Collaboration and team cohesion are essential for higher education institutions (HEIs) to achieve their strategic goals in teaching, research and community service (Akbari et al., 2016; Sharma & Kaushik, 2022). Higher education institutions operate in complex environments that demand adaptability and innovation, which can be effectively addressed through collaboration among academic staff, institutional leaders and external stakeholders. Organisational cohesiveness refers to a group’s tendency to work together to achieve common goals (Hao et al., 2023) and is crucial to institutional effectiveness. However, unhealthy leadership dynamics, such as toxic leadership, can disrupt social integration and group tasks, thereby weakening collective commitment and organisational performance. In the context of higher education, enhancing the professionalism of academic staff requires strong team collaboration (Akhigbe & Alonge, 2024). This is particularly true for curriculum development and instructional design, which are essential for the learning process and require input from higher education professionals (Bravo et al., 2019). On the other hand, academic staff have individual responsibilities for teaching, research, publication and community service. The individual workload faced by faculty members directs their focus towards fulfilling these three aspects. Simultaneously, they must produce high-quality graduates with competitive future career prospects (Bravo et al., 2019). This underscores the need for team cohesion to enhance the curriculum so that it can exceed the profile of graduates capable of competing (Akhigbe & Alonge, 2024; Bravo et al., 2019). Previous studies reveal that toxic leadership is a widespread phenomenon across sectors, such as healthcare (Von Ungern-Sternberg & Becke-Jakob, 2025), corporate organisation (Rasool et al., 2021), government (Hattab et al., 2022) and higher education (Khasawneh et al., 2024). One study has found that over 70% of participants in one study experienced toxic leadership, which is defined as behaviours and actions by leaders that make employees uncomfortable at work. This study discusses how the cultural context shapes leadership, particularly within South East Asian (Gandolfi et al., 2025). Toxic leadership in public organisations impacts employee mental health (Hattab et al., 2022), a crisis of trust in the leader-follower relationship (Ahmed et al., 2024; Baloyi, 2020), and, as a ‘toxic triangle’, in which destructive leadership, vulnerable followers, and a supportive environment undermine organisational integrity (Padilla et al., 2007). In academic contexts, behaviours such as performativity, management practices and mental health challenges exacerbate staff demoralisation (Suchikova et al., 2024; Sutton, 2020; Watermeyer et al., 2025). Meanwhile, structural inequities trigger organisational cynicism and counterproductive behaviours (Ahmed et al., 2024). Addressing this issue requires strategies, such as safe reporting, social support and strengthening ethical leadership (Bublitz-Berg et al., 2024), and a fundamental transition that focuses on transparency, accountability and employee well-being to improve organisational performance sustainably. Numerous empirical studies have identified various factors influencing organisational cohesiveness, such as leadership style (Ben Sedrine et al., 2021) and procedural justice (Nowak, 2021; Xie et al., 2021). In the realm of public organisations, especially public universities. Kivipõld et al. (2020) stated that performance management is necessary to improve organisational goals that impact not only academic staff and students but also society. Therefore, this study aims to determine the factors influencing organisational cohesiveness with toxic leadership as a mediator.

This section discusses the factors that contribute to organisational cohesiveness. In this study, the analysis covers the theories of organisational cohesiveness, toxic leadership, abuse of power and Machiavellianism. The following is a more detailed explanation.

This section delineates the research strategy and design utilised in this study and is structured into distinct sections: participants, research design and procedures, measurements and data analysis. This quantitative study primarily employs structural equation modelling (SEM) to perform statistical tests that examine the relationships among various research variables.

Based on Appendix 1 Table 1-A1, the descriptive analysis and bivariate correlation matrix show that demographic factors, such as gender, have no significant correlations with organisational cohesiveness, Machiavellianism, abuse of power or toxic leadership. Descriptively, all variables had skewness and kurtosis values within ± 2, indicating a near-normal distribution of the data. Significant negative correlations were observed between organisational cohesiveness and Machiavellianism (r = –0.136, p < 0.01) and between organisational cohesiveness and abuse of power (r = –0.439, p < 0.01). In contrast, organisational cohesiveness was significantly positively correlated with toxic leadership (r = 0.515, p < 0.05). In addition, Machiavellianism was significantly positively correlated with abuse of power (r = 0.125, p < 0.01), and abuse of power was also significantly positively correlated with toxic leadership (r = 0.105, p < 0.01).

We used the path analysis method to determine the direct and indirect effects among the variables (Byrne, 2012; Moustaki et al., 2004). At this stage, the hypotheses were tested, and R² was assessed using Smart-PLS version 04 (Ringle et al., 2024). The path analysis coefficients and t-values are shown in Figure 1, and these metrics should be reported for the structural model (Hair et al., 2019). The significance level of the path coefficients can be evaluated using their sign and magnitude (± 1) and the t-statistic. The recommended t-value is 1.96. Appendix 1 Table 2-A1 presents the path coefficient values and t-values for the hypotheses developed in this study, while the R² values indicate the model’s predictive accuracy. Appendix 1 Table 2-A1 indicates the R² values for the dependent variables.

Based on Appendix 1 Table 2-A1, the R² value of .664 (or 66.4%) indicates that toxic leadership can explain 66.4% of the variation in organisational cohesiveness. This means that a quarter of the variation in organisational cohesiveness can be explained by toxic leadership in this model. The remaining 33.6% is explained by factors outside the model or by error variance. Furthermore, Figure 1 shows that path analysis examines the relationship between Machiavellianism, abuse of power, toxic leadership and organisational cohesiveness. It was found that abuse of power had a highly significant direct effect on toxic leadership (β = 0.136, p = 0.000) and a significant direct effect on organisational cohesiveness (β = 0.073, p = 0.003). Similarly, Machiavellianism also showed a highly significant direct effect on toxic leadership (β = 0.124; p = 0.000) and a highly significant direct effect on organisational cohesiveness (β = 0.054; p = 0.000). Although the effect is significant (> 0.05), toxic leadership had a direct effect on organisational cohesiveness (p = 0.032). This suggests that, in addition to the direct effects of Machiavellianism and abuse of power on organisational cohesiveness, toxic leadership acts as a partial mediator of the relationship between them and organisational cohesiveness, despite its path coefficient being very weak. The R² for toxic leadership (0.664) indicates that the exogenous variables Machiavellianism and abuse of power jointly explain 66.4% of the variability in toxic leadership. This result indicates that Machiavellianism and abuse of power play a significant and strong role in predicting toxic leadership in this model. In contrast, the low R² value for organisational cohesiveness (0.096) indicates that the variables in this model only explain a small portion of the variability in organisational cohesiveness, implying that other, more dominant factors are not identified in this model.

Appendix 1 Table 3-A1 presents the model fit evaluation for organisational cohesiveness; overall results indicate that the model meets the fit criteria. The RMSEA is 0.035; although it slightly exceeds the ideal threshold (< 0.05), it is still considered an acceptable indication of fit (Hooper et al., 2008). This is reinforced by the probability RMSEA (p > 0.05) = 0.255, which is greater than 0.05, implying that the model is not significantly poor in fitting the data. Furthermore, the CFI is 0.951, which exceeds the 0.80 criterion and indicates a strong model fit compared to the null model. Although the TLI = 0.842, which is slightly below the stricter ‘good’ threshold (usually 0.90), this value remains within the acceptable range (Hair et al., 2019). Thus, based on these indicators, it can be concluded that the proposed model has an adequate and acceptable fit to explain organisational cohesiveness.

Based on Appendix 1 Table 4-A1, the hypothesis testing results indicate that all direct and indirect path relationships examined leading to organisational cohesiveness are statistically significant. Regarding direct effects, it was found that toxic leadership has a positive and significant effect (β = 0.156, p = 0.032) on organisational cohesiveness, an unusual finding given the theoretical expectation of a negative impact. Furthermore, Machiavellianism showed the strongest and most significant direct effect (β = 0.379, p = 0.000), followed by abuse of power, which is also significant and positive (β = 0.136, p = 0.000). Subsequently, the test for indirect effects confirmed that toxic leadership acts as a significant mediator in the relationship between Machiavellianism and organisational cohesiveness (β = 0.019, p = 0.041), as well as in the relationship between abuse of power and organisational cohesiveness (β = 0.112, p = 0.037). Overall, the model confirms that Machiavellianism and abuse of power, both directly and as mediated by toxic leadership, significantly and positively influence organisational cohesiveness.

This study aims to determine the factors that predict organisational cohesiveness in higher education. Organisational cohesiveness is found to be significantly predicted by toxic leadership, Machiavellianism and abuse of power. In addition, toxic leadership is also significantly predicted by abuse of power and Machiavellianism. Therefore, toxic leadership partially mediates the relationship between Machiavellian-organisational cohesiveness and abuse of power-organisational cohesiveness. This suggests a significant direct and indirect effect on both Machiavellianism-toxic leadership-organisational cohesiveness and abuse of power-toxic leadership-organisational cohesiveness. The findings of this study support a previous study (Bae & Park, 2020), which found a significant contribution of toxic leadership to organisational cohesiveness. In Aboramadan and Dahleez (2020), organisational cohesiveness is defined as an antecedent of team performance. If individuals are in a cohesive work environment, they will be motivated to interact with colleagues, be more engaged with work, and have pride in their work (Aboramadan & Dahleez, 2020; Riisla et al., 2021). These conditions motivate individuals to coordinate activities so that team performance will improve (Aboramadan & Dahleez, 2020; Riisla et al., 2021). Thus, organisational cohesiveness is related to positive team outcomes. In contrast, toxic leadership is characterised by negative behaviour in leaders with a dysfunctional character and harmful actions towards followers (Lipman-Blumen, 2021; Milosevic et al., 2020). Toxic leaders are considered unlikely to motivate followers, and this has an impact on the quality of an organisation (Ibrahim et al., 2024). This is evidenced by the findings in this study, which show a positive coefficient indicating that an increase in toxic leadership is accompanied by an increase in organisational cohesiveness, and vice versa. Although toxic leadership contributes positively to organisational cohesiveness, negative results are seen in the psychological well-being of employees, which affects performance. Riisla et al. (2021) found that leaders who do not pay attention to employee conditions or ethics and who always encourage followers to work harder and achieve more are able to motivate their followers. However, this has an impact on the interpersonal quality and social resources within the team. As leaders, they engage in destructive behaviour that prevents followers from achieving success and accomplishments (Ahmed et al., 2024). Therefore, toxic behaviour by a leader negatively impacts the achievements of employees and organisations (Ahmed et al., 2024). This study revealed significant direct effects of Machiavellianism and abuse of power on organisational cohesiveness. This finding reveals a new aspect: the positive contribution of Machiavellianism and abuse of power to organisational cohesiveness. Damage to trust and fairness is more immediate and directly affects cohesiveness without the need to go through the full manifestation of toxic leadership. Under stressful situations and environments, a sense of camaraderie can develop among members (Cohen & Lansing, 2021; Gerard & Mathewson, 1966; Oswald et al., 2026), encouraging them to ‘band together’ as a collective defence mechanism against external threats (Sorensen et al., 2024). The results of path analysis found that Machiavellianism does significantly predict organisational cohesiveness. Machiavellian leaders exhibit negative behaviours, such as abusive supervision (De Hoogh et al., 2021), which are destructive to employees and organisations (De Hoogh et al., 2021). Machiavellian behaviour is associated with low social skills (Genau et al., 2022), as observed in Song et al. (2025), who found that Machiavellian leaders contribute negatively to the organisation, resulting in low performance. Thus, it can be concluded that organisational cohesiveness will decrease in proportion to the number of Machiavellian leaders. Overall, the finding that abuse of power positively affects organisational cohesiveness aligns with various theories, such as the team cohesion model (Carron et al., 2002) that highlight the role of trust, fairness and psychological well-being as pillars of cohesiveness. Abuse of power creates a culture of distrust, injustice and fear that fundamentally divides and weakens bonds between members, undermining the organisation’s functioning as a cohesive and unified entity. The findings indicating that abuse of power significantly affects organisational cohesiveness are consistent with Batchelor et al. (2023). This illustrates how abuse of power by those in power fundamentally undermines the bonds and sense of community among organisational members. This phenomenon can be explained through the lens of Kramer (2010) on organisational trust. For example, when a leader abuses power through unfair decisions, discriminatory treatment or exploitation of the position for personal gain, it directly erodes subordinates’ trust in the leadership and integrity of the system. Without trust, individuals become sceptical, feel vulnerable and are reluctant to rely on others or the system, thus inhibiting the formation of strong, mutually supportive bonds (Dirks & De Jong, 2022). This creates an environment where individuals focus more on self-protection than collaboration, as Lord Acton warned that ‘power tends to corrupt, and absolute power corrupts absolutely’ (Alsadaan & Alqahtani, 2024). Abuse of power can change individual character, reduce empathy and increase a sense of entitlement, which encourages corrupt actions that undermine trust. Furthermore, an organisational justice perspective is also crucial in understanding this impact. Abuse of power often manifests in injustice, both procedural (e.g. non-transparent or biased decision-making processes) and distributive (e.g. unfair allocation of resources or rewards). When members perceive systematic injustice, this triggers feelings of frustration, anger and powerlessness. The legitimacy of rational-legal authority relies heavily on procedural justice (Anderson et al., 2012). If leaders deviate from established rules and procedures for personal gain, their legitimacy is eroded, which in turn weakens members’ commitment to the organisation itself. An unfair environment will foster conflict and factionalism, where internal groups may form to fight or oppose the source of abuse, instead of uniting for the organisation’s goals. The psychological implications of abuse of power are also highly relevant. An environment characterised by abuse of power can lead to chronic stress and psychological insecurity among employees (Zhao & Ali, 2025). Fear of retaliation or unfair treatment inhibits open communication and cooperation. In this context, group cohesiveness erodes as members tend to withdraw, feel isolated and avoid interactions that could expose them to risks. A toxic work environment because of abuse of power can also lead to high employee turnover rates, as the literature supports that significant work stress can erode psychological well-being. The ongoing loss of members undermines team continuity and weakens the social networks essential for cohesiveness. This study found that Machiavellianism and abuse of power positively contributed to organisational cohesiveness. Mariam et al. (2022) found that leaders with these characteristics could increase team cohesion. Uppal (2022) found similar results, showing that leaders with dark personalities influence team conditions. Followers had various negative experiences at work, enabling them to empathise with their fellow employees. Such destructive leadership imposes real punishment on workers who do not fully meet their targets, which can increase worker unity (Schyns & Schilling, 2013). The stressful situations caused by Machiavellianism and abuse of power directly affect members’ behaviour and perception, quickly boosting the cohesive bond. However, these two characteristics can also negatively impact employee mental health (Mackey et al., 2021). Based on the SEM results in Figure 1, Machiavellianism is found to significantly predict toxic leadership. This supports the previous study by Pfeffer (2021), which found that Machiavellianism affects toxic leadership. The significance of the relationship between Machiavellianism and toxic leadership results indicates that there is a significant relationship between predictors and mediators. This indicates significant indirect effects among Machiavellianism, toxic leadership and organisational cohesiveness. This study is in line with Toker-Gültaş et al. (2021), who indicated a consistent pattern in Machiavellianism and toxic leadership. The findings on the significance of the indirect effect of Machiavellianism on toxic leadership-organisational cohesiveness revealed that Machiavellianism significantly contributes to toxic leadership. Therefore, hypothesis H1d was confirmed. These results corroborate prior studies (Bradley et al., 2017; Coker et al., 2025). Machiavellian leaders manipulate and control subordinates for their own benefit (Coker et al., 2025). Their behaviour impacts the emergence of trust issues among subordinates, causing mental and physical harm and damaging the organisation’s integrity (Coker et al., 2025). Machiavellian leaders exhibit manipulative, selfish and unethical behaviour (Boak, 2021). This type of leadership is damaging to an organisation and is therefore often referred to as destructive leadership (Thoroughgood et al., 2012). The results of this study indicate that abuse of power significantly predicts toxic leadership. This finding confirms hypothesis (H2b) and has important implications for understanding the dysfunctional dynamics of leadership in organisations. This relationship suggests that individuals or groups who tend to abuse power, or systems that enable abuse of power, may foster or exacerbate toxic leadership behaviours. This phenomenon can be analysed through several theoretical lenses. Pelletier (2010) added that when a person or group is given power without adequate accountability, or if they inherently tend to manipulate power for personal gain, this can change their leadership behaviour to become more destructive. In this case, abuse of power is not just an act but rather a precursor that shapes broader patterns of leadership behaviour. In addition, organisational behaviour theory, particularly concerning dysfunctional work environments, offers a further explanation. When abuse of power becomes part of an organisation’s culture or is left without consequences, it can create an environment that is permissive of negative behaviours (Manaa, 2022; Ora et al., 2025). Leaders who witness or engage in power abuse may internalise that such behaviours are acceptable or even effective for achieving goals, even if in a detrimental way. This may lead them to adopt authoritarian, self-centred or condescending leadership styles, which are the hallmarks of toxic leadership (Okray & Hüseyinzade Şimşek, 2020). In the context of this study, the finding that abuse of power predicts toxic leadership highlights that the abuse of power is not an isolated anomaly but rather can be symptomatic of deeper-rooted problems that promote forms of leadership. These point to the importance of addressing individual acts of abuse of power and revisiting power structures, accountability and overall organisational culture to prevent the emergence of toxic leadership. The finding that toxic leadership significantly mediates the relationship between abuse of power and organisational cohesiveness provides crucial insights into dysfunctional dynamics in the work environment. This finding is in line with various theoretical studies, such as Li et al. (2024), that underscore the destructive impact of toxic leadership and abuse of power on the internal functioning of organisations. Theoretically, abuse of power by leaders, which includes authoritarian behaviour, favouritism or exploitation, undermines trust and procedural justice in organisations (Drory et al., 2022). When these behaviours go unchecked and are reinforced through toxic leadership characteristics, such as narcissism, Machiavellianism and subclinical psychopathy, the impact is much more damaging (Tiwari & Jha, 2022). Toxic leadership serves as a catalyst, exacerbating the negative effects of power abuse by creating a climate of fear, insecurity and demoralisation (Reference). In this context, organisational cohesiveness, which refers to the level of bonding, solidarity and sense of community among members (Kidwell et al., 1997), is significantly eroded. Employees under toxic leadership who abuse power tend to feel alienated, unappreciated and reluctant to collaborate or contribute fully, which ultimately weakens the foundation of team cohesiveness and the organisation as a whole (Phillips et al., 2017). These findings confirm that to build a cohesive organisation, it is important not only to address abuse of power but also to neutralise and prevent the emergence of toxic leadership.

This study provides crucial empirical insights into the structural dynamics of leadership behaviours and employee relations within public HEIs. The findings demonstrate that toxic leadership serves as a critical mediating mechanism that explains how darker personality traits, specifically Machiavellianism and the abuse of power, actively deteriorate organisational cohesiveness. While both Machiavellianism and the abuse of power exert significant direct positive impacts on organisational cohesiveness, their destructive capacity is heavily compounded and transmitted through the manifestation of toxic leadership behaviours in the workplace. Furthermore, the empirical evidence confirms that organisational cohesiveness is significantly and positively predicted by the combined presence of toxic leadership and leadership misconduct. In conclusion, mitigating workplace toxicity is not merely an individual concern but a structural necessity; higher education authorities must establish rigorous behavioural compliance systems and transparent monitoring mechanisms to suppress toxic leadership, thereby safeguarding institutional trust, employee well-being and academic performance.