Thursday, January 29, 2015

More on The Travails Of The Intrepid Factual Feminist: Deciphering Hoff Sommer's First Sexual Assault Myth


I feel mean enough to share some of the work that the anti-feminists make me do*.  This example is about the first item Christina Hoff Sommers talks about on her vlog as sexual assault myths.

It juxtaposes the one-in-five figure about college sexual assaults with a new figure which quickly flashes on the screen:  That one in 53 women experience sexual assault during the four years of college.  Before a quick look at a table which provides that number the vlog argues that a new report has come up with the "right" figure for the rate of sexual assaults during women's college experiences.

Got it?  Good.  Then next find out that report.  Read it.  Realize that it doesn't have a table with that one in 52.6 women figure.  Then realize that it doesn't have the other numbers in that table, either.

But...  The table with the calculations argues that its source is that very report.




So what do I do next?  I Google various possible combinations of the words and numbers in that table and find the real source of those calculations: Ashe Schow in the Washington Examiner, who converts the average data from 1995 to 2013 into a prevalence figure by multiplying it by four (to get at four years of college, I guess, though the data applies to all women in post-secondary education).  Others then figured out a way to get that prevalence lower by picking only the years from 2010 to 2013 (not sure where that data came from), because the rates of sexual assault have declined, as have all crimes.  That's how we got the one in 53 women figure!

Sadly, the chain one needs to follow to get the sources was broken there.  The little graph with the one-in-53 figure is not, in fact, based on the report that is cited as its source.

None of this interests you or should interest you.  But the point is that if one is going to call oneself the "factual feminist," then facts should be properly sourced and explained.

Another important point to note here:  Hoff Summers states that the "Justice Department figure" of one in fifty-three (created,it seems,  by Mark J. Perry at the American Enterprise Institute) is the right one and that the other estimates are based on flawed research.

But the report itself doesn't say its figures are the correct ones and the report doesn't calculate that overall rate.  Rather, it gives quite a long explanation of the reasons why different studies arrive at different figures and, incidentally, why the Justice Department figures (the NCVS) tend to be the lowest, even though the data is based on surveys of individuals rather than on police statistics:

The NCVS is one of several surveys used to study rape and sexual assault in the general and college-age population.
In addition to the NCVS, the National Intimate Partner and Sexual Violence Survey (NISVS) and the Campus Sexual Assault Study (CSA) are two recent survey efforts used in research on rape and sexual assault. The three surveys differ in important ways in how rape and sexual assault questions are asked
and victimization is measured. Across the three surveys,
the measurement differences contribute, in part, to varying estimates of the prevalence (the number of unique persons in the population who experienced one or more victimizations in a given period) and incidence (the number of victimizations experienced by persons in the population during a given period) of rape and sexual assault victimization.
 

Although conducted at different times, with different samples and reference periods, both NISVS and CSA produced prevalence rates that were substantially higher than the NCVS victimization and prevalence rates. Based on 2011 NISVS  data, 2% of all females experienced unwanted sexual contact during the prior 12 months.1 The 2007 CSA findings suggested that 14% of females ages 18 to 25 who were enrolled in two colleges and surveyed in the United States had experienced a completed sexual assault since entering college.2 In comparison, in 2010 the NCVS showed that 1% of females age 12 or older experienced one or more rape or sexual assaults in the prior year.3 For the period 2007–13, the NCVS victimization rate was 4.7 per 1,000 for females ages 18 to 24 who were enrolled in post-secondary schools (not shown).

Several of the key measurement differences that contribute to the different estimates include (see Appendix 1):
 Survey context and scope. The NCVS is presented as a survey about crime, while the NISVS and CSA are presented as surveys about public health. The NISVS and CSA collect data on incidents of unwanted sexual contact that may not rise to a level of criminal behavior, and respondents may not report incidents to the NCVS that they do not consider to be criminal.

Definitions of rape and sexual assault. The NCVS, NISVS, and CSA target different types of events. The NCVS definition is shaped from a criminal justice perspective and includes threatened, attempted, and completed
 rape and sexual assault against males and females (see Methodology). The NISVS uses a broader definition of sexual violence, which specifically mentions incidents 
in which the victim was unable to provide consent due 
to drug or alcohol use; forced to penetrate another person; or coerced to engage in sexual contact (including nonphysical pressure to engage in sex) unwanted sexual contact (including forcible kissing, fondling, or grabbing); and noncontact unwanted sexual experiences that do not involve physical contact.4 The CSA definition of rape and sexual assault includes unwanted sexual contact due to force and due to incapacitation, but excludes unwanted sexual contact due to verbal or emotional coercion.5 

ƒQuestion wording. The three surveys use different approaches to asking about experiences with rape and sexual assault.
...
Unlike the NCVS, which uses terms like rape and unwanted sexual activity to identify victims of rape and sexual assault, the NISVS and CSA use behaviorally specific questions 
to ascertain whether the respondent experienced rape or sexual assault. These surveys ask about an exhaustive list of explicit types of unwanted sexual contact a victim may have experienced, such as being made to perform or receive anal or oral sex.

The other two surveys mentioned above (the NISVS and CSA) don't have a criminal focus but a health focus.  That's partly why the questions are different.

I would have been perfectly happy with a criticism of the one-in-five study, by the way, because it is very difficult to get exact data on sexual assaults and because different definitions result in different findings.  But when that figure is just replaced by one-in-fifty-three as The Correct Figure in the manner described above, well, I'm grumpy.

Christina Hoff Sommers has (by the time I'm writing this) four points to make about the "myths" of sexual assaults.  What I've written here is only about the very first point.  Do you see why I get tired of the factual feminist bullshit?

To summarize, the Justice Department did NOT state that the right figure for the rape of women in post-secondary education is one in fifty-three.  Hoff Sommers stated that.

All the digging I've done is worth doing, probably.  For example, another of the myths (the third) Hoff Sommers points out is the idea that colleges are more dangerous places, in terms of sexual assaults,  than not being in college, whereas the reverse is the actual case.

According to Hoff Sommers this focus harms both women in that age group who are not in college (because their greater need goes unmet?) and the men in that age group who are in college (because they are labeled, apriori, as rapists?  or because she believes in lots of false rape accusations by coeds?).  But you might as well argue that it benefits the men who are not in college, too, because they get a pass or whatever.  If you wish to go down that avenue of collective guilt etc and assuming that sexual assault only takes place within the circle of campus or outside that circle.

And indeed, it is true that the sexual assault rates are higher for women between ages 18 and 24 who are not in college.  This graph summarizes the evidence:


Note that all types of crime in that table are more common among non-students than students.  The difference is actually the smallest for rape and sexual assault.

Interpreting what caused the numbers in that table could be tricky.  Maybe being in post-secondary education is correlated with living in safer areas or living a safer lifestyle in general?  What is the role of poverty or low income in these numbers?  Are women in that age group who are not in college poorer, on average, and more likely to be limited to living in dangerous areas?

Flipping everything around, the table might suggest that being in post-secondary education protects women from becoming victims of violent crime, but has less power to protect them from becoming victims of rape and sexual assault.

This does NOT mean that the assertion of higher rates for non-students wouldn't be correct.  But ultimately we would like to make these types of comparisons while holding all other relevant variables constant, including poverty and the impact of living in dangerous areas, to see whether being in post-secondary education per se protects women from violent crime of various types and to what extent.

This post is sort of an example of what it means to study assertions and to look for their parents and grandparents etc.  It's Hard Work, as a famous prior US president stated.
-----
And sometimes feminists, too.  I've written a lot about the difference between the gross gender gap in wages and the net gender gap in wages and also the fact that the gross gender gap is not mostly about paying different amounts for the same work because men and women, on average, do different kinds of jobs in the labor market.  On the other hand, the MRAs tell us that all the difference is just based on men working longer hours, when the net gap takes that already into account and still arrives at a sizable unexplained difference.