Controlling for Individual Heterogeneity in Longitudinal Models, with Applications to Student Achievement

By J. R. Lockwood, Daniel F McCaffrey. 2007

67 citations recorded by the source corpus.

Longitudinal data tracking repeated measurements on individuals are highly valued for research because they offer controls for unmeasured individual heterogeneity that might otherwise bias results. Random effects or mixed models approaches, which treat individual heterogeneity as part of the model error term and use generalized least squares to estimate model parameters, are often criticized because correlation between unobserved individual effects and other model variables can lead to biased and inconsistent parameter estimates. Starting with an examination of the relationship between random effects and fixed effects estimators in the standard unobserved effects model, this article demonstrates through analysis and simulation that the mixed model approach has a ``bias compression'' property under a general model for individual heterogeneity that can mitigate bias due to uncontrolled differences among individuals. The general model is motivated by the complexities of longitudinal student achievement measures, but the results have broad applicability to longitudinal modeling.

Subfields: Management Science and Operations Research, Sociology and Political Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/8eea9661b7b749bacfb3ccf9697b71fd076bd57a

How researchers see this paper

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

In Management Science and Operations Research

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Correcting for Test Score Measurement Error in ANCOVA Models for Estimating Treatment Effects (2014, methodology): “…estimated treatment effects include regression adjustment, propensity score methods (Rosenbaum & Rubin, 1983), hierarchical linear modeling and related random effects models (Lockwood & McCaffrey, 2007; Raudenbush & Bryk, 2002), and econometric panel data methods (Greene, 2003; Wooldridge, 2002).”
Getting Value Out of Value-Added: Report of a Workshop (2010, methodology): “Although Rothstein’s study was intended to test the specification of the econometric models, it has important implications for the interpretation of estimates from statistical models as well, because dynamic classroom assignment would also violate the assumptions that Lockwood and McCaffrey (2007) e”
A Model for Teacher Effects From Longitudinal Data Without Assuming Vertical Scaling (2010, methodology): “Models like the persistence models discussed in this article can remove the potential biases from selection into classes, provided there are sufficiently many tests (Lockwood & McCaffrey, 2007).”
A longitudinal study of principals' activities and student performance (2012, result): “…models, a flurry of publications in recent years have supported its use and, in many cases, confirmed its superiority to alternative models (Briggs & Weeks, 2011; Lockwood & McCaffrey, 2007; Lockwood, McCaffrey, Mariano, & Setodji, 2007; McCaffrey et al., 2004; Sanders, 2006; Tekwe et al., 2004).”