Should I Use Fixed or Random Effects?

By T. Clark, Drew A. Linzer. Political Science Research and Methods, 2014

802 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

The Manipulation of Basel Risk-Weights (2014, neutral): “…model selection relies on Monte Carlo studies, identifying it as the recommended option for small cross-sectional samples with a short time dimension (Petersen, 2009; Clark and Linzer, 2013), but our results are robust to estimating random effects models, and to not clustering the standard errors.”
Allocation of aid for adaptation to climate change: Do vulnerable countries receive more support? (2017, background): “…climate change: Do… 23 random effects and robust standard errors.6 With only a few observations per country, random effects are more stable than fixed effects, since the shrinkage back to the mean stabilizes them and makes them less susceptible to the influence of outliers (Clark and Linzer 2014).”
When Should We Use Unit Fixed Effects Regression Models for Causal Inference with Longitudinal Data? (2019, background): “In spite of this widespread practice, much methodological discussion of unit fixed effects models in political science has taken place from model-based perspectives (often assuming linearity) with little attention to the causal identification assumptions (e.g., Beck, 2001; Wilson and Butler, 2007; B”