Reddit reviews Expert Political Judgment: How Good Is It? How Can We Know?
We found 4 Reddit comments about Expert Political Judgment: How Good Is It? How Can We Know?. Here are the top ones, ranked by their Reddit score.
Princeton University Press
Calibration exercises may not be common outside the rationality sphere, but they certainly didn't start within the rationality sphere either. AFAIK, Philip Tetlock started the trend in Expert Political Judgement when he ran a series of experiments comparing expert predictions to random university undergrads. The experts outperformed, but they could further be divided into foxes and hedgehogs - those who know many things, and those who are quite certain of one thing.
Anyway, Silver quotes Tetlock in his book The Signal and the Noise, so I'm guessing that's where he got the idea.
This book actually goes into that! http://www.amazon.com/Expert-Political-Judgment-Good-Know/dp/0691128715 -
I would recommend it to any Nate Silver fan as I believe this work really put forth a lot of ideas pre-Nate Silver's popularity. It's a lot more rigorous and in-depth than Silver's Signal and Noise though.
As well as this publication made public domain by the CIA that gets into why too much information can lead to worse decisions: https://www.cia.gov/library/center-for-the-study-of-intelligence/csi-publications/books-and-monographs/psychology-of-intelligence-analysis/PsychofIntelNew.pdf
The co-founder of my old fund used to say that "Psychology of Intelligence Analysis" is the #1 book he recommended for trading. I would say that after you learn basic terminology and basic markets info (like what a stock is, the S&P 500, what value investing is), this is absolutely required reading before you read somebody's analysis - and it's now free!
Expert Political Judgment: How Good Is It? How Can We Know?
also by Tetlock
http://www.amazon.com/Expert-Political-Judgment-Good-Know/dp/0691128715/ref=sr_1_1?ie=UTF8&qid=1449363140&sr=8-1&keywords=expert+political+judgment
Re-bonjour,
Pour le premier point, j'ai vu The Dunning-Kruger effect : on being ignorant of one's own ignorance.
50 pages très intéressantes montrant beaucoup d'expériences réalisées sur le sujet, mais demandant un peu de courage pour le lire. Je l'ai lu un peu rapidement, il me faudrait plusieurs lectures.
Page 257, dans une expérience, les personnes qui ont le plus "confiance" ont eu tord. Des personnes ayant tord avaient aussi le même niveau de "confiance" que les personnes ayant eu raison.
Page 277, il cite Expert political judgment de Tetlock, que je pense acheté à l'occasion. Apparemment, il distingue les "renards", ayant une pensée flexible et nuancée des "hérissons", qui font des prédictions basées sur de grandes "théories" et qui ne veulent pas en dévier. Les renards seraient plus précis dans leurs prédictions, et exprimeraient un niveau de confiance moins exubérant et plus approprié que les hérissons.
Page 289, il affirme que l'ignorance engendre une "overconfidence".
Pour le lien de ton article, je ne vais pas le lire. 11$ est un peu cher pour moi sachant que j'ai déjà une petite liste de livre à acheter pour 200€.
Je regarderais tes prochains liens pour le 2) plus tard. Pour la démocratie des crédules, je pense aussi me le commander, cela fait un petit bout de temps que j'en entends parler.