The Truth(s) on Testing for Mediation in the Social and Organizational Sciences

James M. LeBreton, Jane Wu, Mark N. Bing

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Scopus citations

Abstract

One of the principal goals of scientific inquiry is the elucidation of relationships among constructs, such that strong causal inferences may be drawn (Platt, 1964). The realization of this goal involves the use of scientific concepts and methods for the construction and testing of causal systems. In the social sciences, the most basic causal system consists of two unobserved or latent psychological constructs, two observed variables measuring and thus linked to those constructs, a proposition defining the construct-construct linkage, a hypothesis defining the variable-variable linkage, and a statement of the boundary conditions delimiting the circumstances under which our causal system is expected to hold (Bacharach, 1989). Figure 5.1 presents this system, in which the unidirectional arrows linking constructs, measures, and measures to constructs are assumed to be lawful causal relationships.* Specifically, changes in Constructs X and Y are assumed to cause changes in their respective measures, changes in Construct X are proposed to cause changes in Construct Y, and confirmation of this proposition is hypothesized to result in a verifiable statistical relationship between the measures of X and Y. This basic causal system may be thought of as a primary theoretical system (Bacharach, 1989), a principal nomological network (Cronbach and Meehl, 1955), or a basic construct validation framework (Binning and Barrett, 1989).

Original languageEnglish (US)
Title of host publicationStatistical and Methodological Myths and Urban Legends
Subtitle of host publicationDoctrine, Verity and Fable in the Organizational and Social Sciences
PublisherTaylor and Francis
Pages107-141
Number of pages35
ISBN (Electronic)9781135269661
ISBN (Print)9780805862379
DOIs
StatePublished - Jan 1 2010

All Science Journal Classification (ASJC) codes

  • General Psychology

Fingerprint

Dive into the research topics of 'The Truth(s) on Testing for Mediation in the Social and Organizational Sciences'. Together they form a unique fingerprint.

Cite this