The First and the Second Generation of Statistical Methods in Management Accounting Research

Calls for more frequent application of the second-generation statistical methods such as structural equation modeling (SEM) have emerged in the field of management accounting recently. The aim of this article is to compare these statistical methods to the first-generation methods using the real-life example. Specifically, the relationship between the organizational capabilities and perceived nonfinancial performance is investigated. Firstly, the sequential combination of principal component analysis and regression analysis is deployed to the outlined case example. Secondly, partial least squares structural equation modeling (PLS-SEM) is applied to the case example. The comparison of both approaches proves SEM to be more vigilant statistical method for capturing the strength of relationship between latent constructs.