Data-Driven, Statistical Learning Method for Inductive Confirmation of Structural Models

Wolfgang Maaß, Iaroslav Shcherbatyi

In: Hawaii International Conference on System Sciences. Hawaii International Conference on System Sciences (HICSS) January 4-7 Waikoloa Village Hawaii United States University of Hawaii at Manoa 2017.


Automatic extraction of structural models interferes with the deductive research method in information systems research. Nonetheless it is tempting to use a statistical learning method for assessing meaningful relations between structural variables given the underlying measurement model. In this paper, we discuss the epistemological background for this method and describe its general structure. Thereafter this method is applied in a mode of inductive confirmation to an existing data set that has been used for evaluating a deductively derived structural model. In this study, a range of machine learning model classes is used for statistical learning and results are compared with the original model.

Weitere Links

Data-Driven_Statistical_Learning_Method_for_Inductive_Confirmati.pdf (pdf, 1 MB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence