How trust and psychological safety mediate the influence of leadership on workplace flourishing in a high-reliability military context

Introduction

Employee well-being is a cornerstone of organisational sustainability and performance (Kundi et al., 2021). In high-stakes environments like the military, promoting holistic workplace flourishing is vital for personnel resilience and mission success (Rothmann, 2013). Within the Ghanaian context, recent reports indicate a decline in employee well-being (Gallup, 2024), underscoring the urgency of this research. Leadership is widely recognised as a pivotal antecedent to employee well-being (Avolio et al., 2018). Both transformational leadership (TFL), which inspires and motivates followers towards a collective vision (Bass & Riggio, 2006), and authentic leadership (AL), characterised by self-awareness, transparency, and ethicality (Walumbwa et al., 2008, Alhalal et al., 2024), have been linked to positive employee outcomes. However, a critical gap exists in understanding the specific psychological mechanisms that transmit these leadership effects, particularly within the unique socio-cultural and structural context of an African military organisation.

The prevailing narrative, grounded in Social Exchange Theory (Blau, 1964), often positions trust as the primary mediator (Dirks & Ferrin, 2002). Meta-analytic evidence from civilian samples consistently demonstrates that trust significantly mediates the relationship between transformational leadership and employee well-being (Ng, 2017, Peyton et al., 2023), suggesting this pathway is well-established in conventional organisational settings. This trust-mediated pathway may, however, overlook a more foundational condition in hierarchical contexts: psychological safety, defined as the shared belief that one can take interpersonal risks without fear of negative consequences (Edmondson, 1999). In a rigid hierarchy like the military, subordinates may trust a leader’s competence but still feel unsafe to voice concerns or admit mistakes. This distinction suggests that the pathway from leadership to flourishing may be more complex (Hayes, 2018). This study therefore investigates the comparative mediating roles of psychological safety and trust in the relationships between TFL, AL, and workplace flourishing (WPF) within the Ghana Armed Forces (GAF). By employing a mixed-methods design, this paper seeks not only to test a theoretical model but also to provide rich, contextualised explanations for the statistical relationships observed.

The mediating role of trust and psychological safety

The relationship between leadership and employee outcomes is typically transmitted through key psychological mechanisms (Singh & Sarkar, 2019). Trust in leadership is defined as a psychological state comprising the intention to accept vulnerability based upon positive expectations of the leader’s intentions, behaviour, and reliability (Mayer et al., 1995, Iqbal et al., 2020). Psychological safety is the sense that one can speak up with ideas, questions, concerns, or mistakes without fear of being punished or ostracised (Edmondson, 1999; Kahn, 1990). While related, these constructs are conceptually distinct: trust is about confidence in the intentions of another person, whereas psychological safety is about the perceived consequences of taking interpersonal risks in a specific context.

Theoretical basis

This study draws on the Job Demands-Resources (JD-R) Theory (Bakker & Demerouti, 2017) and Social Exchange Theory (SET) (Blau, 1964) to explain the mechanisms through which leadership styles, trust, and psychological safety influence workplace flourishing (Cerfontyne, 2020) and task performance. JD-R emphasizes how transformational and authentic leadership function as job resources that mitigate the impact of high job demands Trust in leadership and psychological safety are considered personal resources that mediate the relationship between leadership styles and outcomes (Moreno et al., 2021, Njoroge, 2024). SET complements this by explaining how employees who experience support, fairness, and authenticity from leaders feel compelled to reciprocate with greater commitment, engagement, and performance (Walumbwa et al., 2008; Edmondson, 1999). The integration of JD-R and SET provides a comprehensive theoretical lens for this study.

Goal of the study

We aimed to examine the relationships between transformational leadership, authentic leadership, and workplace flourishing, and the mediating roles of trust and psychological safety (See Figure 1 for the conceptual framework). Following Figure 1, we tested the following hypotheses:

H1a: Psychological safety mediates the relationship between transformational leadership and workplace flourishing.

H1b: Trust mediates the relationship between transformational leadership and workplace flourishing.

H2a: Psychological safety mediates the relationship between authentic leadership and workplace flourishing.

H2b: Trust mediates the relationship between authentic leadership and workplace flourishing.

H3: The mediating effect of psychological safety on the leadership-flourishing relationship has a statistically larger point estimate than the mediating effect of trust.

H4: Trust and psychological safety sequentially mediate the relationship between leadership styles and workplace flourishing.

Figure 1. Conceptual framework

This study’s significance lies in exploring how employees may create flourishing behaviours in response to leadership. Findings would inform strategies that respect employees’ own values while fostering acceptance of organizational values.

Method

Research design

A two-phase, explanatory sequential mixed-methods design (Creswell & Plano Clark, 2017) was employed to address the research objectives. The rationale for this approach was to use the qualitative data to provide a deeper, context-rich understanding of the statistical relationships identified in the quantitative phase, particularly the nuanced mediating roles of trust and psychological safety.

Participants and setting

Participants were 292 military personnel from the Ghana Armed Forces (Army = 154, Navy = 71, and Air Force = 67), selected using stratified random sampling technique at the 37 Military Hospital (65.5% male, 34.5% female; ages 18–49, M = 32.59, SD = 6.79). After the quantitative analysis, a purposive sampling technique was employed to select 18 participants from the 292 for the interviews. Selection criteria ensured maximum variation in experiences and perspectives based on rank, service branch, and years of service.

Measures

Transformational leadership. Transformational leadership was measured using the 20-item Multifactor Leadership Questionnaire (MLQ-5X) by Bass (2000). The scale captures idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration. A sample item is: “My leader articulates a compelling vision of the future.” Reliability was excellent (α = 0.92).

Authentic leadership. Authentic leadership was assessed using the 16-item Authentic Leadership Questionnaire (ALQ) by Walumbwa et al. (2008). The instrument measures self-awareness, relational transparency, internalized moral perspective, and balanced processing. A sample item is: “My leader says exactly what he or she means.” Reliability was high (α = 0.94).

Psychological safety. Psychological safety was measured using the 7-item team psychological safety scale by Edmondson (1999), adapted to refer to the immediate team/unit within the military context. A sample item is: “It is safe to take a risk on this team.” Reliability was good (α = 0.88).

Trust in leader. Trust in leader was evaluated using a 6-item scale by Adams et al. (2008) assessing employee confidence in and reliance on their direct leader. A sample item is: “I can rely on my leader to keep his/her word.” Reliability was high (α = 0.90).

Workplace flourishing. Workplace flourishing was measured using the 17-item Flourishing-at-Work Scale–Short Form (FWS-SF) by Rautenbach (2015), assessing the tripartite model of well-being encompassing emotional, psychological, and social well-being. A sample item is: “I am engaged and interested in my daily activities at work.” Reliability was excellent (α = 0.95)

Interview. A semi-structured interview protocol was developed to explore participants lived experiences of leadership, trust, psychological safety, and well-being. Example questions included: “Can you describe a situation where you felt comfortable voicing a concern to your superior?” and “How does the military hierarchy influence your willingness to speak up?” Interviews were conducted in a private setting, audio-recorded with permission, and lasted approximately 45–60 min each. They were subsequently transcribed verbatim for analysis.

Procedure

After getting ethical approval from the Ethics Committee for Humanities at the University of Ghana and the Institutional Review Board (IRB) at the 37 Military Hospital (Respective ethics clearance numbers: ECH 178/23-24 and 37MH-IRB/PhD/IPN/890/2024), data collection began with the support of unit commanders on-site. Consent of participants were sought. They were informed about the study’s aim, confidentiality and voluntary participation. Questionnaires were completed during personal time.

Data analysis

Quantitative data were analysed using SPSS 28.0 and AMOS 28.0. After data cleaning and screening for missing values, outliers, and normality, descriptive statistics were done as initial analysis and Pearson’s correlation was conducted to check the relationship between the demographic and main variables. A two-step approach to Structural Equation Modelling (SEM) was employed. First, Confirmatory Factor Analysis (CFA) was used to validate the measurement model and assess discriminant and convergent validity. Second, the structural model was tested to examine direct and indirect relationships. Mediation analyses were conducted using Hayes’ PROCESS Macro (Model 4 for parallel mediation; Model 6 for serial mediation) with 5000 bootstrap samples. Control variables (rank, service branch, and tenure) were initially included but omitted from the final model for parsimony as none had a significant direct effect on workplace flourishing and their inclusion did not alter the significance of hypothesized paths.

To assess the potential influence of common method variance (CMV), both procedural and statistical remedies were employed, following recommendations by Podsakoff et al. (2003). Procedurally, participants were assured of the confidentiality and anonymity of their responses to reduce social desirability bias. Statistically, a marker variable approach was utilized (Lindell & Whitney, 2001). A short form of the Marlowe-Crowne Social Desirability Scale (Strahan & Gerbasi, 1972) was included as a marker variable theoretically unrelated to at least one study construct. This 10-item scale measures the tendency of respondents to present themselves in a socially favourable light (sample item: “I am always willing to admit when I make a mistake”). The scale demonstrated acceptable reliability (α = 0.76). The marker variable exhibited weak and non-significant correlations with all substantive study variables (all r < 0.11, p > 0.05). This pattern suggests that common method variance is not a significant concern in this study and is unlikely to have inflated the observed relationships among the key constructs (Lindell & Whitney, 2001), (See Table 1).

To assess multicollinearity, variance inflation factor (VIF) and tolerance values were examined. All VIF values were below 5.0 and tolerance values exceeded 0.20 (Hair et al., 2010), indicating no multicollinearity concerns (See Table 2).

The measurement model was evaluated using CFA to confirm the distinctiveness of the five latent constructs. The five-factor model demonstrated good fit (χ²/df = 2.11, CFI = 0.96, TLI = 0.95, RMSEA = 0.06), with all factor loadings significant (p < 0.001) and exceeding 0.60, supporting convergent validity (See Table 3).

To establish discriminant validity, a rival model comparison was conducted. The hypothesized five-factor model was compared against a four-factor model where Trust and Psychological Safety were collapsed into a single factor. The five-factor model demonstrated significantly better fit (Δχ² = 184.5, Δdf = 4, p < 0.001), confirming that Trust and Psychological Safety are empirically distinct constructs (Fornell & Larcker, 1981) (See Table 4).

Discriminant validity was further supported using the Fornell-Larcker criterion, as the square root of the Average Variance Extracted (AVE) for each construct exceeded its correlations with all other constructs (See Table 5).

The structural model demonstrated good fit (χ²/df = 2.41, CFI = 0.94, GFI = 0.97, RMSEA = 0.07, SRMR = 0.08) and accounted for 39% of the variance in workplace flourishing (R² = 0.39).

Parallel mediation models were tested with trust and psychological safety as simultaneous mediators between the leadership styles and workplace flourishing (Preacher & Hayes, 2008) (summarised in Table 6). Building on these results, a sequential mediation model was tested to examine whether trust subsequently fosters psychological safety. Figure 2 presents the summarized result of the parallel and sequential mediation in a pictorial form.

Figure 2. Graphical representation of the parallel and sequential mediation effects of psychological safety and trust in the relationship between transformational and authentic leadership and workplace flourishing. Note. *p < 0.05, **p < 0.01, ***p < 0.001.

Multi-group SEM analyses were conducted to assess invariance of mediation paths across gender and rank. Invariance was assessed by comparing constrained and unconstrained models, with evidence for invariance indicated by a change in CFI (ΔCFI) of ≤0.01 (Chen, 2007; Cheung & Rensvold, 2002), (See Table 7).

For the qualitative analysis, transcribed interview data were analysed using thematic analysis following the six-phase approach outlined by Braun and Clarke (2006), with emerging themes discussed and refined between researchers to enhance credibility.

Results

Descriptive and correlational analyses

The means, standard deviations, scale reliability, and intercorrelations for all study variables are presented in Table 8. All key variables were positively and significantly correlated in the expected directions. It is noteworthy that psychological safety demonstrated a stronger bivariate correlation with workplace flourishing (r = 0.66, p < 0.01) than trust did (r = 0.48, p < 0.01), providing an initial indication of its relative importance.

Mediation results

The structural model demonstrated a good fit to the data (χ²/df = 2.41, CFI = 0.94, GFI = 0.97, RMSEA = 0.07, SRMR = 0.08). The predictor variables collectively explained a substantial proportion of the variance in the key endogenous variable. The model accounted for 39% of the variance in Workplace Flourishing (R² = 0.39), indicating that the model has strong explanatory power.

Parallel mediation of trust and psychological safety. We tested parallel mediation models where trust and psychological safety were entered as simultaneous mediators between each leadership style and workplace flourishing. The results reveal a striking pattern for transformational leadership (TFL)-the indirect effect on flourishing via psychological safety was significant, with a point estimate of 0.19 (See Figure 2). Both the bias-corrected 95% confidence interval 0.11, 0.29 and the percentile 95% confidence interval 0.10, 0.28 did not include zero, confirming the mediation effect, thus, H1a was supported. In contrast, the indirect effect via trust was not significant, as both its bias-corrected 95% confidence interval (β = 0.04, 95% CI −0.02, 0.10−0.02, 0.10) and percentile 95% confidence interval (β = 0.04, 95% CI −0.03, 0.09−0.03, 0.09) included zero. Thus, H1b was not supported. This indicates that psychological safety is the sole mediator for TFL, fully explaining its effect on flourishing.

The indirect effect of authentic leadership on flourishing via psychological safety was significant and substantial (β = 0.28). Both the bias-corrected 95% CI 0.18, 0.39 and the percentile 95% CI 0.17, 0.38 excluded zero, therefore H2a was supported. The indirect effect via trust was also significant, as both its bias-corrected 95% CI (β = 0.14, 95% CI 0.05, 0.24) and percentile 95% CI (β = 0.14, 95% CI 0.05, 0.23) excluded zero, though the effect was weaker. Therefore, H2b was also supported. This indicates that both mediators play a role, but psychological safety is the stronger pathway.

Hypothesis 3 proposed that the mediating effect of psychological safety would be stronger than that of trust. This hypothesis was strongly supported. For TFL, psychological safety was the only functioning mediator, as trust’s confidence interval included zero. For AL, the point estimate of the indirect effect through psychological safety (β = 0.28, 95% CI 0.18, 0.39) was twice as large as the effect through trust (β = 0.14, 95% CI 0.05, 0.24), and the confidence intervals did not overlap, indicating a statistically meaningful difference in the strength of these pathways.

Sequential mediation from trust to psychological safety. The analysis revealed a significant sequential indirect effect for Authentic Leadership (β = 0.09). Both the bias-corrected 95% CI 0.04, 0.15 and the percentile 95% CI 0.04, 0.14 excluded zero, confirming the statistical reliability of this effect. Thus, H4 was accepted. This indicates that authentic leaders build trust, and this trust, in turn, creates a climate of psychological safety, which ultimately leads to greater workplace flourishing. However, this sequential pathway was not significant for transformational leadership (β = 0.02). Both the bias-corrected 95% CI −0.01, 0.06−0.01, 0.06 and the percentile 95% CI −0.02, 0.05−0.02, 0.05 included zero. This finding provides an explanation for the parallel mediation results; for authentic leaders, trust is not a dead end—it is a crucial building block that enables the psychological safety necessary for flourishing. For transformational leaders, this sequential process does not hold, and psychological safety operates as a direct and independent mechanism.

Multi-group analysis–Invariance of mediation paths. We tested for measurement invariance and structural path invariance across (a) gender (male vs. female) and (b) rank (Commissioned Officers vs. Non-Commissioned Officers). Invariance was assessed by comparing a constrained model (where paths were forced to be equal across groups) to an unconstrained model (where paths were free to vary). Following established guidelines (Chen, 2007; Cheung & Rensvold, 2002), evidence for invariance is indicated by a change in Comparative Fit Index (ΔCFI) of ≤0.01 between the constrained and unconstrained models.

Invariance across gender. The sample comprised 191 male and 101 female personnel. Configural invariance (the baseline model) showed an acceptable fit for both groups simultaneously. Constraining the mediation paths to be equal across males and females did not result in a significant deterioration of model fit (ΔCFI = 0.004, which is ≤0.01). This indicates that the structural relationships including the mediating roles of trust and psychological safety operate similarly for both male and female personnel in this military context.

Invariance across rank. The sample included 144 Commissioned Officers and 148 Non-Commissioned Officers. Configural invariance was established. Constraining the mediation paths to be equal across ranks also yielded a non-significant change in model fit (ΔCFI = 0.007, ≤0.01). This finding suggests that the mediated pathways from leadership styles to workplace flourishing through trust and psychological safety are consistent and robust across different hierarchical levels within the Ghana Armed Forces.

Interviews

Thematic analysis of the interview data was conducted to provide a context-rich explanation for the quantitative findings, particularly the divergent mediating roles of trust and psychological safety across leadership styles. To ensure the reliability and trustworthiness of the thematic analysis, an inter-coder reliability assessment was conducted. A second coder, independent to the research team, was trained on the coding framework and independently coded a randomly selected subset of the interview transcripts (approximately 30%, n = 6 interviews). The coding outputs from both coders were compared, and Cohen’s kappa (κ) was calculated to assess the level of agreement beyond chance. The analysis yielded a Cohen’s kappa coefficient of κ = 0.85, which exceeds the recommended threshold of 0.80 (Landis & Koch, 1977), indicating excellent inter-coder reliability and strong consistency in the application of the thematic framework.

The demographic characteristics of the interviewees, including pseudonyms, gender, age range, force branch, rank, and years of service, are presented in Table 9. This diverse sample ensured representation across ranks, branches, and experience levels, providing a comprehensive view of leadership dynamics within the GAF.

Three primary themes emerged, which are summarized in Table 10.

Theme 1: The Hierarchy as a dual-faced reality. Participants did not question the necessity of the chain of command; they expressed deep trust in it for operational success. However, this very same structure was identified as the primary inhibitor of psychological safety. This critical distinction explains the quantitative finding that trust was a weak or non-significant mediator. As one Warrant Officer puts it, “I would trust my Major with my life, but I would not feel safe telling him his tactical plan is flawed unless he explicitly asks.” This theme establishes the contextual boundary condition: the hierarchy creates an environment where trust and psychological safety are decoupled. Trust is placed in the person and the rank, but psychological safety, the freedom to speak up with potentially challenging information requires an additional, deliberate act from the leader to be realized.

Theme 2: The authentic leader as an architect of safety. Participants described how certain leaders, predominantly those described with authentic traits, proactively engineered psychological safety. Their methods were behavioral and specific: admitting personal mistakes to model vulnerability, formally instituting “challenge sessions” where rank was temporarily set aside, and holding one-on-one meetings to provide a private channel for feedback. These deliberate actions did more than inspire; they built trust through demonstrated integrity and then that trust is used as a foundation to create a safe space for dialogue. This provides the “why” behind the significant sequential mediation (AL → Trust → Psychological Safety → WPF) found in the quantitative data. As one Junior Officer explained:

“My commander did something simple but powerful. He ended every briefing by asking, ‘What am I missing?’ and then waited in silence. The first time, no one spoke. But he kept doing it. Soon, people started talking. He showed he was fallible, and that made it safe for us to be fallible too” (GAF-SOF-16, Junior Officer).

Transformational leaders were valued for vision and motivation but did not consistently engage in behaviours that actively build psychological safety. In contrast, authentic leadership generally supported openness and trust. However, a deviant case highlights an important boundary condition. A senior officer who practiced transparency by sharing operational challenges found that subordinates interpreted this openness as weakness, leading them to seek guidance elsewhere. What was intended as authenticity was perceived as reduced confidence.

“I thought I was being transparent, you know? I shared the challenges we were facing with the unit—budget cuts, personnel shortages, the whole picture. I wanted them to understand the ‘why’ behind the pressure. But later, I found out that some of the younger soldiers interpreted my openness as weakness or indecisiveness” (GAF-SOF-13, Senior Officer).

This case suggests that the effectiveness of authentic leadership, particularly relational transparency is context-dependent in high-stakes military environments. Expressions of vulnerability must be carefully calibrated and balanced with visible authority and decisiveness. Otherwise, they may undermine, rather than enhance, psychological safety.

The finding refines theory by showing that authentic leadership is not universally effective; its impact depends on follower expectations, hierarchical norms, and situational demands. Leaders must therefore adapt how authenticity is communicated to align with the cultural and operational realities of their context.

Theme 3: Flourishing as intermittent psychological freedom. The qualitative data profoundly refines the understanding of workplace flourishing in this context. It is not a permanent, high-level state of well-being. Instead, it is experienced as intermittent episodes of psychological freedom that occur when the constraints of the hierarchy are temporarily lifted by psychological safety. In these moments, personnel reported experiencing the core dimensions of flourishing: a sense of meaning (their input mattered), engagement (they could focus fully on problem-solving), and positive relationships (genuine collaboration with peers and superiors) (Seligman, 2011). This theme directly links the mediator (PSS) to the outcome (WPF), showing that safety is the trigger for the lived experience of flourishing at work.

Discussion

The study examined the mediating roles of psychological safety and trust in the relationships between transformational leadership, authentic leadership, and workplace flourishing. Combining quantitative and qualitative evidence revealed that leadership development programmes should move beyond inspirational training to building leaders’ capacity to foster psychological safety. The findings contribute to the discourse on the disentanglement of the mediating roles of trust and psychological safety through the integrated lenses of Social Exchange Theory (SET) and the Job Demands-Resources (JD-R) model.

For transformational leadership, the path to flourishing was exclusively through psychological safety, with trust playing no mediating role. The relationship operated through indirect-only mediation, as the direct effect was non-significant while the indirect effect was significant. The JD-R model explains this finding: in a high-demand military hierarchy, transformational leaders’ inspirational motivation remains ineffective unless they first provide psychological safety to buffer against the interpersonal risks of speaking up. Trust alone fails to mediate because it represents confidence in the leader without addressing the situational demand of hierarchical pressure. For authentic leadership, the pathway was more complex, illustrating a positive social exchange cycle. Authentic leadership had both a direct effect on flourishing and significant indirect effects through both trust and psychological safety. The sequential mediation analysis revealed the critical causal chain AL → Trust → Psychological Safety → Flourishing. This pathway exemplifies core SET principles: the authentic leader’s transparency builds follower trust, which enables greater risk-taking, creating psychological safety, ultimately resulting in flourishing. Here, psychological safety acted as complementary mediation (partial mediation), as both direct and indirect effects were significant.

Our qualitative data provides context-specific validation for these explanations. The theme of “The Hierarchy as a Dual-Faced Reality” explains the trust-PSS decoupling through the JD-R lens: soldiers trusted the system, but the hierarchical structure remained an unmet job demand that psychological safety specifically addressed. The theme of “The Authentic Leader as an Architect of Safety” illustrates the SET sequential pathway, showing how leaders use transparent behaviors to build trust and then actively engineer psychological safety. Finally, “Flourishing as Intermittent Psychological Freedom” reframes flourishing as dynamic experiences enabled when psychological safety creates conditions for authentic self-expression.

Our findings both confirm and contradict existing literature. The indirect-only mediation of psychological safety for TFL aligns with Edmondson’s (1999, 2018) work on the importance of safety for team learning but extends it to holistic well-being. However, it contrasts with Liu et al. (2010) and Liu et al. (2023), who found trust to be a significant mediator for TFL. This discrepancy can be attributed to the unique, high-stakes hierarchy of the military, creating a boundary condition for the trust-flourishing relationship. The strong performance of authentic leadership, particularly through the sequential pathway, aligns with Walumbwa et al. (2008) and Kleynhans et al. (2022), providing a mechanistic explanation for why authentic leadership works: it builds trust as a steppingstone to psychological safety. This resonates with Gardner et al. (2021), who assert that relational transparency is critical for effective leadership in military settings.

While our theoretical lense posits that leader behaviors shape follower outcomes, we acknowledge the possibility of reverse causality. This alternative explanation is less likely for several reasons. First, if reverse causality were the primary driver, flourishing would be significantly associated with both leadership styles. However, our results show a differentiated pattern: flourishing was significantly associated with authentic leadership but not with transformational leadership in the structural model. Second, the sequential mediation finding (AL → Trust → Psychological Safety → WPF) provides evidence of a theoretically coherent causal chain, consistent with the proposed direction. As Maxwell and Cole (2007) note, mediation analyses with strong theoretical grounding can provide meaningful evidence for hypothesized directional relationships, even with cross-sectional data. Nevertheless, we acknowledge that bidirectional relationships are possible. Flourishing employees may be more cognitively available to recognize authentic leader behaviors, creating a virtuous cycle where leadership and well-being reinforce each other (Inceoglu et al., 2018).

Implications for research and practice

This study advances leadership theory by demonstrating that psychological safety is a central explanatory mechanism linking leadership to workplace flourishing in high-reliability, hierarchical contexts. This is consistent with Edmondson’s (2018) argument. The results clarify that trust alone is insufficient; rather, psychological safety represents the critical pathway through which leadership behaviours translate into well-being outcomes. This contributes to the affective events framework by highlighting the contextual constraints of hierarchy and the conditional role of mediators in military settings.

Practically, the findings call for a recalibration of leadership development within the Ghana Armed Forces and similar institutions. Training should extend beyond transformational leadership competencies to deliberately build skills in fostering psychological safety, such as non-punitive inquiry, active listening, and leader fallibility. Authentic leadership should be embedded as a core leadership competency, with emphasis on relational transparency and ethical consistency. A key practical recommendation emerging from our qualitative findings is the institutionalization of “challenge sessions”—structured forums where rank is temporarily set aside to encourage open dialogue and feedback.

Limitations and future directions

Despite this study’s novelty, there are some limitations. First, the sample was confined to military personnel stationed at a single tri-service facility rather than a frontline combat or field battalion setting. This may limit the generalizability of the results. Future research should explicitly test the model’s generalizability by replicating this study in frontline combat units, peacekeeping mission contingents, and other high-reliability military contexts to establish the boundary conditions of our findings. Second, given the sensitive nature of research within a hierarchical military institution, we acknowledge the potential risk of reprisal. To mitigate this risk, rigorous data protection protocols were implemented, including anonymization at source, encryption of all digital data, a five-year data deletion timeline, aggregate reporting of findings, and institutional oversight through ethical approval from both the University of Ghana Ethics Committee for Humanities and the 37 Military Hospital Institutional Review Board. Future qualitative research in military contexts should consider even more stringent measures, such as member checking with aggregated findings rather than individual transcripts. Third, the cross-sectional design, while enriched by mixed methods, prevents definitive causal conclusions. Future research should employ longitudinal designs with cross-lagged panel analysis to firmly establish the causal sequence. Fourth, the use of self-report data may introduce common method bias, though procedural remedies were applied and marker variable analysis suggested this was not a significant concern.

Conclusion

This study offers several strengths—the mixed-methods design provided a deeper, context-rich understanding of the statistical relationships identified. The rigorous validation of the measurement model through CFA and the establishment of discriminant validity between trust and psychological safety strengthen the credibility of the findings. Additionally, the multi-group invariance tests confirmed that the mediation model is robust across gender and rank subgroups within the Ghana Armed Forces.

Similarly, this research demonstrates that in the hierarchical context of the Ghana Armed Forces, psychological safety is a more powerful conduit than trust for the influence of leadership on employee flourishing. Transformational leaders, in particular, depend entirely on creating this safety to enhance well-being. For practitioners, the message is clear: developing leaders who can architect psychologically safe climates is not a soft skill, but a strategic imperative for building a resilient and flourishing fighting force.

Also, this study makes a significant contribution to the growing field of African Industrial-Organizational psychology. It moves beyond simply applying Western-developed constructs and theories to an African context and instead demonstrates how that context, in this case, the unique hierarchical structure of a Ghanaian military institution, shapes the relationships between those constructs. The finding that trust is an insufficient mediator challenges a universalist assumption and highlights the need for context-sensitive models. This aligns with the urgent call within African I-O psychology to produce localized knowledge that is both scientifically rigorous and practically relevant for the continent’s unique organizational challenges.

Acknowledgement: The authors gratefully acknowledge the participation of the Ghana Armed Forces personnel and the support of the Military Institutional Review Board.

Funding Statement: The authors received no specific funding for this study.

Author Contributions: Saviour Ayertey Nubuor: Conceptualization, Methodology, Material preparation, Writing—original draft preparation, Reviewing, Editing and statistical analyses. Theresa Esi Bosomtwe: Conceptualization, Methodology, Data collection, Statistical analysis, Writing—reviewing, Editing, Project administration. Kwasi Dartey-Baah: Supervision, Reviewing, Editing. All authors reviewed and approved the final version of the manuscript.

Availability of Data and Materials: Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.

Ethics Approval: Ethical clearance was obtained and approved by the Ethics Committee for Humanities at the University of Ghana (ECH 178/23-24). Additionally, approval was obtained from the Institutional Review Board (IRB) at the Military Organization to proceed with the study (37MH-IRB/PhD/IPN/890/2024). Data collection began with the support of unit commanders on-site. Consent of participants were sought. They were informed about the study’s aim, confidentiality and voluntary participation. Questionnaires were completed during personal time. Participants were assured that their information would be kept confidential. It was clearly stated that the data would only be shared with the research team and used for academic purposes.

Conflicts of Interest: The authors declare no conflicts of interest.

References

Adams, B. D., Waldherr, S., & Sartori, J. (2008). Trust in teams scale, trust in leaders scale: Manual for administration and analyses (DRDC Toronto Report No. CR 2008-090). Toronto, ON, Canada: Defence Research and Development Canada. https://apps.dtic.mil/sti/tr/pdf/ADA488604.pdf. [Google Scholar]

Alhalal, E., Alharbi, J. F., Alharbi, S. T., Alotaibi, S. S., Albagami, N. S., et al. (2024). Impact of authentic leadership on nurses’ well-being and quality of care in the acute care settings. Journal of Nursing Scholarship, 56(2), 189–202. https://doi.org/10.1111/jnu.12978. [Google Scholar] [PubMed] [CrossRef]

Avolio, B. J., Wernsing, T., & Gardner, W. L. (2018). Revisiting the development and validation of the authentic leadership questionnaire: Analytical considerations. Journal of Management, 44(2), 399–411. https://doi.org/10.1177/0149206317739960 [Google Scholar] [CrossRef]

Bakker, A. B., & Demerouti, E. (2017). Job demands-resources theory: Taking stock and looking forward. Journal of Occupational Health Psychology, 22(3), 273–285. https://doi.org/10.1037/ocp0000056. [Google Scholar] [PubMed] [CrossRef]

Bass, B. M. (2000). MLQ Multifactor Leadership Questionnaire for Research. Redwood City, CA, USA: Mind Garden. [Google Scholar]

Bass, B. M., & Riggio, R. E. (2006). Transformational Leadership. (2nd ed.). Mahwah, NJ, USA: Lawrence Erlbaum Associates. [Google Scholar]

Blau, P. M. (1964). Exchange and Power in Social Life. New York, NY, USA: Wiley. [Google Scholar]

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa [Google Scholar] [CrossRef]

Cerfontyne, C. (2020). Workplace flourishing, in-role performance, and organizational citizenship behaviour in an information technology organisation [Master’s thesis]. Potchefstroom, South Africa: North-West University. [Google Scholar]

Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/10.1080/10705510701301834 [Google Scholar] [CrossRef]

Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9(2), 233–255. https://doi.org/10.1207/S15328007SEM0902_5 [Google Scholar] [CrossRef]

Creswell, J. W., & Plano Clark, V. L. (2017). Designing and Conducting Mixed Methods Research. (3rd ed.). Thousand Oaks, CA, USA: SAGE Publications. [Google Scholar]

Dirks, K. T., & Ferrin, D. L. (2002). Trust in leadership: Meta-analytic findings and implications for research and practice. Journal of Applied Psychology, 87(4), 611–628. https://doi.org/10.1037/0021-9010.87.4.611. [Google Scholar] [PubMed] [CrossRef]

Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999;. [Google Scholar] [CrossRef]

Edmondson, A. C. (2018). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Hoboken, NJ, USA: John Wiley & Sons. [Google Scholar]

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104 [Google Scholar] [CrossRef]

Gallup. (2024). State of the Global Workplace: 2024 Report. Retrieved March 10, 2026, from [cited 2026 Jan 1]. Available from: https://www.gallup.com/workplace/349484/state-of-the-global-workplace.aspx. [Google Scholar]

Gardner, W. L., Karam, E. P., Alvesson, M., & Einola, K. (2021). Authentic leadership theory: The case for and against. The Leadership Quarterly, 32(6), 101495. https://doi.org/10.1016/j.leaqua.2021.101495 [Google Scholar] [CrossRef]

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate Data Analysis. (7th ed.). Upper Saddle River, NJ, USA: Prentice Hall. [Google Scholar]

Hayes, T. L. (2018). Relationships between leadership styles and workplace flourishing in the milling industry [Doctoral dissertation, University of Pretoria]. https://repository.up.ac.za/. [Google Scholar]

Inceoglu, I., Thomas, G., Chu, C., Plans, D., & Gerbasi, A. (2018). Leadership behavior and employee well-being: An integrated review and a future research agenda. The Leadership Quarterly, 29(1), 179–202. https://doi.org/10.1016/j.leaqua.2017.12.006 [Google Scholar] [CrossRef]

Iqbal, S., Farid, T., Khan, M. K., Zhang, Q., Khattak, A. et al. (2020). Impact of authentic leadership on organizational citizenship behavior: Mediating role of trust. Journal of Management Development, 39(7/8), 963–978. https://doi.org/10.1108/JMD-05-2019-0167 [Google Scholar] [CrossRef]

Kahn, W. A. (1990). Psychological conditions of personal engagement and disengagement at work. Academy of Management Journal, 33(4), 692–724. https://doi.org/10.5465/256287 [Google Scholar] [CrossRef]

Kleynhans, D. J., Heyns, M. M., Stander, M. W., & De Beer, L. T. (2022). Authentic leadership, trust (in the leader), and flourishing: Does precariousness matter? Frontiers in Psychology, 13, 798759. https://doi.org/10.3389/fpsyg.2022.798759. [Google Scholar] [PubMed] [CrossRef]

Kundi, Y. M., Aboramadan, M., Elhamalawi, E. M. I., & Shahid, S. (2021). Employee psychological well-being and job performance: Exploring mediating and moderating mechanisms. International Journal of Organizational Analysis, 29(3), 736–754. https://doi.org/10.1108/IJOA-05-2020-2204 [Google Scholar] [CrossRef]

Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159–174. https://doi.org/10.2307/2529310 [Google Scholar] [CrossRef]

Lindell, M. K., & Whitney, D. J. (2001). Accounting for common method variance in cross-sectional research designs. Journal of Applied Psychology, 86(1), 114–121. https://doi.org/10.1037/0021-9010.86.1.114. [Google Scholar] [PubMed] [CrossRef]

Liu, J., Siu, O. L., & Shi, K. (2010). Transformational leadership and employee well-being: The mediating role of trust in the leader and self-efficacy. Applied Psychology: An International Review, 59(3), 454–479. https://doi.org/10.1111/j.1464-0597.2009.00407.x [Google Scholar] [CrossRef]

Liu, W., Wang, W., & Yang, S. (2023). Perceived transformational leadership from the coach and athletes’ subjective well-being: A moderated mediated model. Frontiers in Psychology, 13, 1100645. https://doi.org/10.3389/fpsyg.2022.1100645. [Google Scholar] [PubMed] [CrossRef]

Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12(1), 23–44. https://doi.org/10.1037/1082-989X.12.1.23. [Google Scholar] [PubMed] [CrossRef]

Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734. https://doi.org/10.5465/amr.1995.9508080335 [Google Scholar] [CrossRef]

Moreno, A. R., Bravo, M. I. R., García-Guiu, C., Lozano, L. M., Pacheco, N. E. et al. (2021). Effects of emerging leadership styles on engagement—a mediation analysis in a military context. Leadership & Organization Development Journal, 42(5), 665–689. https://doi.org/10.1108/LODJ-06-2020-0242 [Google Scholar] [CrossRef]

Ng, T. W. H. (2017). Transformational leadership and performance outcomes: Analyses of multiple mediation pathways. The Leadership Quarterly, 28(3), 385–417. https://doi.org/10.1016/j.leaqua.2016.11.008 [Google Scholar] [CrossRef]

Njoroge, S. N. (2024). The influence of authentic leadership on employee wellbeing in the banking industry in Kenya [Master’s thesis]. Nairobi, Kenya: Strathmore University. [Google Scholar]

Peyton, T., Gip, H., Paşamehmetoğlu, A., & Guchait, P. (2023). How authentic leadership cultivates trust and desirable workplace behaviors in hotels: Commitment and leader-follower value congruence matters. Journal of Human Resources in Hospitality & Tourism, 22(4), 534–561. https://doi.org/10.1080/15332845.2023.2207987 [Google Scholar] [CrossRef]

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. https://doi.org/10.1037/0021-9010.88.5.879. [Google Scholar] [PubMed] [CrossRef]

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891. https://doi.org/10.3758/BRM.40.3.879. [Google Scholar] [PubMed] [CrossRef]

Rautenbach, C. (2015). Flourishing of employees in a fast moving consumer goods organisation [Doctoral dissertation]. Potchefstroom, South Africa: North-West University. [Google Scholar]

Rothmann, S. (2013). From happiness to flourishing at work: A Southern African perspective. In: M. P. Wissing (Ed.), Well-being research in South Africa (pp. 123–151). Dordrecht, The Netherland: Springer. https://doi.org/10.1007/978-94-007-6368-5_7 [Google Scholar] [CrossRef]

Seligman, M. E. P. (2011). Flourish: A Visionary New Understanding of Happiness and Well-Being. New York, NY, USA: Free Press. [Google Scholar]

Singh, M., & Sarkar, A. (2019). Role of psychological empowerment in the relationship between structural empowerment and innovative behavior. Management Research Review, 42(4), 521–538. https://doi.org/10.1108/MRR-04-2018-0158 [Google Scholar] [CrossRef]

Strahan, R., & Gerbasi, K. C. (1972). Short, homogeneous versions of the Marlowe-Crowne Social Desirability Scale. Journal of Clinical Psychology, 28(2), 191–193. https://doi.org/10.1002/1097-4679(197204)28:2%3C191::AID-JCLP2270280220%3E3.0.CO;2-G [Google Scholar] [CrossRef]

Walumbwa, F. O., Avolio, B. J., Gardner, W. L., Wernsing, T. S., & Peterson, S. J. (2008). Authentic leadership: Development and validation of a theory-based measure. Journal of Management, 34(1), 89–126. https://doi.org/10.1177/0149206307308913 [Google Scholar] [CrossRef]