The Sordid History of Scam Science
Excerpt: Why do scam science and flawed studies so consistently prevail? For starters, the world is a complex place where parsing cause-and-effect is a tricky thing, especially where living creatures are involved. Existence does not easily yield its secrets. Nevertheless, incentives all list toward preliminary studies with big, scary findings because that makes them novel and important and hence newsworthy. Even cub reporters know not to pitch their editors on stories with headlines like “Careful Scientific Study Replicates Previous Work Showing Small, Nuanced Causal Connection.” “Everything Will Be Just Fine If No Action Is Taken” is also a loser because it won’t sell papers or attract pageviews. Retractions of previous errors are also boring so they end up buried when published at all, leaving the impression that the alarmist hot take was correct even when it was clearly not. “Bad Things May Happen in the Future, Unless”-type stories, by contrast, are commercial winners. If adroitly done they do not even elicit backlash, allowing their perpetuation. First, note the weasel word “may.” Next, the amount of possible destruction and the distance of the prediction in time usually vary directly. Finally, the unless provides yet more wiggle room and a segue into policy proposals. When the world doesn’t end in a decade, everyone has forgotten about the article, the reporter is long gone, s/he wrote “may” anyway so s/he wasn’t technically wrong, and besides, one of the policy proposals was kinda sorta implemented so if anything the story “saved us” from Armageddon. [Some of this article is hilarious while being deadly serious at the same time. Several “big names” are mentioned, such as Paul Krugman and Paul Ehrlich–and remind our host, Bob Hall, to tell his Ehrlich story, if you can get him drunk enough. This is worth reading, guys. Ron P.]
No comments:
Post a Comment