Predictive Model Comparison and Selection in Marketing Research Using PLS-SEM
Thursday, January 21, 2027 | 15:00 – 18:30
Isola San Servolo, Venice (Italy)
Presenter: Prof. Dr. Christian M. Ringle
(Hamburg University of Technology, Germany, and James Cool University, Australia)
1 Overview
Session 1 (90 minutes):
- Overview of newest PLS-SEM advances and their usefulness for researchers and practitioners
- Predictive model assessment
Break (30 minutes)
Session 2 (90 minutes):
- Alternative models and results comparison (e.g., information criteria and Akaike weights)
- Predictive model selection (e.g., using the CVPAT)
2 Background
Models simplify reality to reveal key relationships. While they can’t capture all its complexity, their strength lies in clarity and focus. While highlighting meaningful relationships, no single model fully explains a phenomenon. Exploring theoretically plausible alternative models for explaining the phenomenon under study is therefore a crucial step in advancing scientific knowledge in that it promotes transparency and challenges the overreliance on any one framework.
Alternative models typically emerge when considering theories in new contexts with unique variables and effects, or when researchers build conceptual bridges across related streams of inquiry to provide a holistic understanding of the phenomenon. Given a set of alternative models, researchers then try to identify the model that best approximates the data generation process underlying the phenomenon under study. Model selection may focus on explanation and confirmation of theoretical models, on their predictive power, or on an integrated combination of both. The choice depends fundamentally on the specific goals of the research, and understanding this purpose is crucial for appropriate model evaluation and justification. The research and model selection process involves numerous decisions made by researchers, informed by theory, methods, data, and results.
3 Workshop set-up and goals
This workshop focuses on predictive model assess, comparison, and selection. In marketing research, partial least squares structural equation modeling (PLS-SEM) is a particular useful method for this purpose. After an introduction and overview of newest advances in PLS-SEM, this special conference workshops focuses on predictive model assessment using PLSpredict and the cross-validated predictive ability test (CVPAT). On these grounds, we will discuss on how to theoretically establish alternative model and compare their model estimation results. For the predictive model comparison and selection, information criteria and the CVPAT represent useful methods, which we will introduce dan discuss.
All concepts will be illustrated “hands-on,” using a case study and the broadly applied statistical software SmartPLS 4 software. The SmartPLS 4 software output diagnostics and interpretation of the results will be covered. Potential obstacles and “rules-of-thumb” to ensure appropriate application and interpretation of the techniques will be addressed.
4 Who should attend
Researchers wishing to learn more about the predictive estimation and assessment of alternative, their results comparison, and predictive model selection using the PLS-SEM method via the SmartPLS 4 software for their top-tier journal publications.
5 Teaching resources
- Certificate of attendance.
- Comprehensive lecture slides will be provided to all participants
- Bring your laptop computer and a 2 or 3-way power extension lead.
- Download and install the SmartPLS 4 software from https://www.smartpls.com/ before attending the workshop. Participants will receive further instructions and a two-month SmartPLS 4 software license key (shortly before the workshop starts).
6 Instructor and a short bio
Christian M. Ringle is a Chaired Professor of Management and Decision Sciences at the Hamburg University of Technology (Germany), and an Adjunct Professor at the James Cook University (Australia). His research, which has been cited more than 400,000 times (Google Scholar), focuses on management and marketing topics, method development, business analytics, machine learning, and the application of business research methods to decision making. Christian’s contributions have been published in journals such as Industrial Marketing Management, International Journal of Research in Marketing, Information Systems Research, Journal of the Academy of Marketing Science, MIS Quarterly, and Organizational Research Methods. Since 2018, Christian has been included in the Clarivate Analytics’ Highly Researchers list. He is a co-founder and co-developer of SmartPLS (https://www.smartpls.com), a statistical software with a graphical user interface. More information: https://www.tuhh.de/mds/team/prof-dr-c-m-ringle.html
7 Key literature
Prior exposure to PLS-SEM is recommended but not required. In any case, we recommend the following PLS-SEM literature:
- Sarstedt, M., Ringle, C. M., & Hair, J. F. (2025). Partial Least Squares Structural Equation Modeling. In C. Homburg, M. Klarmann, & A. E. Vomberg (Eds.), Handbook of Market Research (pp. 1-56). Cham: Springer.
- Sharma, P. N., Shmueli, G., Sarstedt, M., Danks, N., & Ray, S. (2021). Prediction-oriented Model Selection in Partial Least Squares Path Modeling. Decision Sciences, 52(3), 567-607.
- Sharma, P. N., Liengaard, B. D., Hair, J. F., Sarstedt, M., & Ringle, C. M. (2023). Predictive Model Assessment and Selection in Composite-based Modeling Using PLS-SEM: Extensions and Guidelines for Using CVPAT. European Journal of Marketing, 57(6), 1662-1677.
8 Additional literature recommendations
- Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2027). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM) (4 ed.). Thousand Oaks, CA: Sage.
- Guenther, P., Guenther, M., Ringle, C. M., Zaefarian, G., & Cartwright, S. (2023). Improving PLS-SEM Use for Business Marketing Research. Industrial Marketing Management, 111(May), 127-142.
- Hair, J. F., Sarstedt, M., Ringle, C. M., & Gudergan, S. P. (2024). Advanced Issues in Partial Least Squares Structural Equation Modeling (PLS-SEM) (2 ed.). Thousand Oaks, CA: Sage.
- Ringle, C. M., Sarstedt, M., Sinkovics, N., & Sinkovics, R. R. (2023). A Perspective on Using Partial Least Squares Structural Equation Modelling in Data Articles. Data in Brief, 48, 109074.
- Sarstedt, M., Hair, J. F., Pick, M., Liengaard, B. D., Radomir, L., & Ringle, C. M. (2022). Progress in Partial Least Squares Structural Equation Modeling Use in Marketing Research in the Last Decade. Psychology & Marketing, 39(5), 1035-1064.
- Sarstedt, M., Hair, J. F., & Ringle, C. M. (2022). “PLS-SEM: Indeed a Silver Bullet” – Retrospective Observations and Recent Advances. Journal of Marketing Theory & Practice, 31(3), 261-275.
- Liengaard, B. D., Sharma, P. N., Hult, G. T. M., Jensen, M. B., Sarstedt, M., Hair, J. F., & Ringle, C. M. (2021). Prediction: Coveted, Yet Forsaken? Introducing a Cross-validated Predictive Ability Test in Partial Least Squares Path Modeling. Decision Sciences, 52(2), 362-392.
- Sharma, P. N., Sarstedt, M., Ringle, C. M., Cheah, J.-H., Herfurth, A., & Hair, J. F. (2024). A Framework for Enhancing the Replicability of Behavioral MIS Research Using Prediction Oriented Techniques. International Journal of Information Management, 78, 102805.
- Sharma, P. N., Sarstedt, M., Shmueli, G., Kim, K. H., & Thiele, K. O. (2019). PLS-Based Model Selection: The Role of Alternative Explanations in Information Systems Research. Journal of the Association for Information Systems, 20(4), 346-397.
- Danks, N. P., Sharma, P. N., & Sarstedt, M. (2020). Model Selection Uncertainty and Multimodel Inference in Partial Least Squares Structural Equation Modeling (PLS-SEM). Journal of Business Research, 113, 13-24.
- Rigdon, E., Sarstedt, M., & Moisescu, O. I. (2023). Quantifying Model Selection Uncertainty via Bootstrapping and Akaike Weights. International Journal of Consumer Studies, 47(4), 1596-1608.
Additional literature: https://www.smartpls.com à Resources