Millennial leaders and leadership styles displayed in the workplace

investigate and test the assumption that millennial leaders differ in their leadership styles compared with the older generation leaders in the workplace. This knowledge, pinpointing to how millennials differ in how they lead, and therefore specifying the leadership styles used frequently by millennial leaders could assist in developing leaders of this generation, who are currently the emerging leaders in the workplace. Purpose: The aim of this research was to test the assumption of differences between leadership styles adopted by leaders across generations, as perceived by their subordinates, on the premise that millennials hold different values from other generations, and that these values, in turn, have an impact on their leadership styles. Design/methodology/approach: This research was conducted in South Africa, in a variety of organisations, and by sampling employees across three generations, namely millennials, Generation X and baby boomers. Cross-sectional data were collected with an instrument based on the Pearce leadership typology of leadership styles. After confirming reliability and general factorial validity, mean scores were compared using analyses of variance. Findings/results: The results of the study found practically significant differences between millennial leaders and Generation X leaders in the extent to which they apply empowering and transformational leadership styles, as perceived by their subordinates. In absolute terms, millennial leaders display less leadership behaviour than do those of Generation X, with regard to these leadership approaches. Practical implications: Organisations involved in the development of millennial leaders should be aware that interventions should not blindly align to the stereotypes associated with this generation. Further research is also needed to determine which leadership styles are effective with millennial leaders. Originality/value: The millennial generation has emerged in the workplace as the new leadership pipeline, yet there appears to be a lack of research on how millennials want to lead. This study contributes to a nuanced understanding of and improved development of the millennial generation leaders.


Introduction
Different generations enter the workplace with an understanding and expectation of leadership roles and processes based on their own experiences and on major events that occurred during their upbringing (Andert et al., 2019).They are also influenced by their own personality traits when adopting leadership styles (Sharifi-Rad et al., 2017).The literature suggests that members of the millennial generation have a different view of the relationship between leaders and followers than previous generations did (Ben-Hur & Ringwood, 2017;Medyanik, 2016), having been brought up by parents who took a great deal of interest in protecting and guiding them through direct supervision, care and nurturing (Sledge, 2016).The literature further suggests that millennial leaders have a different workplace leadership style compared to previous generations in the workplace (Fore, 2012), and their approach as leaders, or their leadership traits, encompass their own characteristics and values (Casey, 2015;Medyanik, 2016).Understanding the leadership styles of millennials can assist in creating a work environment where leadership effectiveness is maximised (Chou, 2012).

Literature review
There are currently three generations in the workplace, the oldest being the baby boomer generation (born between 1946 and 1964).Next, is Generation X (born between 1965 and1979), followed by the youngest, Generation Y (born between 1980 and 2000) -also known as the millennial generation (Kaifi et al., 2012).Members of the millennial generation are the youngest leaders to emerge in the workplace, yet their true ability to lead, including their leadership traits and styles, has not been fully explored, resulting in a knowledge gap concerning the behaviour of millennial leaders (Bargavi et al., 2017;Bushardt et al., 2018;Medyanik, 2016).
A leadership style can be defined as the pattern of attitudes that the leader holds and the behaviour that they show (Anderson & Sun, 2017).It might be reasonable to expect that leadership styles differ across generations (Andi, 2018;Faller & Gogek, 2019;Kaifi et al., 2012;Putriastuti & Stasi, 2019).Managers display leadership styles that connect with their views of human nature and with the peers and subordinates with whom they work (Sharifi-Rad et al., 2017), but they may also display a leadership style that resonates with their personality and feels the most natural to them (Long, 2017).
In this study, the leadership typology of Pearce et al. (2003) was adopted as a structure to discuss leadership styles.This typology supports the existence of four leadership styles, namely directive leadership, transactional leadership, transformational leadership and empowering leadership.According to Pearce et al. (2003), these leadership styles are defined as follows: • Directive leadership refers to behaviours where the leader gives orders on how the work needs to be done.• Transactional leadership refers to the behaviours that establish the parameters of the exchange relationship between the leader and the follower.• Transformational leadership refers to those leader behaviours that encourage vision, produce inspiration from their followers and motivate change.• Empowering leadership develops the followers so that they become effective and capable self-leaders.
The literature on leadership styles further expands on the four different leadership styles.Directive leadership informs employees exactly as to what they are supposed to do.The leader tells the employees about their task, what is expected and how it must be done and provides the deadline for the completion of that task (Wachira et al., 2016).
A transactional leadership style occurs in a leader-follower exchange relationship where corrective actions are an exception and followers are rewarded when they have achieved specific goals (Holten & Brenner, 2015).Transactional leaders, therefore, use praise, reward and promise to motivate employees.For corrective action, they will make use of negative feedback, threats or disciplinary action (Solaja & Ogunola, 2016).Transactional leaders will not continually look at their staff's performance, and they will offer assistance only when needed (Cheung et al., 2018).Moreover, they will use contingent rewards to incentivise their employees to share the knowledge that they own (Masa'deh et al., 2016).
Transformational leaders are charismatic, influencing their followers towards a vision through their insistence on moral and ethical standards (Dartey-Baah, 2015).By portraying these powerful visions, they also promote creativity among their followers in the workplace (Banks et al., 2016;Jaiswal & Dhar, 2015).Transformational leaders encourage and inspire their followers to achieve results that exceed expectations (Jauhar et al., 2017), but they also bring their followers into the decision-making process and allow them to develop as individuals (Solaja & Ogunola, 2016;Yahaya & Ebrahim, 2016).Transformational leadership signifies four dimensions, namely idealised influence, intellectual stimulation, inspirational motivation and individualised consideration (Aga et al., 2016).
With empowered leadership, workers are allowed more independence and self-leadership, while managers provide support and encouragement, promote participative decisionmaking and build trust (Liu, 2015).Empowerment programmes in the workplace are generally intended to improve employees' motivation and creativity in their work roles (Amundsen & Martinsen, 2015).

Millennial leaders
With technology having been a regular feature of the millennial generation's upbringing, and also having brought people closer, the millennial leader has become more authentic, democratic and transparent (Au-Yong-Oliveira et al., 2018).Millennial leaders regard teamwork as the most important leadership trait, followed by communication, respect, vision and influence (Graybill, 2014).Millennial leaders consider meaningful relationships as an important element of leadership (Medyanik, 2016) and adopt a twoway communication approach while also valuing reciprocal relationships with their subordinates (Chou, 2012).They also adhere to professional ethics in the workplace, having a dynamic transparent leadership style (Akmalaputri et al., 2018).The millennial leader wants to contribute towards the organisation by using their skills and prefers to be trusted and empowered to lead (Sledge, 2016).
Complementing the aforementioned, the work of Medyanik (2016) on millennial leaders drew the following conclusions on their leadership traits: • Millennial leaders believe in the empowerment of their followers.• Millennial leaders display good listening skills and use positive reinforcement as a primary motivator.• Millennial leaders want to lead by example.
• Millennial leaders look for opportunities to provide feedback to their subordinates and also believe in providing regular recognition.
• Millennial leaders may request firm deadlines and commitments but may also permit their staff to work when and how they want.• Millennial leaders have a team-centred approach and may lack individual accountability.
From this list, tentative hypotheses could be formulated as to the preferred leadership styles that millennial leaders would display.The concept of an ideal leadership style, based on the experience of followers, has received some attention (Andreescu & Vito, 2010;Baker, 2015;Tu et al., 2018).While a lot is written about millennials as a generation, how to lead millennials (Faller & Gogek, 2019;Grubbström & Lopez, 2018;Nolan, 2015), as well as the attitudes of millennial leaders (Churchill, 2019;Medyanik, 2016;Sledge, 2016), little is still known about the behaviour of millennials when acting as leaders.It would be useful to look to millennial leaders' own behaviour in the workplace to better understand their preferred leadership styles.This link between actual behaviour (own leadership style) and preferred behaviour is made according to the assumptions of the cognitive dissonance theory (Festinger, 1957), which suggests that individuals will strive to align behaviour with attitudes.Millennial leaders will not endorse leadership styles for which they have no affection.

Millennials' view of leaders in the workplace
An important assumption in this study is that millennials will lead others, in the manner in which they themselves want to be led -a manner that is aligned with their own values.Since its introduction to the social psychology literature more than 60 years ago, Festinger's cognitive dissonance theory suggests that our behaviour should match our values (Hinojosa et al., 2017).Festinger believed that there is a consistency not only in an individual's opinions and attitudes but also between what a person knows and believes and what that person does (Festinger, 1957).Millennial leaders are therefore assumed to lead according to their own values.
Millennials are social creatures, having been brought up with technology coupled with constant interaction with their parents, and will expect the same close connection with their workplace leaders (Axten, 2015).Millennials prefer leaders who display a leadership style that shows care for their followers, as opposed to being focussed on meeting their own personal agendas (Long, 2017).They prefer leaders who give personal attention to their employees and who get to know them by being more people-orientated than taskorientated (Maier et al., 2015).
The millennial generation does not believe in the notion that hierarchy creates a good leader (Pratama et al., 2019).They prefer not to work under a leadership style with highly autocratic directives as they value empowerment highly and need the opportunity and encouragement to make their own decisions (Maier et al., 2015).
Millennials see leaders as guiding and empowering and also as role models (Lamasan & Oducado, 2018).They expect leaders to be charismatic by providing their workers with a sense of purpose through encouragement (Grubbström & Lopez, 2018).Millennials also expect leaders to provide their followers with challenging tasks that are still within their capabilities, thereby rewarding innovation and tolerating failures (Axten, 2015).Millennials prefer leaders who are inclusive, collaborative and committed in their leadership approach (Maier et al., 2015).
Millennials have high expectations of their leaders revealed in their expectations of immediate feedback, immediate training and immediate recognition (Axten, 2015).Because they do well in work environments that provide progressive career paths, they need reaffirming that they are moving in the right direction, requiring a leader who motivates with rewards, sets goals and mentors their employees (Bodenhausen & Curtis, 2016).
From the literature review, we can infer that millennial leaders have a unique style from other generations in the workplace in terms of the leadership styles they adopt.By applying the leadership typology of Pearce et al. (2003), we can test for whether this uniqueness really exists.

Hypotheses
The general null hypotheses stated were that the leadership styles (all four) displayed by baby boomer leaders, Generation X leaders and millennial leaders do not differ significantly from each other.The following specific null and alternative hypotheses were also set: H1 0 : There is no difference in the empowering leadership style as used by baby boomer leaders, by Generation X leaders or by millennial leaders.
H1 a : Empowering leadership is displayed to a larger extent by millennial leaders than by the other generations' leaders.This hypothesis is based on the literature of Lamasan and Oducado (2018), Maier et al. (2015), Medyanik (2016) and Sledge (2016), suggesting that millennials highly value trust and empowerment in leaders, while also having the opportunity and encouragement to make their own decisions and be a role model to their followers.
H2 0 : There is no difference in the directive leadership style as used by baby boomer leaders, by Generation X leaders or by millennial leaders.
H2 a : Directive leadership is displayed to a lesser extent by millennial leaders than by the other generations' leaders.This hypothesis is based on the literature of Maier et al. (2015) and Pratama et al. (2019), suggesting that millennials will not work well with highly autocratic directive leaders, particularly where leaders have been created through a hierarchy.
H3 0 : There is no difference in the transactional leadership style as used by baby boomer generation leaders, by Generation X leaders or by millennial leaders.
H3 a : Transactional leadership is displayed to a lesser extent by millennial leaders than by the other generations' leaders.
This hypothesis is based on the literature of Axten (2015), Maier et al. (2015) and Medyanik (2016), all of which suggested that millennials as leaders will want to provide constant feedback and personal attention to their subordinates, that they are more orientated towards people and less towards tasks and that where they provide challenging tasks to their followers, they will also tolerate failures.
H4 0 : There is no difference in the transformational leadership style as used by baby boomer leaders, by Generation X leaders or by millennial leaders.
H4 a : Transformational leadership is displayed to a larger extent by millennial leaders than by the other generations' leaders.This hypothesis is based on the literature of Axten (2015), Bodenhausen and Curtis (2016) and Grubbström and Lopez (2018), suggesting that millennials prefer charismatic leaders in the workplace, who provide a sense of purpose through encouragement while also motivating their followers, providing them with challenging work and rewarding innovation.

Method Population and sampling
The target population was employees across generations, and this research was conducted in a variety of organisations in South Africa.The sample consisted of 1140 respondents across 19 South African organisations.In total, nine respondents did not provide their leaders' age, which made the working sample 1131.These organisations included both private and public entities representing, among others, the telecommunication, financial services, media, manufacturing and electronics industries.
Nineteen organisations were identified using the criterion of each having an employee who was a registered master's level student at the Graduate School of Business Leadership of the University of South Africa (GSBL).Entrance to the organisations, and thus access to the respondents, was achieved by leveraging the respective students as fellow researchers.Random samples of 60 employees were drawn from each organisation.

Design
Cross-sectional data were collected with various instruments based on the leadership typology of Pearce et al. (2003), after which the data was segmented according to the three generations.The perceptions of subordinates as to the leadership styles of their leaders comprised the unit of analysis.

Measurement instruments
Data were collected using a questionnaire incorporating various instruments selected based on their ability to measure the four leadership styles, namely empowering, directive, transactional and transformational.Approval for the use of the instruments was obtained from the respective authors.
Empowering leadership was measured using the 10-item instrument of Ahearne et al. (2005).Examples include 'My leader allows me to do my job my way' and 'My leader allows me to make important decisions quickly to satisfy customer needs'.The reliability of the items developed to measure empowering leadership is confirmed by Yoon (2012), reporting a Cronbach coefficient alpha of 0.93.
Directive leadership was measured using six items developed by Pearce and Sims (2002)  The age of the leaders, which was used to create the generational categories, was reported by the respondents.

Statistical analyses
The data analysis was done using IBM SPSS.Frequency analysis was performed to provide a descriptive view of respondent demographics based on sex, race and age, including leader age.This collected data was compared with the population data.
Before considering the data from the scales, reliability and validity were assessed.For the reliability assessment of the scale, Cronbach's alpha coefficients were calculated.While reliability was deemed as satisfactory where the alpha scores were above 0.7 (DeVellis, 2012), alpha scores above 0.8 were regarded as preferable (Pallant, 2020).Factorial validity was assessed using principal components analysis with Varimax orthogonal rotation and Kaizer normalisation (Pallant, 2020), with the aim being to show that the covariance between the items is explainable.The number of factors retained was based on Kaiser's criterion, applying the 'eigenvalues greater than one' rule (Pallant, 2020).
This was followed by calculating the mean scores across generations on the various leadership style scales.The one-way analysis of variance (ANOVA) was performed and indicated whether there were significant differences in the mean scores on the four leadership styles across the three generations.Using the Scheffé test for a post-hoc analysis to determine where the differences lay, an appropriate alpha level of 0.05 was selected.
Cohen d values were calculated to assess the practical strength of the differences, using Cohen's 2008 guidelines (values smaller than 0.2 being negligible, 0.2-0.5 equals small effect, 0.5 up to 0.8 equals medium effect and 0.8 upwards equals large effect).

Demographic variables
Descriptive statistics were calculated for the respondent sex, race and age including leader age, as shown in Table 1, Table 2 and Table 3, respectively.
The sexes were almost equally represented in the total sample.
The majority by race, as represented in the sample, were black respondents at 66.9%, followed by the white population group respondents at 18.2% and the Coloured respondents at 10.0%.The smallest minority population group was the Asian respondents at 4.9%.
From Table 3, it can be observed that two respondents did not provide their own ages.In the group as a whole the respondents ranged from 20 to 64 years of age, and the estimated ages of their leaders were slightly higher, ranging from 23 to 70 years of age.The mean age of the leaders is 44.23 years and within the mid-region of the Generation X age group.

Reliability
In Table 4, the Cronbach's alpha coefficient values of all the factors exceeded 0.70, which is acceptable (DeVellis, 2012), and three met the 0.80 criteria, which is preferable (Pallant, 2020).
All the data met the 0.70 criterion and three quarters the 0.80 criterion.

Factorial validity
The validity of the instruments was assessed through factorial analysis.The instruments were considered separately and individually.When applying Kaiser's criterion of eigenvalues greater than one, the following number of factors that exceed the 60% rule of thumb (Field, 2018) was extracted: • One factor on empowering leadership was extracted, explaining 62.1% of the variance in the data.• Three factors on directive leadership were extracted, explaining 75.8% of the variance in the data.The items loaded on the factors were in line with Pearce and Sims (2002) and Hwang et al. (2015) instruments and the conceptualisation of the construct by Pearce and Sims (2002).When applying the Varimax rotational approach and Kaizer normalisation on all the directive, transactional and transformational leadership items, the results showed that all the measurement items loaded onto their respective factors had no significant cross-loadings. 1The results (not presented in table format here) show support for the factorial validity of the scales used.

Mean scores
Mean scores were calculated on each of the four leadership styles for the millennial leaders, Generation X leaders and baby boomer leaders.The mean scores are depicted in Table 5.
The largest group of leaders identified were from Generation X, totalling 560, followed by the millennial leaders, constituting a total of 423.The smallest group of leaders were the baby boomers, totalling 148.
When comparing the total mean scores on the four leadership styles across the three leader generational groups, the following was noted: • Empowering leadership scored the highest, with a total mean score of 3.647.
1.When adding the empowering leadership items to the rest of the items, and then attempting to extract four, or other combinations of factors, the results were messy.• Transactional leadership scored the lowest, with a total mean score of 2.885.
In general, and thus not considering generational matters, empowering leadership seems to be the dominating leadership style in the workplace.This is reflective of the dominant leadership styles in the workplace but does not shed light on how leaders across generations apply these leadership styles.
To assess the magnitude of the differences between the mean scores across generations (as reported in Table 5), analyses of variance were performed (see Table 6).
In the ANOVA analysis, the large F ratio values for empowering leadership (11.671) and transformational leadership (8.811) indicated variability between the groups, supported by their p-values (both >0.001), supposing the set significance level of 0.05.Generation-based differences across directive and transactional leadership styles were absent.Thus, in Table 6, statistically significant differences in mean scores, across generations, were only observed for empowering and transformational leadership.
The Scheffé post hoc test (see Table 7) was performed to identify which groups differed.No results for directive or transactional leadership are presented here, as generations did not differ from each other in the manner they applied these styles.
In Table 7, only differences between millennials and other generations were of interest, as per the aim of this research.It can be observed (see Table 5) that the largest mean score difference was for empowering leadership, whereas Generation X (mean = 3.784, SD = 0.957) had a statistically significantly higher mean score than that of the millennials (mean = 3.540, SD = 0.997).When the practical significance of the difference is calculated, using the Cohen formula (Cohen, 1988), the Cohen d value is 0.250.Given Cohen's parameters for analysing this result, the difference was small on a practical level.
The only other place where a generation scored differently from millennials was on transformational leadership.Here it can be observed that the second-largest mean score difference was for transformational leadership, where Generation X (mean = 3.690, SD = 0.735) had a statistically significantly higher score than that of the millennials (mean = 3.528, SD = 0.784).When the practical significance of the difference is calculated, the Cohen d value is 0.213.The difference was also small, on a practical level.

Discussion
Discussions of leadership as a topic in the literature are prevalent, yet, given the literature reviewed, little is known concerning how millennial leaders lead or want to lead.The premise is that the millennial generation is different to the older generations, based on their status as millennials, as influenced by the unique circumstances of the period in which they were born and brought up.The very essence of generational theory suggests that the attitudes, values and behaviours of generations differ and that this should influence their leadership styles.
From the literature, hypotheses were formulated on how millennial leaders lead differently compared with the leaders of Generation X and the baby boomer generation.The millennials did not report on their own leadership styles; rather, the research conducted was based on the perspectives of subordinates on the leadership styles displayed by their leaders, which included millennial leaders.The sample could be deemed as representative of the working population in South Africa in as much as the gender distribution (in general), as well as the race composition of the total group, reflected the numbers provided by Stats SA (given Statistics South Africa, 2020 data).When considering the split of the leader age groups, the largest number was of Generation X leaders, followed by the millennial leader age group.The smallest leader age group was that of the baby boomer leaders.These numbers make sense intuitively, but they are also in line with the trend of the baby boomer generation starting to retire (Brack & Kelly, 2012).The millennial generation is presently the largest group in the workforce (Pratama et al., 2019).The collected data confirm this.
The instruments used showed acceptable psychometric characteristics, and the Cronbach alphas were acceptable (0.73-0.94).With regard to factorial validity, the instruments also showed acceptable results across all the instruments used.
Satisfactory factor-analytical data and/or expected relationships, reported by Yoon (2012), support the validity of the instrument of Ahearne et al. (2005) to measure empowering leadership.Satisfactory factor-analytical data and/or expected relationships, reported by Yoon (2012), also support the validity of the instruments of Pearce and Sims (2002) and Hwang et al. (2015) in measuring directive leadership, as well as the instrument of Pearce and Sims (2002) in measuring both transactional and transformational leadership.
The first finding was that, in the present workplace, empowering leadership seems most dominant, and transactional leadership is least applied.Empowering leadership is a leadership style where responsibility is handed down to followers (Sims et al., 2009), which is well reflective of a society where equality among individuals is asserted.
Interesting to note is that transactional leadership is applied the least, even less than in the case of directive leadership.Directive leadership is less aligned with the democratic Zeitgeist of the day.However, this may not be unaligned with the millennial workforce, in which the 'parenting role' of the leader may be acceptable or even asked for.
The results showed that the general null hypotheses, in which millennial leaders are no different from leaders from other generations, could not be rejected in full, as there were statistically significant differences in only two of the four leadership styles when the displayed leadership styles across the three leader age groups were compared.From this, we can conclude that it is irresponsible to make a blanket statement that millennial leaders are seen to lead differently than other generation leaders in the workplace.When considering the practical significance of the differences between the generations leading, these differences were small.Then, considering practical significance, the null hypotheses should not be rejected, and it could be stated that there is no compelling evidence that millennials lead in a manner different to other generations presently in the workforce.
Considering the set alternative hypotheses, the results were contrary to what was proposed.Generation X leaders were shown to be more involved in empowering leadership than millennial leaders.Subordinates under Generation X leaders reported their leaders to be more empowering than did subordinates who reported on their millennial leaders.These results do not align with the literature of Lamasan and Oducado (2018), Maier et al. (2015), Medyanik (2016) and Sledge (2016).From these authors' work, it was deduced that millennials would be more inclined to lead in an empowering way.
Generation X leaders were shown to be more inclined to display transformational leadership than millennial leaders.These results are also not aligned with the literature of Axten (2015), Bodenhausen and Curtis (2016) and Grubbström and Lopez (2018), all of which suggest the reverse to be true.In both the cases where there were differences in empowering and transformational leadership, of importance to note is that the differences may be statistically significant, however considering the effect, small in size.

Conclusion
The study focused on whether millennial leaders adopt leadership styles different from those of the older generational leaders in the workplace.This is on the premise that millennials hold different values from the other generations, and that these values, in turn, have an impact on their leadership styles.
The data revealed statistically significant differences in two of the four leadership styles, empowering leadership and transformational leadership, where millennial leaders differed from their predecessors (Generation X).At a practical level, these differences were small.The lack of statistically significant differences between two of the four leadership styles suggests, firstly, that millennials are no different in the extent to which they apply transactional or directive leadership, which is often, or anecdotally at least, frowned upon by their own generation.The millennials, according to this data and contrary to what was hypothesised, display less empowering leadership and transformational leadership than the leaders from the generation that preceded them.These differences also do not align with the literature, which suggests that millennial leaders should be prone to displaying these types of leadership behaviour.
In many respects, this research turns the matter of generational differences on its head.The set alternative hypotheses were not supported by the data, as the alternative hypotheses findings were not in the direction envisaged.The results necessitate revisiting of the central hypotheses of generational differences or alternatively applying different research strategies to investigate the issue.It should also be considered that other demographics, apart from age, influenced the results.

Limitations and recommendations
A limitation of this study was the convenient sampling of organisations.However, it should be noted that, while the sampling of the respondents in organisations was random, the demographics of the respondents seemed to closely reflect the demographics of the country as a whole.Future researchers are urged to use random sampling of both organisations and respondents.
The study is also limited in being an example of single-source (tapping on the perceptions of subordinates) and singlemethod (using surveys) research.Future researchers are advised to additionally gain information on self-reports from leaders themselves, as well as to use techniques beyond surveys -for example, observations from qualified assessors.However, a focus on differences between generations is encouraged.Much of the research in this field seems to have an emic stance, whereas this research addresses the matter from an etic angle, which provided these interesting findings.Therefore, more quantitative studies, focusing on differences, may be necessary.
It is further recommended that the generational perspective on leadership be revisited.Perhaps generational 'culture' is a much weaker predictor of leadership behaviour than the prevailing workplace 'culture' -and perhaps leadership is just leadership, and the type of leadership displayed in today's workplace is what is required now, irrespective of who fulfils that role.This necessitates future researchers to include variables such as organisational culture, organisational structure and industry in their research.The study was conducted in South African organisations, and future consideration may be merited to not only organisational culture but also culture itself.
Aligned with the statements in the previous paragraph, those involved in the mentoring and training of millennial leaders should take cognisance of this research.Millennial leaders may not be what the textbook tells us, and interventions should not blindly align to the stereotypes associated with this generation.
• Three factors on transactional leadership were extracted, explaining 62.43% of the variance in the data.The items loaded on the respective factors were in line with the Full-Range-Leadership-Model (Avolio & Bass, 2001).• Four factors on transformational leadership were extracted, explaining 68.01% of the variance in the data.The items loaded on the respective factors were in line with the Full-Range-Leadership-Model (Avolio & Bass, 2001).
Hwang et al. (2015)02)ng et al. (2015).'Myleadergivesmeinstructionsabouthow to do my work' and 'My leader identifies specific action steps and accountabilities for me' are examples of these items intended to measure directive leadership.The items developed byPearce and Sims (2002)are confirmed as reliable (Cronbach coefficient alpha of 0.88)(Hinrichs, 2011), whileHwang et al. (2015)confirm the reliability of their items developed to measure directive leadership (Cronbach coefficient alpha of 0.85).
Pearce and Sims (2002)d with an instrument developed byPearce and Sims (2002).Sixteen items addressed transactional leadership, and examples include 'My leader closely monitors my performance for errors' and 'My leader waits until things have gone wrong before taking action'.Transformational leadership was also measured with an instrument developed byPearce and Sims (2002).Twenty statements were utilised to measure transformational leadership, and examples include 'My leader expects me to perform at my highest level', 'My leader is driven by higher purposes or ideals' and 'My leader questions the traditional way of doing things'.Reliability of the items developed to measure transactional and transformational leadership is confirmed byPearce and Sims (2002)(Cronbach coefficient alpha's of 0.87 and 0.72, respectively).

TABLE 3 :
Respondent and leader age (years).

TABLE 6 :
One-way analysis of variance: Mean score differences across generations.

TABLE 5 :
Mean scores per leadership style and generation.

TABLE 7 :
Scheffé post hoc test.Difference with significance levels of less than 0.05. *,