Five, Sex, Seven: Misleading Numbers in The Stanford Daily

[![]( "Daily Math")](
Daily Math
In today’s Stanford Daily in a column titled “Watch Your Language, Young Lady! Part I”, sex columnist Yanran Lu [claims](

According to a U.S. Department of Justice report entitled, “Sexual Assault on Campus: What Colleges and Universities Are Doing About It,” statistics have shown that one in four women have experienced rape in their college career.

That is really quite unsettling. But let’s take a deeper look at the actual report, from 2005. Here’s what that says on page 2:

Just under three percent of all college women become victims of rape (either completed or attempted) in a given 9-month academic year. On first glance the risk seems low, but the percentage translates into the disturbing figure of 35 such crimes for every 1,000 women students… If the percentage is projected to a full calendar year, the proportion rises to nearly 5 percent of college women. When projected over a now-typical 5-year college career, one in five women experiences rape during college.

Ms. Lu’s claim is demonstrably incorrect, but let’s look at those numbers in a little more depth anyway. This is an important issue, so it’s worth trying to get right. In order to get their numbers, the study takes the claim that just under 3% of college women become victims of rape in a 9-month academic year, then projects it over the full calendar year (i.e. .03/912), then projects that number over 5 years (.03/912*5=.2), which is their statistic that claims one in five women experiences rape.

Methodologically, this is rather dubious. First of all, it projects the academic year numbers over the full calendar year without giving any reason to believe that rape is equally likely out of school. Second, they assume a five-year academic career, which (and I am no expert on this) may very well be typical in some sense, but hardly seems like the average. Is there really a student spending six years at college for every one who spends four? Third, the idea that in order to find how many women are raped simply involves taking the number of completed or attempted rapes, and dividing them by the number of women is faulty.

Now, let’s look at the original source of the statistic. The report helpfully provides a footnote, which led me to a 2000 report titled “The Sexual Victimization of College Women.” Here’s their methodology:

NCWSV study results are based on a telephone survey of a randomly selected national sample of 4,446 women who were attending a 2- or 4-year college or university during fall 1996.

So, these numbers Ms. Lu was citing are now 14 years old.  Things change over short periods, but that’s not really a short period–it’s more than 2/3 of my lifetime. Also important, the study does not distinguish between the likelihood of sexual assault at 2- and 4-year schools–one would imagine that there is a distinction there, and that on average students who begin at 2-year schools have shorter academic careers. Here’s the result that the more recent study cites:

2.8 percent of the sample had experience either a completed rape (1.7 percent) or an attempted rape incident (1.1 percent). The victimization was 27.7 rapes per 1,000 students.

Also, there is this:

Of the 123 victims, 22.8% (n=28) were multiple rape victims.

Now, if one were to ask what the best way for measuring how many women will be raped using only the data from one school year, the most intuitive measure would be to simply take how many women were raped in the previous year, and quadruple it. I’ll show my work (123-28)/4,446*4=.085. So, projected differently, one could just as easily draw the conclusion roughly 1 in 12 rather than 1 in 5 women are victims of rape or attempted rape.

This is not to say the report’s methodology is wrong and therefore the smaller number is correct. I sincerely doubt that it is. The point is that these statistics can be manipulated in any number of ways and that until a longitudinal study follows a female cohort across time, all we have is speculation, and the older study says as much (in footnote 18, if you care to look).

Until that happens, let’s not pretend like 1 in 5 (or inexplicably, so far as I can tell, 1 in 4) women are raped, and this is some indisputable fact. It is not. The reason I am so sensitive to this issue is because overestimating the frequency of rape normalizes it. If it is statistically likely that say, there will be one rape victim in every freshman quad in Roble. I don’t have any statistical evidence, but that hardly seems likely. Overemphasizing the frequency of these incidents strains male-female relations (why wouldn’t women be suspicious to the point of complete male avoidance if there was a 25% chance they would be raped in their college career?), forces women to live in an irrational fear, and most importantly makes it seem as if rape is a fact of life to be tolerated and overcome. It is not.

Ms. Lu says regarding rape that, “to emphasize the “rarity” instead of (in context) “often” is counterproductive to my objective.” That is a terrible philosophy. Emphasis should be placed on getting it right rather than fear-mongering. Instead, the column takes a defensive tone (in response to the criticism her previous column received) and its title implies that all or most objections are in some way related to Ms. Lu’s bravery or insistence on talking about sex. That’s just an easy cover. She claims:

Both men and women have been quick in telling me that they/their partners are not like that. In fact, they say, it is my skewed perception of the world that made these things visible to me. While I understand their effort in restoring the name of men made rotten by “a few bad apples,” I insist on my stance on respect and awareness.

Basically, this is saying, people have claimed Ms. Lu was wrong. Ms. Lu is in favor of respect and awareness, and that disagreement with the claims in her columns is in some way related to “respect and awareness”–awareness of what exactly is not entirely clear. That may be true in some cases, but the much larger problem is Ms. Lu’s mangling and misrepresentation of the facts.

Subscribe to the Stanford Review