Interpreting hierarchical linear and hierarchical generalized linear models with slopes as outcomes.(Measurement, Statistics, and Research Design)

ABSTRACT. generally received descriptions of results from hierarchical linear protoplasts (HLM) and hierarchical generalized linear types (HGLM), usually based single on interpretations of individual type parameters, are incomplete in the vicinity of statistically significant and practically important "slope as outcomes" confines in the models. For without fault [i]or[/i] blemish [i]or[/i] flaw description of the resulting interactive protoplasts detailed simple effect descriptions analogous to those drawn out used in factorial ANOVA-based research are emergencyed The required procedure is described and illustrated for 2- and 3-level HLM