Athlete leadership has been defined as an athlete occupying a leadership role (i.e., formal or informal) within a team who influences team members to achieve a common goal (Loughead, Hardy, & Eys, 2006). To date, researchers have demonstrated that athlete leaders tend to be skilled performers (Yukelson, Weinberg, Richardson, & Jackson, 1983), usually occupy central playing positions (Fransen et al., 2016; Glenn & Horn, 1993), have strong interpersonal connections with teammates (Moran & Weiss, 2006; Tropp & Landers, 1979), and are typically veteran players (Tropp & Landers, 1979; Yukelson et al., 1983). In addition, researchers found multiple team members, occupying either a formal or informal leadership role, partake in their team's leadership processes (e.g., Fransen, Vanbeselaere, De Cuyper, Broek, & Boen, 2014; Loughead et al., 2006).

In comparison to other fields of leadership research (e.g., educational psychology and organizational psychology), the study of athlete leadership is still relatively new. As such, the majority of athlete leadership studies have been guided by models from sport coaching research (i.e., multidimensional model of leadership [MML]; Chelladurai, 1978; 1993) and organizational psychology (i.e., full range model of leadership [FRML]; Avolio, 1999). Consequently, the measurement of athlete leadership has also reflected this reality. In particular, athlete leadership has primarily been measured using the Leadership Scale for Sports (LSS; Chelladurai & Saleh, 1980) originally developed to assess coach leadership, the Multifactor Leadership Questionnaire (MLQ-5X; Bass & Avolio, 2004) originally developed to assess organizational leadership, and the Differentiated Transformational Leadership Inventory (DTLI; Hardy et al., 2010) originally developed to assess leadership in a military setting. From a theoretical perspective, the dimensions of leadership behavior assessed by the LSS, MLQ-5X, and DTLI cover what Avolio (1999) would describe as a full range of leadership behaviors. That is, these three inventories measure a wide variety of leadership behaviors. Specifically, Avolio and Yammarino (2002), in discussing the full range leadership theory (FRLT), note three broad categories of leadership behaviors: transformational (motivating others by making them aware of the importance of task outcomes and transcending their own self-interest), transactional (exchange process motivating individuals to comply with leader requests), and laissez-faire (absence of leadership). These dimensions of leadership behavior fall on a continuum ranging from active-passive and effective-ineffective (Avolio, 1999). Laissez-faire is considered to be the most passive and ineffective form of leadership while transformational leadership is viewed as being highly active and effective. Transactional leadership falls between laissez-faire and transformational forms of leadership on the continuum. Avolio indicates that effective leaders display both transformational and transactional leadership behaviors.

The MML (Chelladurai, 1978; 1993) was operationalized through the development of the LSS (Chelladurai & Saleh, 1980), a 40-item inventory that measures five dimensions of leader behavior (i.e., democratic behavior, autocratic behavior, positive feedback, social support, and training and instruction). The MLQ-5X (Bass & Avolio, 2004) was developed to specifically accompany the FRML (Avolio, 1999) and comprises 36 items that measure four dimensions of transformational leadership behavior (i.e., idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration), three dimensions of transactional leadership behavior (i.e., contingent reward, management-by-exception active, and management-by-exception passive), and one dimension of non-leadership behavior (i.e., laissez-faire). Finally, the DTLI (Flardy et al., 2010) is a 26-item inventory built on the theoretical underpinnings of transactional and transformational leadership. It is a composite measure that includes subscales from the MLQ-5X (Bass & Avolio, 2000) and the Transformational Leadership Inventory (TLI; Podsakoff, Mackenzie, Moorman, & Fetter, 1990). Specifically, the DTLI measures six dimensions of transformational leadership behavior (i.e., appropriate role model [TLI], inspirational motivation [MLQ-5X], intellectual stimulation [TLI], individualized consideration [MLQ-5X], fostering acceptance of group goals and promoting teamwork [TLI], and high performance expectations [TLI]) and one dimension of transactional leadership behavior (contingent reward [TLI]).

When the LSS, MLQ-5X, and DTLI have been used, their subscales have shown to be related to a number of individual- and team-level correlates including task and social cohesion (e.g., Callow, Smith, Flardy, Arthur, & Hardy, 2009; Vincer & Loughead, 2010), athlete satisfaction (Paradis & Loughead, 2012), collective efficacy (Price & Weiss, 2011), communication (Smith, Arthur, Hardy, Callow, & Williams, 2013), and motivation (i.e., ability beliefs, motivational orientations, and social orientations; Vidic & Burton, 2011). However, one issue pertaining to previous research is that these measures assess how often the leadership behavior is used; yet, empirical evidence suggests that most leadership behaviors can be overused or underused, and that the optimal frequency of the leadership behavior appears to be a moderate amount (Pierce & Aguinis, 2013). Furthermore, even if exhibiting a leadership behavior more frequently does not reduce the benefits or have negative side effects, spending more time displaying a behavior means that the leader is losing the opportunity to use other types of leadership behavior (Yukl, 2012). In order to advance leadership theory and practice, it is necessary to examine more than the frequency of leadership behaviors (Yukl, 2012). Therefore, the first purpose of the present study was to determine athletes' perceptions regarding the importance for athlete leaders to display the dimensions of leadership behavior as described by the LSS, MLQ-5X, and DTLI. The examination of this purpose is important for two reasons. First, each leadership inventory discussed above was originally developed for use in a context different than athlete leadership (e.g., coach leadership, military leadership). As such, the dimensions of leadership behavior included in the LSS, MLQ-5X, and DTLI may not be entirely important for athlete leaders to display. For instance, Duguay, Loughead, and Munroe-Chandler (2016) recently found no significant mean increase for the dimension of intellectual stimulation following the implementation of an athlete leadership development program with varsity athletes. These authors postulated that this dimension of leadership behavior may be viewed as being the responsibility of a coach or peer mentor rather than an athlete leader because of its focus on problem solving and viewing obstacles from new perspectives. Second, identifying leadership behaviors that are important to the athlete is critical as this information can be used by key stakeholders (e.g., coaches, sport psychology consultants, administrators, athletes) to pinpoint behaviors that ought to be developed and trained (Burke, Stagl, Klein, Goodwin, Salas, & Halpin, 2006). Pain, Flarwood, and Mullen (2012) demonstrated that athletes have the ability to recognize factors that facilitate (e.g., positive team leadership) or impede (e.g., lost composure) their performance. Providing athletes with the opportunity to identify the leadership behaviors they believe are important for their athlete leaders to display offers a unique perspective. Given the associations between many of the leadership behaviors of the LSS, MLQ-5X, and DTLI and various individual- and team-level outcomes (Loughead, Munroe-Chandler, Floffmann, & Duguay, 2014), it was hypothesized that athletes would view a number of leadership behaviors as important for athlete leaders to display.

A second purpose of the current study was to determine the importance athletes assigned to the dimensions of leadership behavior of the LSS, MLQ-5X, and DLTI (as defined in the instruments) based on the tenure of the athlete leader. Janssen (2015) contends that athletes at the intercollegiate level can be classified into leadership categories based on their readiness to assume leadership roles. For instance, within Canadian intercollegiate sport (the current sample) athletes have five years of playing eligibility. That is, athlete leaders may be classified as rookie leaders (i.e., competing in their first year of intercollegiate sport), emerging leaders (i.e., competing in their second or third year of intercollegiate sport), or veteran leaders (i.e., competing in their fourth or fifth year of intercollegiate sport). Athlete leaders who vary in tenns of tenure (i.e., rookie, emerging, and veteran) may share different leadership abilities, responsibilities, expectations, and needs. Identifying whether the importance of leadership behaviors differs based on tenure is critical from a practical standpoint as it has been suggested that leadership development opportunities be customized to accommodate individual needs and characteristics (e.g., Facca & Allen, 2011; McCauley, Kanaga, & Lafferty, 2010). Consequently, it was hypothesized that athletes would rate that it is more important for emerging athlete leaders than rookie athlete leaders to display any of the leadership behaviors. Furthermore, it was hypothesized that athletes would believe it is more important for veteran athlete leaders than emerging and rookie athlete leaders to display any of the leadership behaviors.

Method

Participants

Participants included 133 intercollegiate athletes (94 females and 39 males; [M.sub.age] = 20.76, SD = 1.84) from 19 different sports (basketball, cross country, curling, football, golf, ice hockey, soccer, track and field, volleyball, rugby, baseball, field hockey, rowing, swimming, tennis, Nordic skiing, wrestling, lacrosse, and figure skating) competing in either U Sports or the Canadian Collegiate Athletic Association (CCAA). These two sport associations represent the highest level of intercollegiate sport in Canada. Athletes competing within U Sports or the CCAA have a total of five years of playing eligibility. Of the 133 athletes, 41 (30.8%) were in their first year of competition with their current team, 30 (22.6%) were in their second year, 23 (17.3%) were in their third year, 25 (18.8%) were in their fourth year, 12 (9%) were in their fifth year, and 2 (1.5%) did not indicate their team tenure. In terms of leadership status, 22 (16.5%) participants self-identified as a fonnal leader, 91 (68.4%) as an informal leader, and 17 (12.8%) indicated they did not hold a leadership position. Three athletes (2.3%) did not report their leadership status.

Measures

An online survey to assess intercollegiate athletes' perceptions of the importance of athlete leadership behaviors was created using FluidSurveys, a product of Survey Monkey (Finley & Finley, 1999). The definitions of 13 dimensions of leadership behavior derived from the LSS (Chelladurai & Saleh, 1980), MLQ-5X (Bass & Avolio, 2004), and DTLI (Callow et ah, 2009) were included in the survey. While these inventories measure more than 13 dimensions of leadership behavior when combined, several behaviors are duplicated (e.g., individual consideration) or are conceptually similar (i.e., idealized influence and appropriate role model). Therefore, these leadership dimensions were combined in the current study. It should be noted that the leadership dimensions contingent reward (MLQ-5X and DTLI) and laissez-faire (MLQ-5X) were not included in the current study. As discussed by Judge and Piccolo (2004), contingent reward is resource dependent. As athlete leaders typically do not have the resources necessary to provide tangible rewards to their teammates (e.g., scholarships, equipment, increased playing time) this dimension was removed from the survey. Furthermore, laissez-faire is described as an absence of leadership. As such, it was decided that this dimension was not representative of effective leadership.

The survey package included three parts: (1) a demographic section that asked about age, gender, sport, leadership status, and team tenure; (2) a general importance rating of the 13 dimensions of leader behavior; and (3) an importance rating of the dimensions based on the team tenure of the athlete leader. Specifically, participants were provided with a definition of each dimension of leadership behavior (see Table 1). These definitions were based on how each dimension has been conceptualized within the literature (e.g., Bass & Avolio, 2004, Callow et al., 2009, Chelladurai & Saleh, 1980). In this way, each definition reflected the general meaning of their associated dimension. To assess the content-relevance of the definitions, six expert raters were asked to independently rate the degree to which each definition matched the 13 dimensions of leadership behavior. Ratings were provided on the following 5-point scale: 1 (Poor Match), 2 (Fair Match), 3 (Good Match), 4 (Very Good Match), and 5 (Excellent Match). Aiken's (1985) content-validity coefficient was then used to test the statistical significance of the expert ratings. Values of Aiken's V can range from 0 to 1 with a value of 0 indicating that all judges gave a definition the lowest possible score on the rating scale (i.e., definition-dimension match = 1) and a value of 1 indicating that all judges gave a definition the highest possible score on the rating scale (i.e., definition-dimension match = 5). With the exception of the definition-dimension ratings for democratic behavior (V= 0.67, p > .05), all matches received the highest possible score on the rating scale (i.e., V = 1, p < .01). Follow-up discussions were held with raters who found a discrepancy between the democratic behavior definition and dimension. As a result, minor edits were made to this definition. Raters were in agreement that the edited definition more accurately reflected the dimension of democratic behavior. Taken together, the expert raters felt that the definitions presented in Table 1 strongly reflected their associated dimensions.

To assess the general importance of each dimension of leadership behavior, the participants were asked to rate their level of agreement with each of the 13 dimensions on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The stem, "It is important for athlete leaders to ..." preceded each definition. To rate the importance of the dimensions of leadership behavior based on the tenure of the athlete, the participants were presented with a point-assignment measure where they were provided a fixed number of points and asked to weigh the importance of each dimension in relation to the tenure of the athlete leader. A point-assignment scale was used given that we had three groupings, which reflected a hierarchical nature based on team tenure. Consequently, a point-assignment scale reduces social desirability of responses since participants are required to organize their responses in order of importance (Greer & Dunlap, 1997; Ravlin & Meglino, 1987). In the current study, participants were required to allocate 30 points among rookie athlete leaders (i.e., athlete leaders who were in their first year of team tenure), emerging athlete leaders (i.e., athlete leaders who were in their second or third year of team tenure), and veteran athlete leaders (i.e., athlete leaders who were in their fourth or fifth year of team tenure) according to how important they felt it was for that tenure to display the dimensions of leadership behavior (more points = higher importance). Any combination of point distribution could be created as long as the total summed to 30. For example, points could have been distributed equally to indicate that it is equally important for all leaders, regardless of tenure, to display a particular leadership dimension (i.e., it is important for athlete leaders to inspire teammates by motivating them: rookie athlete leaders: 10 points; emerging athlete leaders: 10 points; veteran athlete leaders: 10 points). Conversely, more points could have been distributed to a specific tenure to indicate that it is more important for that tenure to display a particular leadership dimension (i.e., it is important for athlete leaders to inspire teammates by motivating them: rookie athlete leaders: 0 points; emerging athlete leaders: 5 points; veteran athlete leaders: 25 points). Athletes were reminded that athlete leaders could be those that are formal (an individual who has been appointed or elected to the position by the coach or team [i.e., team captains or assistant captains]) or informal (an individual who emerges as a leader through their interaction with teammates but holds no formal title) in nature.

Procedure

Following the competitive season and upon clearance from the university's Research Ethics Board, athletes competing at the U Sports and CCAA levels were sent an email from the authors that detailed the nature of the study and included a link to the online survey. The link brought the athletes to a welcome page that included a letter of information. At this point, athletes were able to provide informed consent by electronically agreeing to participate in the study. One hundred sixty-six athletes accessed the online survey. Following the removal of incomplete surveys, a total sample of 133 remained, which corresponds to a completion rate was 80.12%.

Results

A summary of the descriptive statistics representing the importance athletes placed on the dimensions of leadership behavior of the LSS, MLQ-5X, and DTLI is reported in Table 2. As shown in Table 2, idealized influence, fostering acceptance of group goals, inspirational motivation, high performance expectations, democratic behavior, social support, positive feedback, and individual consideration were important for athlete leaders to display. Conversely, athletes did not agree that it was important for athlete leaders to display management-by-exception passive behaviors.

A summary of the means and standard deviations representing the importance athletes placed on the dimensions of leadership behavior of the LSS, MLQ-5X, and DTLI based on the tenure of the athlete leader is reported in Table 3. As demonstrated, participants allocated the most points to veteran athlete leaders, followed by emerging athlete leaders, and the least amount of points to rookie athlete leaders across all dimensions. As previously described, a point-assignment scale was used to measure importance by leader tenure. This type of scale, known as an ipsative measure, is one for which the mean for each variable equals the same constant (Greer & Dunlap, 1997). The constant used in the current study was 30. Research has demonstrated that repeated measures ANOVA works well with ipsative data when the epsilon correction for degrees of freedom is used (Greer & Dunlap, 1997). Consequently, a series of repeated-measures one-way ANOVAs using a Greenhouse-Geisser correction were carried out to determine if the mean differences between the leader tenures were significantly different from one another. As shown in Table 4, a significant difference in the points allocated according to athlete leader tenure was found across all dimensions of leadership behavior. Post hoc comparisons using Bonferroni correction indicated that significantly more points were allocated to emerging athlete leaders than to rookie athlete leaders across all dimensions of leadership behavior. Furthermore, significantly more points were allocated to veteran athlete leaders than to emerging athlete leaders and rookie athlete leaders across all dimensions of leadership behavior. All pairwise comparisons were significant at ps < .001. In summary, it was more important for emerging athlete leaders to display all of the 13 dimensions of leadership behavior than rookie athlete leaders. Similarly, it was more important for veteran athlete leaders to display all of the 13 dimensions of leadership behavior than emerging athlete leaders and rookie athlete leaders.

Discussion

The first purpose of the present study w'as to determine athletes' perceptions regarding how important it is for athlete leaders to display the dimensions of leadership behavior as described by the LSS, MLQ-5X, and DTLI. It was hypothesized that participants would rate a number of dimensions as important for athlete leaders to display. Overall, the findings of the current study support this hypothesis. It would appear that it is important for athlete leaders to display the majority of the dimensions of leadership behavior included in the current study.

Overall, the results relating to the first purpose suggest that athletes believe it is important for athlete leaders to display a number of leadership behaviors. This finding is in line with previous research that has shown that athletes fulfill various leadership roles (i.e., task leader, social leader, external leader, and motivational leader; Loughead et al., 2006; Fransen et al., 2014). In fulfillment of these roles, athlete leaders carry out a number of behaviors. A motivational leader, for example, may express confidence or talk enthusiastically leading up to a big game. Similarly, through interviews with ice hockey coaches, Bucci, Bloom, Loughead, and Caron (2012) found that coaches asked their athlete leaders to employ a range of leadership behaviors. The MML and FRLT may help explain these findings. As theorized by the MML and the full range leadership development process model (FRLD; Sosik, 2006), antecedents (e.g., athlete characteristics, situational context) influence a leader's behavior, which in turn impacts various individual- and team-level outcomes (e.g., cohesion, satisfaction). As such, participants may have rated a high number of dimensions of leadership behavior as being important for athlete leaders to display because of the many situations that face a leader in a sport context. In this way, athlete leaders are equipped with a number of leadership behaviors that they can display depending on their personal characteristics, the characteristics of their teammates, and the situational context.

The second purpose was to examine how important participants believed it is for athlete leaders to display the dimensions of leadership behavior as described by the LSS, MLQ-5X, and DLTI based on the tenure of the athlete leader. It was hypothesized that participants would perceive it as more important for veteran athlete leaders, followed by emerging athlete leaders, and then for rookie athlete leaders to display any of the dimensions. The findings supported these hypotheses. Specifically, it was found that athletes significantly allocated the most points to veteran athlete leaders, followed by emerging athlete leaders, and the least amount of points to rookie athlete leaders across all dimensions. The findings support Janssen's (2015) suggestion that intercollegiate athlete leaders who are in different levels of team tenure share different leadership abilities, responsibilities, expectations, and needs. In relation to the current study, findings indicated that participants perceive the importance of displaying athlete leadership behaviors is contingent on the athlete leader's tenure. It was shown that it is more important for veteran athlete leaders to display all of the 13 leadership dimensions, followed by emerging athlete leaders, and then for rookie athlete leaders. These findings may be explained from findings presented by Benson, Evans, and Eys (2015) wherein they interviewed coaches and athletes to explore the nature of socialization in sport. Relevant to the present study, coaches and athletes discussed a contingency-based role progression for new members where athletes anticipated to progress in their role responsibilities with each competitive year. Interestingly, the role of veteran athletes was critical in welcoming newcomers (i.e., rookies) to the team. In particular, athletes and coaches discussed the prominent role veteran athletes fulfill in helping the integration process of new team members. These responsibilities may help explain why athletes believed it is more important for veteran athlete leaders to display each dimension.

The findings of the current study support the contention that leadership development efforts be customized to accommodate individual needs and characteristics (e.g., Facca & Allen, 2011; McCauley et al., 2010). In particular, it has been suggested that leadership development programs would benefit from adjusting for a number of factors including organizational level, high-potential status, and function within the organization (McCauley et al., 2010). Examples of specific accommodations include modifying content, adjusting outcomes, using different methods of delivery, or creating specific groupings of individuals (Facca & Allen, 2011; McCauley et al., 2010). Results of the current study suggest that athlete leadership development efforts may benefit from making program adjustments to accommodate for athlete leader tenure. For instance, it may be beneficial to have athletes participate in leadership development programs with teammates or other athletes in the same tenure as themselves. This would allow athletes with similar leadership tenure to share ideas, challenges, and experiences while also allowing the program facilitator the ability to individualize program delivery, content, and outcomes in order to better target the needs of the specific tenures. For example, a leader who displays the leadership behavior training and instruction helps their teammates' by instructing them in skills, techniques, and tactics of the sport. While veteran athlete leaders may feel comfortable carrying out this behavior provided their advanced experience, rookie athletes may not be prepared to provide such guidance, especially to veteran teammates. In this way, a facilitator may focus on the principles of the leadership behavior with rookie athlete leaders and discuss ways they can comfortably practice the principles within their context (e.g., practice displaying the behavior with younger athletes at sports camps). Conversely, veteran athlete leaders may be encouraged to display training and instruction with the younger athletes on the team in order to help reinforce the coach's instructions. Furthermore, in an intercollegiate context, the facilitator may challenge veteran athlete leaders to discuss how a leadership behavior may be used outside of a sport context to prepare them for leadership opportunities beyond sport. While the example presented here discusses the potential usefulness of athletes participating in leadership development programs with teammates in the same tenure as themselves, it may be equally useful for athletes to participate with teammates of a different tenure. In such a case, athletes may benefit from learning from the experiences of more veteran leaders. It is important to remember, however, that not all situations will be the same and that each team needs to evaluate their members on an individual basis. For instance, there may be a rookie athlete who has significant leadership and playing experience at a high level of competition (e.g., national team). In this situation the facilitator may decide to include this athlete with a different tenure where they can receive a program that is more suited to their developmental needs.

Another important contribution of these findings is noted in the fact that no tenure (rookie, emerging, or veteran) received a point-assignment of zero, indicating that it is important for all leader tenures to display leadership behaviors. Given that athlete leaders can fulfill both formal and informal positions (Loughead et ah, 2006) and that leadership is often distributed within a team (Frasen et ah, 2014), the current study supports research suggesting that all athletes be included in leadership development efforts (e.g., Crozier, Loughead, & Munroe-Chandler, 2013). In this way, coaches and sport psychology consultants not only facilitate the leadership development of athletes who hold formal leadership roles but can also prepare those athletes who already fulfill informal leadership positions or who may find themselves in an informal leadership role at some point in the season.

From a research perspective, there are several future directions gleaned from the current study. First, in relation to the first purpose, results extend previous athlete leadership research by focusing on the importance of the leadership behaviors rather than on how frequently they were displayed. In particular, the results demonstrated that athletes perceived many of the leadership behaviors measured by the LSS, MLQ-5X, and DTLI as important for athlete leaders to display. However, there were several leadership behaviors that may not accurately represent what athlete leaders do (e.g., autocratic behavior, management-by-exception passive). As Avolio (1999) has asserted there are likely aspects of leadership yet to be discovered. Similarly, there may be additional athlete leadership behaviors not yet accounted for by current measures. For instance, in their interviews with ice hockey coaches, Bucci et al. (2012) noted that these coaches believed it was important for athlete leaders to help communicate the coaching stalks message to the team. However, this behavior is not accounted for by the LSS, MLQ-5X, or DTLI. Moreover, using three scales to measure one construct (athlete leadership) is unrealistic and problematic. Future research should explore athlete leadership behaviors in the context of sport and continue to refine its measurement.

In this way, a more complete understanding of athlete leadership behaviors can be acquired. Second, the current study reminded participants that athlete leaders can hold formal or informal positions but did not provide participants the opportunity to rate the importance of behaviors differently for each leadership role. Future research should investigate whether importance ratings differ in the context of formal or informal leadership. Third, it is important for future research to explore why athletes view certain behaviors as more important than others. This information will not only be of interest to athlete leaders who seek to better understand their responsibilities and the expectations of their teammates but also to researchers, coaches, and sport psychology consultants who design and facilitate leadership development programs. Finally, as suggested above, the current study supports previous research that has suggested leadership development opportunities include all athletes and that steps should be taken to individualize development efforts. Research is needed to examine program outcomes in light of these recommendations.

From an applied perspective, the current study provides important information regarding athlete leadership that can be used in the advancement of leadership development programs. Namely, the present findings offer insight into the importance athletes ascribe to the leadership behaviors of the LSS, MLQ-5X, and DTLI both in general and when accounting for athlete leader tenure. Notably, these findings can be used, in combination with previous research, to determine focus areas for the content of leadership development programs. In addition, the findings highlight the importance of individualizing leadership development experiences.

The results of the current study extend our knowledge of athlete leadership in sport by demonstrating that athletes believed it is important for athlete leaders to display a number of leadership behaviors. Furthermore, findings showed that athletes believed it is more important for veteran athlete leaders to display leadership behaviors, followed by emerging athlete leaders, and finally rookie athlete leaders. It is hoped that athletes, coaches, and sport psychology consultants will use the present information to further enhance athlete leadership within teams. In addition, it is hoped that the present study will encourage researchers to further examine athlete leadership behaviors and to investigate the impact of athlete leadership development programs in light of the suggestions that have been presented herein.

References

Aiken, L. R. (1985). Three coefficients for analyzing the reliability and validity of ratings. Educational and Psychological Measurement, 45, 131-142. doi: 10.1177/0013164485451012

Avolio, B. J. (1999). Full leadership development: Building the vital forces in organizations. Thousand Oaks, CA: Sage.

Avolio, B. J., & Yammarino, F. J. (2002). Transformational and charismatic leadership: The road ahead. Greenwich, CT: JAI Press.

Bass, B. M., & Avolio, B. J. (2000). MLQ: Multifactor Leadership Questionnaire (2nd ed.). Redwood City, CA: Mind Garden.

Bass, B. M., & Avolio, B. J. (2004). Manual for the Multifactor Leadership Questionnaire (Form 5X). Palo Alto, CA: Mind Garden, Inc.

Benson, A. J., Evans, M. B., & Eys, M. A. (2015). Organizational socialization in team sport environments. Scandinavian Journal of Medicine & Science in Sports. Advance online publication, doi: 10.1111/sms. 12460

Bucci, J., Bloom, G. A., Loughead, T. M., & Caron, J. G. (2012). Ice hockey coaches' perceptions of athlete leadership. Journal of Applied Sport Psychology, 24, 243-259. doi: 10. 1080/10413200.2011.636416

Burke, C. S., Stagl, K. C., Klein, C., Goodwin, G. F., Salas, E., & Halpin, S. M. (2006). What type of leadership behaviors are functional in teams? A meta-analysis. Leadership Quarterly, 17, 288-307. doi:10.1016/j.leaqua.2006.02.007

Callow, N., Smith, M., Hardy, L., Arthur, C. A., & Hardy, J. (2009). Measurement of transformational leadership and its relationship with team cohesion and performance level. Journal of Applied Sport Psychology', 21, 395-412. doi: 10.1080/10413200903204754

Chelladurai, P. (1978). A contingency model of leadership in athletics (Unpublished doctoral dissertation). University of Waterloo, Ontario, Canada.

Chelladurai, P. (1993). Leadership. In R. N. Singer, M. Murphy & L. K. Tennant (Eds.), Handbook of research on sport psychology (pp. 647-671). New York: Macmillan.

Chelladurai, P. & Saleh, S. D. (1980). Dimensions of leader behavior in sports: Development of a leadership scale. Journal of Sport Psychology, 2, 34-45. doi: 10.1123/jsp.2.1.34

Crozier, A. J., Loughead, T. M., & Munroe-Chandler, K. J. (2013). Examining the benefits of athlete leaders in sport. Journal of Sport Behavior, 36, 346-364.

Duguay, A. M., Loughead, T. M., & Munroe-Chandler, K. J. (2016). The development, implementation, and evaluation of an athlete leadership development program with female varsity athletes. The Sport Psychologist, 30, 154-166. doi: 10.1123/tsp.2015-0050

Facca, T. M., & Allen, S. J. (2011). Using cluster analysis to segment students based on self-report emotionally intelligent leadership behaviors. Journal of Leadership Education. 10, 72-96. doi: 10.12806/v 10/i2/rf4

Finley, R., & Finley, C. (1999). SurveyMonkey Inc [Computer software]. Palo Alto, CA.

Fransen, K., Haslam, S. A., Mallett, C. J., Steffens, N. K., Peters, K., & Boen, F. (2016). Leading from the centre: A comprehensive examination of the relationship between central playing positions and leadership in sport. PLoS ONE, 11, e0168150. doi: 10.1371/ journal.pone.0168150

Fransen, K., Vanbeselaere, N., De Cuyper, B., Vande Broek, G., & Boen. F. (2014). The myth of the team captain as principal leader: Extending the athlete leadership classification within sport teams. Journal of Sport Sciences, 32, 1389-1397. doi: 10.1080/02640414 .2014.891291

Glenn, S. D., & Horn, T. S. (1993). Psychological and personal predictors of leadership behavior in female soccer athletes. Journal of Applied Sport Psychology, 5, 17-34. doi: 10.1080/10413209308411302

Greer, T., & Dunlap, W. P. (1997). Analysis of variance with ipsative measures. Psychological Methods, 2. 200-207. doi: 10.1037//1082-989x.2.2.200

Hardy, L., Arthur, C. A., Jones, G., Shariff, A., Munnoch, K., Isaacs, I., & Allsopp, A. J. (2010). The relationship between transformational leadership behaviors, psychological, and training outcomes in elite military recruits. The Leadership Quarterly, 21, 20-32. doi: 10.1016/j.leaqua.2009.10.002

Janssen, J. (2015). Leadership academy. Retrieved from http://www.janssensportsleadership. com /programs/leadership-academy/

Judge, T. A., & Piccolo, R. F. (2004). Transfonnational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology', 89, 755-768. doi: 10.1037/0021-9010.89.5.755

Loughead, T. M., Hardy, J., & Eys, M. A. (2006). The nature of athlete leadership. Journal of Sport Behavior, 29, 142-158.

Loughead, T. M., Munroe-Chandler, K. J., Hoffmann, M. D., & Duguay, A. M. (2014). Athlete leadership in sport. In M. R. Beauchamp & M. A. Eys (Eds.), Group Dynamics in Exercise and Sport Psychology (pp. 110-127). New York, NY: Routledge.

McCauley, C. D., Kanaga, K., & Lafferty, K. (2010). Leader development systems. In E. Van Velsor, C. D. McCauley, & M. N. Ruderman (Eds.), The Center for Creative Leadership handbook of leadership development (pp. 29-62). San Francisco, CA: Jossey-Bass.

Moran, M. M., & Weiss, M. R. (2006). Peer leadership in sport: Links with friendship, peer acceptance, psychological characteristics, and athletic ability. Journal of Applied Sport Psychology', 18, 97-113. doi:10.1080/10413200600653501

Pain, M. A., Harwood, C., & Mullen, R. (2012). Improving the perfonnance environment of a soccer team during a competitive season: An exploratory action research study. The Sport Psychologist, 26, 390-411. doi: 10.1123/tsp.26.3.390

Paradis, K., & Loughead, T. (2012). Examining the mediating role of cohesion between athlete leadership and athlete satisfaction in youth. International Journal of Sport Psychology', 43. 117-136.

Pierce, J. R., & Aguinis, H. (2013). The too-much-of-a-good-thing effect in management. Journal of Management, 39, 313-338. doi: 10.1177/0149206311410060

Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly, 1, 107-142. doi: 10.1016/10489843(90)90009-7

Price, M. S., & Weiss, M. R. (2011). Peer leadership in sport: Relations among personal characteristics, leader behaviors, and team outcomes. Journal of Applied Sport Psychology, 23, 49-64. doi: 10.1080/10413200.2010.520300

Ravlin, E. C., & Meglino, B. M. (1987). Effect of values on perception and decision making: A study of alternative work values measures. Journal of Applied Psychology, 72, 666-673. doi: 10.1037/0021-9010.72.4.666

Smith, M. J., Arthur. C. A., Hardy, J., Callow, N., & Williams, D. (2013). Transformational leadership and task cohesion in sport: The mediating role of intrateam communication. Psychology' of Sport and Exercise, 14, 249-257. doi: 10.1016/j.psychsport2012.10.002

Sosik, J. J. (2006). Full range leadership: Model, research, extension and training. In C. Cooper & R. Burke (Eds.), Inspiring leadership (pp. 33-36). New York: Routledge.

Tropp, K. J., & Landers, D. M. (1979). Team interaction and the emergence of leadership and interpersonal attraction in field hockey. Journal of Sport Psychology, l, 228-240. doi: 10.1123/jsp. 1.3.228

Vidic, Z., & Burton, D. (2011). Developing effective leaders: Motivation correlates of leadership styles. Journal of Applied Sport Psychology', 23, 1533-1571. doi: 10.1080/1041320 0.3010.546827

Vincer, D. J. E., & Loughead, T. M. (2010). The relationship between athlete leadership behaviors and cohesion in team sports. The Sport Psychologist, 24, 448-467. doi: 10.1123/ tsp.24.4.448

Yukelson, D., Weinberg, R., Richardson, P., & Jackson, A. (1983). Interpersonal attraction and leadership with collegiate sport teams. Journal of Sport Behavior, 6, 28-36.

Yukl, G. (2012). Effective leadership behavior: What we know and what questions need more attention. Academy of Management Perspectives, 26, 66-85. doi:10.5465/ amp.2012.0088

Ashley M. Duguay, Todd M. Loughead, and Krista J. Munroe-Chandler

University of Windsor

Address correspondence to: Ashley M. Duguay, Department of Kinesiology, University Of Windsor, 401 Sunset Avenue, Windsor, Ontario, Canada, N9B 3P4. Email: duguay7@ uwindsor.ca

Table 1

Dimensions of Leadership Behavior and Definitions

Dimension                             Definition
Democratic behavior (1)               Encourage teammates to
                                      participate in decision-making

Autocratic behavior (1)               Stress personal authority by
                                      being independent in their
                                      decision making

Positive feedback (1)                 Praise a teammate's performance

Social support (1)                    Take interest in the welfare of
                                      teammates by building warm
                                      interpersonal relations with
                                      them

Training and instruction (1)          Help their teammates by
                                      instructing them in the skills,
                                      techniques, and tactics of the
                                      sport

                                      Pay special attention to each
Individual consideration (2,3)        teammate's needs for personal
                                      development by acting as a
                                      mentor, teacher, confidant, and/
                                      or counsellor

Inspirational motivation (2,3)        Inspire teammates by motivating
                                      them

Intellectual stimulation (2,3)        Challenge their teammate's way
                                      of thinking by en-couraging
                                      creativity

Idealized influence (2) /             Lead by example (for instance be
Appropriate role model (3)            a role model for teammates)

Management-by-exception active (2)    Monitor their teammates and take
                                      corrective action prior to or
                                      immediately after they deviate
                                      from expectations

Management-by-exception               Wait for mistakes to happen from
passive (2)                           teammates before taking any
                                      corrective measures

Fostering acceptance of group         Encourage their teammates to
goals and promoting teanwork (3)      work together toward the same
                                      team goals

High performance expectations (3)     Expect excellence, quality, and
                                      high performance from teammates

Note. (1) Dimension from the Lbb (Chelladurai & Saleh, 1980);
(2) Dimension trom the MLQ-5X (Bass & Avolio, 2004);
(3) Dimension from the DTLI (Callow et al., 2009).

Table 2

Dimensions of Leadership Behavior--General Importance Ratings for
Athlete Leaders

Dimension                                M      SD     95% Cl
Idealized influence (2) /                4.87   0.40   [4.80, 4.94]
Appropriate role model (3)
Foster acceptance of group
goals ... (3)                            4.76   0.47   [4.68, 4.84]
Inspirational motivation (2,3)           4.65   0.59   [4.55, 4.75]
High performance expectations (3)        4.55   0.57   [4.46, 4.65]
Democratic behavior (1)                  4.51   0.64   [4.40, 4.62]
Social support (1)                       4.44   0.73   [4.31, 4.56]
Positive feedback (1)                    4.15   0.80   [4.01, 4.29]
Individual consideration (2,3)           4.06   0.95   [3.90, 4.22]
Training and instruction (1)             3.93   1.05   [3.75, 4.11]
Intellectual stimulation (2,3)           3.91   0.90   [3.76, 4.07]
Management-by-exception active (2)       3.82   0.98   [3.65, 3.98]
Autocratic behavior (1)                  3.07   1.17   [2.87, 3.27]
Management-by-exception passive (2)      2.16   1.01   [1.99, 2.34]

Note. Ratings were made on a 5-point Likert scale ranging from 1
(strongly disagree) to 5 (strongly agree). (1) Dimensions from the
LSS (Chelladurai & Saleh, 1980); (2) Dimensions from the MLQ-5X
(Bass & Avolio, 2004); (3) Dimensions from the DTLI (Callow et al.,
2009).

Table 3
Dimensions of Leadership Behavior--Importance Ratings by Athlete
Leader Tenure

                                        Rookie       Emerging
                                        Athlete      Athlete
                                        Leader       Leader
Dimension                             M      SD     M      SD

Idealized influence (2) /             5.22   3.28   9.18   2.64
Appropriate role model (3)
Foster acceptance of group            7.95   2.86   9.73   1.38
goals ... (3)
Inspirational motivation (2,3)        7.87   3.06   9.82   1.27
High performance expectations (3)     5.79   2.72   9.38   1.92
Democratic behavior (1)               7.06   3.33   9.65   1.98
Social support (1)                    7.94   3.70   9.69   2.51
Positive feedback (1)                 7.39   3.09   9.78   1.72
Individual consideration (2,3)        4.61   3.64   9.53   2.65
Training and instruction (1)          5.21   3.72   9.63   2.18
Intellectual stimulation (2,3)        7.03   3.39   9.96   2.15
Management by exception               5.10   3.75   9.02   2.75
active (2)
Autocratic behavior (1)               6.67   3.89   9.55   2.81
Management by exception passive (2)   6.36   4.33   9.80   1.93

                                      Veteran Athlete
                                      Leader
Dimension                             M       SD
Idealized influence (2) /             15.37   4.65
Appropriate role model (3)
Foster acceptance of group            12.32   3.61
Goals ... (3)
Inspirational motivation (2,3)        12.30   3.21
High performance expectations (3)     14.53   3.52
Democratic behavior (1)               13.29   4.12
Social support (1)                    12.17   4.39
Positive feedback (1)                 12.83   3.61
Individual consideration (2,3)        15.85   4.59
Training and instruction (1)          15.16   4.55
Intellectual stimulation (2,3)        13.02   3.86
Management by exception               15.88   5.28
active (2)
Autocratic behavior (1)               13.41   5.06
Management by exception passive (2)   13.84   4.78

Note. Participants were given 30 points to distribute amongst the
three tenures. More points represent a higher level of importance.
(1) Dimensions from the LSS (Chelladurai & Saleh, 1980);
(2) dimensions from the MLQ-5X (Bass & Avolio, 2004);
(3) dimensions from the DTLI (Callow et al., 2009).

Table 4

Repeated-Measures ANOVAs--Importance Ratings by Athlete Leader Tenure

Dimension                                df     MS        F

Idealized influence (2) /                1.43   4669.42   180.75
Appropriate role model (1)
Foster acceptance of group
goals ... (3)                            1.18    976.43    49.69
Inspirational motivation (2,3)           1.25    934.82     55
High performance expectations (1)        1.42   3554.34   233.70
Democratic behavior (1)                  1.33   1754.98    72.98
Social support (1)                       1.51    756.86    30.47
Positive feedback (1)                    1.35   1356.13    71.56
Individual consideration (2,3)           1.48   5314.41   190.87
Training and instruction (1)             1.34   4612.09   157.09
Intellectual stimulation (2,3)           1.49   1434.70    68.92
Management by exception active (2)       1.31   5640.58   149.04
Autocratic behavior (1)                  1.42   2088.80    64.82
Management by exception passive (2)      1.26   2630.58    72.90

Dimension                                P       [[eta].sup.2].
                                                     sub.p]
Idealized influence (2) /                <.001        0.58
Appropriate role model (1)
Foster acceptance of group
goals ... (3)                            <.001        0.30
Inspirational motivation (2,3)           <.001        0.32
High performance expectations (1)        <.001        0.64
Democratic behavior (1)                  <.001        0.38
Social support (1)                       <.001        0.19
Positive feedback (1)                    <.001        0.37
Individual consideration (2,3)           <.001        0.61
Training and instruction (1)             <.001        0.56
Intellectual stimulation (2,3)           <.001        0.37
Management by exception active (2)       <.001        0.55
Autocratic behavior (1)                  <.001        0.34
Management by exception passive (2)      <.001        0.38

Note. (1) Dimensions from the LSS (Chelladurai & Isaleh, 1980);
(2) Dimensions from the MLQ-5X (Bass & Avolio, 2004); (3) Dimensions
from the DTLI (Callow et al., 2009). Degrees of freedom were corrected
for each behavior using Greenhouse-Geisser estimates of sphericity.