Exposure to Workplace Violence and Turnover Intention among Medical Professionals: The Mitigating Role of Social Support

Technical Summary: Exposure to Workplace Violence and Turnover Intention among Medical Professionals

Problem Statement
Workplace violence (WPV) in the healthcare sector is a pervasive issue, manifesting as psychological aggression, mobbing, sexual harassment, and cyberbullying. These stressors occur within high-pressure environments characterized by heavy workloads, staff shortages, and constant patient interaction. A critical consequence of such exposure is turnover intention—the conscious decision by employees to leave their organization. While existing literature establishes a correlation between WPV and turnover, there is a significant gap in understanding the psychosocial mechanisms that mitigate these effects. Specifically, prior research has largely focused on direct relationships, often overlooking the role of protective resources like social support in developing economies. This study addresses this deficiency by investigating whether social support acts as a mediating factor that buffers the negative impact of various forms of workplace violence on turnover intention among healthcare professionals.

Methodology
The study employed a quantitative research design utilizing Partial Least Squares Structural Equation Modelling (PLS-SEM) to analyze data from 204 healthcare professionals in Chennai, India. The sample included a diverse range of roles (nurses, physicians, technicians, administrators) across public and private hospitals, with a demographic skew toward early-career professionals (84.7% aged 20–29) and those with less than three years of experience.

Data was collected via structured questionnaires using a five-point Likert scale. The instruments were adapted from validated sources:

• Psychological Aggression & Mobbing: Measured using scales by Einarsen et al. (2011) and Einarsen and Raknes (1997).
• Sexual Harassment: Adapted from McDonald (2012).
• Cyberbullying: Assessed based on digital media behaviors.
• Turnover Intention: Measured using the Mobley et al. (1978) scale.
• Social Support: Evaluated as a mediating construct.

The analysis followed a two-step procedure: first, assessing the measurement model for reliability (Cronbach's alpha, Composite Reliability) and validity (Convergent and Discriminant validity via HTMT ratio); second, evaluating the structural model to test direct and indirect (mediated) hypotheses.

Key Results
The structural model analysis yielded the following findings:

1. Direct Effects: Contrary to the initial hypotheses and some prior literature, the direct paths from specific forms of workplace violence (psychological aggression, mobbing, sexual harassment, and cyberbullying) to turnover intention were not statistically significant.

   • Psychological Aggression: $\beta = 0.239, p = 0.229$
   • Mobbing: $\beta = -0.062, p = 0.645$
   • Sexual Harassment: $\beta = -0.006, p = 0.959$
   • Cyberbullying: $\beta = 0.156, p = 0.483$
   • Social Support: Interestingly, the direct path from social support to turnover intention was positive and significant ($\beta = 0.415, p = 0.004$), though the authors interpret the mediation context as the primary driver.

2. Mediating Effects: The study found strong evidence for full mediation. While workplace violence did not directly predict turnover intention, it significantly influenced turnover intention indirectly through its impact on social support.

   • The indirect paths (Violence $\rightarrow$ Social Support $\rightarrow$ Turnover Intention) were statistically significant for Psychological Aggression, Mobbing, and Cyberbullying.
   • This suggests that workplace violence erodes perceived social support, which in turn increases the intention to leave.

3. Sexual Harassment Exception: Unlike the other forms of violence, sexual harassment did not show a significant indirect effect on turnover intention through social support ($p = 0.089$).

Key Contributions and Significance
The paper claims several theoretical and practical contributions:

• Reframing the Violence-Turnover Link: The study challenges the assumption that workplace violence directly causes turnover intention. Instead, it posits that the relationship is contingent upon the availability of resources, specifically social support.
• Theoretical Integration: By demonstrating that social support acts as a full mediator, the study integrates Conservation of Resources (COR) theory and Social Support theory with the Job Demands–Resources (JD-R) model. It illustrates that social support functions as a critical resource that buffers the depletion of emotional and cognitive resources caused by violence.
• Contextual Insight: The findings highlight that in the studied context (Indian healthcare), professionals may tolerate adverse experiences due to occupational norms or limited alternatives, but the erosion of their support networks is the critical tipping point for turnover.
• Practical Implications: The authors argue that interventions should not focus solely on violence prevention but must prioritize the cultivation of robust social support systems. Strategies such as peer support groups, supervisor training, and transparent reporting mechanisms are identified as essential for retaining staff.

Limitations and Modesty of Claims
The authors explicitly acknowledge the study's limitations, maintaining a modest tone regarding the generalizability of the results. They note that the cross-sectional design prevents causal inferences and that reliance on self-reported data may introduce common method bias. Furthermore, the sample's demographic skew toward younger, less experienced professionals limits the external validity to senior staff. The authors suggest that future research should employ longitudinal designs and multi-source data to capture the dynamic nature of these relationships. The paper concludes that while social support is a vital protective factor, the specific mechanisms regarding sexual harassment require further investigation, as the expected mediation was not observed.