Macartan Humphreys

Alan Jacobs @alanjacobs.bsky.social [ View ] Dec 15, 2023 at 23:44 UTC

This book w/ @macartan.bsky.social – on how to integrate theory with data from diverse empirical strategies – has been many fun years in the making, and we're excited to finally have it out in the world!

See ⬇️ for links to the book at CUP, an open access version, software, and many thank you's!

2 replies 4 reposts 26 likes

Macartan Humphreys @macartan.bsky.social [ View ] Dec 15, 2023 at 19:24 UTC

We try hard to show what you can get out of these models but also when they can lead you astray!

Give it a look and tell us what you think!

So many people to thank:

macartan.github.io/integrated_i...

0 replies 0 reposts 4 likes

Macartan Humphreys @macartan.bsky.social [ View ] Dec 15, 2023 at 19:23 UTC

Along the way the book shows how causal models can guide research design and inference: * how to think about optimal case selection * how to optimally mix methods * how to make inferences when you don't have identification * how to connect theory and empirics 3/4

1 replies 0 reposts 4 likes

Macartan Humphreys @macartan.bsky.social [ View ] Dec 15, 2023 at 19:22 UTC

The basic idea is to use Pearl-style causal models and Gelman-style Bayes methods to justify inferences using within-case (process tracing) evidence, correlational patterns across many cases, or a mix of the two These are powerful, underused tools & they work really well together

1 replies 0 reposts 7 likes

Macartan Humphreys @macartan.bsky.social [ View ] Dec 15, 2023 at 19:21 UTC

@alanjacobs.bsky.social and I are delighted to announce the publication of our new book "Integrated Inferences"

See here: integrated-inferences.github.io for links to:

* The book plus open access version (thanks Cambridge!)
* Accompanying R package on CRAN: CausalQueries

1/4

2 replies 46 reposts 132 likes