Fixed and random effects models for longitudinal data are common in the social sciences. Their primary advantage is that they control for time-invariant omitted variables. However, analysts face several issues when using these models. One is the uncertainty of whether to apply fixed effects (FEM) versus random effects (REM) models. This paper presents a general panel model that includes the standard FEM and REM as special cases. It also presents a sequence of nested models that provide a richer range of models that researchers can easily compare with likelihood ratio tests and fit statistics. Furthermore, researchers can implement our general panel model and its special cases in widely available structural equation models software. An extended empirical example on the cost of motherhood illustrates our results.
Professor Bollen is Director of the Odum Institute for Research in Social Science and the H.R. Immerwahr Distinguished Professor of Sociology at the University of North Carolina at Chapel Hill, where he is also a member of the Statistical Core, a Fellow of the Carolina Population Center, and adjunct professor of Statistics. Bollen is a Fellow of the American Association for the Advancement of Science. The ISI named him among the World’s Most Cited Authors in the Social Sciences. He is coauthor of Latent Curve Models: A Structural Equations Approach (with P. Curran, 2006, Wiley) and author of Structural Equation Models with Latent Variables (1989, Wiley) and over 100 published papers. His primary areas of statistical research are structural equation models, measurement models, and latent growth curve models.