Aug 31, 2012 - Using PLS-SEM â. Issues and Research Opportunities. Academy of Marketing Science. World Marketing Congress. Buckhead, Atlanta.
Using PLS-SEM – Issues and Research Opportunities Academy of Marketing Science World Marketing Congress Buckhead, Atlanta August 31, 2012
Marko Sarstedt, Otto-von-Guericke-University Magdeburg, Germany Christian M. Ringle, Hamburg University of Technology (TUHH), Germany
This presentations is based on the latest publications of the presenters: http://www.imm.bwl.uni-muenchen.de/personen/professoren/sarstedt/sarstedt_publ.html (M. Sarstedt) http://www.tu-harburg.de/hrmo/members/prof-dr-c-m-ringle_11965.html (C.M. Ringle)
Using PLS-SEM – Issues and Research Opportunities Missing data
New algorithms
Model specification and optimization - Mode of measurement model (e.g., CTA-PLS), single items - Nonlinear relationships - Directionality of structural model relationships (Cohen‘s path analysis) - Model optimization
Evaluation
Segmentation
Multigroup
Measurement model invariance
Goodness-of-fit in PLS-SEM (I)
"But PLS does not tell us whether the model is the true model"
Model fit in CB-SEM vs. "Model fit" in PLS-SEM
Proposed by Tenenhaus et al. (2005 – Computational Statistics & Data Analysis): GoF communality R 2
Conceptual issues: Not defined for single item constructs Not defined for formative constructs Does not consider model complexity (R² vs. Adjusted R²)
Henseler, J. / Sarstedt, M. (2012). “Goodness-of-fit Indices for Partial Least Squares Path Modeling,“ Computational Statistics, forthcoming.
Goodness-of-fit in PLS-SEM (II) Pre-specified model
Reference: Henseler and Sarstedt (2012 – Computational Statistics)
Model Diagnostics
Heterogeneity Results for the overall sample
Perceived Usefulness
Result for Segment 1
0.7 Intention to Use
Perceived Usefulness
R²= 0.85
0.40
Perceived Ease of Use Intention to Use
0.30
-0.7
Two equally sized segments
Result for Segment 2
R²= 0.20
Perceived Usefulness
0.1
0.00
Intention to Use
Perceived Ease of Use
R²= 0.50
Perceived Ease of Use
0.7
Unobserved Heterogeneity Latent class approaches
PATHMOX
PLS typological regression approaches
PLS-TPM
REBUS-PLS
Distance-based
FPLS-LCD
FIMIX-PLS
PLS-GAS
PLS-POS
Sarstedt M and Ringle CM. (2010) Treating Unobserved Heterogeneity in PLS Path Modelling: A Comparison of FIMIXPLS with Different Data Analysis Strategies. Journal of Applied Statistics 37: 1299-1318.
Literature
Becker, J.-M., Rai, A., Ringle, C.M., and Völckner, F. (2013). Discovering Unobserved Heterogeneity in Structural Equation Models to Avert Validity Threats, MIS Quarterly, forthcoming.
Hair JF, Sarstedt M, Ringle CM, and Mena, J.A. (2012). An Assessment of the Use of Partial Least Squares Structural Equation Modeling in Marketing Research. Journal of the Academy of Marketing Science 40: 414433.
Hair JF, Sarstedt M, Pieper T, and Ringle, C.M. (2012) Applications of Partial Least Squares Path Modeling in Management Journals: A Review of Past Practices and Recommendations for Future Applications. Long Range Planning, forthcoming.
Rigdon, E.E., Ringle, C.M., Sarstedt, M., and Gudergan, S.P. (2011). Assessing Heterogeneity in Customer Satisfaction Studies: Across Industry Similarities and Within Industry Differences, Advances in International Marketing (22), 169-194.
Ringle, C.M., Sarstedt, M., Schlittgen, R., and Taylor, C.R. (2012). PLS Path Modeling and Evolutionary Segmentation, Journal of Business Research, forthcoming.