The Hamilton students' publication is wrong in at least this respect: the first-ever test of the accuracy of predictions in the media...
Nope; Philip Tetlock investigated the same thing, and reported -- in Expert Political Judgment (2005) -- entirely consistent results. Except that Tetlock also noted that inaccuracy was moreover correlated with added fame and riches for conservative pundits. (Tetlock focused on foxes and hedgehogs rather than liberals and conservatives. It just turned out that hedgehog pundits -- one ideological trick; everything else made to fit it -- were right-wing.)
Reading the full article now, I see they discuss Tetlock in some detail right at the outset, and acknowledge the similarities between his investigation and theirs with respect to the focus on testing accuracy.
How that gets transformed into theirs being "the first-ever test of the accuracy of predictions in the media" in the Executive Summary is unclear to me.
Anyhow, the results of the study are interesting if unsurprising; Cal Thomas is a jackass datapoint in the pathological American media.
"Finally, those prognosticators with a law degree were more likely to be wrong."
ReplyDeleteAlso, "I'm a student of the law myself."
Where do you find this stuff?
ReplyDeleteThe Hamilton students' publication is wrong in at least this respect: the first-ever test of the accuracy of predictions in the media...
ReplyDeleteNope; Philip Tetlock investigated the same thing, and reported -- in Expert Political Judgment (2005) -- entirely consistent results. Except that Tetlock also noted that inaccuracy was moreover correlated with added fame and riches for conservative pundits. (Tetlock focused on foxes and hedgehogs rather than liberals and conservatives. It just turned out that hedgehog pundits -- one ideological trick; everything else made to fit it -- were right-wing.)
Reading the full article now, I see they discuss Tetlock in some detail right at the outset, and acknowledge the similarities between his investigation and theirs with respect to the focus on testing accuracy.
ReplyDeleteHow that gets transformed into theirs being "the first-ever test of the accuracy of predictions in the media" in the Executive Summary is unclear to me.
Anyhow, the results of the study are interesting if unsurprising; Cal Thomas is a jackass datapoint in the pathological American media.
Better yet, he's a "media analyst" on "Fox News Watch."
ReplyDeletei thought the wurst prognosticator was the one who predicts the winner of the sausage races.
ReplyDeleteGroan.
ReplyDelete