MODELS FOR MULTILEVEL RESPONSE VARIABLES WITH AN APPLICATION TO GROWTH CURVES

Abstract

Publisher Summary This chapter discusses the models of multilevel response variables. In the majority of applications of multilevel models, one or more response variables are assumed to be measured at the lowest level of the hierarchy. In a two-level model for students grouped within classrooms, there could be achievement test scores measured on students together with an attitude measure for teachers or even the between student standard deviation of test scores as a measure of class heterogeneity, where the variables at both levels are responses regarded as functions of previous achievements and attitudes. The use of height growth data provides a convenient illustration of the usefulness of the multilevel responses model. The same models can be applied to educational data.