Twin Cities Meetup 2/1/20: Estimation, Collective Wisdom, and Survey Data

post by Drake Thomas (RavenclawPrefect) · 2020-01-29T14:55:31.791Z · ? · GW · 0 comments

Contents

No comments

This week’s topic is about group rationality! In particular, we’re going to try and figure out how to use several people’s estimates of some quantity to arrive at a collective answer. (So, for example, if you hear people guess 50, 100, 180, and 185 for some quantity, should your best guess be the average? The median? Something more complicated?) I’ve made a survey with lots of estimation problems, and by Saturday we should have a dozen or two responses (at least) to each of 40 questions. (Feel free to take it yourself, if you’re so inclined!)

I’ll take this data and turn it into something that’s easily played with via a Python script, but I won’t do much actual analysis of it before the meetup. Instead, we’ll get people together to propose questions and algorithms and other investigations to perform, and anyone around with some programming skills can work on hacking together some numerical data to experiment with whatever gets proposed. Hopefully, the result of this meetup will be some interesting analysis of how to evaluate the collective wisdom of a population, which I’ll try to synthesize into a writeup and associated LW post in the spirit of this call for intellectual work at meetups.

As usual, we’ll be situated in Room 204 in Rapson Hall, 89 Church St SE on the east bank of the UMN campus; ping me at grahamsnumberisbig (at) gmail if you get lost. There will be snacks!

0 comments

Comments sorted by top scores.