Abstract
As the accumulated research base of the behavioral sciences have grown, the amount of actual knowledge discovery has not kept pace as evidenced by an increasing number of disconnected theories and the related problem of construct proliferation. Therefore, integrating social and behavioral sciences across research areas or even disciplines in a meaningful way is imperative. Despite the information systems (IS) discipline’s leadership on creating nomological networks and inter-nomological networks for research integration, a quantitative approach to automatically establish nomological networks from large-scale data is missing. Based on the design science paradigm, we therefore propose a novel natural language processing based approach bringing together these two previous research endeavors. We used a dataset consisting of all the relevant behavioral studies from two tops journal in the IS and psychology fields to evaluate our approach in comparison to human decisions. Finally, the limitations and possible extensions of our approach are critically discussed.
Recommended Citation
Li, Jingjing and Larsen, Kai, "Establishing Nomological Networks for Behavioral Science: a Natural Language Processing Based Approach" (2011). ICIS 2011 Proceedings. 24.
https://aisel.aisnet.org/icis2011/proceedings/knowledge/24
Establishing Nomological Networks for Behavioral Science: a Natural Language Processing Based Approach
As the accumulated research base of the behavioral sciences have grown, the amount of actual knowledge discovery has not kept pace as evidenced by an increasing number of disconnected theories and the related problem of construct proliferation. Therefore, integrating social and behavioral sciences across research areas or even disciplines in a meaningful way is imperative. Despite the information systems (IS) discipline’s leadership on creating nomological networks and inter-nomological networks for research integration, a quantitative approach to automatically establish nomological networks from large-scale data is missing. Based on the design science paradigm, we therefore propose a novel natural language processing based approach bringing together these two previous research endeavors. We used a dataset consisting of all the relevant behavioral studies from two tops journal in the IS and psychology fields to evaluate our approach in comparison to human decisions. Finally, the limitations and possible extensions of our approach are critically discussed.