Utah doesn’t do so great when it comes to its ranking of reported rape. However, as any sexual assault scholar will tell you, most rapes are not reported (and an even smaller fraction lead to a conviction). Low official rape numbers are sort of a Rorschach test, and can be interpreted as evidence of stigma against reporting as much as evidence of low sexual assaults.
Therefore, self-reported victimization (asking people if they’ve been assaulted) is considered a much better way to measure sexual assault. The major survey that has national self-reported victimization data, the National Crime Victimization Survey, generally doesn’t include sub-national level estimates in order to protect confidentiality. However, the Census Bureau has recently released a special public-use file that allows researchers to generate estimates at the Metropolitan Statistical Area (MSA) level.Among the 52 MSAs is the Salt Lake City area, so the first time we have the chance to look at where a Utah location falls in terms of good, self-reported sexual assault data.
One qualification: I’m on the record as pointing out that Utah does not necessarily equal the Church. (This is particularly important when considering the old canard about high Utah porn use, since for some reason the non-Latter-day Saint heavy Utah counties are into paid porn).This is doubly true for the Salt Lake City area, which is even less Latter-day Saint than Utah as a whole.
The NCVS survey is quite complex; thankfully, the provided codebook included the R code for deriving rates. Of course, in a lot of such cases differences between data formats and such make it so that you can’t just copy, paste, and run it, so I had to tweak it, but I was able to successfully and precisely replicate the example figures reported in the codebook. My code is on my Github page (Utah_Vict.R), where people can play with the parameters (changing the years, the crimes involved, etc.). I ran it in a loop and collected figures for each MSA.In order to get enough data to get solid numbers, I used 10 years’ worth of data in this particular run (so 2005-2015).
As you can see, Salt Lake City metropolitan area appears around the median. Even if we assume that Salt Lake City= the Church (which we’re not), there certainly isn’t enough here for a gotcha about how the evil religious patriarchy leads to sexual assault, but neither is there evidence here about how the goodly influence of the Church leads to lower sexual assault. Whatever the case, there is room to improve.
*An earlier version of this post showed the total rate for both men and women combined. At first I thought this was fine since men occasionally experience sexual assault, but then I realized that there actually is non-negligible variation in sex ratios among MSAs, so combining the two might warp the numbers if there are more or less women per average, so here we’re just looking at female victimization.
Thanks for the post. Unremarkable statistics are highly underrated.
Why do you think there’s such a wide variation between the low end (San Antonio/New Braunfels) and the high end (Tucson)? Do you think the data reflects real differences in assault rates, or is it reporting variation all the way across?
I don’t really see any region-specific patterns (e.g. you have blue states and red states, as well as more urban and less urban areas on both ends of the scale), so I suspect that a lot hinges on which areas are included in the MSA. For example, some MSAs might include more lower income areas (not that higher income people don’t also sexually assault) than other MSAs by dint of where they decided to draw the boundary line.