Multiple Regression with Non-Independent Observations: Random-Effects and Fixed-Effects

26 Pages Posted: 3 Apr 2024

Date Written: March 4, 2024

Abstract

Many data sets employed in comparative political science are spatially and/or temporally structured. Spatial and temporal structuring typically lead to statistical dependence that needs to be taken into account for data analysis. This article introduces random-effects models (RE) and fixed-effects models (FE) as analytical tools dealing with dependent observations. This review discusses the problem of statistical dependence and introduces RE and FE models. In addition, it provides decision heuristics and hints for practical application.

Suggested Citation

Ziller, Conrad, Multiple Regression with Non-Independent Observations: Random-Effects and Fixed-Effects (March 4, 2024). Available at SSRN: https://ssrn.com/abstract=4747607 or http://dx.doi.org/10.2139/ssrn.4747607

Conrad Ziller (Contact Author)

University of Duisburg-Essen ( email )

Forsthausweg 2
Duisburg, 47057
Germany

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
33
Abstract Views
101
PlumX Metrics