Enter "the Woman Who Knows" ... and several who don't.

Today I want to discuss one of the binds that men are put in when discussing feminist issues, specifically the idea that if a man disagrees with a woman’s feminist argument, the man is behaving in a sexist manner because a man should “listen” rather than “tell” in feminist issues. Before I go on, let me say that I believe there is a tendency for men to dismiss women’s arguments more readily than they would a man’s. There is no doubt that some men dismiss women because they are women. There is no doubt that men are sometimes overly aggressive in their rhetorical stances in order to intimidate their opponents, and there is no doubt that I’m guilty of that; however, there is also a conundrum that men are put into when discussing feminist issues that makes the discussions sometimes impossible.

As an example of that, I’d like to look at the post here. In the comments section, there is what I guess I’ll call “a discussion” that dealt mainly with rape statistics and whether or not they can be trusted. Frankly, I can’t recommend reading the thread because it boiled down to personal attacks from virtually all involved with the notable exception of Nenena, who tossed in some fairly harmless, frustration-inspired sniping but also included cogent arguments and data to support her opinion that there is no compelling evidence of dramatic drop in the frequency of rapes in the United States from 1973 to the present. To understand the conundrum via this discussion, we have to look at the arguments.

As part of Nenena’s argument she says that the finding that 1 in 4 woman have been raped has held steady for decades. She mentioned the 1 in 4 figure used in the 1970s and then writes the following concerning that number still being used today:

The 1 in 4 number is from the Rennison study which I mistakenly attributed to F/C/T. Full cite is: Rennison, Callie M. "Criminal Victimization 1999: Changes 1998-00 with Trends 1993-99." Bureau of Justice Statistics, U.S. Department of Justice, August 2000. I can't find a copy of the paper online, but several trustworthy sources cite it, including RAINN, the sexual assault wiki, and several "safe campus" and rape crisis centers.

Callie Rennison PhD, the real women who knows

I was able to find and read Criminal Victimization 1999: Changes 1998-00 with Trends 1993-99.  It is a compelling work, and Callie Marie Rennison is an impressive researcher. You can read her remarkable biography here, which becomes more impressive when you see the details of her Curriculum Vitae here. Seriously, take a look at it. It’s ten pages of degrees, professional experience, and papers that mainly deal with crime and specifically rape. I doubt there are more than a handful of people in the world who could know more about the subject than Dr. Rennison. This is a woman who has apparently devoted her life to understanding and eradicating the causes of rape. From what I’ve been able to discover about her, she deserves to be lauded as a champion. Clearly, her influence has been felt by trustworthy sources like RAINN (Rape, Abuse & Incest National Network), the nation’s largest anti-sexual assault organization, and safe campus and rape crisis centers and the wiki page on sexual assault, which I’m sure has been reviewed several times by people in the know. I’d have to be some kind of nincompoop or agenda-driven nut to disagree with Rennison, when she is obviously far more qualified to discuss the trends in rape than I am. I think most of us would agree that Dr. Rennison and her work for the Bureau of Justice Statistics represent a body of knowledge on the subject that dwarfs our own. (She does look suspiciously like Ann Coulter, but I guess I can’t hold that against her.)

Which bring us to some interesting findings in the paper, specifically table 7 on page 10. According to that table, in just the six years from 1993 to 1999, instances of rape/ sexual assault had declined by 32%. Rennison continued to work for the Bureau of Justice Statistics and in 2003, using the same methods as her 2000 report, found in that report that from 1993 to 2002, instance of rape/sexual assault declined by 56.3%.  (Here at page 5, table 3.) That was the last report from the Bureau of Justice Statistics that Rennison worked on, but her research methods were continued by the Bureau of Justice Statistics. In the 2006 edition of the report, also written by a woman, Dr. Shannan M. Catalano, again a woman with far more knowledge of the subject than I, Dr. Catalano found that from 1993 to 2005, instance of rape/sexual assault declined by 68.6%.  (See page 5, table 3.) I will remind you that these reports, written by women, are cited by RAINN and other trustworthy anti-rape organizations just as Nenena said they were.

None of these percentages I listed in the above paragraphs from those reports include the recorded decline of instances of rape from 1973 to 1992 because there was a redesign of the survey made in 1992 (and I’ll discuss that later because there is some disagreement on how that affects how we should look at the decline in rape), but all of the editions of these reports from the Bureau of Justice Statistics recognize a decline in instance of rape from 1973 to 1992, including Criminal Victimization 1999: Changes 1998-00 with Trends 1993-99 cited by RAINN and those trustworthy sources. Despite the redesign, the most prestigious anti-rape organizations cited the reports that included the data that strongly supported the idea that there was about a 40% decrease in the instances of rape from 1973 to 1992 in addition to the approximately 70% drop-off from 1993 to 2005.

Another interesting thing to note about Criminal Victimization 1999: Changes 1998-00 with Trends 1993-99 is that despite what Nenena says, the report does NOT say that 1 in 4 women will be raped in their lifetime. Go ahead and look through it and see if you can find it. I couldn’t. It is possible that someone used the data from the report to calculate the odds at 1 in 4, but there are a few problems with that. The first is that we don’t know how the 1 in 4 figure was arrived at, but let’s assume that the people who came up with the figure used the right math. I see no reason to doubt the fine folks at RAINN. The second problem is much harder to overlook. Let’s begin by quoting Nenena, who used the supposed 1 in 4 figure from Criminal Victimization 1999 and used the figures from RAINN to show that instances of rape haven’t declined much. Nenena wrote about RAINN:

RAINN - 1 in 5 women raped in her lifetime. Down from 1 in 4 in 1973.

This second, harder to overlook problem is that Rennison did NOT say 1 in 4 women will be raped in their lifetime. RAINN does NOT say 1 in 5 women will be raped in their lifetime. RAINN says 1 in 6 will be. You can see it here.

Now, let’s assume that RAINN did say 1 in 4 in 1973 and now uses 1 in 6. A drop from 1 in 4 to 1 in 6 is a 33% drop, which is a bigger drop than Nenena and her supporters (the anti-Rennison/Catalano crowd) have admitted to. I think Nenena went as high as a possible 20% drop, and Ami said that she supported Nenena’s claim that the drop was only 5%. But looking at RAINN’s figures, there was at least a 33% drop.

But those odds are deceptive because if we do a little math, we find the drop is almost certainly larger than 33% using RAINN’s own estimates. I have to believe that when RAINN calculated how many women would be raped in their lifetimes it is not ignoring the women who already lived through the years where the instances of rape were higher. It would be incredibly callous to pretend those rapes didn’t exist and ignore the rapes of the women who were alive in the 70s. Most of those women are still with us today and their rapes were real. Further, it would just be wrong. It would give the impression that far fewer women alive today have been raped than actually have been. A later decrease in the instances of rape does not somehow unrape the women in earlier decades. In addition to that, it would appear to work counter to the goals of RAINN. There doesn’t seem to be any good reason for RAINN to artificially lower the number of women who have been raped by ignoring decades worth of rape when their goal is to help those people. Why would they pretend there are fewer people raped than actually have been? It would be like the American Cancer Society trying to convince you that cancer research was important by downplaying how many people had cancer. It’s the opposite of what they should do. I simply can’t believe the people at RAINN would be so heartless, stupid, and ineffectual that they would come to their calculation in a manner that would ignore the women (and men) still with us. So let’s assume that RAINN is still including all women alive today, not just women of the future, when they calculate the odds at 1 in 6.

I’m going to use some overly simplistic math to demonstrate that the drop in instances of rape would have to more than 33%. As I said, this is overly simplistic and only used to demonstrate a mathematical truth, not to say what the exact drop was or even to be a close estimate. However, if we assume that RAINN included the rapes of women who are still alive today but lived through the earlier decades, that would mean that there would be a group of women who lived the bulk of their lives during the time when the odds were 1 in 4, a later group that lived the bulk of their lives when the odds were 1 in 5, and a third group that lived the bulk of their lives when the odds were 1 in 6. If we average these groups together (assuming, probably erroneously, they are about the same size) to find what the average woman’s odds were of being raped, which is what we can probably assume RAINN’s 1 in 6 figure does, our math would look something like this:

1 in 4 = 25%

1 in 5 = 20%

1 in 6 = 17%

(25% + 20% + 17%)/3 = 20% or still 1 in 5.

Despite the last group having a 33% less chance of being raped than the first group, the average percentage for all women does not drop low enough for the average women to be at a 1 in 6 chance to have raped in her lifetime. The average of those percents is 20% or still 1 in 5. In other words, you need to have the last group (and/or second to the last group) have a far lower than 17% rape rate to get the average for all women down to 1 in 6. In order for this admittedly too simplistic math to work, you need to have it look something like this:

1 in 4 = 25%

1 in 5 = 20%

1 in 17 = 6%

(25 + 20 + 6)/3 = 17% or 1 in 6.

Going from 1 in 4 to 1 in 17 is a 76% drop. Now, I’m not saying that RAINN is using 76% as its figure to represent that drop in the instances of rape. My math is too crude here to say that, but I think it should be clear that RAINN is using a number far greater 33% as the drop in the instances of rape and is using a number that is in line with the figures used by Drs. Rennison & Catalano and the Bureau of Justice Statistics. We can tell this not only because of the math suggested by my demonstration above, but because RAINN actually cites to the Bureau of Justice Statistics and their continually updated Criminal Victimization reports as you can see by looking at the references at the bottom of the page here. Yes, RAINN cites to the report that said that said instances of rape dropped about 40% between 1973 and 1992 and 68.6% from 1993 to 2005.

Now, let’s look at the 1992 redesign of the survey. Not long after the Bureau of Justice Statistics began their National Crime Survey (NCS), there was criticism of it, and so the Bureau looked into the criticism and eventually made changes to their methods. You can see a discussion and analysis of the changes here. Because there were changes in the methodology, it would be inaccurate to compare the results before the redesign to the results after the redesign to say with certainty that a specific percentage of decline in rape was recorded. However, does the redesign prevent us from knowing that there was a significant decline in the instances of rape between 1973 and 1992 or between 1973 and 2005?

Let’s look at what changes were made. The National Academy of Sciences reviewed the NCS and found that the methodology and scope of the NCS that could be improved.

Academy proposed that researchers investigate the following:

 

• an enhanced screening section that would better stimulate respondents’ recall of victimizations, thus reducing underreporting due to forgotten incidents

 

• screening questions that would sharpen the concepts of criminal victimization and diminish the effects of subjective interpretations of the survey questions

 

• additional questions on the nature and consequences of victimizations

Please note that one of the goals of the redesign was to increase the reporting of the crimes to the surveyors. The redesign had specific elements that were implemented to get fuller results in the areas of rape and sexual assault.

More recently, the issue of specifically improving the measurement of sex crimes and domestic violence resulted in the formation of a special committee associated with the American Statistical Association’s Committee on Law and Justice Statistics. The special committee developed enhanced questions and clarification queries on rape, sexual assault, and domestic violence to get better estimates of these crimes that are difficult to measure. The Bureau of the Census subjected the changes recommended by the special committee, as well as those of the redesign consortium, to additional testing. Modifications proving successful in this testing were introduced into the survey.

And the redesign appears to have worked in increasing the reports of rape and sexual assault. In the 1992 NCS, the sample group was split into two equal parts, one receiving the old survey and one receiving the redesigned survey.  The results showed the group getting the redesigned survey were 157% more likely to report rapes than the group that got the old survey. That’s a dramatic difference! This dramatic difference tells us that it would be ridiculous to compare the results of the old survey to the redesigned survey as if they were the same survey. Clearly, they are substantially different.

However, the argument that Nenena and others have used is that because there was a redesign, we can’t be sure there was a decrease in the number of rapes, but this argument flounders when we look at the particulars of the redesign. If the redesign had decreased the number of reported rapes, we could say that the decrease in reported rapes in recent years was caused in part by the redesign and not by an actual decrease in the number of rapes. But the redesign increased the reports of rape. How could a redesign that increases the reports of rape be the cause for a decrease in the reports of rape?

But there are still more problems with comparing the two surveys. The Bureau of Justice Statistics puts it this way:

Rates of rape for 1995 were significantly lower than the adjusted rates for 1973 (figure 3). This finding, however, should be regarded with extreme caution. Before the redesign in 1992 the survey did not ask respondents whether they had been victims of sexual assault other than rape or attempted rape. Some victims of these crimes (sexual assaults) may have reported such victimizations in response to questions about rape or other forms of violence.

 

It is not possible to determine to what extent crimes now categorized as sexual assaults were included in the data as rape or attempted rape in earlier years. To the extent that this occurred, estimates of rape prior to the redesign would not be comparable to those since the redesign. Anomalies in the distribution of male and female victims in the 1992 NCS rape estimates also raise questions about the adjustments of rape estimates.

Far be it from me to ignore a warning of extreme caution from the Justice Dept., so let’s do the logical thing and not compare them. Instead, we’ll look at the old NCS as one survey from 1973 to 1992 and the redesigned NCS as another survey from 1993 to 2005. If we look at the NCS from 1973 to 1992, we see a decline in the instances of rape of about 40%. If we look at the redesigned NCS, we see a decline in the instances of rape from 1993 to 2005 of about 60%. Even if we look at the surveys separately, there is still a dramatic decrease in the instances of rape for both periods.

But there are other things that Nenena suggests invalidate the NCS as a whole:

In this report the definition of "rape" does not non-penetrative rape, drug- or alcohol-fueled rapes, "sexual assault," or statuatory rape. And most amazingly, it excludes victims under the age of 12. The last omission is particularly glaring, because 1 in 3 sexual assault victims are under the age of twelve.

Nenena is correct that the rape statistic does not include non-penetrative rape or sexual assault. However, the NCS does have categories that do include those in the NCSs from 1993 on. The NCS looks at the figures separately and as a single category. Below shows the crime or collection of crimes and the percentage decrease from 1993 to 2005:

Rape/Sexual assault: 68.6%

Rape/Attempted rape: 66.7%

Rape:  71.6%

Attempted rape: 64.5%

Sexual assault: 68.5%

Again, these are only the figures from 1993 to 2005, but even these reductions are significantly greater than those Nenena and others will admit to. Further, given how close these figures are to each other and because the crimes are so similar, I think we can logically conclude that the factors that cause a decrease in one category are likely to cause a decrease in another category. So when we look back at the figures from 1973 to 1992 and see that there was about a 40% decrease in rapes/attempted rapes, we can safely extrapolate a similar decrease in sexual assaults during that time.

Nenena contends that the figures do not include drug- or alcohol-fueled rapes. I don’t know that that is true as I can find no evidence of this. Her contention seems particularly unlikely because Bureau of Justice Statistics does include figures on the perceived drug or alcohol use by offenders in rape/sexual assault as you can see here. Perhaps, when Nenena wrote that she was drug- or alcohol-fueled ... or maybe I am, and that’s why I missed it, but I looked pretty closely and don’t recall taking a drink lately.

Nenena is also correct that the NCS doesn’t include children under 12. However, that doesn’t invalidate the information gathered on people over 12. In the discussion of whether women (adult females) are less likely to be raped and can therefore fear less, the fact that there might not be a decline in childhood rape or that there might even be an increase in childhood rape, does not change the odds of women being raped because with very few exceptions, women are not plunged back in time to their childhoods and raped. Moreover, her criticism seems to suggest that while there was a decrease in rape and the related crime of sexual assault, there was an increase in the instances of the related crime of the rape of children. While that is possible, I don’t see why that would be true. It seems unlikely that the drop in the instances of rape can be explained by rapists changing their preference from raping adults to raping children.

Nenena and others suggested that changes in ethnicity or income of the respondents might account for the changes, but if you look at the figures here (page 6, table 4), you’ll see that all ethnic groups and income levels reported less violent crime committed against them from 1993 to 2005. Moreover, the NCS appears to be influenced by census data in a rather complex algorithm that helps to prevent errors in calculating for ethnic and incomes groups, and they used the data collected from the redesign to adjust the earlier pre-redesign figures to take race into account, which you can see here on page 7, figure 8. In other words, National Academy of Sciences, American Statistical Association, Bureau of Justice Statistics, and Drs. Rennison & Catalano were not idiots who weren’t aware of the affects of ethnic and income groups. Drs. Rennison and Catalano have done extensive research comparing the rates of crime among the various ethnic and economic groups -- more research, I suspect, than Nenena or anyone commenting on the thread. I would be astonished to find that these two highly respected specialists didn’t bring their knowledge of those groups to their analysis of the crime data of the NCS, especially since we can see that the Bureau of Justice Statistics did adjust the figures, but if anyone has evidence to discredit them, I’m willing to hear it.

Nenena also suggested that since the rapes have to be reported to the surveyors that perhaps women stopped telling the surveyors. While that is theoretically possible, there doesn’t seem to be any evidence to support that theory. However, there is evidence to support that women have become more likely to report rapes to the police. (See here page 11, figure 8. 28.8 of rapes and sexual assaults were reported to the police in 1993, 38.3 were in 2005. That’s a 33% increase in reporting the crime to the police, while there was a drop of 67% in people telling the NCS surveyors they were raped/sexually assaulted.) Nenena agrees that women have been reporting rapes to the police more often, but for some reason, still thinks they might be telling the NCS surveyors less often. That just doesn’t make sense to me. Why would they be more likely to report rapes to the police, when that involves all sorts of potentially unpleasant aspects related to dealing with the legal system and maybe the press, but be simultaneously less likely to tell the NCS surveyors of the rapes, for which there are relatively few ramifications? The theory is both counterintuitive and unsupported by evidence.

This isn’t to say that I don’t know that rape is the most underreported crime in the U.S. It is. Ironically, the NCS, which Nenena and company are claiming is such a flawed study, is the very study cited by anti-rape organizations like RAINN to show how underreported rape is. That’s right. The NCS is the most definitive study to show that rape is underreported. It was groundbreaking. The anti-Rennison/ Catalano crowd’s arguments suggest that despite having no evidence that instances of rape aren’t dramatically decreasing, they’re sure that rape is the one crime for which there hasn’t been a dramatic decrease. (See here, page 5, table 3.) The only one. For instance, if we look at murder rates, we’ll see that there has been a fairly steady decrease in the rate of murder among women from 1976 to 2005. One thing we can say conclusively about the calculations of the homicide rate is that it is not dependent on the victims reporting the crime, no matter what we might think from watching Medium and Ghost Whisperer. There has been a 72% drop in the murders of women in those years. (During that same time there was only a 27% drop in the murder of men.) What Nenena and company would have us believe is that despite the fact that there was unquestionably a decrease in the rate of violence specifically against women in the form of murder, which can not possibly be an error caused by women not reporting the crime, what appears to be a similar decrease in the rate of violence against women the form of rape is caused by women being less likely to report the crime to the Bureau of Justice Statistics now even as they are more likely to report it to the police, and we should believe even though they have no evidence to support their theory. We should believe this baseless, counterintuitive theory because they are women telling us this; even though, other women, Drs. Rennison & Catalano, are telling us there has been a decrease in rapes and these women are experts in the field with enormous amounts of evidence to support their claims and peer reviewed studies that are cited by rape organizations that the skeptical women of the first group claim are trustworthy organizations.

There were other attempts to find some other reason for the drop, but all of them suggest that National Academy of Sciences, American Statistical Association, Bureau of Justice Statistics, and Drs. Rennison & Catalano don’t know what they are talking about while internet fangirls (and boys) do. These arguments ignored the one very obvious possibility for there being fewer rapes and sexual assaults reported to the NCS, i.e. there were fewer rapes and sexual assaults. It’s as if they are dead set against using Occum’s razor and prefer to use a Remington Smooth & Silky Women's Shaver because it’s the razor that has been marketed to them.

Which brings us back to the beginning; and the question, how are men supposed to talk about feminist issues with women? Specifically, in the thread I mentioned above, Kadymae responded to me saying that the despite the fact statistics show that murder is the most prevalent cause of death among pregnant women, it is not true that statistics show that pregnant women are more likely to be murdered than anyone else with this:

Oh look, THE MAN WHO KNOWS has arrived
Armed with his statistics

And when *women* point out that his statistics are like bikinis, that is, they show a lot but conceal what is vital? Or they counter with other statistics?

Well, he's the man who knows, you see.

I work at one of the largest Academic Libraries in the Western US. I worked the reference desk until I took a job that was a transfer/promotion. And somehow, despite knowing how to do academic/scholarly research that shows he's full of excrement, I (and my sisters) will never be anything but a hysterical woman, and he, he will always be the MAN who knows, here to save us wimmins from the overwrought error of our ways despite how many times we point out the flaws in his cited statistics.

Kadymae suggests I that I just won’t believe what women are telling me, but my beliefs are based on the research by women, Drs. Rennison & Catalano. So who am I supposed to believe? If I say I think rape is declining, I’m disagreeing with Nenena, but if I say I agree with Nenena that rape is not declining, I’m disagreeing with Drs. Rennison & Catalano, and apparently, I’d be claiming that they were hysterical women. There’s no way to win or even break even.

And yes, I was armed with statistics, but Kadymae seemed fine with citing statistics when they said that the most common cause of death among pregnant women was murder; even though, the statistic does not say that pregnant women were more likely to be murdered than any other group (and according to one study, they may be less likely to be murdered than non-pregnant women of the same age group.) So are statistics usable or aren’t they? Are women allowed to cite statistics in feminist discussion but men aren’t? Are statistics only misleading when men use them?

In the case of the NCS rape statistics, Kadymae is correct that they are statistics, but in this case, “statistics” means “a collection and analysis of an enormous amount of data done by female, renowned experts in the field using, peer reviewed methods developed in concert with National Academy of Sciences and the American Statistical Association’s Committee on Law and Justice Statistics, which have been cited by virtually all the major anti-rape organizations.” And what did the other people on the thread come armed with? Here is an example from Fox in the Stars:

To stand here and say to women (who've shown themselves more thoughtful, well-informed, and capable of understanding the data than you, MORON that you are to draw a conclusion across a major redesign of the metric) who are concerned about this pervasive naturally-occurring-so-to-speak institutionalised terrorism that they are four-fifths full of shit?? That if women fear said pervasive terrorism, this act of dehumanisation and domination, that they're four-fifths stupid?? Because this is the most belittling ignorant---no STUBBORNLY HEAD-UP-ASS STUPID response IMAGINABLE!

With a brain or a lick of decency you'd be half human. Please fuck yourself and spare the rest of us. (And if you want to reply to me, don't bother, I'll be in the bathroom throwing up.)