Andrew Gelman

The document doesn't say who that is.

Andrew Gelman, a Columbia statistics and political science professor who has criticized Haidt's work, wrote this week that he was never hired by Facebook to go after Haidt.

He said he had taken NoString's money from Meta before, but was never asked to look at social media and teen health.

Meta was approached for comment, but for now, this is a record of what staff discussed, not proof of what the company did.

And one of the blogs that I, two, three blogs that I picked up on reading, which are all related to data and stats.

One of them I still read, it's by this guy called Andrew Gelman.

He's a professor at Columbia.

He used to write about statistical modeling, causal inference and social science.

So I'm asking

what else.

These kind of like, but on the other hand, these softwares have a huge impact on how how you how you can how you can process information how you think and things like that let me give you a very simple example right for all my so i started my journey in data science in terms of programming data stuff, actually, in 2008, I'll say, like, well, formally got into it, post Goldman Sachs, but like 2008, I started, I think, that's when from Andrew Gelman's blog, I found out about R and about how you can load up data in just data to a whole bunch of statistical analysis and things like that.

So this was before this package called Pandas, which everybody uses for Python, that was that was invented the following year and things like that.

Joining me are regular panelists, John Bailer, chair of Miami Statistics Department, and Richard Campbell, former chair of media, journalism, and film.

Our guest today is Andrew Gelman.

Gelman is a professor of statistics and political science, as well as director of the Applied Statistics Center at Columbia University.

He's received the Outstanding Statistical Application Award three times.

So as political scientists, we have to kind of separate the ideological content from the empirical content because it's not kind of our job to judge someone's views.

It's more that we're trying to describe them.

You're listening to Stats and Stories, and today we're talking with Columbia University's Andrew Gelman.

Andrew, this is going to come out a few weeks before the election.

The ease with which unverified figures were absorbed into mainstream political narratives highlights a systemic breakdown in media vetting and data aggregation, Wedgwood said in its statement.

Andrew Gelman, a professor of statistics and political science at Columbia University, questioned Wedgwood's polling statement.

that it had always been intended to be an experiment.

Maybe so, but it's kind of suspicious that they happened to release the statement on the very day that they got caught.

And then you also had some questioning if this was really an experiment at all.

With, for example, Andrew Gelman, a professor at Columbia University, writing in a blog post, it's kind of suspicious that they happened to release this statement on the very day that they got caught.

But regardless of whether it was an experiment or a strategy, you had people concerned about the kind of consequences that these fake numbers might have had.

And there, what you've seen is that the person who's been the most vocal is Michigan State Senator Mallory McMorrow, who's a Democrat who dropped out of the Senate primary back in July.

So I actually completed my degree and everything and then just fell in love with like forecasting and stuff like that.

I then went from undergrad to Columbia to do a master's in their quantitative methods in the social sciences program that Andrew Gelman helped run.

Um, and at that time I took a class on GPU computing and deep learning because I was very pretentious at the time.

And I said, oh, well I'm going to take these classes just so I can talk about them at parties and tell people how stupid they are.

Statistical forecasting.

Nate Silver's model, which descends from the old 538 model, the Economist model, built with Andrew Gelman and Elliot Morris.

These are Bayesian models.

They take poles, they take fundamentals like the economy, They estimate how wrong the polls usually are, and they simulate the election tens of thousands of times.