just unwanted kissing is a form of assault. We don’t take it seriously because it’s expected and because it’s gendered. Just like we used to not take seriously sexual harassment at work, until the feminist movement successfully mobilized around the issue.


      3. “But even just unwanted kissing is a form of assault.”

        Not it’s not. And I think it’s contradictory to say that I’m “overly dichotomizing the outcomes” and then defend the span of the interval that I complained was too wide.


      4. You’re right, it might be battery. IANAL, as they say on the internet. Though in a context of common sexual assaults and rapes (e.g. the college party scene), unwanted kissing certainly could create “apprehension of an imminent harmful or offensive contact with a person” (that is, assault).


      5. “Though in a context of common sexual assaults and rapes (e.g. the college party scene), unwanted kissing certainly could create ‘apprehension of an imminent harmful or offensive contact with a person’ (that is, assault).”

        Your statement here demonstrates my point that fears over campus rape are driven by the increased risk of exposure that people fear has increased as a result of campus hookup culture — which doesn’t exist.


  7. From Armstrong et al. in Contexts:

    Young people today are not having more sex at younger ages than their parents. The sexual practices of American youth changed in the 20th century, but the big change came with the Baby Boom cohort who came of age more than 40 years ago. The National Health and Social Life Survey—the gold standard of American sexual practice surveys—found that those born after 1942 were more sexually active at younger ages than those born from 1933-42. However, the trend toward greater sexual activity among young people appears to halt or reverse among the youngest cohort in the NHSLS, those born from 1963-72. Examining the National Survey of Family Growth, Lawrence B. Finer, Director of Domestic Research for the Guttmacher Institute, found that the percent of women who have had premarital sex by age 20 (65-76 percent) is roughly the same for all cohorts born after 1948. He also found that the women in the youngest cohort in this survey—those born from 1979-1984—were less likely to have premarital sex by age 20 than those born before them. The Centers for Disease Control, reporting on the results of the National Youth Risk Behavior Survey, report that rates of sexual intercourse among 9th-12th graders decreased from 1991-2007, as did numbers of partners. Reports of condom use increased.


    1. You’re confusing the lack of an increase with something being uncommon. But that’s the point of the social movements discussion above – such assaults were always common (at least, in the past few decades). What’s changed is our awareness. It’s possible we’ve become aware even in a context where there has been a small decline. But that’s still consistent with pervasiveness.

      To go back to the sexual harassment analogy: sexual harassment (or at least, the behaviors that term names) predated the social movement to name and prevent it. Sexual harassment may even have been declining as that movement was taking off. But that doesn’t mean sexual harassment wasn’t common.

      Beyond that, stats about sexual activity and hookup culture are largely irrelevant here. The data from Armstrong et al don’t speak to the question of how common sexual assaults on campus are (but lots of other data do, and the answer is ‘common’).


  8. I think we’re both now repeating our main points and have clarified the hypotheses on either side, and explored some ways to study the issue further empirically. Pretty ideal discussion, but maybe not super productive to keep going. Good luck out at Brown.


  9. Just checked Dan’s link: : “Measuring Rape and Sexual Assault: Successive Approximations to Consensus”.

    I find much of what is in there outrageous. The National Intimate Partner and Sexual Violence Survey is deliberately designed to yield false positives and lacking in intellectual honesty, for instance it automatically assumes that sex when drunk or high is rape and puts it under the heading: “Completed alcohol/drug facilitated penetration”.

    Accordingly, every youthful night of “sex, drugs and rock’n’roll”, no matter how enjoyable and consensual, is rape? Get out of here!

    This type of nonsense trivialises rape. It may even contribute indirectly to the never ending parade of “Jackie/ Rolling Stone” melodramas that appear to be the modern equivalent of medieval phantom pregnancies.

    A sociologist could profit from examining how the ideology of rape crisis intersects with the material self-interests of the professional classes that constitute the rape crisis industry.


    1. Here’s the text from the National Intimate Partner and Sexual Violence Survey questionnaire:

      “Sometimes sex happens when a person is unable to consent to it or stop it from happening because they were drunk, high, drugged, or passed out from alcohol, drugs, or medications. This can include times when they voluntarily consumed alcohol or drugs or they were given drugs or alcohol without their knowledge or consent. Please remember that even if someone uses alcohol or drugs, what happens to them is not their fault.

      When you were drunk, high, drugged, or passed out and unable to consent, how many people have ever…?”

      Obviously, there are large disagreements about how to define and measure rape. But I don’t think it’s fair to characterize this question (with its emphasis on being unable to consent) as counting “every youthful night of “sex, drugs and rock’n’roll”, no matter how enjoyable and consensual” as rape. It seems to do quite the opposite.


      1. “When you were drunk, high, drugged, or passed out and unable to consent, how many people ever…had vaginal sex with you?”

        The question is at best poorly worded if it is meant to capture only sex without consent. The “unable to consent” part might be read as applying to the entire list (“drunk, high, drugged, or passed out”) but could easily be read as applying to only “passed out.”

        Notice, by the way, that the remainder of the list items for that introduction you cite include the word “made” (e.g., “made you perform…”). But not the above item. I can’t think of a good reason for that variation.


      2. Dan, as you say, the wording is: “When you were drunk, high, drugged, or passed out and unable to consent, how many people have ever …”

        There is nothing ambiguous about this sentence, the comma before the “or passed out” means “unable to consent” belongs to that part of the sentence. This is straightforward English grammar. The sentence therefore means:

        (A) Whenever you were drunk, high, drugged,

        (B) or passed out and unable to consent

        This could result in false positives.


  10. I have not seen the film, but I did try to use the rise in awareness to have students apply what they learned about structure and agency in my 200-level class to propose potential solutions (mostly through bureaucracy — not through external pressure). One thing that really struck me as my students were doing the project and I was reading up on it some myself is that there is a real critical lack of evidence on just about every point of debate.

    Sexual assault statistics come from sources that should not be used to generalize to entire populations. Most studies fail to define an appropriate comparison group (i.e., we should look at sexual assault on college campuses relative to the incidence in the same age population outside of college campuses — I think Yoffe makes that point in one of her earlier articles). There are large estimation problems that come from non-linearities in the data. And most of the data is extremely problematic even when we do use them because institutional reports of various kinds form the basis of much of the data and are simply unreliable and potentially invalid.

    If DOE really wanted to do something, I think that they should field a comprehensive study on victimization on college campuses that they design to statistically generalize to the college population of the United States. Ideally, they would work with BJS to field a comparable measure to non-college sample of the same age for comparison. They would support linked qualitative studies that provide context to the survey findings and how the process works (or doesn’t) in different places.

    Liked by 1 person

    1. Fantastic points. But getting better info on distributions of victimization don’t suggest, without some enormous theoretical leaps, how to interdict perpetrators. So we need to apply the same battery of victimization and perpetration questions to men and women, and try to match them, in order to understand the mechanism. It could be the case that 90% of college women are sexually assaulted, and that 0.05% of college men are responsible.

      Liked by 1 person

  11. Hmm. My comment disappeared. Anyway, its pretty clear that some unjustified claims are being made about sexual assault and rape based on overly expansive definitions.

    Liked by 1 person

Comments are closed.