A 2021 article in The Lancet combines data from several sources to get estimates of the number of people killed by police from the 1980s onward. It summarizes: "as our analysis shows, fatal police violence rates and the large racial disparities in fatal police violence have remained largely unchanged or have increased since 1990." Here's a table giving their estimates of age-standardized mortality rates per 100,000 for (non-Hispanic) blacks, Hispanics, and (non-Hispanic) whites:
Black Hisp White All
1980s .81 .44 .15 .25
1990s .66 .35 .18 .25
2000s .60 .32 .19 .26
2010s .71 .34 .28 .34
So the overall rate has increased, but racial and ethnic disparities have declined--the black/white ratio was over 5 in the 1980s, then about 3.5, 3, and 2.5 in the following decades. The Hispanic/white ratio was almost 3 in the 1980s and about 1.2 in the 2010s. The numbers I give above aren't from a reanalysis of the data--they are taken straight from the paper. And they were in a major table, not hidden away in a footnote or the supplemental materials. So how did the authors make (and the reviewers miss) an obvious mistake in summarizing their results? "Disparity" has a connotation of unfairness, so a reduction in disparity sounds like a good thing. But in this case, things didn't get better--the rate for blacks and Hispanics stayed about the same, and the rate for whites got worse. As a result, describing the changes as a "reduction in disparities" doesn't sound right, even though it is by the dictionary definition. The term "disparities" seems to have become widely established in public health--I don't know when or how this happened--even when the data just involve differences (as is the case here).
Although they used four different data sets, only one (the National Vital Statistics System) went back before 2005. Last August, I did an analysis of the NVSS data going back to 1960. I mentioned that the NVSS data seemed to under-report the number of killings by police, and that the rate of under-reporting could change over time. It seemed plausible that reporting would tend to become more complete, which would create the appearance of an increase even if there was no real change. So has there really been an increase in the overall rate of police killings? I'll look at that in my next post.
The term "disparities" is used to hide reality. For example, last year in the Seattle Times, the person who writes the statistical column published a simple table of household income by race (four rows for race, one column for data), then tisk-tisked the public for "racism". Of course his simple table was hiding the monsters. The lowest income group was about a third the *household* income of the highest income group. It also has *by FAR*: 1) the lowest level of education; 2) the highest rate of single parent families; 3) the highest rate of both one and zero workers in the household; 4) and the highest rate of teen pregnancy (a whopping TEN TIMES the rate of teen pregnancy of the highest income group).
ReplyDeleteSo if you adjust for all these factors - that is, if you compare households that are equally educated with equal numbers of employed people working equal hours - the income "disparity" between the top and bottom groups would become almost imperceptible. The "healthcare community" - e.g., structural racism advocates - are well aware that an equal comparison undermines their theory. So rather than using the term "inequality" - which is often false when similar things are compared - they prefer to compare things that aren't similar, and use the term "disparity" to hide the fact that the term "inequality" is false.
But I like your data. It puts the lie too the "police racism" claim. 1990s-2010s, police killings of blacks rose 7.6% (0.66-0.71). In the same period, police killings of whites rose a whopping 36% (0.25-0.34)! So much for increasing racism of cops
Moreover, although the death rate for blacks is a bit more than double the death rate for whites in the most recent decade CDC data from 2015 show that the general homicide rate for blacks is almost eight times that for whites. Indeed, it appears the police are showing extraordinary restraint with the black population.
Last but not least, note in the CDC data that homicide rates increased in all groups in 2014, but the increase was only minor for whites and huge for blacks.
CDC data:
https://www.cdc.gov/mmwr/volumes/66/wr/mm6631a9.htm
I would like to see an account of the rising use of "equity" and "disparities." With "equality" and "inequality", there's a long tradition of debate about when and how much inequality is justified, what kinds of equality are or are not important, etc. With the newer language, that disappears: it's taken for granted that equity is good and disparities are bad. I'm not sure about the extent to which terminology is the cause, but it certainly goes along with a naive way of thinking.
ReplyDelete