The authors respond to a paper about partial least squares and other statistical research methods. Topics include a distinction between parameterization and correct parameterization, distinguishing between latent constructs and composite variables, and information on Eigenvalues and squared loadings.
The article presents an introduction to two papers on partial least-squares modeling published in the current issue including "Using PLS Path Modeling for Assessing Hierarchical Construct Models: Guidelines and Empirical Illustration," and "Assessing Between-Group Differences in Information Systems Research: A Comparison of Covariance- and Component-based SEM."
The article offers information and guidelines when using partial least squares (PLS) modeling. The authors discuss the importance of researchers proposing consistent models with available theoretical knowledge. Researchers should also perform data screening as well as examinations of the psychometric properties of all variables in the model. They discuss some of the factors which should be explored by researchers when choosing appropriate sample sizes with PLS modeling.