Turnover intentions (TI) stand as an insidious problem that impacts on the functioning of organizations and the well-being of their members. Currently, there is a growing interest in identifying the explanatory mechanisms of TI, in order to strengthen and retain valued employees for organizations. In line with this trend, the aim of the present study was to test an integrative serial multiple mediation model that examined the possible mediating role of burnout and job satisfaction in the relationships between organizational justice and TI. To achieve this objective, a cross-sectional empirical study was carried out on a multi-occupational sample of 408 Argentine employees (219 women and 189 men). Participants completed a self-report questionnaire comprising previously validated measures for the target population. Structural equation modeling showed that perceptions of distributive, procedural, and interpersonal justice have negative indirect effects on TI through burnout and job satisfaction, while perceptions of informational justice exert such effects on TI only through job satisfaction. These results indicate that distributive, procedural, and interpersonal justice perceptions relate to lower levels of burnout, which in turn promote greater job satisfaction and lower TI among employees. In addition, informational justice perceptions are positively related to job satisfaction, leading to a decrease in employees’ TI. Findings are discussed in light of their theoretical and practical implications. Managers and human resource professionals could consider the research results in their attempts to design and implement talent retention strategies within organizations.

El objetivo del presente estudio fue validar y analizar las propiedades psicométricas de la Escala de Rendimiento Laboral Individual de Koopmans et al. (2013). Se ejecutó un estudio piloto con 31 estudiantes de posgrado para analizar la equivalencia operacional de la escala. Se elaboró un estudio principal a efectos de analizar las propiedades psicométricas de la escala puesta a punto en el estudio piloto. En esta ocasión se trabajó sobre una nueva muestra seleccionada en forma no probabilística, la que quedó integrada por 434 trabajadores. Sobre los datos recabados se ejecutaron análisis factoriales (exploratorio y confirmatorio). Sendos análisis indicaron una estructura trifactorial con adecuada consistencia (rendimiento en la tarea: ? = .76, comportamientos contraproducentes: ? = .76, y rendimiento en el contexto: ? = .72), confiabilidad compuesta (CR mayores a .70), y validez convergente (AVE mayor a .50). La versión argentina del instrumento quedó conformada por 13 ítems, demostrando ser una medida confiable y parsimoniosa del rendimiento laboral percibido.

No abstract