Exploring Talent Loss in Private Enterprises from a Psychological Game Perspective Using Factor Analysis and Multiple Logistic Regression Methods

Factor analysis was used to condense 20 indicators of talent surveys in private enterprises into 4 factors, with a variance explanation rate of more than 72%. The results showed that the weights of the "salary and system" and “training and promotion” factors were in the top two. Confirmatory factor analysis shows that the correlation between factors is strong, with strong convergence effectiveness and fitting effectiveness. Multiple logistic regression analysis revealed that the prediction accuracy of the model reached 61.08%, and there was a negative correlation between monthly income, job matching, and turnover intention, with correlation coefficients of -0.942 and -0.889, respectively, indicating a strong correlation. The process and factor analysis of the psychological game between enterprises and talent, as well as the results of multiple logistic regression analysis, are consistent and confirmed, indicating that job hopping is the result of breaking the game balance. To reduce the turnover rate of talent, enterprises should actively make concessions in the psychological game of labor management relations, including differentiated compensation, smooth promotion channels, and full authorization of talent; additionally, these enterprises should make appropriate compromises and adjustments, with the responsibility of improving their own abilities and serving society, establishing loyalty to the enterprise and responsibility to society. This study further expands the psychological research in the field of talent loss and provides important references for coping strategies for talent loss in private enterprises.