Abstract
Partial least square structural equation modelling (PLS-SEM) provides researchers with a predictive tool for theory building. In an attempt to understand deviant behaviour which can potentially become a criminal offence, PLS-SEM opens up a valuable mean to analyse latent constructs are designed from a composite of indicators. At its basic, this is called a first-order variable. Using the first-order variable in a basic predictor-criterion research model illuminates in-depth structure on how each component of these variables affects each other. However, as an analysis moves to a more complex level, the first-order variable poses a great challenge to the researchers. This is especially true when the main focus of a study is to look at a general predisposition of a group of related first-order variables with the criterion. Nevertheless, driven by sufficient theories and validated by appropriate statistical tests, related first-order latent variables can be funnelled into a higher-order latent construct. This in turn, helps to reduce complexity of the overall research model allowing more interpretable output and concise discussions.
Document Type
Conference Proceeding
Date of Publication
2015
Faculty
Faculty of Business and Law
Publisher
Elsevier
School
School of Business
RAS ID
20223
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Comments
Roni, S. M., Djajadikerta, H., & Ahmad, M. A. N. (2015). PLS-SEM Approach to Second-order Factor of Deviant Behaviour: In Constructing Perceived Behavioural Control. Procedia Economics and Finance, 28, 249-253. 7th International Conference On Financial Criminology, 13-14 April 2015,Wadham College, Oxford University, United Kingdom.
https://doi.org/10.1016/S2212-5671(15)01107-7