The purpose of this dissertation is to draw attention to a long neglected, yet very important issue in the statistical modeling of longitudinal data. The issue can arise in any analysis in which one variable, measured at a particular time, is modeled as a predictor or cause of another variable, measured at some later time. The problem is that the magnitude of the variable's effect can vary with the amount of time that elapses between the measurements, or the lag. The dissertation is divided into the following sections: 1) a brief discussion of the issue of causality in models for longitudinal data; 2) an examination of the fundamental role of time lag in any model for longitudinal data in which variables are depicted as predictors or causes; 3) a review of the existing solutions regarding the choice of lags for longitudinal models; 4) the introduction of an alternative strategy to addressing the lag issue: the lag as moderator (LAM) approach; and finally, 5) a demonstration of the potential of the LAM approach by applying it to the analysis of simulated and empirical data.
The University of Kansas prohibits discrimination on the basis of race, color, ethnicity, religion, sex, national origin, age, ancestry, disability, status as a veteran, sexual orientation, marital status, parental status, gender identity, gender expression and genetic information in the University’s programs and activities. The following person has been designated to handle inquiries regarding the non-discrimination policies: Director of the Office of Institutional Opportunity and Access, IOA@ku.edu, 1246 W. Campus Road, Room 153A, Lawrence, KS, 66045, (785)864-6414, 711 TTY.