The NYT sinks to a new low in political coverage
Statistical Modeling, Causal Inference, and Social Science 2024-09-05
OK, this is really embarrassing: “Harris or Trump? The Prophet of Presidential Elections Is Ready to Call the Race.” The video includes a tacky horserace-themed graphic.
No, I do not think the method described at that link is useful, for reasons explained at this post from a few years ago. The short answers are: (a) Some elections are not close at all and any prediction method will get them right, (b) Some elections are so close that for a prediction method to pick the winner is just chance, like picking a coin flip, (c) There is information in the vote margin that is being thrown away if you just try to predict the winner; additional information is being thrown away by using true/false questions. “The Prophet of Presidential Elections” . . . this is just magical thinking, my dude.
It’s bad social science—ok as an amusing feature story, I guess, in the same way that you might get a funny news item about astrologers or Elvis sightings or whatever—but to be featured in this way in the country’s leading newspaper . . . this is really embarrassing, sinking to the level of NPR’s science coverage or stories about aircraft taking off in ferocious tailwinds.
So, yeah, news media outlets can get fooled by promoters, but, jeez, the Times has some serious political reporters—couldn’t someone have run this by them first? And, sure, clickbait works (I linked to them above!), but . . . reputation counts for something, no? This really makes me sad for everyone else who works for this newspaper. They must feel kind of like how I felt after learning about Columbia faking its U.S. News numbers. (You might be interested in this story too.)
And this just took the Times’s reputation down one notch for me. Before, I’d have characterized them as being occasional suckers. Now I’d say they’re active promulgators of junk science.
The sad thing is, I’m pretty sure that most of the other media outlets out there are much worse. History Channel, anyone?