Francisco CeronFeb 15, 2023
Fabrizio Bernardi

How to select your control variables?

https://doi.org/10.1093/esr/jcac078

Start your week well and read this paper just published on ESR.

And tell your PhD students to read it...

Control variable selection in applied quantitative sociology: a critical review

Abstract. A review of all research papers published in the European Sociological Review in 2016 and 2017 (N = 118) shows that only a minority of papers clearly

OUP Academic
Feb 13, 2023 at 7:48amWeb
Show threadFabrizio BernardiFeb 13, 2023

This paper adds to a lists of papers published on ESR that (self)reflect on how we do empirical sociological research (in Europe) and provide indication on how to improve our research practices

Logit and marginal effect
https://doi.org/10.1093/esr/jcp006

The abuse of significance testing https://doi.org/10.1093/esr/jcw047

Multilevel modelling
https://doi.org/10.1093/esr/jcv059
https://doi.org/10.1093/esr/jcv090
https://doi.org/10.1093/esr/jcy053

[continues]

Logistic Regression: Why We Cannot Do What We Think We Can Do, and What We Can Do About It

Abstract. Logistic regression estimates do not behave like linear regression estimates in one important respect: They are affected by omitted variables, even wh

OUP Academic
Show threadJonathan HorowitzFeb 13, 2023

@FabrizioBernardi I believe you select them based on the following criteria:
1) You brainstorm a list of famous articles in your area and find all of their controls
2) You add them to the model.
3) When reviewers tell you to add more variables, you do it.

Wait, is this how researchers *actually* select our control variables or how we *should* select them?