Through idealized influence, a leader understands the behaviors of followers and instills pride in them ( Djourova et al., 2020). Idealized influence involves the capacity to analyze and replicate the paths of successful leaders, which may involve mentorship, fostering relationships, and developing leadership abilities ( Howard, 2016). Ideale-influenced leaders promote confidence and maturity among followers, nurturing their abilities to foster innovative thinking and bringing advanced ideas ( Schiuma et al., 2022).

Research has produced inconsistent results regarding the impact of idealized influence on employees’ innovative behavior. Shafi et al. (2020) found that idealized influence significantly influenced employee creativity and innovation. According to Gong et al. (2009) idealized influence has a positive relationship with EIB, while Sethibe and Steyn (2017) revealed a negative relationship between these two variables.

On the other hand, a study conducted in North America by Zdaniuk and Bobocel (2015) illustrated the relationship between idealized influence and OC. According to Y. A. S. ALmahasneh et al. (2022a) the relationship between idealized influence and OC is positive. H1:

Idealized influence positively relate with EIB in Ethiopian public research Universities

Leaders with intellectual stimulation (IS) encourage novel techniques in task completion and discourage traditional thinking ( Al-Harthy & Yusof, 2016). This means that intellectual stimulation encourages followers to question the status quo ( Masa’deh et al., 2016). Leaders who demonstrate IS within their organizations are believed to enhance team members’ responsiveness to diverse situations, particularly when encountering challenges in the workplace ( Tepper et al., 2018). Studies of Avolio and Bass (1995) and Yukl (2013) suggest that TLs, through IS, challenge followers to re-examine problems and find new solutions, leading to increased innovation within the organization. This means that TL captures followers’ attention by encouraging creativity and innovation through IS ( Tajasom et al., 2015). Additionally, studies results ( Mittal & Dhar, 2015; To et al., 2015) have demonstrated that through IS, TL can foster a sense of innovation in employees.

In addition, TL has been demonstrated to exert a beneficial influence on OC ( Y. A. S. ALmahasneh et al., 2022b). According to Martins and Terblanche (2003), OC has an impact on intellectual stimulation, as it cultivates creativity and innovation. According to Yukl (2013) intellectual stimulation from TLs promotes open communication, which enhances innovation. H3:

Intellectual stimulation positively relate with EIB in Ethiopian public research Universities

When employees are motivated by a leader’s vision and enthusiasm, they are more likely to explore novel ideas and solutions ( Jun & Lee, 2023). Studies in different industries, such as manufacturing ( Toseef et al., 2022), banking ( Tayal et al., 2018), and government sectors ( Ameen et al., 2021) have consistently shown that inspirational motivation enhances employees’ innovative work behavior.

On the other hand the research of Y. A. S. ALmahasneh et al. (2022a) has revealed that inspirational motivation has a positive impact on OC. Leaders who employ inspirational motivation techniques foster optimism and enthusiasm among their employees and motivate them ( Lee et al., 2020). This practice enhances employee morale and creates a positive workplace, ultimately contributing to the culture of engagement. H5:

Inspirational motivation positively relate with EIB in Ethiopian public research Universities

Individualized consideration has been identified as a key aspect of leadership behavior and a potential antecedent of innovative behavior ( López-Domínguez & Enache, 2009). In addition, individualized consideration has been explored to positively influence innovative work behavior ( Situmorang et al., 2023). Additionally, Eisenbeiß and Boerner (2010) found that leaders who provide individualized consideration are more effective in encouraging their followers to contribute to innovative ideas.

On the other hand, a study showed that individualized consideration positively influences OC ( Y. A. S. ALmahasneh et al., 2022b). In addition, individualized consideration of OC can be likened to the understanding of individual cultural profiles. This type of culture is better equipped to handle change and respond to external challenges because employees feel supported and empowered to adapt to new situations ( Van Knippenberg & Sitkin, 2013). H7:

Individualized consideration positively relate with EIB in Ethiopian public research Universities

There is a substantial relationship between OC and innovative behavior displayed by employees within university establishments. It is imperative to foster a positive OC that promotes innovation to cultivate innovative behavior among employees in the realm of HE ( Pavlova, 2023; Zhu & Engels, 2014). H9:

OC positively relate with EIB in Ethiopian public research Universities

Studies have demonstrated that TL positively influences OC ( Hosseini et al., 2020; Lasrado & Kassem, 2021). OC can either promote or hinder innovation ( Naranjo-Valencia et al., 2016). OC influences employees’ innovative behavior ( Afsar & Umrani, 2020; Al Ahmad et al., 2019; Özdaşlı et al., 2023). Similarly, studies indicate that robust OC serves to mediate the relationship between leadership and innovative work behavior ( Indrayanti & Ulfia, 2022). Khan et al. (2020) also demonstrated that OC acts as a mediator between TL styles and innovative work behavior. However, the previous study did not use OC as a single mediating variable, whereas this study specifically focused on OC as the sole mediator. H10:

OC mediates the relationship between the dimensions of TL and employees’ innovative behavior in Ethiopian Universities.

Figure 1 illustrates a conceptual framework that delineates the relationship between transformational leadership and innovative employee behavior, encompassing both the direct and indirect effects mediated by organizational culture. The framework identifies four critical dimensions of leadership as essential contributors to fostering innovation: idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration. Furthermore, it posits that organizational culture serves as a mediating variable, indicating that a supportive cultural environment amplifies the influence of leadership on innovation. The model underscores the notion that effective leadership must be complemented by a robust organizational culture to facilitate innovation.

A standardized questionnaire was used to conduct the study. Accordingly, TL was measured as described in previous studies of Avolio and Bass (2004) and Bass (2006). This survey assessed four TL dimensions with five items per dimension, totalling 20 items. Sample items from each dimensions are “my leader always talks about my most important values and beliefs”, “my leader always Articulates a compelling vision of future”, “my leader always Re-examines the critical assumptions and questions whether they are appropriate”, and “my leader always Spends time coaching and teaching his/her followers”. To measure EIB, the survey included 15 items, with five items for each of the three dimensions. Three items for each dimension were adapted from De Jong and Den Hartog (2010) and Qi et al. (2019), whereas the remaining two items were adapted from Lukes and Stephan (2017). Sample of items from each dimensions are “I always search for new working method, techniques or instrument”, “I always make important organizational members enthusiastic about innovative ideas”, and “I always systematically integrate innovative ideas into work practices”.

To measure OC, 20 items were utilized, with five items allocated to each dimension. These items were adapted from previous studies ( Denison, 2000; Denison et al., 2003; Denison & Mishra, 1995). Sample of items were from each dimensions are “In my university Employees always demonstrate high level of engagement in their professional responsibilities”, “In my university Our leaders practice what they preach”,” In my university the way things are done is very flexible and easy to change”, and “In my university there is shared mission or goals”. Close-ended questions based on the Likert five-point scale were used to rate all questionnaire items: Strongly Agree = 5, Agree = 4, Neutral = 3, Disagree = 2, and Strongly Disagree = 1.

Ethiopia has eight research universities in total. From these, using a simple random sample, three universities, JU, AAU, and HU, were selected. This is because simple random sampling ensures that each sample represents a subgroup/strata ( Kalsbeek & Heiss, 2000). From these three universities based on Yamane (1973), the sample size for each university, JU, AAU has 376 samples, and HU, have 361, 376, and 347 samples, respectively. A total of 979 samples were analyzed for data collection.

From these three research universities, data were collected from academic and administrative staff. Accordingly, a total participants of 979 were distributed, and 726 returned and analyzed, with a return rate of 74.2%.

The analysis was conducted using structural equation modelling (SEM) in two stages: the measurement model and the structural model. The hypothesized measurement model was evaluated using confirmatory factor analysis (CFA) with SPSS version 26 and AMOS 17.0. Structural Equation Modelling (SEM) was a robust statistical method extensively employed in the field of education research to examine intricate hypotheses that encompass latent variables, multiple indicators, and measurement errors ( Jobst et al., 2023). The full raw data can be found under underlying data ( Mamo et al., 2025).

Table 1 which contains the descriptive statistics for the study variables, including mean, standard deviation, skewness, and kurtosis for transformational leadership dimensions (Idealized Influence, Inspirational Motivation, Intellectual Stimulation, Individual Consideration), employee innovative behavior dimensions (Idea Generation, Idea Promotion, Idea Implementation), and organizational culture dimensions (Involvement, Consistency, Adaptability, Mission).

Examination of the descriptive statistics showed that participants generally gave positive ratings to all variables, with mean scores ranging narrowly from 17.23 18.63. Idea Implementation had the highest average, implying that it is the most strongly perceived or practiced aspect, while adaptability had the lowest, pointing to a potential area for enhancement. The standard deviations showed moderate variability, indicating individual differences in the responses. All variables displayed negative skewness, meaning responses were concentrated towards the higher end of the scale, reflecting overall favorable perceptions. Furthermore, higher kurtosis values for Idea Implementation and Mission indicate more peaked distributions, indicating that the responses for these dimensions were more consistently near the average. These patterns suggest strengths in innovation-related behaviors, particularly in implementing and aligning with a mission, while also uncovering areas such as adaptability, which may require further attention.

The measurement model was evaluated using criteria such as construct reliability, construct validity, and model fitness, as demonstrated by Youngs (2020). Therefore, to assess the model’s accuracy, the reliability and validity of the variables were first evaluated, along with tests for convergent validity using average variance extracted (AVE), and then divergent validity was tested using discriminant validity.

In Table 2, reliability and validity tests are conducted to evaluate the measurement model. To confirm the reliability of the construct, Cronbach’s alpha must be greater than 0.7. As shown in Table 2, all Cronbach’s alpha coefficients were greater than 0.7, indicating that the internal consistency of the variables was acceptable. On the other hand, according studies of ( Ghazali & Nordin, 2019; Sovey et al., 2022) reliability assessment is conducted through the use of Composite Reliability (CR) that has to be greater than 0.7, which is the value of CR of this study is above 0.70, which is deemed necessary for sufficient reliability. To confirm convergence validity, the average variance extracted (AVE) for each construct must be greater than the threshold value of 0.5 ( Hair Jr et al., 2021). Accordingly, the findings of this study indicate robust convergent validity, as all values of the constructs are greater than 0.5.

As illustrated Table 3, the rule of thumb for assessing discriminant validity was developed in previous studies ( Chin, 1998; Fornell & Larcker, 1981). They stated that discriminant validity can be evaluated by comparing the square root of the Average Variance Extracted (AVE) for each construct with the correlations between that construct and other constructs in the model. Specifically, if the square root of a construct’s AVE is greater than its diagonal element correlations with other constructs, adequate discriminant validity is present. In this study, since the square root of AVE for every construct is greater than its highest correlation with other constructs, discriminant validity was confirmed.

According to Table 4, in the study by Marsh and Hocevar (1985), a CMIN/df value of ≤ 5 indicates a good fit. In this study, the degrees of freedom and chi-square yielded a CMIN/df value of 3.599, which is considered acceptable as it is below the generally recognized upper threshold of 5.0. Furthermore, the majority of the fit indices (GFI, CFI, TLI, RMSEA, and SRMR) fell within acceptable ranges.

The goodness-of-fit indices (GFI, CFI, and TLI) were all greater than 0.90, suggesting that the model accurately captured the data. Studies by Hu and Bentler (1999) and Kline (2023) confirmed that the acceptable values for these indices should be greater than 0.90. According to Awang (2012), the RMSEA value should be less than 0.08. In this study, the RMSEA value was 0.06, which is considered acceptable. Additionally, an SRMR value of 0.0361 indicates a good fit, as it is below the 0.06 threshold, which is also deemed acceptable according to Hu and Bentler (1999).

However, the model was not completely free from residuals, as indicated by the relatively high RMR value of 0.186, suggesting some residual error in the model. Nonetheless, this does not necessarily imply a poor fit because there is no strict threshold for acceptable RMR values.

After verifying the accuracy and credibility of the measurement model, a structural model is developed see Figure 2.

Figure 2 above of the structural equation model diagram delineates the interrelationships among the dimensions of TL, OC, and EIB within Ethiopian public research universities. Each dimension of TL—namely idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration—demonstrates varying degrees of impact on both OC and EIB. Specifically, idealized influence and individualized consideration are found to exert strong positive effects on organizational culture; however, their direct influence on employee innovative behavior is negative. Conversely, intellectual stimulation and inspirational motivation are associated with both direct and indirect positive effects on employee innovation. The model further substantiates the assertion that OC serves a crucial mediating function in amplifying the impact of leadership on innovation, as evidenced by its robust positive correlation with innovative behavior. In summary, the diagram reinforces the notion that TL is more conducive to fostering innovation when underpinned by a strong and adaptive OC.

The Table 5 above analysis of the regression that multifaceted relationship between the dimensions of TL and EIB, with organizational culture acting as a critical mediating factor.

The Table 6 above analysis the comprehensive relationship between the TL and EIB dimensions, highlighting both direct and indirect pathways mediated by OC.

This study underscores the mediating role of OC on TL and EIB. Establishing a supportive OC is essential for translating leadership influence into innovative behavior. This mediating approach is particularly important in the context of higher education and has been underexplored in previous studies. Cameron and Green (2019) argue that positive OC can enhance employee innovation. Also supported by the study of Wu and Dulebohn (2018), TL dimensions directly relate to OC, which in turn influences EIB.

The findings of this study indicate that strong and positive OC is critical for the promotion of innovative behavior among employees. TL in Universities must be encouraged to prioritize the development and reinforcement of a culture that is conducive to innovation. Such a culture may include fostering trust, promoting open communication, and acknowledging employees’ contributions to innovation. Schein (2010) emphasized that OC influences employee behavior. In addition, as Zheng et al. (2016) stated, the involvement of employees in decision-making processes has led to the development of supportive OC that helps enhance innovative behavior.

The current study may have limited diversity in terms of participating institutions, restricted to public research universities. This could affect the generalizability of the findings across different institutions or cultural contexts within developing countries, such as Ethiopia. According to Etikan et al. (2016), restricted diversity can diminish generalizability. Therefore, future studies need to expand the study to include a more diverse sample from various types of institutions (e.g., private universities and comprehensive and applied types of universities).