Comparing Two-Stage Approaches to Detect Continuous Manifest Moderating Effects on Construct Relationships

Mei-Fang Chen ； Cherng G. Ding

《Asia Pacific Management Review》 14卷2期 (2009/06) Pp. 137-157

https://doi.org/10.6126/APMR.2009.14.2.03

並列摘要

In this study, seven two-stage approaches in detecting continuous manifest moderating effects on construct relationships are compared in terms of bias and power by using Monte Carlo simulation. The simulation is conducted for each of the conditions formed by different skewness, kurtosis and intercorrelations for exogenous constructs and the moderator, different factor loadings, error variances, and sample sizes under specified models. The results indicate that different approaches tend to produce little bias difference but lead to power difference, varying with different situations. Recommendations for the proper use of the two-stage approaches are provided based on the simulation results obtained.

並列關鍵字

Bias ； confirmatory factor analysis ； moderating effects ； structural equation modeling ； moderated regression analysis ； power

參考文獻

Algina, J.,Moulder, B.C.(2001).A note on estimating the Jöreskog-Yang model for latent variable interaction using LISREL 8.3.Structural Equation Modeling.8(1),40-52.

Grapentine, T.(2000).Path analysis vs. structural equation modeling.Marketing Research.12(3),12-20.

Acito, F.,Anderson, R.D.(1986).A simulation study of factor score indeterminacy.Journal of Marketing Research.23(2),111-118.

Aguinis, H.,Stone-Romero, E.F.(1997).Methodological artifacts in moderated multiple regression and their effects on statistical power.Journal of Applied Psychology.82(1),192-206.

Algina, J.,Moulder, B.C.(2001).A note on estimating the Jöreskog-Yang model for latent variable interaction using LISREL 8.3.Structural Equation Modeling.8(1),40-52.