Developing an Artificial Intelligence–Based Organizational Behavior Model in Iranian Sports Federations: A Mixed-Methods Approach Based on Psychological Capital Theory and Data Mining

Abstract

The present study aimed to develop and validate an artificial intelligence–based organizational behavior model in three selected Iranian sports federations, drawing on psychological capital theory and its four dimensions of hope, self-efficacy, resilience, and optimism. The study employed a sequential exploratory mixed-methods design (QUAL → QUAN). In the qualitative phase, semi-structured interviews were conducted with 18 experts selected through purposive snowball sampling until theoretical saturation was achieved, and the data were analyzed using open, axial, and selective coding. In the quantitative phase, data were collected from 174 employees selected through proportionate stratified random sampling using standardized psychological capital and organizational behavior questionnaires and were analyzed using decision tree, random forest, artificial neural network, support vector machine, K-Means, and Apriori algorithms. The qualitative analysis yielded 82 open codes, 21 subcategories, and six main categories. Cluster analysis identified three behavioral profiles: “Hopeful Resilients” (38%), “Optimistic Self-Efficacious Employees” (42%), and “Employees at Risk of Deviant Behavior” (20%). The Apriori algorithm generated 28 rules with a minimum confidence of 0.70; the strongest rule indicated that the combination of low self-efficacy and low hope was associated with a high level of deviant behavior with a probability of 88%. The artificial neural network achieved the best predictive performance, with an accuracy of 88.6% and an AUC of 0.905 on the test set. Significant differences were observed among the three federations in self-efficacy, hope, organizational commitment, organizational trust, job satisfaction, and organizational citizenship behavior. The final model consists of three operational components—“if–then” rules, standardized behavioral profiles, and a prediction algorithm—and can serve as a decision-support tool for human resource management in sports federations, particularly for the early identification of employees at risk.