There have been rapid and continuous changes in the business world in the last 20–30 years. These changes have started to increase especially in crises such as the global financial crisis experienced in 2008–2009 and the COVID-19 pandemic that emerged in 2019 [1, 2]. VUCA environments are uncertain environments where rapid changes are experienced, which become more complex with the processes of digital transformation and globalization [3–5], and where accurate predictions cannot be made for the future. In the VUCA world, business environments are intertwined with volatility, uncertainty, complexity, and ambiguity. In these environments, leadership perceptions also vary, traditional leadership skills are insufficient, and radical leadership changes are experienced [6–8]. Success in business environments with volatility, uncertainty, complexity, and ambiguity will be achieved primarily by understanding VUCA conditions [9, 10]. It is believed that organizations in VUCA environments can overcome their problems with agility [11–18]. VUCA threats can be turned into opportunities with agility [19–21].

VUCA means to the volatile and chaotic work, economic, and physical environment that everyone faces [22]. Each of these factors introduces its complexity and uncertainty. VUCA components need to be addressed one by one to be ready for the VUCA environment and to understand VUCA. VUCA is an acronym made up of the first letters of four words (volatility, uncertainty, complexity, ambiguity). Variability is the variability of facts or events, observing unexpected or unstable values, and not knowing how long the situations will last. It is foreseen that this changing environment will be overcome with vision [23–25]. Uncertainty is the difficulty of predicting future situations and values ​​and the high probability of realization of the situation called a surprise. It is foreseen that the environment of uncertainty will be overcome with understanding. Complexity is a situation in which too many variables are directly related to each other while analyzing the current situation and the future, and these relationships are not cause-effect relationships. It is foreseen that the complex situation will be solved by restructuring [26]. Ambiguity is a situation where the states and values ​​cannot be read clearly, and the accuracy of the available data is open to discussion. It is foreseen that the ambiguous environment will be overcome with agility [27]. All four of the components that make up the VUCA are often used together in the business world or in managers and are considered synonymous. Although the word meanings of these components are similar, there are differences between them that are very valuable for leaders to understand [28]. In the VUCA environment, it will be very useful for a leader to determine strategies according to the environment with the help of these components [25]. In times of great uncertainty, managers are expected to set the right strategies, protect the core business in their organization, be open to innovation, and prepare their organizations for the future [29].

Being able to manage the VUCA environment, and additionally shape the future is possible if the leader knows how to make decisions in this environment [25]. At this point, many leadership theories and models have been developed [30–31] within the expanding leadership literature [32]. The changing world conditions over time have revealed the need for a new leadership concept, "agile leadership" [33]. Agile leadership is the ability to lead effectively under rapid change and complex conditions [34]. The concept of "effectiveness" in the definition means that the leader can anticipate and respond to changes [35]. The ability of organizations/organizations to survive in a challenging competitive environment and to continue their development in the face of change largely depends on the leadership skills of managers [36]. The authors [37–38] discussed that new skills, approaches, and behaviors are needed to manage the four VUCA threats in business. VUCA represents several challenges that individuals, managers, and organizations may face [25]. Managing teams in the business environment requires being aware of changes [39–40]. The VUCA environment makes people indecisive, anxious, and unmotivated and hinders career moves. It takes a lot of time and effort to struggle with VUCA. It causes people to make wrong decisions. It can put long-term projects at risk. It harms internal communication. Therefore, businesses should obtain up-to-date information for their sustainability in the VUCA environment, and review their working styles and human resources management [25, 41]. The most difficult aspect of managing VUCA is teamwork that resists change. The unpredictability of VUCA makes the traditional institutional structure obsolete. For this reason, the inflexible, autocratic leadership style should be abandoned. Instead, there is a need for an agile approach that has the ability and flexibility to adapt to change quickly and efficiently. The agile approach has directed leaders in managerial positions toward change. Managing change has revealed agile leaders [25, 42]. Whether an organization can adapt to future challenges and opportunities depends on the agility of the leaders [43].

In the history of humanity, especially since the industrial revolution, institutions/businesses are going through a period of the fastest change/transformation [44, 45]. When we look at the change-transformation today, it is seen that we are faced with a very rapid transformation in the education/training system [46–49]. In the face of this transformation, leaders have to perform more effective management [50]. Because the two most important features that emerge in the face of transformation and are very difficult to manage are uncertainty and complexity. In the face of this transformation, leadership is also transforming [51]. The change experienced has required creativity and innovation in organizations [52]. Organizational agility is required for organizations to quickly adapt to external environmental conditions and change [53–54]. In the VUCA world, the need for agile leaders comes to the fore for success and corporate sustainability [55]. Agile leaders are leaders who act by knowing how to prepare for and deal with each element of VUCA for a successful management strategy in an unpredictable business world. Agile leaders must be able to identify the situation created by the four components of VUCA and come up with solutions [56]. Each situation should identify its unique causes and solutions. The leaders should not resist change but should develop a clear and shared vision of the future of teamwork. They should understand the ambiguity. They should make analysis and interpretation a priority, should evaluate performance [57, 58], should communicate clearly and openly with their employees [59]. For the success of the organization, the leader's words and actions must be compatible [60]. The leaders should instill self-confidence in their employees and encourage them to cooperate, should create self-renewing teams that can work effectively in the case of VUCA [61]. Brain drain is increasing among countries in the world. Leaders should be able to retain talented and qualified personnel in a rapidly changing world [62]. The leaders are also expected to have a frugal understanding to bring innovations to their institutions feasibly [63]. Leaders should promote flexibility, adaptability, and agility in their organizations [22, 48]. Although it may not be beneficial in all circumstances [64], leaders should reward employees who demonstrate vision and agility [48]. The leaders should be creative [65], and should be able to give fast but also accurate reactions, manage the reactions, and put them into practice. Leaders should also be accountable and honest [66]. Leadership processes in the organizational context [67] depend on the acquisition and use of knowledge [68]. Leaders should be open to learning in this direction [69] and should do their job with enthusiasm. Many employees expect their jobs to add meaning to their lives [70]. At this point, leaders should be able to provide employees with job satisfaction and adapt to the institution they work for [22, 45, 71]. According to the organizational support theory, the words and actions of leaders are not only the product of their own will but also a reflection of the perspective of the organization [72]. Leadership behavior has a significant impact on employee behavior, motivation, performance, and well-being [73, 74]. Managerial innovations and strategies create positive effects on organizational renewal and performance [75, 76]. Leaders should not punish their subordinates unjustly [77]. Provided that they do not pacify their visibility [78], should not have sole authority in their organization [57]. They should be able to include subordinates in decision-making processes [79]. Leader candidates who can become leaders in the future should be able to emerge from among the staff raised by the leaders [80]. The leaders should not take the standard point of view and should be open to change.

As the employees' perception of variability increases, the perception of agile leadership skills increases [81]. However, the emergence of intolerance to uncertainty among leaders, top managers, and staff (a component of VUCA), their poor cognitive flexibility, and a low tolerance for ambiguity result in the weakening of agile leadership in the VUCA environment to varying degrees. We can think of universities as businesses in terms of examining agile leadership in VUCA environments [25]. The authors [82–87] studied on management in public universities. Buchashvili et al. [88] discussed the need for updated leadership skills in the VUCA reality and the role of higher education in stepping forward to transform VUCA disruptions into VUCA possibilities. Yaşar [89] determined that the variables of gender, marital status, seniority, and time spent in the institution were effective on the servant leadership perceptions of academicians. Koçoğlu et al. [90] determined the differences in innovation abilities of the demographic factors of the employees (such as gender, marital status, and working time). Gender differences affect leader behaviors [91, 92]. The authors [93–95] have conducted studies examining leadership in the workforce from a gender perspective.

Providing psychological empowerment, especially in the public sector, increases organizational performance [96]. Awareness and self-leadership strategies can also improve psychological capital and increase job engagement [97]. Organizations struggle to cope with rapidly changing environments and more complex and interconnected information sets. In this environment, a cyclical interaction through learning from experience, reasoning, and engaging in a mutual learning process becomes increasingly important for organizational learning within organizations [98]. To effectively manage planned change and to understand the differences in leaders' responses to it, it is important to understand how change is cognitively represented by organizational members [99]. Employees at all organizational levels have influence over their subordinates, colleagues and even their bosses [100]. Decades of research have shown that people can reach completely different perceptions within the group hierarchy in the same social situations [101]. Different perceptions need to be managed and examined effectively [102]. Most of the research on development in organizations has been done on promoting development between managers and employees [103]. Although there are studies in which managers are evaluated from the eyes of the employees, there is no study in which this evaluation is made on the agility of the leader. When the leadership literature is examined, few studies deal with the concepts/issues of VUCA and agile leadership together. A study examining agile leadership in the university sample could not be found in the references accessed. In this study, unlike other studies carried out in the literature, the concepts of VUCA and agile leadership were examined together in a university sample. Addressing a business concept from the perspective of the university in the administrative dimension makes the research unique. It is thought that determining the effect of leadership activities carried out in the VUCA world on the academic, administrative, and administrative system, and creating a perspective on the academic world will contribute to the literature.

Factors such as increasing digitalization, rapidly developing technology, post-pandemic transformation in education, and global competition force universities to operate in an ever-changing environment. At this point, university administrators, academics, and student leaders should develop an awareness of how to act in an environment of uncertainty and possess agile leadership skills. Since the business world of the future will be shaped under VUCA conditions, adopting an agile leadership approach in universities will support students by preparing them for a successful business life. In order for universities to adapt quickly to changing conditions, increase the quality of education, and remain competitive at the international level, they need to adopt an agile management approach.

Considering that events with VUCA features such as COVID-19 or earthquakes have occurred in the region where the University is located in the last few years, the negative effects on the efficiency of the functioning and the flow of work at the university subject of the study have revealed the need for this study. In addition, this research was conducted to prevent the negative impact on the education and academic achievements of the university within the superior-subordinate relationship of the personnel. Thanks to this research, within the scope of a subject that has been little studied in the literature before, the working efficiency of the staff and the university can be increased with the agile leadership perspective in the VUCA environment in public & academia and harmonious action between the staff can be ensured. This study can guide future VUCA studies in the public & university field. The theoretical justification for this research is that leadership approaches need to be rethought in VUCA environments, agile leadership has come to the fore in recent years, and yet this relationship has not been sufficiently researched in the university context. In this direction, this study aimed to determine the VUCA perceptions of the staff working in the academic and administrative staff at a state university in Turkey, regarding the working environments they are in, and to determine the leadership skills in the VUCA environment from the perspective of the employees. In addition, it is aimed to increase the level of knowledge and awareness about VUCA.

Since this study was conducted in a state university in Adana, Turkey, for the academic and administrative staff working in the spring term of the 2020–2021 academic year, there is a time and sample limitation in the research. In line with the aim determined in the study, a questionnaire was prepared using appropriate measurement tools. The data of the research were obtained from the survey application via e-mail. At the same time, the survey questions were submitted to the survey application via e-mail. In this application, the principle of voluntariness and confidentiality is observed. For the survey consisting of 90 questions in total, the participants were asked to read all the questions carefully and answer them completely. Participants were informed that it would take an average of 10–14 minutes to answer the questions. The questionnaire form used in the research consists of 7 parts and a total of 90 questions. The first part includes the principle of voluntariness and confidentiality. In the 2nd part of the questionnaire, there is a personal information form, in the 3rd, 4th, 5th, and 6th parts of the questionnaire, there are resistance to change, intolerance of uncertainty, cognitive flexibility, and ambiguity tolerance scales to measure the four components of the VUCA, and in the 7th part, the agile leadership scale.

In this study, general questions were asked about gender distribution, (married/single) marital status, birth dates, education levels, working hours in the institution, staff qualifications in the institution (civil servant/contractual), and (academic/administrative) managerial characteristics of the 187 employees who participated in the survey. In addition to these, questions that can be evaluated within the scope of agile leadership/VUCA were asked of the employees. First of all, the mean scores of resistance to change, inhibitory anxiety, prospective anxiety, cognitive flexibility, tolerance to ambiguity, and then the standard deviations of all of the answers given to these questions were calculated.

In this study, the scale adapted to Turkish culture to measure attitudes toward change [104] is the “Resistance to Change” scale developed by Oreg [105]. (7) strongly agree – (1) strongly disagree seven-rated Likert Scale consists of a total of 15 items, three dimensional (affective, behavioral, cognitive resistance) and four reverse coded (RTC₃, RTC₁₀, RTC₁₄, RTC₁₅). The scale adapted to Turkish culture to measure intolerance to uncertainty is the “Intolerance of Uncertainty” scale developed by Carleton et al. [106]. There are 12 items in the scale, which consists of five scales and a two-dimensional prospective & inhibitory anxiety structure with the ends of "(5) completely suitable for me" and "(1) not suitable for me at all". The scale adapted to Turkish culture to measure cognitive flexibility is the "Cognitive Flexibility Scale" developed by Martin & Rubin [107]. This scale consists of 12 items in total. Four items (IOU₂, IOU₃, IOU₅, IOU₁₀) of this six-point Likert scale, scored between “6 (strongly agree)” – “1 (strongly disagree)”, are reverse coded. The scale adapted to Turkish culture to measure ambiguity tolerance is “The Multiple Stimulus Types Ambiguity Tolerance Scale–II (MSTAT–II)” developed by McLain [108]. This scale consists of 13 items in total. Nine items (AT₁, AT₂, AT₃, AT₄, AT₅, AT₆, AT₉, AT₁₁, AT₁₂) of this Likert-type scale, which has five ratings scored between "5 (strongly agree)" and "1 (strongly disagree)", are reverse coded.

In this study, the "Agile Leadership Scale" developed by the author [55] was used to measure the agility dimension of leaders. To develop this scale, data were obtained from a total of 679 volunteers from three sample groups consisting of individuals over the age of 18 working in companies operating in the Aegean and Marmara regions of Turkey and operating in the production sector, and these were analyzed. It was concluded that this agile leadership scale, which was developed according to all the results, is a valid and reliable scale that can be used for institutions and organizations operating in different sectors.

This scale, consisting of 32 questions, was designed as a 5-point Likert scale in our study. The answers to the scale are "strongly agree" (5), "agree" (4), "neither agree nor disagree" (3), "disagree" (2), and "strongly disagree" (1). The existence of multivariate extreme values ​​in the data of 187 employees constituting the sample of the study was examined by calculating the "Mahalonobis distance" before the factor analysis applications to be used to ensure the construct validity of the scales. In this direction, the data of 12 participants, which were determined to be multivariate extreme values, were removed from the data set, and the data of the remaining 175 employees were evaluated in the analyses. In line with the data obtained, in this study, the relations between the perceptions of the employees towards the VUCA components and the perceptions of the managers towards the agile leadership skills were examined. In addition, it was also determined whether the perceptions of VUCA components and agile leadership skills differ according to the gender and marital status of the employees. The purposive sampling method was used in this study to select individuals, groups, or situations that could provide the most information about VUCA. This aimed to capture the opportunity to collect rich data from individuals relevant to the research topic. Purposive sampling was used to reach staff at a public university who could provide in-depth information about VUCA. This method was used to select individuals who would be most relevant to the research question or provide the most information, rather than just anyone. The types illustrate the selection logic. The hypotheses created in this framework are presented in Table 1. The results of the analysis of the survey questions were used in the evaluation of these hypotheses. Analysis results with significant correlations are presented in Table 3–8 with "*", "**", and "***" signs. For easier reading of the article, the title "Investigation of Structural Validity of Agile Leadership Scale via CFA & EFA" is presented as an Annex-1.

Table 1. Codes and relationship properties.

Tables 4 and 5 present correlations between the sub-dimensions of agile leadership and VUCA components. Table 4 demonstrates that all sub-dimensions of agile leadership are complementary and emerge together. Considering Table 5, there is no significant correlation between the VUCA components and the sub-dimensions of agile leadership. The only significant correlation is between cognitive flexibility and speed. The results of the correlation analysis of the relationships between the VUCA components are as follows: Statistically significant and positive correlations were found between resistance to change and inhibitory anxiety (r = 0.355) and prospective anxiety (r = 0.338) scores. Statistically significant and negative correlations were found between cognitive flexibility (r=-0.231, p = 0.002) and ambiguity tolerance (r=-0.445) scores. A statistically significant and positive relationship was found between inhibitory anxiety and prospective anxiety scores (r = 0.531), and a statistically significant and negative relationship was found between ambiguity tolerance scores (r=-0.574). A statistically significant and negative correlation was found between prospective anxiety and ambiguity tolerance scores (r=-0.294). A statistically significant and positive correlation was found (r = 0.323, p < 0.01) between ambiguity tolerance and cognitive flexibility scores (Table 4). The results of the correlation analysis of the relations between the sub-dimensions of agile leadership are as follows:

Statistically significant and positive relationships were found between the perception of result orientation and other perceptions, respectively: Perception of team orientation (r = 0.907), perception of competence (r = 0.837), perception of flexibility (r = 0.785), with the perception of swiftness (r = 0.815) and the perception of change orientation (r = 0.749). Statistically significant and positive relationships were found between the perception of team orientation and other perceptions, respectively: Perception towards competence (r = 0.858), perception towards flexibility (r = 0.836), perception towards speediness (r = 0.826), and perception towards change orientation. (r = 0.795). Statistically significant and positive relationships were found between the perception of competence and other perceptions, respectively: Perception of flexibility (r = 0.873), perception of speediness (r = 0.877), and perception of change orientation (r = 0.875). Statistically significant and positive correlations were found between the perception of flexibility and the perception of speediness (r = 0.834) and the perception of change orientation (r = 0.799). A statistically significant and positive relationship was found between the perception of speed and the perception of change orientation (r = 0.870) (Table 4). There was no statistically significant relationship between resistance to change, inhibitory and prospective anxiety, ambiguity tolerance scores and perceptions of results, team orientation, change orientation, competence, flexibility, and speediness (p > 0.05). (Table 5).

Table 4. Correlation analysis results between VUCA components and agile leadership sub-dimensions.

Table 5. Correlation analysis results between VUCA components and agile leadership perception.

An Independent Samples T Test was applied to determine whether employees' perceptions of VUCA components differ according to gender (Table 6). There is no significant difference between men and women in terms of resistance to change, inhibitory anxiety, anticipatory anxiety, and cognitive flexibility. There is a significant difference in the tolerance for ambiguity variable. The mean score for men (mean = 35.75) is higher than for women (mean = 33.43). This result suggests that men are more able to tolerate uncertain and ambiguous situations than women. It can be said that female participants, in particular, have a higher perception of results-orientedness than men, while there are no significant gender differences in other dimensions.

It has been determined that the mean scores of ambiguity tolerance according to the gender of the employees show a statistically significant difference (p < 0.05). Accordingly, it is seen that women's perceptions of result-orientedness are higher than men's (p < 0.05). Women perceive their managers as more result-oriented leaders than men. The H7_e hypothesis is supported by research findings. It is seen that men have a higher tolerance for ambiguity than women. It was determined that the mean scores of resistance to change, inhibitory and prospective anxiety and cognitive flexibility did not differ according to gender (p > 0.05). Independent Sample T Test was applied to determine whether the employees' perception of agile leadership differs according to gender. According to the gender of the employees, it was determined that the perceptions of result orientation, team orientation, competence, speediness, flexibility and change orientation did not differ statistically (p > 0.05). The H9_a hypothesis is supported by research findings. It is seen that women's perceptions of result-orientedness are higher than men's. It was determined that the perception of team-oriented, change-oriented, competence, speediness, flexibility, and change orientation did not differ significantly according to gender (p > 0.05). (Table 6).

Table 6. T-test results of employees' perceptions of VUCA components and agile leadership according to gender.

The differences in participants' scores for the VUCA components by marital status (t-test) are presented in Table 7. This result indicates that married participants have higher future concerns than single participants. Regarding the differences in participants' perceptions of agile leadership by marital status, married individuals appear to have higher scores in agile leadership than single individuals. This can be interpreted as suggesting that married life can instill in individuals a greater sense of responsibility, cooperation, and the ability to adapt to change.

It was determined that the prospective anxiety score averages of the employees differed statistically significantly according to their marital status (p < 0.05). The H8_c hypothesis was supported by research findings. It is seen that the future anxiety of married people is higher than that of singles. It was determined that the mean scores of resistance to change, inhibitory anxiety, cognitive flexibility, and ambiguity tolerance did not differ significantly according to marital status (p > 0.05). To determine whether the perception of the employees towards agile leadership differs according to marital status, the independent sample t-test was applied. According to the marital status of the employees, it was determined that the perceptions of result-oriented, team-oriented, change-oriented, and flexibility differ statistically significantly (p < 0.05). H10_a,b,d,f supported by research findings. It is seen that the perceptions of result-oriented, team-oriented, change-oriented, and flexibility of married people are higher than singles. It was determined that the perception of competence and speed did not differ significantly according to gender. (p > 0.05) (Table 7).

Table 7. The t-test results of the difference between the employees' perception of VUCA and agile leadership according to marital status.

The results of the research revealed that there was no relationship between the VUCA components of the employees (resistance to change, inhibitory and prospective anxiety, tolerance of ambiguity) and their perceptions of the 6 sub-dimensions of agile leadership. On the other hand, only cognitive flexibility levels and perceptions of agile leadership's speediness dimension were determined to be correlated. First (H1_a,b,c,d,e), second (H2_a,b,c,d,e), third (H3_a,b,c,d,e), fourth (H4_a,b,c,d,e), fifth (H5_a,b,c,e) and sixth (H6_a,b,c,d,e) hypotheses were not supported by the research results (Among these, only the H5_d was supported as an exception). Based on the results obtained, it is possible to say that employees with cognitive flexibility need agile leaders. In other words, employees in this category want to see their leader as fast, determined, and stable while analyzing the current situation and the future. In a complex situation, they expect their leader to be effective at the point of strategic decision-making. In general, when the perspective of the employees of the institution towards the leader is examined from the perspective of VUCA, it is determined that the employees see the leader as a leader who can make quick decisions and share this with her stakeholders. This has created the perception that the leader can quickly solve even the seemingly unsolvable events and approach the problems with creative solutions. On the other hand, situations such as change, uncertainty, and ambiguity do not affect the personnel to see their leaders as result-oriented, team-oriented, change-oriented, competent, fast, and flexible.

Indeed, fast leaders are needed in adapting to technology, identifying and meeting customers' needs and wants, and working on a production, communication, and team basis. Leaders who have the speed to learn about changes and teach their teams and use this knowledge in production/service are preferred [110]. Therefore, in the context of the need for fast agile leaders in the university examined in the study, seeing their leaders in this orientation will provide an advantage to the institution in VUCA environments.

Within the scope of the H7_a,b,c,e hypothesis, the relationship of employees' perceptions of VUCA components according to gender did not show a significant difference. In the H7_d hypothesis, the tolerance for ambiguity shows a significant difference in the male employee. In other words, in case of an increase in ambiguity, the tolerance of ambiguity of male employees increases more than females, and in case of a decrease, the tolerance of ambiguity decreases. In this context, it can be said that female employees are perhaps more stressed and skeptical in ambiguous environments. When the differences in the perceptions of agile leadership by gender are examined, no significant relationship has emerged in team orientation, change orientation, competence, flexibility, and speediness within the scope of the H9_b,c,d,e,f hypothesis. In the H9_a hypothesis, a significant relationship is observed for male employees in results orientation, which is one of the sub-dimensions of agile leadership. Men, more than women, perceive their managers as results-oriented leaders.

Within the framework of H8_a,b,d,e hypotheses, the relationship of employees' perceptions of VUCA components according to their marital status did not show a significant difference. In the H8_c hypothesis, a significant relationship is observed in the perception of prospective anxieties of married employees. It is seen that the future anxiety of married employees is higher than that of single employees. It is seen that married employees perceive their managers as more result-oriented, team-oriented, change-oriented, and flexible leaders compared to singles. It can be said that single employees do not like surprises and they want to know and organize everything beforehand. When the difference in agile leadership perception according to marital status is examined within the scope of the H10_c,d hypothesis, no significant relationship has emerged between competence and speed. In the H10_a,b,e,f hypothesis, there is a significant relationship among married employees in terms of results orientation, team orientation, change orientation, and flexibility, which are sub-dimensions of agile leadership.

In this study, how VUCA is perceived in the university environment and how the agile leadership approach is reflected in academic and administrative leadership processes were examined. It has been revealed that cognitive flexibility is linked to the speed dimension in agile leadership and that there are areas that need to be improved in university administrators' adaptation to technology and change management. In this context, suggestions are presented for universities to improve their current leadership understanding and adapt to changing environmental conditions. The study's results and the suggestions presented by comparing previous studies are as follows.

The scale developed by Oreg [105] was applied as a preliminary study on a sample of 73 people. The Turkish validity and reliability of this scale were determined by [104]. Item analysis was applied to this scale and the total test correlations of all items were examined. Item-total correlation reveals the relationship between each item in the scale and the total score. The high correlation of the total test scores of each item indicates the consistency of that measurement tool. The minimum value required for the item-total test correlation to be sufficient is specified as 0.30 [149]. In the Kline [149] hypotheses, RTC₃, RTC₁₀, RTC₁₄, and RTC₁₅ inverse-coded items were excluded from the scale because their total correlation values ​​were below 0.30. As a result of the repeated analysis, the internal consistency of this scale was calculated as 0.934. According to Kline [150], the reliability coefficient should be higher than 0.70.

The KMO sample adequacy value was found to be 0.911, and the Barlett sphericity test was found to be significant as a result of the EFA applied to test the structural validity of the scale on a research sample consisting of 175 people (χ2(55) = 1458.457, p < 0.001). The fact that the KMO value is > 0.60 and the Bartlett test is p < 0.05 indicates that factor analysis can be applied to the sample data [151, 152]. It is understood that the data obtained according to the KMO and Bartlett test results are suitable for factor analysis. After determining that the data related to the scales were suitable for EFA analysis, the Maximum Likelihood Method was used for factor extraction. As a result of the analysis, the scale showed a single-factor structure in its Turkish form [104]. It was determined that the factor loads of the items ranged between 0.857 (highest) − 0.510 (lowest). The rate of variance explaining the overall scale was found to be 58.426 (Table 8, No.1). Pallant [152] stated that the factor load limit value that an item should have is 0.30. Streiner [154] also stated that the total variance explained in EFA should be at least 30% in one-dimensional scales, in terms of behavioral sciences.

It is seen that the chi-square value showing the general fit of the model is (χ2) 244.214, and the degree of freedom is significant at the (df) = 44, p = 0.001 level and χ2/df = 5.55 (Table 8). Although the χ2 test is expected to be meaningless, χ2 is usually significant in practice, since it is a value that is highly affected by the number of samples. In this case, to evaluate the goodness of fit of the general model, it is accepted that the ratio of χ2 value to (df) degrees of freedom (χ2/df) is taken into consideration and will give more accurate results. Therefore, a significant p-value is tolerated in many studies [155–158]. It is seen that the χ2/df ratio is at an acceptable threshold value. This value shows that the model is acceptable.

Table 8. EFA results in resistance to change, intolerance, cognitive flexibility, and tolerance for ambiguity scale.

CFA was used to test the confirmation of the single-factor structure obtained as a result of EFA. Let's examine the GFI values ​​obtained as a result of Single Factor CFA applied to the Resistance to Change Scale. CFI, NFI, TLI and IFI values ​​from the comparative GFI showed an acceptable fit. However, it was determined that the RMSEA value, the χ2/df value indicating the general fit of the model, and the absolute fit indices (GFI, AGFI) were not within the acceptable limit values. It has been determined that the χ2/df value, which shows the general fit of the model, is quite close to the acceptable level. To correct these fit indices, the model was improved. During the improvement, the modification indices (MI) values ​​formed covariance in the presence of high errors. Afterward, it showed a perfect fit for χ2/df value in the renewed fit index calculations. In addition, accepted values ​​were obtained for RMSEA, GFI, and AGFI fit indices. Accordingly, it is seen in Table 4 that the TLI value and SRMR value also show a perfect fit. These findings show that the single-factor model is compatible with the data and is acceptable. As a result of the single-factor CFA model, it was determined that 11 items were related to the scale structure. It is seen that the factor loads of the items vary between 0.518 and 0.844 and all correlation relations are significant (p < 0.001). The structural model of the Resistance to Change Scale, whose single-factor structure on the sample was confirmed with 11 items, is presented in Fig. 2(a).

Table 9. CFA Fit Indices of the Single Factor Model of the Resistance to Change Scale [159–161].

Table 10. Model results on resistance to change, intolerance of uncertainty, cognitive flexibility, tolerance of ambiguity, and single/multi-factor CFA of agile leadership scale.

Figure 2. Structural models for the agile leadership scale.

The Turkish validity and reliability of the scale, which was developed by Carleton et al. [106] and applied as a preliminary study on a sample of 73 people, was carried out by [104]. Item analysis was applied to this scale and the total test correlations of all items were examined. As a result of the analysis, it was determined that there was no item below 30. The internal consistency of the scale was calculated as 0.889. The KMO sample adequacy value was found to be 0.904, and the Barlett sphericity test was found to be significant (χ2(66) = 894.215 p < 0.001) as a result of the EFA applied to test the construct validity of the scale on a research sample consisting of 175 people. KMO and Bartlett test results show that the obtained data are suitable for EFA. After it was determined that the data related to the scales were suitable for EFA, the Maximum Likelihood Method was used for factor extraction. It is seen that the χ2 value, which shows the general fit of the model, is significant at the level of 69.189, df = 43, p = 0.009, and χ2/df = 1.58. The χ2/df value indicates a good fit of the model.

As a result of the analysis, it was determined that this scale [106] showed a two-factor structure in its Turkish version [104]. Then, Varimax was chosen as the rotation method, as was done in the study in which the scale was developed and in its Turkish form. It was determined that the factor loads of the items ranged from 0.783 (highest) to 0.360 (lowest). On the other hand, CF₄ and CF₂ items were excluded from the scale due to their accumulation in more than one factor, resulting in the overlap. As a result of repeated analysis, the KMO value was calculated as 0.889. Barlett test of sphericity was found to be significant (χ2(45) = 713.496, p < 0.001). There was no factor load below 0.30 and no overlapping item. The rate of variance explaining the overall scale was found to be 50.170%. Streiner [154] stated that the total variance of all factors in EFA should be at least 50% in multidimensional scales. The original contribution of the explained variance of a factor should be at least 5% [162]. The χ2 value, which indicates the general fit of the model, was found to be insignificant at the level of 32.576, df = 26 and p = 0.175 (χ2/df = 1.36) (Table 8, No.II).

Let's examine the GFI values ​​obtained as a result of Multi-Factor CFA applied to the Intolerance of Uncertainty Scale. The χ2/df value indicating the general fit of the model, the GFI value from the absolute fit indices and the CFI, TLI, IFI values ​​from the comparative goodness of fit indices showed perfect fit (Table 9, No.II). These findings show that the single-factor model is compatible with the data and is acceptable. As a result of the multi-factor CFA model, it was determined that 10 items and two factors were related to the scale structure. It was observed that the factor loads of the items varied between 0.432 and 0.892 and all correlation relations were significant (p < 0.001) (Table 9, No.II). The structural model of the "Intolerance of Uncertainty Scale", whose two-factor structure on the sample was confirmed with 10 items, is presented in Fig. 2(b).

The Turkish validity and reliability of the scale developed by Martin & Rubin [107] and applied as a preliminary study on a sample of 73 people was carried out by [104]. Item analysis was applied to this scale. As a result of the analysis, the reverse-coded items CF₂, CF₃, and CF₅ were excluded from the scale, since the total correlation values ​​were below 0.30. In the repeated analysis, the internal consistency of the scale was calculated as 0.842. As a result of the EFA applied to test the construct validity of the scale on a research sample consisting of 175 people, the KMO sample adequacy value was found to be 0.871, the Barlett sphericity test was significant (χ2(36) = 636.008, p < 0.001). KMO and Bartlett test results demonstrate that the obtained data are suitable for EFA. After it was determined that the data related to the scales were suitable for EFA, the scale design was developed by using the Maximum Likelihood Method in the study. In its Turkish form [104], it was evaluated by limiting one factor. As a result of the analysis, it was revealed that the EFA loads of the items ranged from 0.790 (highest) to 0.370 (lowest). The rate of variance explaining the overall scale was determined as 43.288%. It is seen that it is significant at the level of χ2 = 72.536, df = 27, p = 0.001, which shows the general fit of the model (χ2/df = 2.68). The value of χ2/df reveals that the model has a good fit (Table 9, No.III).

Let's examine the goodness of fit values ​​obtained as a result of Single Factor CFA applied to the Cognitive Flexibility Scale. The χ2/df value indicating the general fit of the model, the GFI value from absolute fit indices, and the SRMR value showed perfect fit. Comparative goodness of fit indices also showed acceptable fit (Table 9, No.III). These findings demonstrate that the single-factor model is compatible with the data and is acceptable. As a result of the single factor CFA model, it was determined that 9 items were related to the scale structure. The factor loads of the items vary between 0.370 and 0.790. All correlation relationships are significant (p < 0.001). The structural model of the Cognitive Flexibility Scale, whose single factor structure on the sample was confirmed with 11 items, is presented in Fig. 2(c).

The Turkish validity and reliability of the scale, which was developed by McLain [108] and applied as a preliminary study on a sample of 73 people, was conducted by Koç et al. [104]. In this study, item analysis was applied. As a result of the analysis, the reverse coded item AT₉ and item AT₁₃ were excluded from the scale, since the total correlation values ​​were below 0.30. In this way, as a result of repeated analysis, the internal consistency of the scale was calculated as 0.812. In this study consisting of 175 people, the KMO sample adequacy value was found to be 0.806, and the Barlett sphericity test was found to be significant (χ2(55) = 517.076, p < 0.001) as a result of the EFA applied to test the structural validity of the scale on the research sample. KMO and Bartlett test results demonstrate that the obtained data are suitable for EFA. After it was determined that the data related to the scales were suitable for EFA, the scale design was developed by choosing the Maximum Likelihood Method. In this study, the Turkish version of this method [104] was evaluated by limiting one factor. As a result of the analysis, the factor loads of the items ranged from 0.717 (highest) to 0.307 (lowest). The rate of variance explaining the overall scale was determined as 29.534%. χ2 = 134.303, df = 44, which shows the general fit of the model, are significant at the p = 0.001 level (χ2/df = 3.05). The value of χ2/df reveals that the model has a good fit (Table 8, No.IV).

Let's examine the GFI values ​​obtained as a result of Single Factor CFA (Table 10, No.IV) applied to the Ambiguity Tolerance Scale. χ2/df value showing the general fit of the model, absolute fit indices (GFI, AGFI), Standardized root mean square residual (SRMR), and CFI and IFI values ​​from comparative goodness-of-fit indices show acceptable fit. However, it has been determined that the RMSEA value is not within the acceptable limit values. To correct this fit index, the model was improved. During the improvement, the modification indices (MI) values ​​formed covariance in the presence of high errors. In the later, renewed fit index calculations, χ2/df and GFI values, which provided the accepted values ​​for the RMSEA index, also showed a perfect fit (Table 9, No.IV). These findings demonstrate that the single-factor model is compatible with the data and is acceptable. As a result of the single-factor CFA model, it was determined that 11 items were related to the scale structure. The factor loads of the items varied between 0.303–0.723. All correlation relationships are significant (p < 0.001). The structural model of the Cognitive Flexibility Scale, whose single factor structure on the sample was confirmed with 11 items, is presented in Fig. 2(d).

To verify the six-factor structure of the Agile Leadership Scale in the research sample, Multi-Factor CFA was conducted. As a result of CFA, it was determined that 32 items and six factors were related to the scale structure. The factor loads of the items varied between 0.584 and 0.948. All correlation relationships are significant (p < 0.001). The values ​​for the fit of the model are also at an acceptable level (Table 10, No.V). The Cronbach Alpha values ​​for the scale were calculated as follows: Result-oriented dimension 0.942, team-oriented dimension 0.960, competence dimension 0.907, flexibility dimension 0.900, speediness dimension 0.909, change-oriented dimension 0.901. The structural model of the multi-factor CFA of the "Agile Leadership Scale", whose six-factor structure was confirmed with 32 items, is presented in Fig. 3.

Figure 3. Model for the Multi-Factor CFA of the Agile Leadership Scale (Table 10, No.V).