Measurement and modeling practices in business research – problems and solutions
Main funder
Funder's project number: 311309
Funds granted by main funder (€)
- 270 650,00
Funding program
Project timetable
Project start date: 01/09/2017
Project end date: 31/12/2019
Summary
Research in social sciences relies heavily on quantitative research, where theories are tested with statistical analyses. To do this, researchers constantly develop new analysis techniques. However, due to the social nature of these endeavors, failures are unavoidable sometimes leading to adoption of questionable methods or practices. The project contributes to two heated methodological discussions about formative measurement and partial least squares path modelling, both of which have surged in popularity but at the same time face increasing criticism from methodologists, including the principal investigator. The project contributes to the debate by 1) studying when (if ever) these techniques are viable modeling approaches in business research social sciences more generally, 2) by proposing alternative modeling approaches for the scenarios where these two techniques have been previously used, and 3) by releasing software components to support the first two objectives.
Principal Investigator
Primary responsible unit
Related publications and other outputs
- An Updated Guideline for Assessing Discriminant Validity (2022) Rönkkö, Mikko; et al.; A1; OA
- Eight Simple Guidelines for Improved Understanding of Transformations and Nonlinear Effects (2022) Rönkkö, Mikko; et al.; A2; OA
- On ignoring the random effects assumption in multilevel models : review, critique, and recommendations (2021) Antonakis, John; et al.; A2; OA
- Polynomial Regression and Measurement Error : Implications for Information Systems Research (2020) Aguirre-Urreta, Miguel I.; et al.; A1; OA
- A cautionary note on the finite sample behavior of maximal reliability (2019) Aguirre-Urreta, Miguel I.; et al.; A1; OA
