The Use of Partial Least Squares Path Modeling and Generalized Structured Component Analysis in International Business Research: A Literature Review

摘要

Structural equation modeling (SEM) is by far the best known and most widely used path modeling technique in the international business literature. However, recently international business researchers have begun to use a lesser known path modeling technique called partial least squares (PLS). PLS offers some advantages over SEM such as lower sample size requirements, easier testing of moderating relationships, and built-in capability to handle formative indicators which may explain the increased use by international business researchers. We examine the use of PLS in the international business literature, and the potential of a new path modeling technique called generalized structured component analysis (GeSCA) for international business research. We find mixed support for some of the commonly cited reasons for using PLS over SEM in the international business literature, but do find support for the use of PLS when sample sizes are not large enough for the use of SEM. Finally, we discuss GeSCA s ability to handle multi-group analysis which may make it an attractive alternative over PLS for international business researchers who are using small sample sizes with data from multiple countries.