Wednesday, June 11, 2014 at 10:00 a.m.
Neville Scarfe Building, Room 308A
Title: Comparison of the Approaches to Assessing Statistical Interactions: An Application to Risk Factors for Adolescent Problem Behaviour.
Examining Committee:
Dr. Nand Kishor, Research Supervisor
Dr. Bruno Zumbo, Research Committee
Dr. Amery Wu, Research Committee
ABSTRACT
The purpose of the current project was to utilize and compare several approaches to assessing interactions among continuous variables. The approaches used in the project were: (a) multiple regression, (b) unconstrained mean-centered approach (Marsh, Wen, & Hau, 2004), (c) orthogonalizing approach (Little, Bovaird, & Widaman, 2006), and (d) latent moderated structural equations approach (LMS; Klein & Moosbrugger, 2000). The last three approaches utilize the latent variables framework, and they address some of the limitations of multiple regression technique related to the assumption that the predictors in regression models are measured without error. All selected approaches were applied to a problem from the psychology domain concerned with adolescent problem behaviour. Specifically, the interactions among certain risk factors relevant for adolescent delinquency (i.e., low self-control, family risk, and neighbourhood risk) were assessed. The International Youth Survey data collected from 3114 students in grades 7 to 9, in the city of Toronto, were utilized in the study. The results obtained by different approaches were compared in regard to agreement of the results with the theoretical propositions, direction and magnitude of the obtained regression coefficients, magnitude of the standard errors, and the model fit indices. According to the results of the comparison, there was a considerable consistency among the results of the different approaches. However, some differences were also noted. Obtained differences are of importance to the applied researchers as they may affect researchers’ conclusions in regard to the substantive problems of interest. The current study provided number of highlights that may be of interest to the researchers focused on methodological, as well as those concerned with applied aspects of assessing interactions.