For the Berlin Bayesians meetup, organized by Eren Elçi:

Causal Inference and Generalizing from Your Data to the Real World

Andrew Gelman, Department of Statistics and Department of Political Science, Columbia University

Learning from data involves three stages of extrapolation: from sample to population, from treatment group to control group, and from measurement to the underlying construct of interest. Discussions of causal inferences focus on the second of these steps, but all three are important. The first and third steps are relevant for external validity: taking a causal inference from study A and applying it to scenario B. To solve these problems, we use measurement errors, interactions, and multilevel regression and poststratification. We discuss in the context of applied problems in political science, psychology, and pharmacology. This work is in collaboration with Lauren Kennedy.

Location is ResearchGate Berlin, address Chausseestrasse 20.

Wow! Really cool! any chance of recording it? You should also visit Copenhagen!

Hi Andrew,

Just curious to know why your talk will not be covering strategies like this one:

https://rss.onlinelibrary.wiley.com/doi/abs/10.1111/rssa.12357

CK:

What makes you so sure you know what I will not be discussing in my talk?? The abstract is just a starting point. I have no idea what I will say during this hour.

But, just to be sure, if there is anything that you think is super important that you are concerned I won’t cover, I recommend contacting the Berlin Bayesians directly and volunteering to give a talk on it. That’s what I did!

yea – glad this is catching on. See https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3035070 and chapter 11 in https://www.wiley.com/en-us/Information+Quality%3A+The+Potential+of+Data+and+Analytics+to+Generate+Knowledge-p-9781118890653