From Digital Leadership to Performance Through Organisational Agility in Open Education Institutions

Underpinning Theory

This study draws on Transformational Leadership Theory (Bass & Avolio, 1994) and Dynamic Capabilities Theory (Teece et al., 1997) to explain how digital leadership may shape Institutional Performance in ODDE settings. Transformational Leadership Theory views leadership as a pattern of behaviours through which leaders set direction, motivate followers, encourage new ways of thinking, and support individual development. In digitally mediated education, these dimensions help explain how leaders communicate a shared digital vision, encourage innovation, support staff in adapting to new technologies, and model the values needed for organisational change.

Leadership behaviours alone may not be sufficient to explain performance in settings marked by ongoing technological and organisational change. Dynamic Capabilities Theory offers a useful extension by shifting attention from leadership behaviour itself to the organisation’s ability to sense emerging opportunities, respond to them, and reconfigure its resources accordingly (Teece, 2007).

Within this study, Organisational Agility is positioned as the mechanism through which leadership behaviours are translated into performance outcomes.

The Connection Between Individualised Consideration, Organisational Agility & Institutional Performance

Prior research indicates that Individualised Consideration can support Institutional Performance, particularly when its effect operates through Organisational Agility (Awashreh, 2025). Leaders who provide personal guidance, show empathy, and attend to employees’ developmental needs are more likely to build a committed and capable workforce (Alhosani & Ahmad, 2024). In turn, this form of support can strengthen the institution’s ability to adapt, learn, and respond to change, which are central features of Organisational Agility (Jaafar et al., 2025; Ozan & Özarslan, 2025). The contribution of Individualised Consideration to Institutional Performance may therefore depend less on immediate performance gains and more on its role in shaping organisational conditions that support responsiveness and resilience.

In digitally changing higher education settings, Organisational Agility therefore provides a plausible mechanism through which a leader’s attention to individual growth may be translated into improved institutional outcomes, including student success and operational effectiveness (Osman et al., 2025a).

The Connection Between Idealised Influence, Organisational Agility & Institutional Performance

The literature indicates that Idealised Influence is associated with Institutional Performance, particularly when it strengthens Organisational Agility (Samodien et al., 2024; Dimple & Tripathi, 2024). In higher education settings, leaders who act as credible role models can strengthen shared purpose and reinforce commitment to institutional direction, including the digital future of the institution (Tayyab & Sharif, 2025). This kind of credibility may be particularly important in periods of change, when staff are required to accept uncertainty, adopt new practices, and engage with innovation. A trusting and cohesive environment can make it easier for employees to take calculated risks and support changes that enhance organisational adaptability (Mohamed Alzeyoudi & Nik Abdullah, 2025). The contribution of Idealised Influence to Institutional Performance appears to operate largely through Organisational Agility. When leadership strengthens trust, alignment, and openness to change, the organisation may become better able to respond to technological shifts and pedagogical demands.

In ODDE institutions, where adaptation is central to continued effectiveness, Organisational Agility thus provides a plausible mechanism linking Idealised Influence to performance outcomes such as operational effectiveness and student success (Osman et al., 2025b).

The Connection Between Inspirational Motivation, Organisational Agility & Institutional Performance

Inspirational Motivation has been proved that may contribute to stronger performance in higher education, with Organisational Agility providing an important pathway through which this effect can occur (Litsulitsa et al., 2024; Dimple & Tripathi, 2024). Leaders who display Inspirational Motivation tend to communicate a clear and compelling sense of direction, which can help staff align with institutional goals and move beyond individual interests in support of shared objectives (Umar et al., 2025; Mahmutoğlu et al., 2025). Such communication may strengthen a collective sense of purpose and encourage the proactive and adaptive behaviours associated with Organisational Agility (Musundire, 2025; Antwi-Boampong et al., 2025).

The role of Inspirational Motivation may extend beyond boosting morale, as it can also help institutions become more responsive to technological change, pedagogical innovation, and shifting external demands. For ODDE institutions, Organisational Agility therefore appears to be a plausible mechanism through which a motivating vision is translated into performance outcomes such as student success, operational effectiveness, and longer-term institutional competitiveness.

The Connection Between Intellectual Stimulation, Organisational Agility & Institutional Performance

Prior research has identified Intellectual Stimulation (IS) as a leadership behaviour that can enhance Institutional Performance, particularly in contexts that require innovation and adaptation (Sazly et al., 2025; Goraya et al., 2024). Leaders who demonstrate IS tend to question assumptions, encourage new ideas, and support experimentation among faculty and staff, thereby fostering a culture of critical thinking and innovation (Musundire, 2025; Mohamad & Osman, 2025; Karunanayaka et al., 2021). Such an environment, marked by intellectual curiosity and psychological safety, can strengthen the dynamic capabilities associated with OA. It encourages employees to reflect critically on existing practices, recognise emerging educational trends and technological change, and develop responses to complex challenges (Alnasser et al., 2025; Soomro & Soomro, 2024). In this context, OA helps explain how Intellectual Stimulation is linked to stronger performance outcomes. By supporting a more adaptive and responsive organisation, IS may contribute to educational quality, operational resilience, and continued relevance in a changing digital higher education environment.

Research Method

This study examines how academic and non-academic staff perceive the direct and indirect effects of Inspirational Motivation, Individualised Consideration, Intellectual Stimulation, and Idealised Influence on Institutional Performance, with Organisational Agility acting as a mediator in ODDE higher institutions. A quantitative cross-sectional survey design was adopted. The study follows an explanatory and predictive approach to test the relationships among the research constructs. Partial Least Squares Structural Equation Modelling (PLS-SEM) was used to examine the hypotheses and evaluate the proposed model. Data were collected through a survey instrument adapted from established studies identified through a comprehensive literature review.

Research Design or Model

Figure 1 presents the research model. Notes: IM = Inspirational Motivation; IC = Individualised Consideration; IS = Intellectual Stimulation; II = Idealised Influence; OA = Organisational Agility; IP = Institutional Performance.

Figure 1

Data Collection Tools

Digital leadership was represented through four transformational leadership behaviours adapted from Bass and Avolio (1994): Inspirational Motivation (4 items), Individualised Consideration (5 items), Intellectual Stimulation (5 items), and Idealised Influence (4 items). In the ODDE context, digital leadership refers to leaders’ ability to communicate a shared digital vision, encourage innovation, provide individualised support, and model values that help the organisation adapt to digitally mediated environments. Organisational Agility (OA), the mediator, was measured using four items developed for this study on the basis of Dynamic Capabilities Theory (Teece et al., 1997). The dependent variable, Institutional Performance (IP) in ODDE, was measured using four items adapted from Richard et al. (2009) to capture employees’ overall assessments of institutional effectiveness. Institutional Performance was assessed through employees’ perceptions rather than archival data, as comparable KPI records were not consistently available across the participating ODDE institutions in Malaysia. Differences in reporting systems and the lack of standardised indicators across institutions made perceptual measures an appropriate way to capture staff views of institutional effectiveness in this context.

All constructs were measured on a five-point Likert scale ranging from strongly disagree to strongly agree. The questionnaire is presented in Appendix A.

Sampling or Research Group

Three ODDE higher education institutions in Malaysia including Open University Malaysia (OUM), Wawasan Open University (WOU), and Asia e-University, were purposively selected because they are established providers operating in digitally mediated teaching, learning, and administrative settings relevant to this study. The participants were academic and non-academic staff who were currently employed at these institutions and involved in institutional processes within the ODDE context. As a complete population list was unavailable, the survey was distributed by email to eligible employees through organisational channels, and participation was voluntary. This sampling approach was used to reach informed respondents who were directly relevant to the research objectives, as random sampling was not feasible in this setting (Creswell & Creswell, 2018; Etikan et al., 2016; Patton, 2015). All respondents provided informed consent before completing the survey. Of the 551 questionnaires distributed, 439 were returned, for a response rate of 79.7%. After data screening, cases with substantial missing data, formulaic responses, or irregular completion were excluded. This left 413 valid responses for final analysis, sufficient for PLS-SEM.

In terms of demographics, 60.0% (n = 248) of the survey participants were male and 40.0% (n = 165) were female. The majority of participants were between 31 and 60 years old, accounting for 84.7% of the total sample. The oldest age group was 41–50 years old (41.6%, n = 172), followed by 31–40 years old (22.3%, n = 92) and 51–60 years old (20.8%, n = 86). Participants under 30 years old accounted for 6.8% (n = 28), while those over 60 years old accounted for 8.5% (n = 35). Consistent with age characteristics, survey participants reported significant work experience, with nearly 60% having 11 to 20 years of experience (11–15 years: 31.0%, n = 128; 16–20 years: 28.3%, n = 117). Smaller proportions reported 6–10 years (14.5%, n = 60) or 21–25 years (12.3%, n = 51), while those with less than 5 years (5.6%, n = 23) or more than 26 years (8.3%, n = 34) accounted for a relatively small proportion. In terms of professional roles, lecturers constituted the majority in the survey sample (67.6%, n = 279), while non-teaching staff accounted for 32.4% (n = 134), ensuring representation from both teaching and administrative functions within the organisation.

Research Procedures

The data analysis used a well-organised, multi-step methodology. The measurement model was initially assessed for internal consistency reliability, convergent validity, and discriminant validity. The structural model was then assessed to examine the hypothesised direct and indirect relationships through bootstrapping. Next, predictive performance was analysed using PLSpredict and the Cross-Validated Predictive Ability Test (CVPAT). To provide additional managerial insight, Importance–Performance Map Analysis (IPMA) was also conducted to identify areas requiring greater managerial attention. All analyses were carried out in SmartPLS 4, which is well suited to complex predictive models and has been widely applied in PLS-SEM research (Ringle et al., 2022).

Validity and Reliability Measures

Following the procedure recommended by Kock (2015), full collinearity variance inflation factor (VIF) values were assessed. All inner VIF values reported in Table 1 were below the prudential cutoff threshold of 3.3, suggesting that general methodological bias is unlikely to be a significant concern in this study.

As shown in Table 2, the measurement model demonstrates acceptable reliability and convergent validity in accordance with the thresholds suggested by Hair et al. (2019). All constructs achieved satisfactory internal consistency, as both Cronbach’s alpha and composite reliability values exceeded 0.70. Although a few indicators recorded loadings slightly below the commonly preferred threshold of 0.70, they were retained because construct-level convergent validity remained satisfactory. For these constructs, composite reliability and AVE still met the recommended thresholds, so the items were retained to preserve content coverage relevant to the ODDE context. All constructs also recorded AVE values above 0.50. Discriminant validity was then assessed using the Heterotrait–Monotrait (HTMT) ratios reported in Table 3, following Henseler et al. (2015).

Structural Model

Following Hair et al. (2017), the structural model was evaluated based on path coefficients (β), coefficients of determination (R²), and the significance of the hypothesised relationships. A bootstrapping procedure with 5,000 subsamples was then used to test both direct and indirect effects in the model. Table 4 presents the results for all hypotheses, including the path coefficients, t-values, p-values, and confidence intervals, indicating that the four leadership dimensions are linked to institutional outcomes through both direct and indirect paths.

Individualised Consideration did not have a significant direct effect on Institutional Performance (H1: β = 0.078, t = 1.336, p = 0.182, CI [–0.033, 0.194]), but it showed a significant positive effect on Organisational Agility (H2: β = 0.141, t = 2.069, p = 0.039, CI [0.003, 0.269]) and a significant indirect effect on Institutional Performance through Organisational Agility (H3: β = 0.057, t = 2.059, p = 0.040, CI [0.003, 0.113]).

Idealised Influence did not have a significant direct effect on Institutional Performance (H4: β = 0.036, t = 0.703, p = 0.482, CI [–0.062, 0.138]), but it showed a significant positive effect on Organisational Agility (H5: β = 0.176, t = 3.214, p = 0.001, CI [0.067, 0.280]) and a significant indirect effect on Institutional Performance through Organisational Agility (H6: β = 0.072, t = 2.800, p = 0.005, CI [0.026, 0.126]).

Inspirational Motivation showed significant direct and indirect effects on Institutional Performance, with a significant direct path to Institutional Performance (H7: β = 0.229, t = 4.766, p < 0.001, CI [0.128, 0.317]), a significant effect on Organisational Agility (H8: β = 0.099, t = 2.116, p = 0.034, CI [0.006, 0.189]), and a significant indirect effect through Organisational Agility (H9: β = 0.040, t = 2.106, p = 0.035, CI [0.004, 0.081]).

Intellectual Stimulation showed a positive direct effect on Institutional Performance (H10: β = 0.171, t = 2.987, p = 0.003, CI [0.060, 0.283]) and the strongest effect on Organisational Agility among the four leadership dimensions (H11: β = 0.272, t = 4.402, p < 0.001, CI [0.152, 0.388]). Organisational Agility also exerted a strong positive effect on Institutional Performance (H12: β = 0.408, t = 8.805, p < 0.001, CI [0.317, 0.497]), while the indirect effect through Organisational Agility was also supported (H13: β = 0.111, t = 4.004, p < 0.001, CI [0.062, 0.170]).

The model explained 31.5% of the variance in Organisational Agility (R² = 0.315) and 51.6% of the variance in Institutional Performance (R² = 0.516), indicating a moderate level of explanatory power overall, with stronger explanatory strength for Institutional Performance. Organisational Agility also emerged as the key mechanism in the model, as it directly affected Institutional Performance and transmitted the effects of the four leadership dimensions to institutional outcomes. The mediation is full for Individualised Consideration and Idealised Influence, but partial for Inspirational Motivation and Intellectual Stimulation.

Effect Sizes (F²)

Effect sizes (f²) were interpreted based on Cohen’s (1992) criteria, with 0.02, 0.15, and 0.35 indicating small, medium, and large effects, respectively. The results in Table 5 show that Organisational Agility (OA) has a medium effect on Institutional Performance (IP) (f² = 0.236). Intellectual Stimulation (IS) has a small effect on OA (f² = 0.056), while Inspirational Motivation (IM) has a small effect on IP (f² = 0.082). The direct effects of Individualised Consideration (IC) and Idealised Influence (II) on Institutional Performance are negligible (f² = 0.005 and f² = 0.001, respectively). This suggests that their contribution to performance is not primarily direct, but operates mainly through Organisational Agility, which is consistent with the significant indirect effects reported in Table 4.

PLSpredicts

Following the PLSpredict procedure proposed by Shmueli et al. (2016, 2019), the model’s out-of-sample predictive performance was assessed. All Q²predict values were above zero, indicating predictive relevance. When compared to the standard linear model (LM), the PLS-SEM model yielded lower RMSE values for six of the eight indicators, while the LM model performed slightly better for IP1 and IP4. These results in Table 6 suggest the model has average predictive ability rather than superior predictive accuracy across all indicators.

Cross-validated Predictive Ability Test (CVPAT)

According to Hair et al. (2022) and Liengaard et al. (2021), CVPAT was conducted to test the model’s out-of-sample predictive ability. Specifically, the negative average loss differences reported for Institutional Performance (IP: –0.152), Organisational Agility (OA: –0.101), and the overall model (–0.127) show that the PLS-SEM model yields lower prediction error than the standard benchmark. The significance of the reported t-values and p-values (p < 0.001) supports the model’s predictive validity in this ODDE context. The CVPAT results are presented in Table 7.

Importance-performance Map Analysis (IPMA)

Organisational Agility (OA) was the most important predictor of Institutional Performance, with the highest total effect (0.408), yet it recorded the lowest performance score (62.284). This suggests that OA should be treated as the main area for improvement. Intellectual Stimulation (IS) and Inspirational Motivation (IM) also showed relatively high importance (0.282 and 0.270, respectively), while their performance levels were moderately high.

Idealised Influence (II), showed the lowest importance (0.108) and a moderate performance score (67.003), suggesting that its current contribution to Institutional Performance is more limited in this setting.

A more relevant approach would be to strengthen the connection between II and the model’s key mediator, Organisational Agility. In practice, this may involve leaders more visibly demonstrating digital transformation behaviours, such as actively using new teaching technologies and showing clear commitment to the institution’s digital vision. The IPMA results are presented in Table 8.