Advocacy mis-measurement syndrome?

It comes across as just another social justicey rant complaining about bias and stereotypes and discrimination, etc. but it isn't quite that. This is one of those storms in a teacup that just don't have much of a clear picture of what is going on.

I see some cryptic allusions here and there about some alumni or student video out of Wisconsin's flagship university University of Wisconsin Madison. It is one of the ironies of modern life that so many claims of abject racism arise from universities where the entire milieu, student, professors and administrators, are notably and materially to the left of the rest of the nation. For enterprises which seem so strongly committed to anti-racism, the claims of systemic racism are surprisingly frequent.

I think it is simply a product of philosophical monomaniacalism. People see what they want to see and if you put a bunch of left leaning, social justice enthused postmodernists into one place, of course they are going to see what they hope to see.

After several sightings of this as a controversy, I finally read the NYT's account of the issue, In a Homecoming Video Meant to Unite Campus, Almost Everyone Was White by Julie Bosman, Emily Shetler and Natalie Yahr.

I had gleaned enough from the allusions to know that the core issue was supposed to be an absence of people of color in the two minute video, so I had my eyes open, ready to count.

Well, yes. It is pretty white. But so is Wisconsin. Is it disproportionately white? Could be, but there were a couple of other things which stood out to me first.

Not only is the video clearly very white but it is also disproportionately female. Across the entire university population in the US, 55-60% are female and that is what appeared to be the case in the video. I would have guessed perhaps 55%. However, apparently the student body at UWI is pretty balanced at 51% female, 49% male. So we have a sex mystery as well as a race mystery. The video looks much whiter and female than the population at large. What gives?

But there was something even more interesting going on and that had to do with the structure of the video itself.

I had rather expected that the video would be a collage of various student events - clubs, sports, dining halls, etc. But that wasn't the case. About half the 2 minute video consisted of unstaged large-scale public events. Panning around the football stadium for example, a ten second segment of the runners and walkers by the lakeshore, wide-angle scenes from around campus. These certainly conveyed overwhelming whiteness but since they are unstaged and public, they don't really indicate any sort of bias. They are filming what is there and what is there is pretty white.

The remaining minute of time in the video is divided between two remaining categories of scenes. About half of them are selected shots of 10-20 people doing one thing or another such as escorting prospective students in groups around campus, doing some sort of UWI choreography, etc. Still fairly sizeable groups of people and, because they are pretty fast pans of scenes, kind of hard to do any sort of quick racial census.

The last category are the closer in shots of small groups of 2-10 people. A little easier in that case to pick out someone who might be Asian, someone who might be Hispanic, someone who looks like they might be African-American.

But the upshot is that the whole video is much more random selections of shots from around campus of large groups of people which makes it less likely that there is any inherent editorial bias. The problem appears not to be that they deliberately (or accidentally) shot a video with no African-Americans. The problem appears to be that there just aren't that many African-Americans at UWI Madison.

One more observation before I move off the video. A few months ago I had the opportunity to watch the commissioning of a class of newly minted US Marine Lieutenants, perhaps six hundred of them out on a parade field. One of the striking things was similarly how white the candidate body seemed. Why would that be? The Marines have a pretty aggressive outreach program, tend to be pretty aware and conscientious of gathering the best of the best wherever they can find them. So why the disparateness? Familial traditions of military service? Random chance? Something else?

My suspicion was that part of the issue was simply one of optics. At a few hundred feet distance, White, Hispanics, and Asians, particularly when in common and standard uniforms, can all appear to be closely alike on a color continuum scale. Under the circumstances only distinctly dark African-Americans stood out as different in color. Perhaps the Marines were more diverse than they appeared merely because of optics.

Well, that was part of it. Native Americans and Asian-Americans appear to be represented in rough proportion to their population (source). African-Americans are distinctly underrepresented at 8% compared to their general population position of 13%. But when you consider college degree requirements and absence of criminal convictions and other similar requirements, 8% might not be all that much out of line with the population who can actually meet those requirements.

The point being that, at a distance, a crowd can appear to be much whiter than it is. Perhaps that is part of what is happening at UWI?

But that can't be the totality of the explanation. I wouldn't think.

What is the African-American population of Wisconsin? The NYT says 6.7% but Wikipedia and the Census says 6.3%.

What percent of UWI student population is African-American? From the NYT article, though they do not calculate the percentage, the numbers are such that 3.3% of the student body is African American. So three percentage points less than the population at large. But, just as with the Marines, you would have to adjust that to identify the population with the minimum required test scores. For the Marines, they can get about 2/3 of the African-American candidates they want. If we were to apply the same rough rule of thumb, we might estimate that UWI is getting 3.3% of their student body to be African-American when they could expect to get 4%. So, definitely a shortfall but on a very small population base.

The upshot is that comparing Wisconsin demographics and UWI, particularly taking into account the effect of test scores, UWI doesn't appear to be too far out of whack on its demographics. Its just a pretty white state.

Given that the video is mostly large panoramas of large groups of random people, it doesn't appear that there is actually much of an issue of deliberate editing to omit populations. Given that male and female are 49:51 in the student body but in the video appear more like 45:55, there is clearly something going on in which random selections of public groups are simply not aligning with the numeric statistics in a fashion that is not intentional.

The NYT and Twitter commenters are working hard to make there be an issue. They can't really say what the issue is. And eyeballing the video, it doesn't look like there actually is much of one. It is just a function of large numbers and random chances.

It almost looks like the racial grievance advocates are really complaining about their minority status per se rather than complaining about any actual malicious acts or bad faith.

But who knows? There is always a backstory. Perhaps there is more to this than meets the eye. But I suspect what is going on here is much more a function of optics and random chance and small numbers than it has to do with any particular element of identitarianism.