It’s the Quantitative Research Methods Workshop, 12:00-1:15 p.m. in Room A002 at ISPS, 77 Prospect Street

Slamming the sham: A Bayesian model for adaptive adjustment with noisy control data

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

It is not always clear how to adjust for control data in causal inference, balancing the goals of reducing bias and variance. In a setting with repeated experiments, Bayesian hierarchical modeling yields an adaptive procedure that uses the data to determine how much adjustment to perform. We demonstrate this procedure on the example that motivated this work, a much-cited series of experiments on the effects of low-frequency magnetic fields on chick brains, as well as on a series of simulated data sets. We also discuss the relevance of this work to causal inference and statistical design and analysis more generally.

This is joint work with Matthijs Vakar: http://www.stat.columbia.edu/~gelman/research/unpublished/chickens.pdf

I might even use a few slides!

P.S. Originally I was going to send them the announcement below, but then I thought it would be better to go specific, as above. In recent years, I’ve usually given pretty general abstracts and drilled into particular examples during the talk. This time I’ll frame the talk around a particular example and use that as a launching pad for more general discussions.

Anyway, here’s the title and abstract I decided not to use:

Things to Worry About

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

We will also discuss one or more of the following topics, depending on audience interest:
– Theoretical Statistics is the Theory of Applied Statistics: How to Think About What We Do
– Evidence-Based Practice is a Two-Way Street
– Holes in Bayesian Statistics
– How the Replication Crisis Has Made Me Aware of Problems in My Own Research
– Political Science and the Replication Crisis