“Rising death rates among white Americans caused by misperceived threat to their dominant social status, study shows,” blared the headline, and indeed that’s the central claim the study makes. The problem is that the study doesn’t even measure whites’ sense of status threat, much less prove that status threat causes anything else.
Instead, it looks at how counties changed their voting behavior between the 2000 and 2016 presidential elections — and assumes that shifts toward the GOP (i.e., toward Trump) indicate rising status threat among whites in those counties. Areas that moved toward the GOP tend to be places where white mortality is rising, even after controlling for some other changes in the counties’ characteristics; therefore, racial status threat causes white deaths.
This logic is, of course, ridiculous. It’s certainly true that counties with high white mortality went for Trump — the Washington Post pointed this out before the primaries were even over — but the study gives us no good reason to dismiss the Occam’s Razor explanation that left-behind places where whites are dying in droves voted for Trump simply because he spoke to them directly and tried to win them. Of course, you can find research suggesting that Trump voters were motivated in part by racial anxieties (and his own rhetoric hardly acquits him of this charge), but it’s utterly absurd to treat Trump voting as an outright proxy for such anxieties, because much else is at work here. The study fails to grapple with the evidence that import competition from China played a role in counties’ moving toward Trump, for example, and with the way the opioid epidemic targeted different groups of people to different degrees for a host of reasons.
And even if you were to put all these problems to the side, a finding that places with rising racial fears also have rising mortality would hardly prove that abstract worries about whites’ status cause people to put guns in their mouths or overdose on fentanyl. It could just as easily be the case that drug and suicide epidemics in poor communities leave people looking for scapegoats.
There are lots of theories as to why white mortality rates have started climbing, and another recent study says other racial groups have begun heading in that direction as well. I imagine that a lot of factors are at play, and that the role of white people suddenly discovering that the country will be minority-majority in a few decades will turn out to be pretty small.
Read more from the statistician Andrew Gelman here.