Return
How organizational identification and relative deprivation predict transportation workers' job strain. Combination of decision-tree algorithms on hierarchical risk clusters
G
G
DOI:10.1016/j.trf.2026.103598.png)
Abstract
En 中文
Transport workers routinely face psychosocial risk factors within their work environment, leading to elevated job strain, which is further associated with adverse individual and organizational outcomes. However, the pathways that connect workplace psychosocial factors to employees' job strain are not fully understood. In particular, contextual and social dimensions have gained less attention-compared to individual differences and isolated work factors-and have not been adequately integrated into predictive models. This cross-sectional study sought to elaborate a step-by-step approach for predicting transport workers' job strain levels while testing the integration of socio-psychological variables such as organizational identification and relative deprivation into the predictive framework. Data were collected through self-report questionnaires (COPSOQ-III) from a convenience sample of 439 Greek transport sector employees. Job strain risk scores were calculated, and the participants were classified into three job strain groups (low, medium, and high risk) using Hierarchical Clustering. Next, Decision Tree algorithms (CHAID and CART) were applied to predict risk-group membership. The results highlighted organizational and contextual characteristics as the main psychosocial hazards. Organizational identification emerged as the most significant predictor within the produced models, while Work-Life Conflict and Relative Deprivation were also identified as predictors. The discussed findings support the relevance of socio-psychological mechanisms in occupational strain and suggest meaningful implications, both practical-for organizational interventions-and theoretical.
Keywords:
Workplace psychosocial risks
Transport sector
Decision-tree
Job strain
Organizational identification
Relative deprivation
Journal
IF:
4.4
Papers:
3.3K
Citations:
1.3W
