Treating unobserved heterogeneity in PLS path modelling: A comparison of FIMIX-PLS with different data analysis strategies

Publication Type:
Journal Article
Journal of Applied Statistics, 2010, 37 (8), pp. 1299 - 1318
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2009000675OK.pdf239.99 kB
Adobe PDF
In the social science disciplines, the assumption that the data stem from a single homogeneous population is often unrealistic in respect of empirical research. When applying a causal modeling approach, such as partial least squares path modeling, segmentation is a key issue in coping with the problem of heterogeneity in the estimated causeeffect relationships. This article uses the novel finite-mixture partial least squares (FIMIX-PLS) method to uncover unobserved heterogeneity in a complex path modeling example in the field of marketing. An evaluation of the results includes a comparison with the outcomes of several data analysis strategies based on a priori information or k-means cluster analysis. The results of this article underpin the effectiveness and the advantageous capabilities of FIMIX-PLS in general PLS path model set-ups by means of empirical data and formative as well as reflective measurement models. Consequently, this research substantiates the general applicability of FIMIX-PLS to path modeling as a standard means of evaluating PLS results by addressing the problem of unobserved heterogeneity.
Please use this identifier to cite or link to this item: