Partial least squares (PLS) path modeling is a variance-based structural equation modeling technique that is widely applied in business and social sciences. It is the method of choice if a structural equation model contains both factors and composites. This chapter aggregates new insights and offers a fresh look at PLS path modeling. It presents the newest developments, such as consistent PLS, confirmatory composite analysis, and the heterotrait-monotrait ratio of correlations (HTMT). PLS path modeling can be regarded as an instantiation of generalized canonical correlation analysis. It aims at modeling relationships between composites, i.e., linear combinations of observed variables. A recent extension, consistent PLS, makes it possible to also include factors in a PLS path model. The chapter illustrates how to specify a PLS path model consisting of construct measurement and structural relationships. It also shows how to integrate categorical variables. A particularly important consideration is model identification: Every construct measured by multiple indicators must be embedded into a nomological net, which means that there must be at least one other construct with which it is related. PLS path modeling results are useful for exploratory and confirmatory research. The chapter provides guidelines for assessing the fit of the overall model, the reliability and validity of the measurement model, and the relationships between constructs.
|Title of host publication||Partial Least Squares Path Modeling|
|Subtitle of host publication||Basic Concepts, Methodological Issues and Applications|
|Editors||Hengky Latan, Richard Noonan|
|Place of Publication||Cham|
|Number of pages||21|
|Publication status||Published - 2017|