A new structural equation modeling based method for conditional process analysis

Wai CHAN, Lok Yin Joyce KWAN, Cherry Yik Ting CHOI

Research output: Contribution to conferencePaper

Abstract

In social science research, the analysis of moderation, mediation, and conditional process is an important methodological question that goes beyond multiple regression. Generally speaking, moderation occurs when the relationship between the independent variable (X) and the dependent variable (Y) varies as a function of a third variable (moderator). Mediation occurs when the effect of X on Y is transmitted through an intervening variable (mediator). Conditional process, collectively, refers to models that combine both mediation and moderation processes. These models are important because they allow researchers to understand, describe, and explain complex social phenomenon and human behavior in a more accurate way. Despite their theoretical usefulness, applied researchers always find it difficult to execute the analysis appropriately due to both technical and conceptual reasons. In this study, a new structural equation modeling (SEM) based method for conditional process analysis is proposed. This method has several advantages as compared to the traditional regression-based approach in terms of its flexibility in model specification, efficiency in parameter estimation, and ability in assessing overall model goodness-of-fit. Computationally, a new computer software program, VS, for implementing the proposed method is introduced. Real examples will be given to demonstrate how VS can be used to examine models that involve conditional process.
Original languageEnglish
Publication statusPublished - Jun 2015

Fingerprint

process analysis
mediation
regression
moderator
flexibility
social science
efficiency
ability

Citation

Chan, W., Kwan, J. L.-Y., & Choi, C. Y.-T. (2015, June). A new structural equation modeling based method for conditional process analysis. Paper presented at the 7th International Conference on Probability and Statistics, Smolenice Castle, Slovakia.