Creating Construct Distance Maps with Machine Learning: Stargazing Trust

2020 | Americas Conference on Information Systems | Citations: 0

Authors: Larsen, Kai; Gefen, David; Petter, Stacie; Eargle, David

Abstract: Within psychometrics, whether in Psychology or the Management Information Syste ... Expand

Abstract: Within psychometrics, whether in Psychology or the Management Information Systems (MIS) discipline, a generalizable distance metric within and between constructs from different studies has not emerged. This paper takes a first step towards creating such a metric by developing and testing the Revelation of Nomological Network (RONIN) algorithm. RONIN uses a combination of machine learning approaches and algorithms to map how construct measurement items are empirically inter-related. The result is an objective, semi-automated, numeric-based tool to develop nomological maps as graphical aids for literature reviews. We apply the RONIN algorithm to the construct of trust within MIS journals. In contrast to some seminal papers about trust outside of MIS, with RONIN we learn that trust within MIS is less about risk, but rather is largely about social uncertainty and is integrally linked to social identification. The implications of RONIN and its potential for transforming research within MIS are also discussed. Collapse

Topics: open source IT risk management privacy database system

Methods: computational algorithm machine learning data transformation XGBoost natural language processing

Theories: trust theory