Title: A novel approach to estimating and testing specific mediation effects in educational research: explication and application of Macho and Ledermann's (2011) phantom model approach
Authors: Harsha N. Perera
Addresses: Faculty of Education, University of Southern Queensland, West Street, Toowoomba QLD 4350, Australia
Abstract: This paper addresses the issue of estimating and testing specific indirect effects in recursive structural equation models by considering Macho and Ledermann's (2011) novel approach to testing specific mediation hypotheses. Mediating processes are critical to understanding the underlying pathways through which predictor variables influence criterion variables. Notwithstanding the utility of assessing mediation for elucidating underlying psychological processes, only recently have educational researchers become increasingly concerned with performing empirical significance tests of specific mediation hypotheses. Following an overview of statistical mediation and consideration of the limitations of widely-used structural equation modelling (SEM) programs in handling tests of specific indirect effects, this paper argues for the use of Macho and Ledermann's (2011) phantom model approach among applied educational researchers to test specific indirect effect hypotheses.
Keywords: indirect effects; mediation effects; educational research; intervening variables; bootstrapping; bootstrap; confidence intervals; mediation analysis; structural equation modelling; SEM; Macho and Ledermann; phantom model.
International Journal of Quantitative Research in Education, 2013 Vol.1 No.1, pp.39 - 60
Received: 04 Apr 2012
Accepted: 16 Jan 2013
Published online: 29 Jul 2013 *