Issues and opinion on structural equation modeling

In the past few years, the IS field has seen a substantial increase in the number of submissions and publications using structural equation modeling (SEM) techniques. Part of the reason may be the increase in software packages to perform such covariance-based (e.g., LISREL, EQS, AMOS, SEPATH, RAMONA, MX, and CALIS) and componentbased (e.g., PLS-PC, PLS-Graph) analysis. Viewed as a coupling of two traditions -an econometric perspective focusing on prediction and a psychometric emphasis that models concepts as latent (unobserved) variables that are indirectly inferred from multiple observed measures (alternately termed as indicators or manifest variables) -SEM has allowed social scientists to perform path analytic modeling with latent variables (Lvs), which in turn has led some to describe this approach as an example of “a second generation of multivariate analysis” (Fornell 1987, p. 408).