Ordinal Logistic Regression Model for Predicting Employee Satisfaction from Organizational Climate

The article introduces ordinal logistic regression as an alternative method for modelling the relationship between predictor variables and job satisfaction. It emphasizes the importance of comprehending job satisfaction factors to enhance organizational performance. The study employs a quantitative approach to predict job satisfaction levels among operational staff in the textile industry, using a test devised by Sonia Palma, consisting of 7 dimensions and 36 items for job satisfaction assessment. Additionally, a 50-items test measures the work climate. By applying a logistic model, the study categorizes job satisfaction into "low - medium"or "high"levels. The dataset encompasses socio-demographic variables and questions from the work climate test CL-SPC (Work Climate - Satisfaction, Productivity and Commitment), which includes five dimensions. Significant factors for the logistic regression model are identified through exploratory factor analysis. These include commitment, autonomy at work, leadership, interpersonal relationships, learning and personal development, clarity of job expectations, motivation, and performance. The analysis unveils associations between these factors and the likelihood of predicting job satisfaction levels. Motivation, job performance and clarity of job expectations emerge as influential predictors. The article recommends fostering a culture of commitment, empowering decision-making, and clearly defining job responsibilities to improve job satisfaction in the textile industry. In conclusion, ordinal logistic regression analysis deepens our understanding of job satisfaction factors in the textile industry, enabling organizations to implement strategies to increase job satisfaction and overall performance. The results of the study enrich our knowledge of job satisfaction and work climate in the textile industry, offering practical guidance to professionals responsible for talent management. © 2023 IEEE.