ASSESSING THE EFFECTIVENESS OF A TRANSFORMATIONAL COACHING
WORKSHOP FOR CHANGING YOUTH SPORT COACHES’ BEHAVIOURS
by
Sarah Victoria Clewes Lawrason
A thesis submitted to the Graduate Program in Kinesiology and Health Studies
in conformity with the requirements for
the degree of Master of Science
Queen’s University
Kingston, Ontario, Canada
(June, 2018)
Copyright Sarah Lawrason, 2018
ii
Abstract
There is a need for coach development programs that target coaches’ interpersonal
behaviours (Lefebvre et al., 2016) and that are informed by behaviour change techniques (Allan
et al., 2017). In addition, current and future coach development programs could benefit from
comprehensive evaluations to determine their impact on the coaches and athletes involved
(Evans et al., 2015). Recently, Turnnidge and Côté (2017a) developed the Transformational
Coaching workshop to fill these gaps in the literature. Informed by the full-range leadership
model (Bass & Riggio, 2006) and the Behaviour Change Wheel (Michie et al., 2011), the
workshop offers coach education on a range of interpersonal behaviours using a leadership
perspective. Considering that the Transformational Coaching workshop has yet to be evaluated,
the purpose of this study was to evaluate its impact by systematically observing coaches’
behaviours before and after their involvement in the workshop.
Participants included eight male head coaches of youth competitive soccer teams with an
average of 14.25 years of coaching experience. Systematic observation and coding using the
Coach Leadership Assessment System (CLAS; Turnnidge & Côté, 2016b) was employed pre-
and post-workshop participation to examine the duration of coaches’ leadership behaviours.
Paired samples t-tests, bootstrapped confidence intervals, and effect sizes indicated that
idealized influence (p = .067, d = .76), inspirational motivation (p = .087, d = .70), and
intellectual stimulation (p = .132, d = .60) behaviours had confidence intervals that did not cross
zero, thus approaching statistical significance. However, from a pragmatic perspective, the
workshop significantly influenced these behaviours considering the medium to large effect sizes
observed. Furthermore, following the workshop, coaches displayed less neutral behaviours (p =
.007, d = 1.34). Coaches also spent significantly more time, after the workshop, displaying
iii
leadership behaviours through instruction/feedback (p = 0.013, d = 1.17) and significantly less
time displaying leadership behaviours through organization (p = .001, d = 1.90). There were no
significant differences in other leadership behaviours, modifiers, or recipients. These findings
offer detailed descriptions of how coaches’ behaviours changed after participating in the
Transformational Coaching workshop. Practical implications for the workshop and future
research opportunities are discussed.
iv
Co-Authorship
The co-supervisors, Dr. Jean Côté and Dr. Luc Martin, are co-authors on this thesis.
Collaboration with Dr. Jennifer Turnnidge has also led to her co-authorship for the entirety of the
document. Drs. Côté, Martin, and Turnnidge all offered guidance on study design, data
collection, analysis, and provided feedback on the writing stages of the thesis. Sarah Lawrason
held the primary responsibility for study design, data collection, data analysis, coding, and
drafting and revising the written document.
v
Acknowledgements
First and foremost, I would like to thank my supervisors for guiding me through the
completion of my Master’s degree. Jean, I am so grateful for your expertise in coaching and
positive youth development. You encouraged me to read (and re-read) literature and asked
questions that fostered my ability to think about how my work fits in the “bigger picture”. Luc, I
am so thankful to have had you as my undergraduate thesis supervisor. Your continued support
throughout my Master’s degree equipped me with the skills to navigate the academic and
personal challenges in graduate school. Jen, you have been like a third co-supervisor (and great
friend) to me. Our weekly/daily chats about my project, TFL, or general life philosophies helped
me to conceptualize my ideas and generally made me feel more normal. Jean, Luc, and Jen – you
are all incredible, transformational leaders and I would be lucky to adopt even half of the
qualities you have in the future.
I would be remiss if I did not thank the people that make the PLAYS lab. Everyone gave
up their own time to contribute to my project in some way or another (e.g., helping with data
collection, providing feedback, assisting with coding). Even in less-than-ideal situations – like
when it was freezing cold and pouring rain during filming, or cabbing between practice locations
– you all had a positive attitude that made working fun. Our friendship outside of the lab was
also so valuable – please visit Kelowna so that we can have more coffee breaks, Bachelor nights,
and dancing. From the bottom of my heart, thank you.
Mom, Dad, Peter, Emily, Floyd, and Macey – during the most challenging times, your
encouragement and empathy was what helped me stay connected when Kingston felt far away.
Thomas, thanks for letting me ramble on about things without any context, and then pretending
to understand me to make me feel better. To my family – thank you for your love and care.
vi
Thanks to my proposal committee members constructive criticism that improved my
project (and statistics knowledge). Thank you to my defense committee members for taking time
out of your busy schedule.
Finally, I would like to thank everyone from the soccer club who was involved in this
project. Coaches, athletes, parents, directors, administrators, volunteers – thank you for your
time, dedication, and enthusiasm.
vii
Table of Contents
Abstract............................................................................................................................................ii
Co-Authorship Statement...............................................................................................................iv
Acknowledgements..........................................................................................................................v
List of Tables..................................................................................................................................xi
List of Figures................................................................................................................................xii
List of Abbreviations....................................................................................................................xiii
Chapter 1 Introduction.....................................................................................................................1
Chapter 2 Literature Review............................................................................................................5
2.1 Leadership Approaches to Coaching.............................................................................5
2.1.1 Mediational model of coaching behaviours....................................................6
2.1.2 Motivational model of the coach-athlete relationship....................................8
2.2 Observational Studies on Coaching Behaviours..........................................................11
2.3 Qualitative Studies on Coach Leadership....................................................................12
2.4 The Full-Range Leadership Model..............................................................................14
2.4.1 Toxic and laissez-faire..................................................................................15
2.4.2 Transactional.................................................................................................16
2.4.3 Transformational leadership (i.e., TFL)........................................................16
2.4.4 Evidence for follower outcomes...................................................................17
2.5 Teaching TFL Through Interventions..........................................................................19
2.6 Incorporating TFL into CDPs......................................................................................21
2.6.1 Behaviour Change Wheel.............................................................................22
2.7 Evaluating TFL-Informed CDPs..................................................................................23
viii
2.7.1 Previous methodological techniques.............................................................24
2.7.2 Development of new measures.....................................................................25
2.8 Transformational Coaching Workshop........................................................................25
2.9 Purpose.........................................................................................................................27
Chapter 3 Methods.........................................................................................................................28
3.1 Design..........................................................................................................................28
3.2 Participants...................................................................................................................28
3.3 Measures......................................................................................................................28
3.3.1 Coach Leadership Assessment System.........................................................28
3.4 Procedure.....................................................................................................................31
3.4.1 Pre-test procedure.........................................................................................32
3.4.2 Workshop......................................................................................................33
3.4.2.1 Workshop fidelity...........................................................................33
3.4.3 Post-test procedure........................................................................................34
3.5 Data Cleaning...............................................................................................................34
3.6 Data Analysis...............................................................................................................35
3.6.1 Pre-workshop analyses..................................................................................35
3.6.2 Pre-post-test analyses....................................................................................36
3.6.2.1 Descriptive statistics......................................................................36
3.6.2.2 Comparing pre-post means............................................................36
Chapter 4 Results...........................................................................................................................38
4.1 Workshop Fidelity.......................................................................................................38
4.2 Pre-Workshop Analyses...............................................................................................38
ix
4.2.1 Descriptive statistics.....................................................................................38
4.2.2 Test re-test reliability....................................................................................38
4.2.3 Differences between pre-workshop behaviours............................................38
4.3 Pre-Post Workshop Analyses.......................................................................................39
4.3.1 Descriptive statistics.....................................................................................39
4.3.2 TFL behaviours.............................................................................................39
4.3.3 Non-TFL behaviours.....................................................................................40
4.3.4 Content modifiers..........................................................................................42
4.3.5 Recipient modifiers.......................................................................................44
Chapter 5 Discussion.....................................................................................................................45
5.1 Transformational Leadership.......................................................................................45
5.1.1 Idealized influence........................................................................................47
5.1.2 Inspirational motivation................................................................................48
5.1.3 Intellectual stimulation..................................................................................49
5.1.4 Individualized consideration.........................................................................50
5.2 Transactional, Neutral, Laissez-Faire, and Toxic Leadership.....................................51
5.2.1 Transactional, laissez-faire, and toxic behaviours........................................51
5.2.2 Neutral behaviours........................................................................................53
5.3 Content Modifiers........................................................................................................54
5.3.1 Instruction/feedback and organization..........................................................54
5.3.2 General communication................................................................................55
5.4 Recipient Modifiers.....................................................................................................56
5.5 Theoretical and Practical Implications.........................................................................57
x
5.6 Limitations and Future Directions...............................................................................59
Chapter 6 Conclusion.....................................................................................................................62
References......................................................................................................................................65
Appendix A: General Coding Guidelines......................................................................................84
Appendix B: Research Ethics Board Letter of Approval...............................................................93
Appendix C: Coach Letter of Information and Consent Form......................................................94
Appendix D: Coach Information Sheet..........................................................................................98
Appendix E: Implementation Checklist and Workshop Fidelity.................................................100
Appendix F: Mean, Standard Deviation, and Pearson Correlation (r) for Leadership Behaviours
for Pre 1 and Pre 2 Observations (n = 5) ....................................................................................103
Appendix G: Mean, Standard Deviation, and Pearson Correlation (r) for Content Modifiers for
Pre 1 and Pre 2 Workshop Observations (n = 5) ........................................................................104
Appendix H: Mean, Standard Deviation, and Pearson Correlation (r) for Recipient Modifiers for
Pre 1 and Pre 2 Workshop Observations (n = 5) ........................................................................105
Appendix I: Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for
Leadership Behaviours for Pre 1 and Pre 2 Workshop Observations (n = 5) .............................106
Appendix J: Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for
Content and Recipient Modifiers for Pre 1 and Pre 2 Workshop Observations (n = 5) .............107
xi
List of Tables
Table 1. The Coach Leadership Assessment System: Leadership Dimensions.............................30
Table 2. Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for 4 I’s for
Pre- and Post-Workshop Observations.........................................................................................40
Table 3. Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Full-
Range Leadership Behaviours for Pre- and Post-Workshop Observations..................................41
Table 4. Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Content
Modifiers for Pre- and Post-Workshop Observations...................................................................43
Table 5. Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for
Recipient Modifiers for Pre- and Post-Workshop Observations...................................................44
Table F1. Mean, Standard Deviation, and Pearson Correlation (r) for Leadership Behaviours for
Pre 1 and Pre 2 Observations (n = 5).........................................................................................103
Table G1. Mean, Standard Deviation, and Pearson Correlation (r) for Content Modifiers for Pre
1 and Pre 2 Workshop Observations (n = 5)...............................................................................104
Table H1. Mean, Standard Deviation, and Pearson Correlation (r) for Recipient Modifiers for
Pre 1 and Pre 2 Workshop Observations (n = 5)........................................................................105
Table I1. Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for
Leadership Behaviours for Pre 1 and Pre 2 Workshop Observations (n = 5)............................106
Table J1. Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for
Content and Recipient Modifiers for Pre 1 and Pre 2 Workshop Observations (n = 5).............107
xii
List of Figures
Figure 1. Schematic representation of the full-range leadership model adapted for the sport
context (Bass & Riggio, 2006; Turnnidge & Côté, 2016b)...........................................................15
Figure 2. The Behaviour Change Wheel framework (Michie et al., 2011)...................................23
Figure 3. Procedure for assessing coaches’ leadership behaviours pre- and post-workshop........32
xiii
List of Abbreviations
4 I’s: Idealized Influence, Inspirational Motivation, Intellectual Stimulation, and
Individualized Consideration
AGT: Acheivement Goal Theory
CBAS: Coach Behaviour Assessment System
CDP: Coach Development Program
CET: Coach Effectiveness Training
CLAS: Coach Leadership Assessment System
COM-B: Capability-Opportunity-Motivation-Behaviour
DTLI: Differentiated Transformational Leadership Inventory
DTLI-YS: Differentiated Transformational Leadership Inventory – Youth Sport
GC: General Communication
IC: Individualized Consideration
II: Idealized Influence
IM: Inspirational Motivation
IND: Individual
IS: Intellectual Stimulation
LF: Laissez-Faire
MAC: Mastery Approach to Coaching
MLQ: Multifactor Leadership Questionnaire
MMCOS: Multidimensional Motivational Climate Observation System
NEU: Neutral
ORG: Organization
xiv
SDT: Self-Determination Theory
TC: Transactional Leadership
TFL: Transformational Leadership
TOX: Toxic Leadership
TTQ: Transformational Teaching Questionnaire
1
Chapter 1
Introduction
Coaches play a critical role in fostering positive youth sport experiences (Petitpas,
Cornelius, Van Raalte, & Jones, 2005). Previous research suggests that effective coaching
involves professional, interpersonal, and intrapersonal knowledge and behaviours that result in
athletes’ positive developmental outcomes (Côté & Gilbert, 2009). One way through which
coaches can learn the three types of knowledge and behaviours required for effective coaching is
through coach development programs (CDPs). CDPs can be defined as “an encompassing term
to describe learning activities applied systematically through education, social interaction, and/or
personal reflection with the goal of changing … coach behaviours” (Evans, McGuckin,
Gainforth, Bruner, & Côté, 2015, p. 871). Accordingly, it is important to understand the features
of CDPs that can facilitate changes in coach behaviours, and ultimately help foster positive sport
experiences for athletes.
Although an extensive body of literature exists pertaining to the development and
understanding of CDPs, recent review articles highlight several shortcomings warranting future
attention. First, few CDPs focus on coaches’ interpersonal behaviours (Lefebvre, Evans,
Turnnidge, Gainforth, & Côté, 2016). Second, although the overarching objectives of most
CDPs is to change coach behaviours, few CDPs incorporate the use of behaviour change theories
(Allan, Vierimaa, Gainforth, & Côté, 2017). Third, CDPs are often not evaluated properly and
rarely in a systematic and contextualized manner (Evans et al., 2015). Consequently, there is a
need to be more deliberate about designing, implementing, and evaluating CDPs that will help
coaches develop interpersonal behaviours to increase the likelihood that athletes have positive
experiences in sport.
2
The full-range leadership model (Bass & Riggio, 2006) is a valuable approach for
examining how leaders, such as coaches, can use interpersonal skills to influence followers’ (i.e.,
athletes’) outcomes. The model includes a spectrum of passive to active leadership behaviours
that lead to adaptive or maladaptive follower outcomes (Bass & Riggio, 2006). The most
effective style is Transformational Leadership (TFL; Bass, 1998), where leaders exhibit
behaviours that motivate followers to seek challenges and reach their full potential while
providing a shared vision for the group (Bass & Riggio, 2006). TFL has four dimensions: (a)
idealized influence (i.e., acting as a positive role model), (b) inspirational motivation (i.e.,
inspiring followers through shared, challenging goals), (c) intellectual stimulation (i.e., involving
followers in the learning process, and (d) individualized consideration (i.e., understanding each
follower’s unique contributions and needs; 4 I’s; Bass & Riggio, 2006). Coaches who are
perceived by athletes as demonstrating the 4 I’s tend to have athletes who experience greater
team cohesion, motivation, performance, and personal growth (e.g., Callow, Smith, Hardy,
Arthur, & Hardy, 2009; Charbonneau, Barling, & Kelloway, 2001; Smith, Arthur, Hardy,
Callow, & Williams, 2013; Vella, Oades, & Crowe, 2013).
Although coaches who are perceived as being more transformational can facilitate
positive athlete outcomes, few attempts to teach transformational behaviours to youth sport
coaches have been undertaken. Whereas there is evidence supporting the effectiveness of TFL
training on follower outcomes in a variety of contexts (e.g., military; Hardy et al., 2010;
organizations; Barling, Weber, & Kelloway, 1996; education; Beauchamp, Barling, & Morton,
2011; healthcare; Kelloway, Barling, & Helleur, 2000), only one study has looked at the impact
of a TFL-based intervention in sport. Vella and colleagues (2013) found that athletes of coaches
who had received a TFL intervention experienced greater positive developmental outcomes.
3
However, the TFL intervention was an adaptation of a CDP that was not specifically designed to
teach TFL behaviours. The intervention was also not explicitly informed by behaviour change
techniques and thus, did not identify the conditions that specify why, when, and how coaches
should portray transformational behaviours (Allan et al., 2017). Consequently, there is a need
for evidence-informed TFL coaching workshops that are specifically designed to teach TFL
behaviours, and that are guided by behaviour change theories (Allan et al., 2017).
Another limitation of previous interpersonal CDPs is that their effectiveness has not been
adequately evaluated (Evans et al., 2015). Previous interpersonal CDPs, including TFL
interventions (e.g., Vella et al., 2013) have typically employed quantitative questionnaires to
assess changes in perceptions of behaviours pre-and post-workshop. However, while
questionnaires that assess TFL behaviours provide valuable insight into athletes’ and coaches’
perceptions of TFL behaviours, questions remain pertaining to how these behaviours are actually
manifested in the sport context (Turnnidge & Côté, 2016a). It is thus important to complement
the questionnaire findings with observational methodologies to assess changes in the actual
behavioural profiles of coaches following their participation in a TFL-based CDP. Observational
methods can also contextualize coaches’ behaviours to understand the content and recipient of
behaviours, details that often differentiate between effective and less effective coaching (e.g.,
Erickson et al., 2011). Assessing the effectiveness of CDPs by measuring actual coaches’
behaviours in specific contexts will help identify key barriers for replication and translation of a
TFL-informed coaching workshop.
In response to calls for evidence-informed interpersonal coaching workshops, Turnnidge,
Barling, et al. (2016) and Turnnidge and Côté (2017a) developed a youth sport TFL workshop
informed by previous TFL interventions (Barling et al., 1996; Beauchamp et al., 2011) and the
4
Behaviour Change Wheel framework (Michie, van Stralen, & West, 2011). Within the
Behaviour Change Wheel, the Capability, Opportunity, Motivation-Behaviour (COM-B) model
stresses that there are three elements required to change a person’s behaviour: (a) capability, an
individual’s physical or psychological ability to execute the preferred behaviour; (b) opportunity,
the physical or social environment that can enable an individual’s behaviour; and (c) motivation,
an individual’s internal automatic or reflective processes that direct these behaviours (Michie et
al., 2011). The Transformational Coaching workshop uses the COM-B model to inform the use
of specific behaviour change techniques, such as action planning and goal setting. However, the
workshop has not been assessed based on its ability to effectively change coaches’ behaviours.
Considering the current dearth of studies involving the evaluation of CDPs, and that the
proposed workshop is informed by TFL theory, an assessment of the Transformational Coaching
workshop through systematic observation of TFL behaviours is warranted.
The purpose of this study is to examine the effectiveness of a TFL workshop with youth
soccer coaches using a novel observational instrument. Pre- and post-workshop assessments will
measure changes in the duration of coaches’ observed leadership behaviours and the content and
recipient of these behaviours. This study has the potential to improve future TFL workshops and
to provide coaches with interpersonal skills training that may ultimately improve the quality of
youth’s sport experiences.
5
Chapter 2
Literature Review
Provided that the coach-athlete relationship is a central tenet in the definition of coaching
leadership (Vella, Oades, & Crowe, 2010), the purpose of this literature review is to provide an
overview of the leadership behaviour research on coaching and TFL. Therefore, this section will
(a) review relevant frameworks of leadership approaches to coaching that have informed
observational tools and CDPs, (b) examine observational and qualitative studies that have
contextualized leadership behaviours, (c) discuss how the full-range leadership model and TFL
theory may be useful for informing interpersonal CDPs, and (d) review methodological
approaches for evaluating interpersonal CDPs and TFL interventions.
Leadership Approaches to Coaching
It is widely recognized that coaches play an important role in fostering quality sport
experiences for youth athletes (e.g., Côté, Strachan, & Fraser-Thomas, 2008; Fraser-Thomas,
Côté, & Deakin, 2005; Petitpas et al., 2005). Indeed, effective coaches consistently apply three
types of knowledge and behaviours: (a) professional (i.e., sport-specific physical, technical, and
tactical skills), (b) interpersonal (i.e., managing relationships), and (c) intrapersonal (i.e.,
reflexivity; Côté & Gilbert, 2009). When coaches perform these behaviours appropriately in
different contexts, they are more likely to facilitate important athlete outcomes (i.e., competence,
confidence, connection, character; Côté & Gilbert, 2009). Given the predominant focus on the
training of coach’s professional behaviours, researchers have advocated that there is a need to
enhance our understanding of coaches’ interpersonal behaviours and its potential influence on
athlete development (Côté & Gilbert, 2009; Evans et al., 2015).
6
One way through which researchers have studied coaches’ interpersonal behaviours is by
adopting a leadership approach. Although there has been debate about the nature of coach
leadership, Vella and colleagues (2010) defined coach leadership as a “process of interpersonal
influence that is dependent upon the relationship between coach and athlete, and is used to
facilitate the athlete outcomes of competence, confidence, connection, and character” (p. 431).
Not only do Vella and colleagues (2010) situate the definition of coach leadership within Côté
and Gilbert’s (2009) coaching effectiveness definition, they also suggest that leadership in
coaching is most related to coaches’ interpersonal abilities. Two prominent coaching
frameworks that employed a leadership approach (as per Vella and colleagues’ [2010] definition)
to conceptualizing coaches’ interpersonal behaviours include: (a) the mediational model of
coaching behaviours (Smoll & Smith, 1989; Smoll, Smith, Curtis, & Hunt, 1978), and (b) the
motivational model of the coach-athlete relationship (Mageau & Vallerand, 2003). These
frameworks, along with the related theory-informed observational tools and CDPs, will be
reviewed in the following section. In addition, influential observational and qualitative studies
on coaching behaviours that add contextual detail on coaching leadership will be reviewed.
Mediational model of coaching behaviours. The mediational model is a
comprehensive framework that considers how situational factors, athletes’ cognitive processes,
and individual difference variables (e.g., age) mediate the relationships between leader
behaviours and athlete outcomes (Smoll & Smith, 1989; Smoll et al., 1978). To assess the
validity of the model, coaches’ behaviours were evaluated using observational techniques (i.e.,
the Coach Behaviour Assessment System [CBAS]; Smith, Smoll, & Hunt, 1977). The CBAS
provides 12 behavioural categories, divided into eight reactive behaviours (i.e., coach responses
immediately preceding athlete behaviours, such as reinforcement) and four spontaneous
7
behaviours (i.e., coach behaviours that are not a response to a discernible preceding event, such
as general technical instruction; Smith et al., 1977). Combined with questionnaires and
qualitative interviews, the results using the CBAS generally demonstrate that coaches’ use of
instructive (e.g., general technical instruction) and supportive (e.g., reinforcement) behaviours
predict athletes’ high level of self-esteem (e.g., Smoll et al., 1978) and enjoyment in sport (e.g.,
Smith, Zane, Smoll, & Coppel, 1983). Conversely, coaches’ use of punitive behaviours (e.g.,
punishment) have been associated with negative athlete outcomes, such as less affinity for the
coach (Smoll et al., 1978).
Based on the mediational model, Smith, Smoll, and Curtis (1979) developed the Coach
Effectiveness Training (CET) program, which evolved into the Mastery Approach to Coaching
(MAC; Smith, Smoll, & Cumming, 2007). The MAC is a 2.5-hour workshop targeting
cognitive-behavioural strategies for effective coaching (Smith et al., 2007). The intention of the
MAC is to teach coaches effective behaviours to create a mastery-oriented (i.e., providing high
levels of supportive and instructive behaviours and focusing on athlete effort and improvement)
and positive (i.e., using positive reinforcement and avoiding mistake-contingent punishment)
motivational climate (Smoll, Smith, & Cumming, 2007). Several studies using questionnaires
have linked coaches’ participation in MAC and CET training to improvements in athletes’
outcomes (i.e., higher enjoyment; Cumming, Smoll, Smith, & Grossbard, 2007; self-esteem;
Smoll et al., 2007; return intentions; Barnett, Smoll & Smith, 1992; and reduced performance
anxiety; Smith et al., 2007) in team sports. Furthermore, the MAC and CET training is effective
in changing athletes’ perceptions of coaches’ behaviours to be more supportive (e.g., using
reinforcement) and instructive and less punitive (Smith et al., 1979; Smoll, Smith, Barnett, &
Everett, 1993). It is also noteworthy that Smith and colleagues’ (1979) assessed coaches’
8
observable behaviours using the CBAS, but found that the CET program only improved coaches’
use of reinforcement behaviours.
Although the work by Smith, Smoll, and colleagues has provided vast evidence-informed
knowledge on coach behaviours, the mediational model and MAC predominantly focus on
coaches’ instructional actions (e.g., using positive feedback more than negative punishment).
While instruction is certainly critical, the mediational model and the MAC offer education on the
quantity and content of instructional behaviours rather than on the quality and delivery of the
instruction (e.g., the leadership tone). Likewise, the MAC does not identify how coaches should
deliver athlete-directed communication. As an example, the MAC does not include information
pertaining to the recipient of coaches’ behaviours, the context in which coaches’ behaviours
should be used, or the tone of coaches’ behaviours. It may also be useful to build upon Smoll
and colleagues’ (1979) work by continuing to use the CBAS to evaluate coaches’ observable
behaviours. Furthermore, there have been less favourable results from the MAC in the realm of
individual sports, highlighting the need to gain further evidence in a variety of youth sport
contexts (Coatsworth & Conroy, 2006; Erickson & Gilbert, 2013). Thus, the MAC may benefit
from more detailed evaluations of coaches’ interpersonal behaviours to develop a holistic set of
effective coach behaviours.
Motivational model of the coach-athlete relationship. One model that has focused on
key components of Vella and colleagues’ (2010) coach leadership definition (e.g., coaches’
interpersonal ability to form coach-athlete relationships) is the motivational model of the coach-
athlete relationship (Mageau & Vallerand, 2003). Grounded in self-determination theory (SDT),
the motivational model of the coach-athlete relationship suggests that three factors – (a) coaches’
personal orientation, (b) the coaching context, and (c) coaches’ perceptions of athletes’
9
behaviour and motivation – determine the extent of coaches’ autonomy-supportive behaviours.
Autonomy-supportive behaviours involve acknowledging athletes’ feelings and avoiding
controlling behaviours to provide a climate where athletes perceive high levels of autonomy (i.e.,
choice and control; Mageau et al., 2009). The model suggests that coaches who are autonomy-
supportive have a positive effect on their athletes’ perceptions of competence, autonomy, and
relatedness, which, in turn, determines athletes’ level of motivation (Mageau & Vallerand,
2003).
To understand the legitimacy of this relationship, coaches’ behaviours were assessed
using the Multidimensional Motivational Climate Observation System (MMCOS; Smith et al.,
2015). Using a global potency scale, the MMCOS draws upon the work of the motivational
model along with prominent motivational theories (e.g., Achievement Goal Theory [AGT;
Nicholls, 1989], SDT; Deci & Ryan, 1985) to provide seven environmental dimensions (e.g.,
autonomy-supportive) that assess coaches’ empowering (i.e., autonomy-supportive and task-
involving) and disempowering (i.e., controlling and ego-involving) behaviours. When combined
with athlete questionnaires, the results demonstrated that coaches’ autonomy-supportive, task-
involving, and instructional behaviours predicted athlete autonomy, relatedness, and competence,
leading to autonomous motivation (Smith et al., 2015; Smith et al., 2016). In contrast, coaches’
controlling and relatedness-thwarting behaviours were associated with athletes’ low levels of
perceived competence and amotivation (Smith et al., 2015; Smith et al., 2016). Although coders
can identify empowering and disempowering behaviours via the MMCOS, this system relies on
the use of a global rating scale to evaluate the potency of coaches’ behaviours (Smith et al.,
2015). Importantly, while potency measures through the MMCOS can inform researchers of the
10
impact of coaches’ behaviours, it does not identify detailed accounts of coaches’ behaviours and
the contexts in which they are performed (e.g., to whom or how behaviours are conveyed).
Notably, Duda (2013) integrated the aforementioned findings into a 6-hour, Empowering
Coaching workshop. This CDP teaches coaches how to create empowering coaching
environments—task-oriented, autonomy supportive, and socially supportive—to enhance
athletes’ psychological need satisfaction and optimal well-being (Appleton & Duda, 2016).
While there is extensive detail on the protocol of evaluating the Empowering CoachingTM
workshop (Duda et al., 2013), there is little information on the effectiveness of Empowering
CoachingTM. Preliminary evidence from conference proceedings using questionnaires and the
MMCOS suggests that the Empowering CoachingTM workshop improved coaches’ motivational
climate, leading to positive athlete outcomes (Smith, Tzioumakis, et al., 2015; Quested et al.,
2015). Qualitative interviews were also conducted with coaches who participated in the
workshop, and the results found that coaches’ competence to create an empowering motivational
climate increased after the workshop (Larsen et al., 2015).
Though the Empowering CoachingTM workshop has been influential in making athletes’
well-being the most important outcome of coaches’ relational behaviours, there are a range of
nuanced leader behaviours that are not covered in the framework and workshop (e.g., elements
of non-leadership behaviours, such as laissez-faire). Furthermore, the Empowering CoachingTM
workshop could benefit from a more comprehensive, objective evaluation. Additional empirical
evidence using the MMCOS is required to understand if and how actual coach behaviours have
changed after participating in the workshop (i.e., ecological validity). Thus, the MMCOS and
Empowering CoachingTM workshop could benefit from observing coaches’ behaviours in a
11
detailed and contextualized manner and incorporating less favourable leadership behaviours to
conceptualize a wider range of coaching leadership behaviours.
Observational Studies on Coaching Behaviours
As identified in the previous section, coaching workshops could gain from employing
observational techniques to assess changes in coaches behaviours after a training intervention.
Additionally, while the aforementioned observation systems (i.e., the CBAS and MMCOS)
identify what coaching behaviours look like, they lack contextual detail on how coaching
behaviours can convey information to athletes. Nevertheless, other observational methods have
been used to examine the context of coaches’ behaviours with respect to the general content,
tone, and/or the recipient of effective and less effective coach behaviours.
With regards to the content of coaches’ behaviours, earlier observational studies found
that encouragement and positive feedback in addition to high levels of instruction may
differentiate between effective and less effective coaches (e.g., Bloom, Crumpton, & Anderson,
1999; Claxton, 1988; Lacy & Goldston, 1990). In addition to instructing and providing feedback
to athletes about sport-related tasks, effective coaches also spend more time communicating with
athletes about non-sport specific information, such as their hobbies, families, or school activities
(e.g., Erickson & Côté, 2016; Turnnidge, Côté, Hollenstein, & Deakin, 2014). Conversely, less
effective coaches spend more time structuring the mechanics of a practice and organizing
athletes (Erickson, Côté, Hollenstein, & Deakin, 2011). Although this work tells researchers
what effective coaches are doing, it is important to understand how coaches can direct their
behaviours to be effective.
One way to conceptualize how coaches display effective coaching behaviours is by
acknowledging the recipient of the behaviours and other contextual factors (e.g., emotional tone)
12
found in more recent observation instruments (e.g., Assessment of Coach Emotions; Allan,
Turnnidge, Vierimaa, Davis, & Côté, 2016; Assessment of Coaching Tone; Erickson & Côté,
2015; the Coach Analysis Intervention System; Cushion, Harvey, Muir, & Nelson, 2012). For
example, Erickson and colleagues (2011) found that effective and less effective coaches did not
differentiate with regards to the content of their behaviours (e.g., instruction/feedback), but that
targeting individuals was the salient element in regulating the efficacy of the behaviours.
Similarly, other observational studies have found that individualized coaching behaviours are
linked to effective coaching and positive sport environments for athletes (Erickson & Côté, 2016;
Turnnidge et al., 2014). In addition, coaches who elicit input from athletes in a calm manner are
more likely to promote athlete learning and less likely to have athletes that exhibit anti-social
behaviours (Allan & Côté, 2016; Partington & Cushion, 2013). Consequently, directing positive
instructive/feedback behaviours towards individuals, including athletes in the coaching process,
and showing interest in athletes’ lives beyond sport may determine coaches’ effectiveness. Thus,
if coaching workshops are assessed by observing coaches’ behaviours in a contextualized
manner, it may be possible to understand how coaches become more effective, in ways that are
not possible with questionnaires and qualitative interviews.
Qualitative Studies on Coach Leadership
Complementing the observational coaching research, there has been valuable qualitative
studies on a range of specific, effective coaching behaviours. Uniquely, Partington and Cushion
(2013) bridged observational and qualitative techniques to understand why coaches were
performing certain instructional (e.g., due to available space) and interpersonal (e.g., to make
athletes feel good) behaviours. Using qualitative methods, Becker (2009) highlighted the
importance of consistent coach behaviours and expectations, while going above and beyond the
13
required coach role. Keegan, Harwood, Spray, and Lavallee (2014) uncovered similar qualities,
as elite athletes were found to prefer coaches that were trustworthy, who incorporated athletes’
ideas, and explained their decision-making process. Meanwhile, other qualitative studies
emphasize the value of coaches who support and accommodate individual athlete needs (Bloom,
Durand-Bush, & Salmela, 1997; Côté & Sedgwick, 2003). Fittingly, Becker (2012) summarized
the qualitative coaching literature to establish the qualities of certain coaching behaviours that
make them more effective than others. Consistent with the aforementioned observational
research (e.g., Erickson et al., 2011), coaching behaviours are more likely to be effective when
they involve qualities that are positive, supportive, individualized, fair, appropriate, clear, and
consistent (Becker, 2012).
Similarly, Vallée and Bloom (2005) used qualitative interviews to explore the leadership
characteristics of five expert coaches working with university athletes. The researchers found
that three components enabled coaches’ successful leadership skills: (a) personal attributes that
allowed them to display context-appropriate behaviours, (b) the ability to foster athletes’
individual growth, and (c) organizational skills that prepared coaches for practices and games.
These elements were reinforced by athletes’ who bought into the coaches’ vision (i.e., goals,
philosophies) for the program. Vallée and Bloom (2005) provided a unique link between their
findings and the salient constructs of TFL theory, suggesting that TFL could be a useful
leadership approach to understanding coaches’ interpersonal behaviours. Chelladurai (2007)
echoed this sentiment by suggesting that TFL may be a useful framework for developing a
comprehensive set of coaching leadership behaviours. Therefore, these findings reinforce that
TFL may be an appropriate lens for conceptualizing coaches’ effective interpersonal behaviours.
14
It is important to highlight that some qualitative studies also suggest that coaches display
a wide range of behaviours that include less effective outcomes, and this work largely draws
from motivational theories (e.g., SDT; Ryan & Deci, 1985). For example, athletes who perceive
coaches to criticize without praise tend to experience lower motivation and thwarted autonomy,
and these athletes acknowledge the necessity to consider the context and tone of the coach
(Keegan et al., 2014). Furthermore, using a case-study approach, Krane, Greenleaf, and Snow
(1997) linked the coach’s ego-involved behaviours (i.e., those that make an athlete’s self-worth
dependent upon performance) with the gymnast’s depression and disordered eating. Both
examples illustrate the deleterious influence of a coach-centered sport environment, rather than
an athlete-centered or person-centered sport environment, on athletes’ well-being. Thus, the
wide range of leadership behaviours have the potential to be examined along a spectrum between
ineffective and effective approaches. The full-range leadership model (which includes TFL),
then, may be a useful framework for understanding a continuum of coaching behaviours.
The Full-Range Leadership Model
The full-range leadership model conceptualizes a spectrum of leadership behaviours
along two axes: ineffective to effective, and active to passive (Bass & Riggio, 2006). The three
leadership styles included in the original model are laissez-faire, transactional, and TFL (Bass &
Riggio, 2006). Toxic leadership was added to the full-range leadership model when it was found
to be relevant for certain contexts (i.e., military; Pelletier, 2010; sport; Turnnidge & Côté,
2016a). See Figure 1 for a schematic representation of the model in the sport context.
15
Figure 1. Schematic representation of the full-range leadership model adapted for the sport
context (Bass & Riggio, 2006; Turnnidge & Côté, 2016b).
Toxic and laissez-faire. The ineffective forms of leadership are toxic and laissez-faire,
the former being active and the latter being passive (Bass & Riggio, 2006). While not included
in the original full-range leadership model, toxic leadership was added when Turnnidge and Côté
(2016b) drew upon abusive leadership literature to adapt the model to the sport context. Toxic
leadership involves abusive behaviours wherein coaches exhibit negative attitudes or frustrations
towards athletes (Turnnidge & Côté, 2016a). Originally identified in military settings (Pelletier,
2010), toxic leaders contribute to a host of negative follower outcomes including lack of trust,
reduced effectiveness, and low retention (Haslam, Reicher, & Platow, 2010). Contrastingly,
laissez-faire behaviours are characterized by leaders’ absence and lack of involvement, which
results in followers’ role conflict and ambiguity (Burns, 1978; Skogstad, Einarsen, Torsheim,
Aasland, & Hetland, 2007) and lower well-being (Barling & Frone, 2017). Therefore, both
engaged (toxic) and disengaged (laissez-faire) behaviours can contribute to negative athlete
outcomes.
Passive Active
Ineffective
Laissez-
Faire
(Disinterest)
Toxic
(Anger and
Hostility)
Transformational
(4 I’s)
Effective
Neutral
Transactional
(Rewards and
Punishments)
16
Transactional. Transactional leadership involves behaviours that offer rewards and
punishments for meeting or not meeting objectives to develop motivation and performance in
followers (Eagly, Johannesen-Schmidt, & Van Engen, 2003). Leaders provide rewards or
punishments to followers after observing followers’ positive performance or mistakes, and can
be done in an active or passive manner (Bass & Riggio, 2006). Specifically, three components
of transactional leadership include: contingent reward (providing rewards for successful follower
performance), active management by exception (attending to followers’ mistakes to meet
standards), and passive management by exception (waiting until followers’ mistakes become
severe problems before attending to them; Bass & Riggio, 2006). In contrast to the ineffective
behaviours within the toxic and laissez-faire dimensions, transactional behaviours can both
positively or negatively impact followers’ performance and satisfaction (Howell & Avolio,
1993). As an example, transactional leaders may enhance individual performance (Hamstra, Van
Yperen, Wisse, & Sassenberg, 2014; Lord, Brown, & Freiberg, 1999), but as a result, emphasize
individual as opposed to collective goals, and reduce team cooperation and performance (Kahai,
Sosik, & Avolio, 2003). Thus, transactional leadership behaviours can lead to both effective and
ineffective outcomes.
Transformational leadership (TFL). Transformational leaders exhibit behaviours that
individually motivate followers to seek challenges and reach their full potential while providing
a shared vision for the group (Bass & Riggio, 2006). The term followers encapsulates all
individuals whom the transformational leader is attempting to develop in a given context (e.g.,
patients, recruits, students, employees, athletes; Turnnidge & Côté, 2016a). Transformational
leaders typically achieve higher performances and have more committed and satisfied followers
17
due to establishing high expectations and aligning objectives and goals between the individuals,
leader, group, and the larger organization (Bass & Riggio, 2006).
One conceptualization of TFL suggests that it is comprised of four dimensions: (a)
idealized influence, (b) inspirational motivation, (c) intellectual stimulation, and (d)
individualized consideration (Bass & Riggio, 2006). Idealized influence refers to leaders that
gain the followers’ admiration, trust, and respect by acting as a positive role model. Inspirational
motivation is demonstrated when leaders inspire followers’ through providing meaning to a
shared goal and holding high expectations. Intellectual stimulation proposes that
transformational leaders encourage followers to think critically and creatively in order to address
problems and find solutions. Finally, individualized consideration involves behaviours that
recognize individual follower needs and differences for achievement and growth.
Evidence for follower outcomes. TFL may be an appropriate lens for creating an
interpersonal-focused CDP as it provides tested determinants (i.e., the 4 I’s) that lead to positive
follower outcomes in many environments. For instance, in military settings (US Navy and
Marine Corps midshipmen), TFL behaviours significantly predicted a range of follower
outcomes such as extra effort toward the task, satisfaction, and perceptions of leader
effectiveness (Atwater & Yammarino, 1989). These results were further supported by Bass,
Avolio, Jung, and Berson (2003) who noted that TFL behaviours of sergeants and platoon
leaders correlated with unit cohesion and performance in challenging conditions. TFL also has
implications in the workplace. For example, Corrigan, Diwan, Campion, and Rashid (2002)
found that in mental health settings (e.g., hospitals), team leaders’ transformational behaviours
were associated with a positive organizational culture and low staff burnout. Similarly, in the
healthcare system, supervisors’ transformational behaviours positively influenced workers’
18
psychological well-being (Arnold, Turner, Barling, Kelloway, & McKee, 2007). Together, these
findings support the augmentation hypothesis, which states that transactional behaviours provide
the foundation for effective leadership, but TFL builds upon this foundation to enhance optimal
follower development (Bass, 1998). Therefore, leaders who exhibit transformational behaviours
are effective in producing comprehensive follower outcomes including cohesion, performance,
and well-being in a variety of settings and contexts.
In addition to the support generated with adult populations, TFL has also shown to be
effective for youth followers. In school settings, the ability of principals to exert TFL behaviours
can influence students’ organizational commitment and citizenship behaviour, which in turn
improves students’ academic performance (Koh, Steers, & Terborg, 1995). These findings align
with Côté and Gilbert’s (2009) effective coaching definition, considering that certain outcomes
(e.g., citizen behaviour) are associated with demonstrating prosocial behaviour (i.e., developing
character). In a more recent study, Beauchamp and colleagues (2010) found that physical
education (PE) teachers’ use of TFL behaviours fostered students’ autonomous motivation and
positive affect through both individual interactions and the development of a motivational
climate. Wilson and colleagues (2012) added to this literature as they identified that students’
psychological needs satisfaction mediated the relationships between students’ transformational
teaching perceptions and their self-determined motivation and physical activity engagement.
Further evidence suggests that when PE teachers exhibit transformational behaviours, their
students are more likely to enjoy physical activity and are thus more physically active not only
within the classroom but also during leisure time (Beauchamp et al., 2014). Given the evidence
of promoting youth followers’ positive outcomes in physical education contexts by focusing on
19
principals’ and teachers’ interpersonal and leadership behaviours, TFL has implications for
studying effective behaviours in youth sport coaches and their influence on athlete outcomes.
Researchers have extended these findings from the PE classroom into organized sport
contexts by investigating the relationships between coaches’ TFL behaviours and athletes’
performance, participation, and personal development. TFL in coaching is effective in young
athletes by improving sports performance via the mediating effects of intrinsic motivation
(Charbonneau, Barling, & Kelloway, 2001). Additionally, athletes who perceive their coaches to
use TFL behaviours have more positive perceptions of coach effectiveness, satisfaction, and
extra effort over and above coaches who demonstrate transactional behaviours (Rowold, 2006).
Coaches’ TFL behaviours also positively influence task and social cohesion in team sports
(Callow et al., 2009). Finally, TFL behaviours have recently been linked with developmental
outcomes in youth athletes such as the growth of personal, social, and cognitive skills; goal
setting abilities; and initiative (Vella et al., 2013). Coaches’ TFL behaviours can foster positive
youth development outcomes such as leadership, performance, and quality coach-athlete
relationships. Overall, TFL offers a set of behaviours based on the 4 I’s that have been shown to
be associated with positive outcomes in followers. Taken together, these findings in sport and
other contexts suggest that TFL is an appropriate lens for developing an evidence-based CDP.
Teaching TFL Through Interventions
Another key reason for the potential value of TFL as a framework for developing an
interpersonal CDP is because TFL behaviours can be enhanced through interventions that
incorporate evidence-based recommendations. Kelloway and Barling (2000) recommend that
TFL interventions should (a) incorporate TFL principles (i.e., the 4 I’s), (b) provide examples of
TFL behaviours, (c) create environments to practice TFL behaviours, and (d) offer feedback on
20
those TFL behaviours. The use of these components in military, organizational, and educational
training contexts have resulted in improvements in TFL behaviours (Kelloway & Barling, 2000).
For example, Barling and colleagues (1996) demonstrated that TFL training of bank managers
led their employees to perceive them as exhibiting higher rates of transformational behaviours.
The workers also felt more committed to working for their company (Barling et al., 1996). TFL
training has also demonstrated success for PE teachers. Teachers with TFL training were rated
as significantly higher on transformational teaching as perceived by their students, and these
students had higher levels of self-determined motivation, self-efficacy, and intentions to be
physically active (Beauchamp et al., 2011). Thus, not only can training improve TFL leader
behaviours and follower outcomes, there are also evidence-based guidelines for the development
of future TFL theory-informed CDPs.
One TFL intervention for youth sport coaches was also developed by Vella and
colleagues (2013), whereby they administered a 2-hour workshop and monthly follow-up phone
calls to a group of soccer coaches and compared athlete outcomes to a control group. One year
after the workshop, athletes perceived higher levels of their cognitive skills and goal setting
abilities, and perceived higher rates of transformational leadership in coaches who had received
the TFL intervention. In addition to the successful results, the authors highlight that future TFL
coaching workshops should incorporate case studies and videos of TFL coaching behaviours,
provide personalized leadership development guides, and identify strategies and barriers to
adopting TFL behaviours.
That said, this was a pilot intervention that was heavily based on the MAC training
program and was not necessarily informed by the recommendations put forth by Kelloway and
Barling (2000). The workshop was also informed by an adapted form of TFL (e.g., did not
21
explicitly use idealized influence; see Vella, Oades, & Crowe, 2012 for details), and thus, certain
outcomes may not necessarily have been derived by the 4 I’s. In addition, like most
interpersonal CDPs, this TFL coaching workshop was not explicitly guided by behaviour change
theories (Allan et al., 2017).
Incorporating TFL into CDPs
Based on the evidence that TFL behaviours can potentially foster positive athlete
outcomes and be developed through training programs, the full-range leadership model provides
a valuable vantage point for designing an interpersonal-focused CDP. While there are many
different types of learning opportunities for coaches (i.e., formal, nonformal, informal; Mallett,
Trudel, Lyle, & Rynne, 2009), CDPs provide one formal method (i.e., learning is guided by a
specific curriculum from knowledgeable leader in a controlled workshop setting) for coaches to
advance their interpersonal behaviours. Unfortunately, earlier CDPs have not provided a critical
overview of all three types of coaching behaviours (i.e., professional, interpersonal, and
intrapersonal). Recently, Lefebvre and colleagues (2016) conducted a systematic review
classifying 285 CDPs into higher-order and lower-order domains. The results indicated that
most programs focused on coaches’ professional knowledge and behaviours (i.e., skill and
performance, health and well-being; n = 261), while little attention was given to developing
coaches’ interpersonal knowledge and behaviours (n = 18; Lefebvre et al., 2016). Given that
interpersonal behaviours are an important element of effective coaching (Côté & Gilbert, 2009)
and are essential to coach leadership (Vella et al., 2010), it is critical to bridge the gap between
theory and practice to provide greater opportunities for interpersonal development.
The aforementioned CDPs—the MAC, Empowering Coaching, and Vella and
colleagues’ (2013) TFL workshop—are examples of evidence-informed, interpersonal-focused
22
CDPs. Results from these workshops demonstrate that coaches can be taught leadership skills
through theory-based CDPs, and that trained coaches can influence athletes’ short-term
developmental outcomes. However, the MAC, Empowering Coaching, and Vella and
colleagues (2013) TFL workshop were not explicitly informed by behaviour change theories and
could benefit from more rigorous evaluation using different methodologies (Allan et al., 2017;
Evans et al., 2015). It is important to build from these CDPs to gain a greater understanding of
the depth and delivery of coaches’ leadership behaviours. Consequently, there have been calls to
develop coaches’ interpersonal behaviours through training using strong interpersonal theoretical
frameworks (i.e., TFL) and behaviour change theories (Allan et al., 2017), and to evaluate these
workshops using systematic observation (Arthur, Bastardoz, & Eklund, 2017; Turnnidge, Evans,
Vierimaa, Allan, & Côté, 2016).
Behaviour Change Wheel. The use of behaviour change theories in interpersonal-
focused CDPs is important for ensuring that coaches adopt the recently-taught behaviours after
the workshop. One helpful tool in conceptualizing the multitude of behaviour change theories is
the Behaviour Change Wheel framework (Michie et al., 2011). The Behaviour Change Wheel
(Figure 2) is a framework that informs the use of context-specific and targeted programs for
changing a broad spectrum of behaviours (Michie et al., 2011). At the centre of the wheel are
the sources of behaviour (i.e., the COM-B model), encircled by intervention functions and then
policy categories (Michie et al., 2011). According to the COM-B model, behaviour occurs as an
interaction between three necessary conditions: capability (i.e., psychological or physical ability
to enact the behaviour), motivation (i.e., reflective and automatic mechanisms that initiate or
inhibit behaviour), and opportunity (i.e., the physical and social environment that enables the
behaviour; Michie et al., 2011). The COM-B model is particularly relevant for developing coach
23
workshops aimed at changing coaches’ interpersonal behaviors (Allan et al., 2017; Michie et al.,
2011), as this tool could lead to consistency in behaviour change theory use among CDPs (Allan
et al., 2017). Furthermore, the COM-B model can inform relevant and salient behaviour change
techniques for use in CDPs (Michie et al., 2013). Thus, a full engagement of the COM-B model
holds potential utility in influencing coaches to adopt recently taught behaviours from a TFL
coaching workshop.
Figure 2. The Behaviour Change Wheel framework (Michie et al., 2011).
Evaluating TFL-Informed CDPs
Although leaders can learn TFL through interventions, another limitation of previous
interpersonal-focused CDPs and TFL interventions alike is that they often lack the
implementation of adequate evaluation procedures (Evans et al., 2015). For instance, self-report
questionnaires are typically used to assess athlete outcomes, whereas actual coach behaviours are
not being measured to evaluate affects of the CDP on the coach (Arthur et al., 2017; Evans et al.,
2015). Similarly, coaches’ behavioural evaluations are lacking contextual evidence, which limits
24
our ability to understand how they are being used on a daily basis. When CDPs are not
appropriately evaluated, researchers and coaches cannot be certain that evidence-informed
workshops produce positive long-term changes in coaches’ behaviours.
Previous methodological techniques. Although some TFL interventions have been
evaluated (e.g., Beauchamp et al., 2011; Vella et al., 2013), their evaluation approaches have
been limited. The most commonly used methods are quantitative questionnaires that are
completed by followers after an intervention. A prevailing questionnaire used across research
contexts is the Multifactor Leadership Questionnaire (MLQ), which is a validated instrument for
assessing transformational, transactional, and non-leadership behaviour (Bass & Avolio, 1990).
As an example, Barling and colleagues’ (1996) TFL training study used the MLQ to assess bank
supervisors’ behaviours. That said, there are issues with the psychometric validity of the MLQ
(Tejeda, Scandura, & Pillai, 2001) that has warranted caution.
Additional examples of TFL questionnaires used in sport settings include the
Differentiated TFL Inventory (DLTI; Callow et al., 2009), the Differentiated TFL Inventory-
Youth Sport (DTLI-YS; Vella, Oades, & Crowe, 2012), and the Transformational Teaching
Questionnaire (TTQ; Beauchamp et al., 2010). These questionnaires offer athletes’ subjective
perceptions of coaches’ behaviours, where athletes may be biased to other factors not included in
the questionnaire (e.g., coaches’ expert knowledge) and thus maintain consistency in coaches’
ratings (known as common-source, common-method variance bias; Arthur et al., 2017). In
addition, the sport-related TFL questionnaires do not account for a broad variety of other
leadership behaviours (e.g. laissez-faire style), and specifically, the DTLI-YS assesses an
adapted form of TFL (i.e., does not include all 4 I’s; Vella et al., 2012). It is thus unclear how
interventions may be influencing leaders’ real-time application of behaviours according to the
25
full-range leadership model. Consequently, researchers have advocated for expanding beyond
the accumulation of participant perceptions of leadership by including methodologies that assess
leaders’ actual behaviours (rather than follower outcomes) detailing the content and recipient of
the behaviours (Eberly, Johnson, Hernandez, & Avolio, 2013; Hoffman & Lord, 2013;
Lehmann-Willenbrock, Meinecke, Rowold, & Kauffeld, 2015).
Development of new measures. In response to calls for diverse measurement tools
assessing coach leadership, the Coach Leadership Assessment System (CLAS) was developed as
a novel observational coding system (Turnnidge & Côté, 2016b). The CLAS provides an
objective evaluation of coaches’ leadership behaviours, capturing several lower-order
dimensions of toxic, laissez-faire, transactional, and transformational behaviours drawing from
the full-range leadership model. The CLAS also allows researchers to continuously code for the
frequency and duration of four components (i.e., 4 I’s) of TFL, affording the possibility of
identifying which dimensions are more salient for fostering positive outcomes in youth athletes.
Furthermore, coders can account for the overarching content through which behaviours are
conveyed (e.g., instruction/feedback) and the target of the behaviours (e.g., individual). Thus,
the CLAS provides valuable contextual information about the nature of coaching leadership.
Overall, the CLAS complements the self-report research by offering a unique opportunity to
assess the effectiveness of a TFL workshop through providing real-time, in-depth descriptions of
the full range of coach leadership behaviours.
Transformational Coaching Workshop
Addressing several of the previous gaps in CDPs and TFL interventions, Turnnidge,
Barling, et al. (2016) and Turnnidge and Côté (2017a) developed the evidence-based
Transformational Coaching workshop guided by the TFL coach intervention recommendations
26
(Kelloway & Barling, 2000; Vella et al., 2013). The Transformational Coaching workshop
informs coaches of the effects of TFL behaviours on positive athlete outcomes and the impact of
other leadership behaviours (e.g., laissez-faire) with less positive athlete outcomes, with the
intention of influencing coaches’ persistence and effort in adopting a new style of leadership
(Turnnidge et al., 2016). Given the evidence for the importance of coach-athlete relationships
and coaches’ interpersonal behaviours for effective coaching, Turnnidge, Barling, et al. (2016)
and Turnnidge and Côté’s (2017a) workshop may be influential in teaching and instilling change
in coaches’ interpersonal behaviours. By placing greater emphasis on coaches’ interpersonal
rather than professional behaviours, this TFL workshop responds to recommendations for
improving the scope of CDPs (Lefebvre et al., 2016). The workshop is also informed by the
Behaviour Change Wheel (Michie et al., 2011), employing behaviour change techniques to target
coaches’ capability, opportunity, and motivation to perform TFL behaviours. Thus, the
Transformational Coaching workshop addresses the gap in previous TFL coaching workshops as
it is informed by the COM-B model and explicitly integrates behaviour change techniques
throughout the workshop (Allan et al., 2017).
However, it is unknown whether the Transformational Coaching workshop (Turnnidge,
Barling et al., 2016; Turnnidge & Côté, 2017a) is effective in changing coaches’ behaviours. A
first step in this evaluation process is to utilize the CLAS to assess the duration of coaches’
observable leadership behaviours before and after a group of coaches attend the workshop. The
CLAS can also be used to assess and contextualize coaches’ behaviours according to the content
and recipient modifiers, providing a detailed account of how coaches are using leadership
behaviours following the workshop.
27
Purpose
The purpose of this study is to evaluate the effectiveness of the Transformational
Coaching workshop by examining potential changes in coaches’ observable leadership
behaviours (specific to the full-range leadership model) and any differences in the context in
which the behaviours are performed (according to content and recipient modifiers). Relative to
baseline measures, it is hypothesized that coaches’ who participate in the workshop will exhibit
(with respect to duration of behaviours):
H1: Increased TFL behaviours,
H2: Decreased transactional, laissez-faire, and toxic behaviours,
H3: Increased use of instruction/feedback and general communication and decreased use
of organization content modifiers, and
H4: Increased use of individual and decreased use of team recipient modifiers.
28
Chapter 3
Methods
Design
This study followed a one-arm, pre- and post-test design. All participating coaches took
part in the Transformational Coaching workshop and coaches’ leadership behaviours were
evaluated through systematic observation (i.e., CLAS) at two time-points (i.e., before and after
the workshop). Of primary interest was to examine the potential for coaches’ behavioural
plasticity (i.e., intraindividual change) within a sample of coaches as this is a central tenet in
human development (Lerner & Busch-Rossnagel, 1981).
Participants
Eight male head coaches (Mage = 41.86; SD = 13.78) with 5 to 25 years of coaching
experience (M = 14.25, SD = 6.18) participated in this study. One coach did not report his age.
Coaches led athletes between 12 to 17 years of age, and five teams were composed of female
athletes whereas three were composed of male athletes. The inclusion criteria required all
coaches to spend a minimum of three hours per week with their teams (e.g., three practices
and/or games per week). This was to ensure that coaches spent ample time with their athletes to
prospectively allow for the development of leadership behaviours to occur after the workshop
and establish a level of contextual similarity between coaches (e.g., same frequency of
interaction, similar competitive level).
Measures
Coach Leadership Assessment System. Coach leadership behaviours were assessed
through the CLAS observation coding system, which has been validated and found to be reliable
for youth sport coaches (Turnnidge & Côté, 2016b). The CLAS has also been previously used in
29
the youth sport coaching context (Lefebvre & Côté, 2017; Turnnidge & Côté, 2017b). The
CLAS was developed to directly code the video recording of coaches’ full range of leadership
behaviours during practices and games using a software observational system (e.g., Noldus
Observer XT). This tool is intended for continuous coding and each behaviour is mutually
exclusive (i.e., must fall into one category). The coder reviewed the videos of coaches, pausing
the video to code for every behaviour exhibited. The behaviours were coded according to the
individual who initiated the behaviour (e.g., coach) and the leadership style of the behaviour (i.e.,
the primary code). Each of the leadership behaviours was also coded for the type of content
modifier, including: (a) instruction/feedback, (b) organization, or (c) general communication;
and the recipient of the behaviour (e.g., individual athlete, team). The context (e.g. warm-up) of
the behaviour was coded as a separate dimension (Turnnidge & Côté, 2016b).
Coaches’ behaviours were coded according to one of 17 lower-order dimensions (see
Table 1) that corresponds with one of eight higher-order leadership dimensions. TFL is
categorized into four higher-order dimensions: (a) individualized consideration (b) intellectual
stimulation (c) inspirational motivation and (d) idealized influence. The other higher-order
leadership dimensions available for coding include: transactional, neutral laissez-faire, toxic and
N/A (uncodable; Turnnidge & Côté, 2016b). The eight dimensions and 17 lower-order
dimensions of the CLAS are displayed in Table 1 and a full overview of the general coding
guidelines can be found in Appendix A.
30
Table 1
The Coach Leadership Assessment System: Leadership Dimensions
Higher-order dimension Lower-order dimension
Idealized influence 1 – Discussing/modelling pro-social values or behaviours
2 – Showing vulnerability/humility
Inspirational motivation 3 – Discussing goals/expectations
4 – Expressing confidence in athlete(s)’ capabilities
5 – Implementing a collective vision
6 – Providing meaningful and challenging tasks and roles
Intellectual stimulation 7 – Eliciting athlete input
8 – Sharing decision making/leadership responsibilities
9 – Emphasizing the learning process
Individualized consideration 10 – Showing interest in athletes’ needs
11 – Recognizing individual roles/contributions
Transactional 12 – Discussing rewards/penalties
13 – Searching for/responding to errors
Neutral 14 – Neutral
Laissez-faire 15 – Showing disinterest
Toxic 16 – Expressing anger/hostility
17 – Discussing and modelling anti-social values or behaviours
N/A X – Uncodable
Coding was conducted by one coder who had participated in 2.5 months of structured
coder training to ensure familiarity, skill, and knowledge of the CLAS. The coder was required
to meet a 75% reliability with the CLAS developer on timing, frequency, and duration of
behavioural codes before being allowed to code video that were used in the study analysis
(Hollenstein, Granic, Stoolmiller, & Snyder, 2004). The coder maintained these standards in two
subsequent reliability checks throughout the coding process (inter-rater reliability: M = 77.52%,
31
kappa M = 70.00%; intra-rater reliability: M = 88.44%, kappa M = 83.00%). All videos were
reviewed after initial coding was completed to ensure the accuracy of the codes.
Procedure
Once ethical approval was granted by the primary researcher’s academic institution (see
Appendix B), coaches were recruited through contact with the technical director of a youth
competitive soccer league in Ontario, Canada. When introducing the project, the researcher
discussed the confidentiality of the observation videos and demographic data collected with the
participants. Parents of athletes were notified of the video observation and that athletes were not
included in the observational data. All eight coaches were asked to read a letter of information
and complete a consent form (Appendix C) along with a coach information sheet (Appendix D)
that included basic demographic and coaching experience information to ensure that eligibility
criteria were met. Although the content of coaches’ practices was not dictated, all observational
sessions included a warm-up, drills, team scrimmages, and a cool-down period. This coherence
between practice content allowed for consistent opportunities to record coaches’ leadership
behaviours in a variety of practice contexts. However, it is important to note that the recording
focused on practice-based interactions and did not capture interactions that may have occurred in
other situations (e.g., arrival period before practice, post-practice team briefing).
All videos were recorded in midseason during the summer months (end of April – early
June). All coaches had been with their team since September of the previous year, which
allowed for consistent athlete familiarization with the coaches. The workshop was delivered one
week following Time 1 data collection, enabling sufficient time for the researcher to gather data
and prepare the workshop. Time 2 data collection occurred three to four weeks following the
32
workshop, allowing researchers to gauge the sustainability of the behaviour change. Figure 3
provides a visual representation of the procedure timeline.
1 week 3-4 weeks
Figure 3. Procedure for assessing coaches’ leadership behaviours pre- and post-workshop.
Pre-test procedure. The pre-test data collection took place between the end of April and
beginning of May 2017. Each participating coach was videotaped using a video camera to
capture the coaches’ leadership behaviours with individuals and the team. Coaches’
verbalizations were captured with an omni-directional wireless microphone worn by the coach
that was operated by a research assistant. Two practices were sought to be videotaped for each
participating coach; however, three coaches (C1, C2, and C3) were only filmed once: C1 and C2
experienced several practice cancellations due to inclement weather, and C3 only consented for
one practice to be filmed. All videos of each coach were used for data analysis, for a total of 13
observations ([5 coaches X 2 videos per coach] + [3 coaches X 1 video per coach]). The average
Transformational
Coaching Workshop
Time 1 Assessment
2 Observations (C4-C8)
1 Observation (C1-C3)
Time 2 Assessment
2 Observations (C4-C6, C8)
1 Observation (C1-C3, C7)
33
duration for the first practice was 1 hr 25 min (SD = 14 min 1 s) and the average duration of the
second pre-workshop practice was 1 hr 19 min (SD = 38 min 20 s).
Workshop. The coaches participated in the workshop one week following the pre-test
data collection in early May. The Transformational Coaching workshop (Turnnidge, Barling et
al., 2016; Turnnidge & Côté, 2017a) involved one group session lasting approximately three
hours in duration. The workshop occurred in the evening during the week. Two co-authors (JT
and JC) served as the principal workshop facilitators while the primary author provided
assistance (e.g., facilitating group discussions) throughout the workshop (along with three other
research assistants). The workshop facilitators educated coaches on specific behaviours related
to the full-range leadership model through lecture, video, and group discussion (Turnnidge,
Barling et al., 2016; Turnnidge & Côté, 2017a). By equipping coaches with a workbook that
they could use during and after the workshop, coaches engaged in behaviour change techniques,
such as behavioural practice (i.e., role playing), action planning (e.g., leadership calendar), goal
setting (e.g., Action Cards), demonstration of behaviour (e.g., video examples), and giving
feedback (e.g., from research assistant during discussion; Michie et al., 2013). For more details
on the development and content of the workshop, please see Turnnidge and Côté (2017a).
Workshop fidelity. During the workshop, four researchers completed an implementation
checklist (see Appendix E for one example) to ensure that the workshop was delivered as
intended. The items in the implementation checklist are systematized according to the following
sections: organization (e.g., materials organized and ready), procedure (e.g., presenting all topics,
such as the full-range leadership model), activities (e.g., Best/Worst Coach), debrief/discussion
(e.g., after the Stand By Your Quote activity), and monitoring participant engagement (e.g.,
participants completed all activities).
34
Post-test procedure. Post-test data collection occurred three or four weeks after the
workshop in early June. Each coach was videotaped using the same video camera and
microphone as the pre-test observation. Again, two practices were sought for observation for
post-test data collection; however, four coaches were only filmed once: C1, C2, and C7
experienced several cancellations due to weather, and C3 consented to be filmed once. All other
coaches were filmed twice, and only the second filmed practice was analyzed. Thus, a total of
eight post-test observations (8 coaches X 1 video per coach) were coded using the CLAS. The
average duration of the post-test practice was 1 hr 27 min (SD = 7 min 42 s). In total, 21
practices were coded using the CLAS (13 pre-test videos, 8 post-test videos).
Data Cleaning
Prior to data analysis, certain data cleaning steps were taken. All behaviours using the
observation content modifier (i.e., when coaches are silently observing athletes) and other target
modifier (i.e., when coaches are communicating with non-athletes, such as assistant coaches)
were subtracted from the original dataset. Although these behaviours are important, the purpose
of the present thesis was to explore the differences in coaches’ leadership behaviours when they
were actively communicating with athletes. Behaviours where coaches are silently observing
have intentionally not been coded in previous observational literature as observing behaviours
are difficult to interpret during video observation and recording (e.g., Webster et al., 2013).
Similarly, other observational systems have not coded for coaches’ communication with non-
athletes as non-athletes were similarly not the focus of their study (e.g., More & Franks, 1996).
Uncodable behaviours remained in the dataset to account for errors due to audio or video
equipment malfunctions but are not discussed in the pre-post-test analyses as they rarely occured.
35
Following the removal of the observation and other modifiers, all data were adjusted to
reflect a standard practice duration of one hour (3600 seconds) using equivalent ratios (i.e.,
[3600 seconds*duration of behaviour [s]/total observation duration [s]). Although coaches were
coded according to the lower-order dimensions (e.g., showing vulnerability), the data were
summed and categorized into the higher-order dimensions (e.g., idealized influence; see Table
1). Given that the 4 I’s are conceptually distinct factors, the measures were not averaged into a
global, single-score index for overall TFL (Antonakis & House, 2014; Arthur et al., 2017).
Lastly, the data were organized into the total content (e.g., instruction/feedback) and recipient
(e.g., individual) modifiers observed per practice across all five leadership behaviours.
Data Analysis
Data were analyzed using the duration (rather than frequency) of behaviours as this
provides a more representative account of coaches’ practice when a continuous coding scheme is
employed. The frequency of behaviours may be biased as some coaches may have fluctuated
between behaviours (providing higher frequencies) more than other coaches, even when the
duration of those behaviours was shorter.
Pre-workshop analyses. Prior to all pre-post-test analyses, the two pre-workshop
observations for the five participants were subjected to tests of normality, outliers, and
descriptive statistics using SPSS to understand the overall representation and normality of the
behaviours and modifiers. Next, bivariate correlations were computed to assess the test re-test
reliability for all leadership dimensions and modifiers. Although certain textbooks (e.g., Litwin,
2002) argue that test re-test reliability is acceptable when r > .70, it is important to recognize that
few standards exist for judging the minimum reliability estimate given the nature of the sample
and interval time between tests (Crocker & Algina, 2006). This is especially true for
36
observational research and the variability of human behaviour; thus, for the purposes of this
study, acceptable test re-test reliability will be when r > .50. Considering the small sample size
(n = 5), further testing was warranted to explore the differences between the two observations.
Paired samples t-tests were conducted according to duration for the (a) higher-order
leadership dimensions and (b) modifiers to determine whether significant differences in
behaviours and modifiers between practices existed before participants participated in the
workshop. These analyses were conducted so that the differences between pre- and post-
workshop behaviours could be attributed to the workshop, and not time, as a pseudo-control
group. The average values of the two pre-workshop observations were subsequently used for the
pre-post-test analyses for the five participants. Using average values can assist in normally
distributing the data for the pre-post-test analyses (Oberfield & Franke, 2013).
Pre-post-test analyses.
Descriptive statistics. All pre-post-test analyses for the leadership behaviours and
modifiers were subjected to tests of normality, outliers, and descriptive statistics. Non-normal
distributions were identified and thus, bootstrapping techniques were used in subsequent
statistical analyses (Efron & Tibshirani, 1993). Bootstrapping is a nonparametric method that
estimates the properties of the sampling population based on the bootstrap sample of 5000 (Efron
& Tibshirani, 1993).
Comparing pre-post means. To investigate the hypotheses of the study, a paired samples
t-test was conducted for each variable of interest. First, paired samples t-tests were used to
assess the changes in coaches’ leadership behaviours according to the higher-order leadership
dimension (e.g., idealized influence, transactional). Second, paired samples t-tests were used to
assess the changes in the content used to display behaviours (e.g., instruction/feedback) and the
37
recipient of coaches’ behaviours (e.g., individual) regardless of leadership behaviour. For each
pairwise comparison, bias corrected and accelerated bootstrapped (BCa) 95% confidence
intervals are presented as they provide a more robust point estimate (Efron & Tibshirani, 1993).
Importantly, a difference between the pre- and post-test means exists when the confidence
interval does not cross zero (Efron & Tibshirani, 1993). Last, the effect sizes are reported
according to the Cohen’s d, where .20 is considered to be a small, .50 is a medium, and .80 is a
large effect (Cohen, 1988).
38
Chapter 4
Results
Workshop Fidelity
The workshop was delivered as intended (Appendix E) on a weeknight by two presenters
(JT and JC) and four research assistants (one of whom was the primary author). The facilitators
engaged coaches throughout the workshop, and completed all lectures, videos, and activities
related to the full-range leadership model.
Pre-Workshop Analyses
Descriptive statistics. No outliers were identified, and the data were normally
distributed for all leadership behaviours and modifiers.
Test re-test reliability. The results from the bivariate correlations indicated that all
leadership behaviours and modifiers had good test re-test reliability (r > .500), with the exception
of idealized influence (r = -.478), inspirational motivation (r = .216), general communication (r
= -.134), and instruction/feedback (r = -.379). The variables that did not achieve good test re-test
reliability had high standard deviations. All results from the pre-workshop correlations can be
found in Appendices F through H.
Differences between pre-workshop behaviours. Results from the paired samples t-
tests demonstrated that there were no significant differences between the pre-workshop practices
for all leadership behaviours or content and recipient modifiers, except for the duration of
laissez-faire behaviours between time 1 and time 2 of the pre-workshop practices t(4) = 2.82, p =
0.048. Full results can be found in Appendices I and J.
39
Pre-Post Workshop Analyses
Descriptive statistics. In terms of leadership behaviours, two potential univariate
outliers were identified (i.e., falling between 1.96 and 2.57 standard deviations from the mean):
post-workshop laissez-faire behaviours (case 2) and post-workshop toxic behaviours (case 1).
These potential outliers remained in the data set. Similarly, the data were not normally
distributed and had a significantly positive skew. There were no outliers with regards to the
content and recipient modifiers, and all variables were normally distributed. Because
bootstrapping techniques do not rely on the assumption of normally distributed data,
bootstrapping with a sample of 5000 was applied for the paired samples t-tests.
TFL behaviours. Table 2 displays the percentage proportion, mean, and standard
deviation according to duration (s) for the 4 I’s by practice (pre- and post-workshop). For both
practices, when coaches used TFL behaviours, they predominantly displayed inspirational
motivation and individualized consideration. The duration of all TFL behaviours increased after
coaches participated in the workshop. The results from the paired samples t-test revealed that the
increases in coaches’ time spent using of idealized influence, t(7) = -2.16, p = .067, d = .76, and
inspirational motivation, t(7) = -1.99, p = .087, d = .70, were approaching significance and
exhibited medium to large effect sizes after participating in the workshop. However, increases in
coaches’ time spent using intellectual stimulation, t(7) = -1.70, p = .132, d = .60, and
individualized consideration, t(7) = -0.95, p = .374, d = .34, were not significant but exhibited
medium to large and small to medium effect sizes, respectively. Although idealized influence,
inspirational motivation, and intellectual stimulation variables did not reach significance, the
associated BCa 95% confidence intervals do not include zero and thus, there are observed
differences between these pre- and post-workshop behaviours.
40
Table 2
Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for 4 I’s for Pre-
and Post-Workshop Observations
Practice
Pre Post
Behaviour % M SD % M SD BCa 95% CI p
II 1.77 63.57 25.56 3.24 116.70 83.96 [-114.03, -7.64] 0.067
IM 11.93 429.54 186.90 15.44 555.79 239.83 [-265.42, -2.92] 0.087
IS 5.49 197.77 92.97 7.89 283.98 160.03 [-171.51, -1.25] 0.132
IC 10.79 388.51 153.98 12.95 466.15 182.06 [-252.48, 101.21] 0.374
Total 29.98 39.52
Note: II = Idealized Influence, IM = Inspirational Motivation, IS = Intellectual Stimulation, IC =
Individualized Consideration. This table, split between pre- and post-workshop practices,
displays the duration of coaches’ leadership behaviours according to the 4 I’s when actively
interacting with athletes. Means are represented in rate per hour format (e.g., 63.57 s per hour).
Non-TFL behaviours. Table 3 shows the percentage proportion, mean, and standard
deviation according to duration(s) for transactional, neutral, laissez-faire, and toxic behaviours
by practice (pre- and post-workshop). For both practices, coaches spent most of their time using
neutral behaviours, and the least amount of time displaying toxic behaviours. The duration of
transactional, neutral, and laissez-faire behaviours decreased while toxic behaviours remained
the same after coaches participated in the workshop. The results from the paired samples t-tests
indicate that there were significant decreases in coaches’ time spent using of neutral behaviours,
t(7) = 3.80, p = .007, d = 1.34, after participating in the workshop, exhibiting a large effect.
However, there were no significant changes in coaches’ time spent using transactional, t(7) =
0.72, p = .494, d = .25, laissez-faire, t(7) = 1.45, p = .190, d = .51, and toxic, t(7) = -0.23, p =
.825, d = .08, behaviours, and small, medium, and no effects were exhibited, respectively.
41
Table 3
Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Full-Range Leadership Behaviours for Pre- and Post-
Workshop Observations
Practice
Pre Post
Behaviour % M SD % M SD BCa 95% CI p
Transactional 6.27 225.53 142.58 5.47 196.91 105.85 [-30.68, 99.17] 0.494
Neutral 58.15 2093.01 202.05 50.84 1830.11 225.63 [163.29, 369.64] 0.007
Laissez-faire 4.58 165.08 138.10 3.09 111.19 136.89 [-23.50, 132.27] 0.190
Toxic 0.79 28.53 31.82 0.83 29.74 33.27 [-13.01, 10.46] 0.825
Total 69.79 60.23
Note: This table, split between pre- and post-workshop practices, displays the duration of coaches’ full-range of leadership behaviours
when actively interacting with athletes. Means are represented in rate per hour format (e.g., 225.53 s per hour).
42
Content modifiers. Table 4 shows the percentage proportion, mean, and standard
deviation according to duration(s) for the content modifiers by practice. For both practices,
coaches spent the majority of their time conveying leadership through organization and
instruction/feedback, while little time was spent using general communication to interact with
athletes. After coaches participated in the workshop, the duration of instruction/feedback and
general communication modifiers increased while organization modifiers decreased. The results
from the paired samples t-tests demonstrate that there were significant increases in coaches’ time
spent using of instruction/feedback, t(7) = -3.32, p = 0.013, d = 1.17, to convey leadership after
participating in the workshop, exhibiting a large effect. There were significant decreases in
coaches’ time spent using organization, t(7) = 5.38, p = .001, d = 1.90, to convey leadership after
participating in the workshop, exhibiting a large effect. There were no significant differences in
coaches’ use of general communication to interact with athletes, t(7) = -0.18, p = .863, d = .06,
exhibiting no effect.
43
Table 4
Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Content Modifiers for Pre- and Post-Workshop
Observations
Practice
Pre Post
Modifier % M SD % M SD BCa 95% CI p
General
Communication
7.09 242.57 97.19 7.34 255.27 193.77 [-146.45, 114.40] 0.863
Instruction/Feedback 43.37 1484.51 342.25 50.96 1773.19 320.66 [-468.10, -91.26] 0.013
Organization 49.54 1695.82 281.50 41.70 1450.92 280.62 [167.48, 316.49] 0.001
Total 100.00 100.00
Note: This table, split between pre- and post-workshop practices, displays the duration of the content modifiers through which
leadership behaviours are conveyed. Means are represented in rate per hour format (e.g., 242.57 s per hour)
44
Recipient modifiers. Table 5 presents the percentage proportion, mean, and standard
deviation according to duration(s) for the recipient modifiers by practice. For both practices,
coaches spent more time directing their behaviours towards the team rather than individuals.
After coaches participated in the workshop, coaches spent more time directing behaviours
towards individuals and less time directing behaviours towards the team. The results from the
paired samples t-tests demonstrate that there were no significant increases in coaches’ time spent
directing behaviours towards individuals, t(7) = -0.66, p = .528, d = .23, after participating in the
workshop, exhibiting a small effect. There were no significant decreases in coaches’ time spent
directing behaviours towards the team, t(7) = 0.15, p = .886, d = .05, after participating in the
workshop, exhibiting no effect.
Table 5
Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Recipient
Modifiers for Pre- and Post-Workshop Observations
Practice
Pre Post
Modifier % M SD % M SD BCa 95% CI p
Individual 38.13 1306.62 235.73 39.54 1375.68 337.85 [-244.62, 86.64] 0.528
Team 61.87 2120.55 224.25 60.46 2103.49 311.92 [-185.41, 274.32] 0.886
Total 100 100
Note: This table, split between pre- and post-workshop practices, displays the duration of the
content modifiers through which leadership behaviours are conveyed. Means are represented in
rate per hour format (e.g., 1306.62 s per hour).
45
Chapter 5
Discussion
The overarching purpose of this thesis was to assess the effectiveness of the
Transformational Coaching workshop by observing coaches’ leadership behaviours before and
after their participation in the workshop. The following section will address the four hypotheses
put forth with regard to the overall objective of the study. First, the discussion will begin by
examining the changes in coaches’ TFL behaviours according to the 4 I’s. Second, the
differences in coaches’ non-TFL leadership behaviours will be reviewed. Third, the changes in
the content modifiers through which coaches convey leadership behaviours will be discussed,
and the last section will contextualize coaches’ behaviours according to the changes in recipient
modifiers.
Transformational Leadership
Hypothesis one stated that coaches would spend more time engaging in TFL behaviours
according to the 4 I’s after participating in the workshop. This hypothesis was partially
supported, as the results demonstrated that idealized influence (p = .067) and inspirational
motivation (p = .087) behaviours only approached statistical significance. Considering that p
values rely on sample size, and that this study was a pilot test with a small sample, it is likely that
the sample was not large enough for the idealized influence and inspirational motivation
variables to reach significance (e.g., Coe, 2002; Cohen, 1994; Gardner & Altman, 1986).
Indeed, large differences in small sample sizes may lead to non-significant results (du Prel,
Hommel, Röhrig, & Blettner, 2009) and thus, future tests of effectiveness with more coaches
may enhance the likelihood of reaching statistical significance. That said, p values are only one
binary and somewhat simplistic measure of statistical significance (du Prel et al., 2009), and
46
bootstrapped confidence intervals offer a robust range of values estimating precise differences in
pre-post workshop behaviours (Efron & Tibshirani, 1993). Considering that the bootstrapped
confidence intervals for the differences in coaches’ pre-post idealized influence, inspirational
motivation, and intellectual stimulation behaviours did not cross zero, it is assumed that the
workshop had a statistical influence on these behaviours.
Furthermore, statistical significance is not the only measure of importance especially in
pilot tests (e.g., Cohen, 1994). With small sample sizes, and in line with the literature on TFL
training evaluation (Hochholdinger, Rowold, & Schaper, 2008), practical significance (i.e.,
“whether the result is useful in the real world”; Kirk, 1996, p. 746) using effect sizes (typically
between a medium to large effect; Kirk, 1996) should be used to determine the success of the
intervention and importance of differences between two groups (Coe, 2002; Cohen, 1994; du
Prel et al., 2009). Evidently, the comprehensive use of p values, bootstrapped confidence
intervals, and effect sizes can be used to conduct null hypothesis significance testing (Banjanovic
& Osborne, 2016). Thus, moving forward, it will be interpreted that the workshop had a
practically significant effect on increasing coaches’ idealized influence and inspirational
motivation behaviours that would have reached statistical significance with a larger sample size.
In addition, there is partial support for the statistically (i.e., the confidence interval did not
include zero) and practically (i.e., a medium to large effect size) significant increase in
intellectual stimulation behaviours, but a has a non-significant p value (p = .132). Last, although
there were no significant differences in coaches’ individualized consideration behaviours, the
mean duration increased after coaches’ participated in the workshop, exhibiting a small effect.
In general, this study reinforces the fact that TFL workshops can have an effect on increasing
47
coaches’ TFL behaviours (e.g., Beauchamp et al., 2011; Vella et al., 2013) and that CDPs can be
effective (e.g., Larsen et al., 2015; Smith et al., 1983).
Idealized influence. Similar to previous studies (e.g., Turnnidge & Côté, 2017b),
coaches spent the least amount of time engaging in idealized influence behaviours, even after
participating in the workshop. Thus, it is valuable that the Transformational Coaching workshop
removed some barriers to using idealized influence behaviours and assisted coaches in acting as
a positive role model. According to the COM-B model, the workshop may have enhanced
coaches’ capability to perform idealized influence by equipping them with practical knowledge
and tools for everyday use (Michie et al., 2011). However, more research is required to
understand the barriers and facilitators that coaches face for using idealized influence
behaviours, and to assess the workshop using the COM-B model constructs.
Interestingly, the workshop had a large effect (d = .76) on increasing coaches’ time spent
using idealized influence behaviours. This is consistent with previous research, as a TFL
workshop improved athletes’ perceptions of coaches’ use of appropriate role modeling (Vella et
al., 2013). Although other TFL training programs (e.g., Abrell, Rowold, Weibler, &
Moenninghoff, 2011; Barling et al., 1996) discovered that it was difficult to change leaders’
idealized influence behaviours, it is worthy to note that the content of these training programs
was focused on changing intellectual stimulation and individualized consideration. The present
workshop used behaviour change techniques (e.g., scenarios) to provide education for all 4 I’s,
which may explain the practically significant idealized influence findings. In addition, Abrell
and colleagues’ (2011) and Barling and colleagues’ (1996) training programs were directed for
leaders of adult bank employees. Perhaps it is easier for leaders of youth populations to adopt
positive role modelling behaviours (e.g., Vellat et al., 2013) due to the unique interpersonal
48
context of coach-athlete dyads that involve characteristics of friendship (e.g., share news), work
relationships (e.g., instructional support), and parental relationships (e.g., trust; Jowett & Clark-
Carter, 2006).
Inspirational motivation. Coaches spent the majority of their time displaying
inspirational motivation when they used TFL behaviours, regardless of their participation in the
workshop. This may be explained by the fact that inspirational motivation behaviours lend
themselves to be more similar to traditional, professional behaviours of coaches (e.g., goal
setting) than interpersonal (Turnnidge & Côté, 2017b).
Additionally, the results demonstrate that the Transformational Coaching workshop had a
medium to large effect (d = .70) on increasing the amount of time coaches spent using
inspirational motivation behaviours. This finding is supported by previous literature which
found that TFL education improved employees’ perceptions of supervisors’ charisma (i.e.,
providing a vision and enhancing followers’ self-expectations; Barling et al., 1996). Importantly,
increasing coaches’ inspirational motivation behaviours (e.g., expressing confidence in athletes)
may have created a task-involving and empowering motivational climate (Beauchamp et al.,
2010; Smith et al., 2015). Thus, it is possible that athletes’ feelings of relatedness to their coach
and athletes’ intrinsic and autonomous motivation to participate in sport was enhanced
(Beauchamp et al., 2010; Beauchamp et al., 2011; Smith et al., 2015). Although Vella and
colleagues (2013) did not witness improvements in coaches’ inspirational motivation following
TFL training, the workshop was informed by an adapted version of TFL and did not incorporate
behaviour change techniques. Thus, the theoretical underpinnings for the Transformational
Coaching workshop and inclusion of relevant behaviour change techniques may be salient
elements in determining the efficacy of changing coaches’ inspirational motivation behaviours
49
(Michie & Johnston, 2012). Future research is required to identify specific workshop
components and behaviour change techniques implemented that aid increasing coaches’
inspirational motivation behaviours.
Intellectual stimulation. In comparison to other TFL behaviours, coaches spent little
time using intellectual stimulation behaviours, both before and after participating in the
workshop. These results mirror findings from previous observational studies of coaches
(Partington & Cushion, 2013) and studies employing TFL questionnaires (e.g., Barling et al.,
1996).
Although there was no statistically significant increase in coaches’ intellectual
stimulation behaviours after the workshop, there remains some practical significance as a
medium to large effect was exhibited (d = .60). Similarly, changing coaches’ intellectual
stimulation behaviours in the coaching and TFL literature appears to be mixed. TFL workshops
that have evaluated coaches’ intellectual stimulation behaviours through questionnaires have
consistently demonstrated significant increases in these behaviours (Abrell et al., 2011; Barling
et al., 1996; Vella et al., 2013). However, through observation and qualitative interviews,
Partington and Cushion (2013) found that although coaches want to display behaviours that align
with intellectual stimulation (e.g., questioning), coaches continue to coach in a traditional,
prescriptive manner. Indeed, the ideology that coaches are decision-makers and gatekeepers of
knowledge is difficult to change in practice (Cope, Partington, Cushion, & Harvey, 2016). Thus,
according to the COM-B model (Michie et al., 2011), the Transformational Coaching workshop
may have influenced coaches’ motivation but not social opportunities (i.e., norms and coaching
culture) to perform intellectual stimulation behaviours. Enhancing coaches’ motivation may
have resulted in the practical significance observed (i.e., medium to large effect size), but it is
50
possible that without changing coaches’ social opportunities, no statistical significance was
observed. By including intervention techniques such as enablement (e.g., group follow-up
sessions with coaches that discuss and practice intellectual stimulation behaviours) in future
workshops, it may be possible to foster norms, and ultimately social opportunities that encourage
coaches’ use of intellectual stimulation behaviours (Michie et al., 2011). In turn, additional tests
of effectiveness of the Transformational Coaching workshop may witness statistically significant
long-term changes in coaches’ intellectual stimulation behaviours.
Individualized consideration. Behind inspirational motivation, coaches spent the
second-longest time displaying individualized consideration behaviours when they perform TFL.
This finding aligns with previous observational studies that where similar behaviours (e.g.,
reinforcement, relatedness-supportive) were used more often (Smith et al., 1977; Smith, Quested,
Appleton, & Duda, 2017). This is likely because individualized consideration behaviours are
similar to traditional notions of feedback and praise in coaching (Turnnidge & Côté, 2017b).
The results from the present thesis revealed that the duration of coaches’ individualized
consideration behaviours did not significantly increase after participating in the workshop, but a
small to medium effect (d = .34) was exhibited. Vella and colleagues (2013) had similar findings,
as coaches’ perceived individualized consideration behaviours did not improve after training. In
contrast, there is support for enhancing followers’ perceptions of bank supervisors’
individualized consideration behaviours after TFL training (Barling et al., 1996), and observable
differences in coaches’ reinforcement (similar to individualized consideration) behaviours after
the CET program (Smith et al., 1979). That said, in the present study, coaches displayed
relatively high levels of individualized consideration before the workshop (i.e., 36% of all TFL),
which may have made these behaviours less susceptible to changing after participating in the
51
workshop (i.e., the ceiling effect). Nonetheless, coaches may have also been performing
individualized consideration behaviours in other contexts with their athletes (e.g., in the dressing
room, at games; Lefebvre & Côté, 2017), when the camera was not capturing coaches’
behaviours. Thus, further research examining coaches’ behaviours in a variety of sport contexts
may provide greater understanding of the impact of the Transformational Coaching workshop on
coaches’ individualized consideration behaviours.
Transactional, Neutral, Laissez-Faire, and Toxic Leadership
The second hypothesis stated that coaches would spend less time using non-TFL
behaviours after participating in the workshop. This hypothesis was partially supported, as there
were significant decreases in coaches’ use of neutral behaviours, but no significant differences in
coaches’ use of transactional, laissez-faire, and toxic leadership behaviours. However, the mean
durations of coaches’ transactional and laissez-faire behaviours decreased after the workshop,
while the mean duration of toxic behaviours remained the same. Overall, these findings
demonstrate that the Transformational Coaching workshop can encourage coaches to use TFL
behaviours as a replacement for less effective coach behaviours (i.e., neutral) to improve coach-
athlete interactions.
Transactional, laissez-faire, and toxic behaviours. Both pre- and post-workshop
observations revealed that coaches spent little time performing transactional, laissez-faire, and
toxic behaviours as compared to neutral and TFL behaviours. This is consistent with previous
literature analyzing coaches’ observable leadership behaviours (e.g., Smith et al., 2017).
Furthermore, the results from this study reveal that there were no statistically significant
differences in coaches’ transactional, laissez-faire, or toxic behaviours between pre- and post-
workshop, and that small (d = .25), medium (d = .51) and no (d = .08) effects were exhibited,
52
respectively. This finding may partially be explained by the fact that laissez-faire and toxic
behaviours often occur due to specific situations (e.g., poor performance) and thus were
displayed less consistently than positive behaviours (Côté, Yardley, Hay, Sedgwick, & Baker,
1999). Similarly, coaches were already displaying such low levels of laissez-faire and toxic
leadership that there may have not been room to decrease these behaviours to a point that would
reach significance (i.e., the floor effect). Regardless, it is important to report on coaches’ non-
TFL behaviours as coaches are often not educated on these behaviours in other TFL workshops
and subsequently, changes in non-TFL behaviours are not reported (Abrell et al., 2011;
Beauchamp et al., 2011; Vella et al., 2013), even when they are assessed using the MLQ (e.g.,
Barling et al., 1996).
Additionally, the full-range leadership model and ultimately, the Transformational
Coaching workshop, acknowledges that coaches will not use TFL behaviours for an entire
practice, and that there are some instances when coaches will understandably use non-TFL
behaviours (Bass & Riggio, 2006; Kirkbride, 2006; Turnnidge & Côté, 2017a). Indeed, previous
TFL research suggests that effective leaders will still engage in moderate levels of transactional
leadership and low levels of laissez-faire and toxic leadership styles (Bass & Avolio, 1994;
Kirkbride, 2006). Regardless, it is important to recognize that rare occurrences of laissez-faire
and toxic behaviours may have a negative impact on youth’s sport experiences (e.g., Baumeister,
Bratslavsky, Finkenauer, & Vohs, 2001). In a similar vein, Becker (2012) argued that coaches
who display inconsistent, negative behaviours are ineffective and can lead to detrimental youth
outcomes (e.g., dropout). Thus, more research is required to understand how the
Transformational Coaching workshop can change ineffective leadership behaviours to improve
athlete experiences.
53
Neutral behaviours. The results from this study show that coaches spent the majority of
their time using neutral behaviours to communicate with athletes both pre- and post-workshop.
Similar behavioural profiles of coaches were observed in research utilizing the CLAS (Turnnidge
& Côté, 2017b; Lefebvre & Côté, 2017).
Although coaches still communicate with athletes through neutral means of
communication for the majority of the practice, a unique and important finding is that coaches
spent significantly less time engaging in neutral behaviours after participating in the workshop.
It is clear that the workshop aided coaches in substituting some of their pre-workshop neutral
behaviours with TFL behaviours post-workshop. This finding is valuable given that
questionnaires used to assess previous TFL training programs (e.g., TTQ; Beauchamp et al.,
2010) do not allow followers to reflect on a full range of behaviours and thus do not provide
details on the ways in which leaders changed their behaviours post-workshop. Evidently, the
Transformational Coaching workshop was successful in shifting coaches’ use of less effective
non-leadership behaviours to effective TFL behaviours (i.e., idealized influence, inspirational
motivation, and to a lesser extent, intellectual stimulation). These findings can be integrated into
the workshop for future participants to demonstrate evidence that TFL behaviours do not require
extra time or resources, and can be substituted for less effective behaviours. Thus, greater
inclusion of activities that allow coaches to reflect on current behaviours and practice effective
TFL behaviour replacements (Michie et al., 2011) may be warranted to further improve the
balance of coaches’ leadership behaviours.
Content Modifiers
The third hypothesis stated that coaches would spend more time conveying leadership
using instruction/feedback and general communication and spend less time conveying leadership
54
through organization. This hypothesis was partially supported, as there were significant
increases in coaches’ use of instruction/feedback and significant decreases in coaches’ use of
organization to express leadership, both exhibiting large effects (d = 1.17, d = 1.90,
respectively). However, there was no difference in coaches’ use of general communication with
athletes, and no effect was observed (d = .06). Nevertheless, the mean duration of general
communication trended towards increasing.
Instruction/feedback and organization. In this study, coaches used similar levels of
instruction/feedback and organization to convey leadership. Previous research using the CLAS
found that coaches’ expressed TFL behaviours more often through instruction/feedback than
organization, and that the opposite was true for neutral behaviours (Turnnidge & Côté, 2017b).
Furthermore, effective coaches provide athletes with more instruction and feedback (e.g., Bloom
et al., 1999), while less effective coaches spend more time organizing athletes (e.g., Erickson et
al., 2011). Given that coaches used more effective (i.e., higher TFL, lower neutral) behaviours
after the workshop, it is not surprising that there were significantly higher levels of
instruction/feedback and lower levels of organization (e.g., Cushion & Jones, 2001; Erickson et
al., 2011).
It is interesting to note that the workshop influenced the content of coaches’ leadership
behaviours when instruction/feedback and organization was not explicitly covered in the
workshop. This is unlike the CET and MAC training programs, where one of the main goals was
to emphasize the importance of instruction and praise (Smith et al., 1983; Smith et al., 2007).
Indeed, while the CET and MAC have been successful in enhancing athletes’ perceptions of
coaches’ instruction and praise, they have not achieved such success when observing coaches’
behaviours (Smith et al., 1979). Thus, the present thesis demonstrates objective evidence that the
55
Transformational Coaching workshop successfully changed what leadership behaviours coaches
used, and how coaches used these behaviours to effectively teach athletes sport-specific skills.
Ultimately, by targeting and enhancing coaches’ leadership behaviours, the workshop also
enhanced the communication of coaches’ professional behaviours, equipping them with
knowledge and skills of an overall effective coach (Côté & Gilbert, 2009).
General communication. Both before and after the workshop, coaches displayed low
levels of general communication in their interactions with their athletes. Erickson and Côté
(2016) and Turnnidge and colleagues (2014) observed similarly low levels of general
communication among their participants. Interestingly, the workshop did not improve coaches’
levels of general communication, even though coaches’ spent more time displaying effective
behaviours, and that more effective coaches discuss matters outside of sport more often with
their athletes (Erickson & Côté, 2016). It is possible that coaches felt more comfortable
changing typical coaching behaviours (i.e., sport-specific), and that coaches may still be
concerned with professional identity and the time required to talk about non-sport specific topics.
Greater education and training is required for coaches to integrate general communication into
their practice. However, coaches may have integrated general communication more often when
the cameras were not filming (e.g., travelling to away games), and thus some behaviours may not
have been accounted for in the current study. Nonetheless, future research should investigate the
ways in which coaches can integrate general communication throughout practice as it can lead to
athletes’ enjoyment, commitment, and prolonged participation (Stuntz & Spearance, 2010; Visek
et al., 2015).
56
Recipient Modifiers
The last hypothesis stated that coaches would spend more time directing leadership
towards individuals and less time directing leadership towards groups (i.e., team). This
hypothesis was not supported as there were no significant increases in coaches’ expression of
behaviours towards individuals and no significant decreases in coaches’ expression of
behaviours towards groups. Nevertheless, the mean duration of interacting with individual
athletes trended towards increasing with a small effect size (d = .23) while the mean duration of
interacting with more than one athlete trended towards decreasing with no effect size (d = .05).
For both practices, coaches directed their behaviours towards individual athletes (i.e.,
individual) less often than they directed their behaviours towards more than one athlete (i.e.,
team). Given that the workshop increased coaches’ effective (i.e., TFL) behaviours, it is
interesting that coaches did not increase their time spent interacting with individuals as effective
coaches tend to use more individualized behaviours (e.g., Becker, 2012; Gould, Dieffenbach, &
Moffett, 2002). One reason may be that some relationships exist between leadership behaviours
and recipients of behaviours (e.g., individualized consideration and individual athlete recipients)
that partially explain why changes in recipients were not observed. In addition, it may be
necessary for soccer coaches to continue to communicate with groups of athletes considering the
nature of team sport, where the focus is on the group working together for performance (Jowett,
Paull, & Pensgaard, 2005). Conversely, the results may be different in an individual sport
setting, where the coach’s focus is on individual athlete development and progression (Jowett et
al., 2005). However, it is important to further investigate how the Transformational Coaching
workshop can promote greater individualized coaching behaviours and the effectiveness of
57
individual compared to team-directed TFL behaviours in sports with varying levels of
interdependence.
Theoretical and Practical Implications
Overall, this study was the first to test the effectiveness of the Transformational Coaching
workshop, and employed a unique methodological technique for its assessment. Observing
coaches’ leader behaviours through the CLAS, rather than perception of leader behaviours
through questionnaires, provided detail on the development of coaches’ actual leader behaviours
before and after participating in the workshop. As previously discussed, transformational
coaches could facilitate positive athlete experiences (e.g., Callow et al., 2009), but the actual
TFL behaviours underpinning this development process had yet to be assessed. By identifying
differences in actual leader behaviours, it was possible to pinpoint potential areas of
improvement for the future implementation of the Transformational Coaching workshop (i.e.,
greater integration of behaviour change techniques that target social opportunity [e.g.,
enablement] for enhancements in coaches’ intellectual stimulation behaviours, promoting ways
to individualize coaches’ behaviours, and techniques on how to discuss non-sport specific topics
with athletes).
Furthermore, this project represents one of the first attempts at employing an
observational technique as an evaluation tool of an interpersonal-focused CDP. Specifically, the
use of the CLAS diversified the research methods used in TFL workshop evaluations to provide
real-time accounts of a full range of coaches’ leadership behaviours. Objectively evaluating
behaviours enhanced the understanding of the complexity of using various leadership styles in
actual sport practices. Theoretically, this project offered additional tests of validity and
reliability for the CLAS by having new coders evaluate different coach behaviours that aligned
58
with the CLAS guidelines. The successful application of the CLAS sheds light on various
dimensions of coaches’ leadership behaviours (e.g., content and recipient) and encourages the
use of diverse methods for future CDP evaluations. In addition, this thesis offers support for
using the full-range leadership model as a way of conceptualizing, teaching, assessing, and
reporting coaches’ behaviours. By focusing on the full-range of coaches’ behaviours (not simply
TFL), this project offered a full understanding of leadership styles beyond that of previous
literature (e.g., Vella et al., 2013).
Lastly, this study was important as researchers collaborated with community sport
organizations to conduct real-world research. This study represents a field experiment that
tested the Transformational Coaching workshop’s effectiveness rather than efficacy. Efficacy
studies occur in controlled conditions and are often removed from the realities and complexities
of youth sport coaching (Gartlehner, Hansen, Nissman, Lohr, & Carey, 2006). Coaches were
able to change their behaviours from neutral to transformational in their everyday sport practice.
Given the integration of the COM-B model (Michie et al., 2011) and tailoring the workshop to
the sport organization, coaches may be more likely to maintain these newly adopted behaviours
over a sustained period of time (Michie et al., 2011). Since the workshop was informed by the
Behaviour Change Wheel (Michie et al., 2011), coaches were able to understand why, when, and
how to use TFL behaviours (Allan et al., 2017). Theoretically, this project aided in
understanding the appropriate combination of behaviour change techniques to be used for
coaching contexts. These behaviour change techniques align with the COM-B model which
indicates that individiauls must be provided with the capability, opportunity, and motivation to
perform leadership behaviours (Michie et al., 2011). By partnering with a community sport
organization, it was possible to provide coaches with valuable and needed education on
59
interpersonal skills (e.g., Lefebvre et al., 2016) that resulted in actual changes in coaches’
leadership behaviours.
Limitations and Future Directions
Despite the numerous strengths associated with this study, there are also several
limitations that must be addressed. Notably, this study was conducted with a small number of
participants as observational research requires significant time and resources (Frick, Barry, &
Kamphaus, 2010). For example, in the present thesis, each 1.5-hour sport practice required 10
hours of coding using the CLAS, after completing the 2.5 month coder training program. In
addition, the sample size limited the ability to conduct more robust statistical analyses, and
although basic repeated measures tests have been used with TFL interventions comprised of
small sample sizes to maintain statistical power (e.g., Abrell et al., 2011), there are more
omnibus tests that should be conducted with larger sample sizes to control for potential
covariates (Ma, Mazumdar, & Memtsoudis, 2012). Due to the small sample size, it is difficult to
know how confounding variables may have affected the current results. However, the use of
bootstratpping confidence intervals enhanced confidence in the present findings as it is a robust
resampling method that allows precise estimations in the differences between the means. The
inclusion of effect sizes also provides important conclusions on the magnitude and relevance of
differences between means. Furthermore, although time-consuming, future research should
consider recruiting larger numbers of coaches (e.g., n > 30) to participate in observational tests of
effectiveness as more robust statistical analyses will be available.
In addition, the small sample size limited the heterogeneity and generalizability of the
results to the current sample of male youth sport coaches. There is reason to believe that the
workshop may have affected female coaches’ behaviours differently, given the fact that females
60
display higher levels of TFL than males (e.g., Eagly et al., 2003). It may also be beneficial to
understand how the Transformational Coaching workshop influences coaches’ behaviours in
other youth sport contexts (e.g., different levels of interdependence; Evans et al., 2012). Thus,
future research should diversify the sample of participants beyond male youth soccer coaches to
understand how the Transformational Coaching workshop works in other sport settings and with
coaches of different gender. Furthermore, considering that coaches’ behaviours were the focus
of the present study, it may also be important to test the impact of the workshop on athletes’
outcomes in a variety of sport contexts.
Another limitation with the present research is that there was no true control group to
compare the effects of the Transformational Coaching workshop. Although examining the
differences between the pre-workshop practices helped to determine that the differences in
coaches’ behaviours was in fact due to the Transformational Coaching workshop and not history
or maturation (Marsden & Torgerson, 2012; Torgerson & Torgerson, 2008), there remain threats
to causal validity. Differences in coaches’ behaviours may be due to test effects (i.e., that
improvements in behaviour were a result of being familiarized with filming during practice), and
thus a Solomon four group design (pre- and post-test no intervention, pre- and post-test with
intervention, post-test only no intervention, and post-test only with intervention) should be
considered for future research (Shadish, Cook, & Campbell, 2002).
Finally, this study tested the effectiveness of the Transformational Coaching workshop by
assessing coaches’ behaviours after a period of three to four weeks post participation in the
workshop. Thus, it is not possible to determine whether coaches maintained their behaviour
change well after (e.g., > 6 months) the Transformational Coaching workshop. Given the need
for more comprehensive evaluations of CDPs (Evans et al., 2015), future research should assess
61
coaches’ behaviours multiple times after participating in the workshop. This will allow
researchers to understand if and how the workshop is affecting coaches’ leadership behaviours
over time. Similarly, it may be important to assess the associated COM-B constructs to further
identify areas of improvement for the Transformational Coaching workshop and mechanisms for
overcoming barriers to coaches’ behaviours (Allan et al., 2017). For example, future research
can examine coaches’ capability, opportunity, and motivation to perform TFL behaviours pre-
and post-workshop to identify the specific behavioural constructs that the Transformational
Coaching workshop is targeting.
62
Chapter 6
Conclusion
The Transformational Coaching workshop was recently developed to address the gaps in
coach education by providing an evidence-informed interpersonal-skills workshop for youth
sport coaches (Turnnidge & Côté, 2017a). While other TFL-informed workshops have relied on
questionnaires for evaluation (e.g., the DTLI-YS; Vella et al., 2012), the purpose of this study
was to test the effectiveness of the Transformational Coaching workshop through systematic
observation. From 21 observations across eight coaches, the current thesis provides a detailed
description of the changes in coaches’ leadership behaviours after participating in the
Transformational Coaching workshop.
Notably, coaches spent more time displaying idealized influence, inspirational
motivation, and to a lesser extent, intellectual stimulation behaviours and significantly less time
displaying neutral behaviours after participating in the workshop. In addition, coaches spent
significantly more time conveying leadership through instruction/feedback and significantly less
time conveying leadership through organization after participating in the workshop. Although
no differences were observed for the other variables, most behaviours and modifiers trended in
the hypothesized direction, and may be found to be significant with a larger sample size.
Ultimately, this is the first study to have assessed the effectiveness of a an interpersonal
workshop through systematic observation of coaches’ behaviours.
Irrespective of the significant findings, it is clear that there is room for improvement for
the Transformational Coaching workshop and ultimately, coaches’ behaviours. Importantly,
coaches are still displaying high levels of neutral behaviours when compared with TFL
behaviours. Future implementations of the workshop should emphasize evidence that neutral
63
behaviours can be substituted with TFL behaviours and include greater practice activities. In
addition, intellectual stimulation behaviours may be difficult to change because of coaching
norms (Cope et al., 2016). Thus, the workshop may need to further target coaches’ social
opportunity (i.e., social norms) to perform intellectual stimulation behaviours by including
enablement techniques (i.e., follow-up sessions and group discussions; Michie et al., 2011).
Regardless of coaches’ participation in the workshop, coaches rarely communicated with athletes
about non-sport specific matters. Coaches also did not display individualized behaviours for
longer after the workshop. Given the relationship between coaches’ use of individualized and
general communication with athletes’ positive outcomes (e.g., Erickson & Côté, 2016; Visek et
al., 2015), researchers should explore the ways in which the Transformational Coaching
workshop can further integrate and encourage non-sport specific communication to specific
athletes.
Overall, this study offers additional support for providing TFL-theory informed education
to youth sport coaches. In accordance with several previous TFL studies (e.g., Barling et al.,
1996; Vella et al., 2013), it is clear that TFL can be taught to youth sport coaches. Furthermore,
this research provides support for the full-range leadership model as a framework for identifying
a spectrum of leadership behaviours through systematic observation. By complementing
previous CDP and TFL-workshop evaluations that employ quantitative questionnaire measures,
this study adds support for the use of systematic observation in variety of research contexts (e.g.,
intervention research). Further investigation into a variety of observable contexts (e.g., coach-
athlete dyads; Erickson et al., 2011) may be warranted to understand the effects of the
Transformational Coaching workshop on the complexities of sport relationships and behaviours.
It is hoped that this study will generate further interest in the use of systematic observation as a
64
method of evaluating CDPs and TFL-workshops, and continue to provide coaches with the tools
to perform effective behaviours that promote positive athlete development.
65
References
Abrell, C., Rowold, J., Weibler, J., & Moenninghoff, M. (2011). Evaluation of a long-term
transformational leadership development program. German Journal of Human Resource
Management, 25(3), 205-224. doi:10.1177/239700221102500307
Allan, V., & Côté, J. (2016). A cross-sectional analysis of coaches’ observed emotion-behavior
profiles and adolescent athletes’ self-reported developmental outcomes. Journal of
Applied Sport Psychology, 28(3), 321-337. doi:10.1080/10413200.2016.1162220
Allan, V., Turnnidge, J., Vierimaa, M., Davis, P., & Côté, J. (2016). Development of the
Assessment of Coach Emotions systematic observation instrument: A tool to evaluate
coaches’ emotions in the youth sport context. International Journal of Sports Science &
Coaching, 11(6), 859-871. doi:10.1177/1747954116676113
Allan, V., Vierimaa, M., Gainforth, H.L., & Côté, J. (2017). The use of behaviour change
theories and techniques in coach development programmes: A systematic review.
International Review of Sport and Exercise Psychology.
doi:10.1080/1750984X.2017.1286514
Antonakis, J., & House, R. J. (2014). Instrumental leadership: Measurement and extension of
transformational-transactional leadership theory. The Leadership Quarterly, 25(4), 746-
771. doi:10.1016/j.leaqua.2014.04.005
Appleton, P. R., & Duda, J. L. (2016). Examining the interactive effects of coach-created
empowering and disempowering climate dimensions on athletes’ health and functioning.
Psychology of Sport and Exercise, 26(1), 61-70. doi:10.1016/j.psychsport.2016.06.007
Arnold, K. A., Turner, N., Barling, J., Kelloway, E. K., & McKee, M. C. (2007).
Transformational leadership and psychological well-being: The mediating role of
66
meaningful work. Journal of Occupational Health Psychology, 12(3), 193-203.
doi:10.1037/1076-8998.12.3.193
Arthur, C. A., Bastardoz, N., & Eklund, R. (2017). Transformational leadership in sport: Current
status and future directions. Current Opinion in Psychology, 16, 78-83.
doi:10.1016/j.copsyc.2017.04.001
Atwater, L., & Yammarino, F. J. (1989). Transformational leadership among midshipmen
leaders at the United States Naval Academy (Technical report //ONR-TR6). Washington,
DC: Office of Naval Research.
Banjanovic, E. S., & Osborne, J. W. (2016). Confidence intervals for effect sizes: Applying
bootstrap resampling. Practical Assessment, Research & Evaluation, 21(5). Available
online: http://pareonline.net/getvn.asp?v=21&n=5.
Barling, J., & Frone, M. R. (2017). If only my leader would just do something! Passive
leadership undermines employee well-being through role stressors and psychological
resource depletion. Stress & Health, 33(3), 211-222. doi:10.1002/smi.2697
Barling, J., Weber, T., & Kelloway, E.K. (1996). Effects of transformational leadership training
on attitudinal and financial outcomes: A field experiment. Journal of Applied
Psychology, 81(6), 827 – 832. doi:10.1037/0021-9010.81.6.827
Barnett, N. P., Smoll, F. L., & Smith, R. E. (1992). Effects of enhancing coach-athlete
relationships on youth sport attrition. The Sport Psychologist, 6(2), 111-127.
doi:10.1123/tsp.6.2.111
Bass, B.M. (1998). Transformational leadership. Mahwah, NJ: Erlbaum.
Bass, B. M., & Avolio, B. J. (1990). Transformational leadership development: Manual for the
multifactor leadership questionnaire. Palo Alto, CA: Consulting Psychologists Press.
67
Bass, B. M., & Avolio, B. J. (1994). Transformational leadership and organizational culture. The
International Journal of Public Administration, 17(3), 541-554.
doi:10.1080/01900699408524907
Bass, B.M., Avolio, B.J., Jung, D.I., & Berson, Y. (2003). Predicting unit performance by
assessing transformational and transactional leadership. Journal of Applied Psychology,
88(2), 207-218. doi:10.1037/0021-9010.88.2.207
Bass, B.M., & Riggio, R.E. (2006). Transformational leadership (2nd ed.). New York, NY:
Psychology Press.
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than
good. Review of General Psychology, 5(4), 323-370. doi:10.1037/1089-2680.5.4.323
Beauchamp, M. R., Barling, J., Li, Z., Morton, K. L., Keith, S. E., & Zumbo, B. D. (2010).
Development and psychometric properties of the transformational teaching questionnaire.
Journal of Health Psychology, 15, 1123-1134. doi:10.1177/1359105310364175
Beauchamp, M.R., Barling, J., & Morton, K.L. (2011). Transformational teaching and adolescent
self-determined motivation, self-efficacy, and intentions to engage in leisure time
physical activity: A randomized controlled pilot trial. Applied Psychology: Health and
Well-Being, 3(2), 127 – 150. doi:10.1111/j.1758-0854.2011.01048.x
Beauchamp, M. R., Liu, Y., Morton, K. L., Martin, L. J., Wilson, A. H., Wilson, A. J., …
Barling, J. (2014). Transformational teaching and adolescent physical activity: Multilevel
and mediational effects. International Journal of Behavioral Medicine, 21(3), 537-546.
doi:10.1007/s12529-013-9321-2
68
Becker, A. J. (2009). It’s not what they do, it’s how they do it: Athlete experiences of great
coaching. International Journal of Sports Science & Coaching, 4(1), 93-119.
doi:10.1260/1747-9541.4.1.93
Becker, A. (2012). Quality coaching behaviours. In P. Potrac, W. Gilbert, & J. Denison (Eds.),
The Routledge of sports coaching (pp. 184-195). London, UK: Routledge.
Bloom, G. A., Durand-Bush, N., & Salmela, J. H. (1997). Pre- and post competition routines of
expert coaches of team sports. The Sport Psychologist, 11(2), 127-141.
doi:10.1123/tsp.11.2.127
Bloom, G. A., Crumpton, R., & Anderson, J. E. (1999). A systematic observation study of the
teaching behaviors of an expert basketball coach. The Sport Psychologist, 13(2), 157-170.
doi:10.1123/tsp.13.2.157
Burns, J. M. (1978). Leadership, 1978. New Yorker: Harper & Row.
Callow, N., Smith, M.J., 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(4), 395 – 412.
doi:10.1080/10413200903204754
Charbonneau, D., Barling, J., & Kelloway, E.K. (2001). Transformational leadership and sports
performance: The mediating role of intrinsic motivation. Journal of Applied Social
Psychology, 31, 1521 – 1534. doi:10.1111/j.1559-1816.2001.tb02686.x
Chelladurai, P. (2007). Leadership in sports. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook
of sport psychology (3rd edn., pp. 113-135). New York, NY: John Wiley and Sons.
69
Claxton, D. (1988). A systematic observation of more and less successful high school tennis
coaches. Journal of Teaching in Physical Education, 7(4), 302-310.
doi:10.1123/jtpe.7.4.302
Coatsworth, J. D., & Conroy, D. E. (2006). Coach training as a strategy for promoting youth
social development. The Sport Psychologist, 20(2), 128-144. doi:10.1123/tsp.20.2.128
Coe, R. (2002, September). It’s the effect size, stupid: What effect size is and why it is important.
Paper presented at the Annual Conference of the British Educational Research
Association, University of Exeter, England.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ:
Erlbaum.
Cohen, J. (1994). The earth is round (p < .05). American Psychologist, 49(12), 997-1003.
doi:10.1037/0003-066X.49.12.997
Cope, E., Partington, M., Cushion, C. J., & Harvey, S. (2016). An investigation of professional
top-level youth football coaches’ questioning practice. Qualitative Research in Sport,
Exercise and Health, 8(4), 380-393. doi:10.1080/2159676X.2016.1157829
Corrigan, P. W., Diwan, S., Campion, J., & Rashid, F. (2002). Transformational leadership and
the mental health team. Administration and Policy in Mental Health, 30(2), 97-108.
doi:10.1023/A:1022569617123
Côté, J., & Gilbert, W. (2009). An integrative definition of coaching effectiveness and
expertise. International Journal of Sports Science & Coaching, 4(3), 307-323.
doi:10.1260/174795409789623892
Côté, J., & Sedgwick, W. A. (2003). Effective behaviors of expert rowing coaches: A qualitative
investigation of Canadian athletes and coaches. International Sports Journal, 7(1), 62-77.
70
Côté, J., Strachan, L., & Fraser-Thomas, J. (2008). Participation, personal development, and
performance through youth sport. In N. L. Holt (Ed.), Positive youth development
through sport (pp. 34-46). London, UK: Routledge.
Côté, J., Yardley, J., Hay, J., Sedgwick, W., & Baker, J. (1999). An exploratory examination of
the coaching behavior scale for sport. Avante, 5(2), 82-92.
Crocker, L., & Algina, J. (2006). Introduction to classical and modern test theory. New York:
Cengage Learning.
Cumming, S. P., Smoll, F. L., Smith, R. E., & Grossbard, J. R. (2007). Is winning everything?
The relative contributions of motivational climate and won-lost percentage in youth
sports. Journal of Applied Sport Psychology, 19(3), 322-336.
doi:10.1080/10413200701342640
Cushion, C., Harvey, S., Muir, B., & Nelson, L. (2012). Developing the Coach Analysis and
Intervention System (CAIS): Establishing validity and reliability of a computerised
systematic observation instrument. Journal of Sports Sciences, 30(2), 201-216.
doi:10.1080/02640414.2011.635310
Cushion, C. J., & Jones, R. L. (2001). A systematic observation of professional top-level youth
soccer coaches. Journal of Sport Behavior, 24(4), 354-376.
Deci, E. L., & Ryan, R. M. (1985). The general causality orientations scale: Self-determination
in personality. Journal of Research in Personality, 19(2), 109-134. doi:10.1016/0092-
6566(85)90023-6
du Prel, J.-B., Hommel, G., Röhrig, B., & Blettner, M. (2009). Confidence interval or p-value?
Part 4 of a series on evaluation of scientific publications. Deutsches Ärzteblatt
International, 106(19), 335-339. doi:10.3238/arztebl.2009.0335
71
Duda, J. L. (2013). The conceptual and empirical foundations of Empowering Coaching™:
Setting the stage for the PAPA project. International Journal of Sport and Exercise
Psychology, 11, 311-318. doi:10.1080/1612197X.2013.839414
Duda, J. L., Quested, E., Haug, E., Samdal, O., Wold, B., Balaguer, I., ... Cruz, J. (2013).
Promoting Adolescent health through an intervention aimed at improving the quality of
their participation in Physical Activity (PAPA): Background to the project and main trial
protocol. International Journal of Sport and Exercise Psychology, 11(4), 319-327.
doi:10.1080/1612197X.2013.839413
Eagly, A. H., Johannesen-Schmidt, M. C., & Van Engen, M. L. (2003). Transformational,
transactional, and laissez-faire leadership styles: A meta-analysis comparing women and
men. Psychological Bulletin, 129(4), 569-591. doi:10.1037/0033-2909.129.4.569
Eberly, M. B., Johnson, M. D., Hernandez, M., & Avolio, B. J. (2013). An integrative process
model of leadership: Examining loci, mechanisms, and event cycles. American
Psychologist, 68(6), 427-443. doi:10.1037/a0032244
Efron, B., & Tibrshirani, R. (1993). An introduction to the bootstrap. New York, NY: Chapman
& Hall.
Erickson, K., & Côté, J. (2015). The intervention tone of coaches’ behaviour: Development of
the Assessment of Coaching Tone (ACT) Observational Coding System. International
Journal of Sports Science & Coaching, 10(4), 699-716. doi:10.1260/1747-9541.10.4.699
Erickson, K., & Côté, J. (2016). A season-long examination of the intervention tone of coach-
athlete interactions and athlete development in youth sport. Psychology of Sport and
Exercise, 22(1), 264-272. doi:10.1016/j.psychsport.2015.08.006
72
Erickson, K., Côté, J., Hollenstein, T., & Deakin, J. (2011). Examining coach-athlete interactions
using state-space grids: An observational analysis in competitive youth sport. Psychology
of Sport and Exercise, 12(6), 645-654. doi:10.1016/j.psychsport.2011.06.006
Erickson, K., & Gilbert, W. (2013). Coach-athlete interactions in children’s sport. In J. Côté &
R. Lidor (Eds.), Conditions of children’s talent development in sport (pp. 139-156).
Morgantown, WV: Fitness Information Technology.
Evans, M. B., McGuckin, M., Gainforth, H. L., Bruner, M. W., & Côté, J. (2015). Coach
development programmes to improve interpersonal coach behaviours: A systematic
review using the re-aim framework. British Journal of Sports Medicine, 49, 871-877.
doi:10.1136/bjsports-2015-094634
Fraser-Thomas, J., Côté, J., & Deakin, J. (2005). Youth sport programs: An avenue to foster
positive youth development. Physical Education and Sport Pedagogy, 10(1), 19 – 40.
doi:10.1080/1740898042000334890
Frick, P. J., Barry, C. T., & Kamphaus, R. W. (2010). Clinical assessment of child and
adolescent personality and behaviour (3rd ed.). New York, USA: Springer.
Gardner, M. J., & Altman, D. G. (1986). Confidence intervals rather than p values: Estimation
rather than hypothesis testing. British Medical Journal (Clinical Research Education),
292, 746-750. doi:10.1136/bmj.292.6522.746
Gartlehner, G., Hansen, R. A., Nissman, D., Lohr, K. N., & Carey, T. S. (2006). A simple and
valid tool distinguished efficacy from effectiveness studies. Journal of Clinical
Epidemiology, 59(10), 1040-1048. doi:10.1016/j.jclinepi.2006.01.011
73
Gould, D., Dieffenbach, K., & Moffett, A. (2002). Psychological characteristics and their
development in Olympic champions. Journal of Applied Sport Psychology, 14(3), 172-
204. doi:10.1080/10413200290103482
Hamstra, M. R. W., Van Yperen, N. W., Wisse, B., & Sassenberg, K. (2014). Transformational
and transactional leadership and folowers’ achievement goals. Journal of Business and
Psychology, 29(3), 413-425. doi:10.1007/s10869-013-9322-9
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(1), 20-32.
doi:10.1016/j.leaqua.2009.10.002
Haslam, S. A., Reicher, S. D., & Platow, M. J. (2010). The psychology of leadership: Identity,
influence, and power. New York, NY: Psychology Press.
Hochholdinger, S., Rowold, J., & Schaper, N. (2008). Approaches to training and transfer
evaluation. In J. Rowold, S. Hochholdiner, & N. Schaper (Eds.), Evaluation and
transfersicherung operational exercises (pp. 29-53). Göttingen, Germany: Hogrefe.
Hoffman, E. L., & Lord, R. G. (2013). A taxonomy of event-level dimensions: Implications for
understanding leadership processes, behavior, and performance. The Leadership
Quarterly, 24(4), 558-571. doi:10.1016/j.leaqua.2013.03.009
Hollenstein, T., Granic, I., Stoolmiller, M., & Snyder, J. (2004). Rigidity in parent-child
interactions and the development of externalizing and internalizing behavior in early
childhood. Journal of Abnormal Child Psychology, 32, 595-607.
doi:10.1023/B:JACP.0000047209.37650.41
74
Howell, J. M., & Avolio, B. J. (1993). Transformational leadership, transactional leadership,
locus of control, and support for innovation: Key predictors of consolidated-business-unit
performance. Journal of Applied Psychology, 78(6), 891-902. doi:10.1037/0021-
9010.78.6.891
Jowett, S., & Clark-Carter, D. (2006). Perceptions of empathic accuracy and assumed similarity
in the coach-athlete relationship. British Journal of Social Psychology, 45(3), 617-637.
doi:10.1348/014466605X58609
Jowett, S., Paull, G., & Pensgaard, A. M. (2005). Coach-athlete relationship. In: J. Taylor & G.
S. Wilson (Eds.), Applying sport psychology: Four perspectives (pp. 153-170).
Champaign, IL: Human Kinetics.
Kahai, S. S., Sosik, J. J., & Avolio, B. J. (2003). Effects of leadership style, anonymity, and
rewards on creativity-relevant processes and outcomes in an electronic meeting system
context. The Leadership Quarterly, 14(4-5), 499-524. doi:10.1016/S1048-
9843(03)00049-3
Keegan, R. J., Harwood, C. G., Spray, C. M., & Lavallee, D. (2014). A qualitative investigation
of the motivational climate in elite sport. Psychology of Sport and Exercise, 15(1), 97-
107. doi:10.1016/j.psychsport.2013.10.006
Kelloway, E.K., & Barling, J. (2000). What we have learned about developing transformational
leaders. Leadership & Organization Development Journal, 21(7), 355 – 362.
doi:10.1108/01437730010377908
Kelloway, E. K., Barling, J., & Helleur, J. (2000). Enhancing transformational leadership: The
roles of training and feedback. Leadership and Organizational Development Journal,
21(3),145-149. doi:10.1108/01437730010325022
75
Kirk, R. E. (1996). Practical significance: A concept whose time has come. Educational and
Psychological Measurement, 56(5), 746-759. doi:10.1177/0013164496056005002
Kirkbride, P. (2006). Developing transformational leaders: The full-range leadership model in
action. Industrial and Commercial Training, 38(1), 23-32.
doi:10.1108/00197850610646016
Koh, W. L., Steers, R. M., & Terborg, J. R. (1995). The effects of transformational leadership on
teacher attitudes and student performance in Singapore. Journal of Organizational
Behavior, 16, 319–333. doi:10.1002/job.4030160404
Krane, V., Snow, J., & Greenleaf, C. A. (1997). Reaching for gold and the price of glory: A
motivational case study of an elite gymnast. The Sport Psychologist, 11(1), 53-71.
doi:10.1123/tsp.11.1.53
Lacy, A. C., & Goldston, P. D. (1990). Behaviour analysis of male and female coaches in female
basketball. Journal of Sport Behavior, 13(1), 29-39.
Larsen, T., Van Hoye, A., Tjomsland, H. E., Holsen, I., Wold, B., Heuzé, J.-P., ... Sarrazin, P.
(2015). Creating a supportive environment among youth football players: A qualitative
study of French and Norwegian youth grassroots football coaches. Health Education,
115(6), 570-586. doi:10.1108/HE-04-2014-0054
Lefebvre, J., & Côté, J. (2017). A systematic observation of coach leadership behaviours in
youth sport (Unpublished Master’s thesis). Queen’s University, Kingston, ON.
Lefebvre, J., Evans, M. B., Turnnidge, J., Gainforth, H. L., & Côté, J. (2016). Describing and
classifying coach development programmes: A synthesis of empirical research and
applied practice. International Journal of Sports Science & Coaching, 11(6), 887-899.
doi:10.1177/1747954116676116
76
Lehmann-Willenbrock, N., Meinecke, A. L., Rowold, J., & Kauffeld, S. (2015). How
transformational leadership works during team interactions: A behavioral process
analysis. The Leadership Quarterly, 26(6), 1017-1033. doi:10.1016/j.leaqua.2015.07.003
Lerner, R. M., & Busch-Rossnagel, N. A. (1981). Individuals as producers of their development:
Conceptual and empirical bases. In R. M. Lerner & N. A. Busch-Rossnagel (Eds.),
Individuals as producers of their development: A life-span perspective (pp. 1-36). New
York, NY: Academic Press, Inc.
Litwin, M. S. (2002). How to assess and interpret survey psychometrics. Thousand Oaks, CA:
Sage Publications.
Lord, R. G., Brown, D. J., & Freiberg, S. J. (1999). Understanding the dynamics of leadership:
The role of follower self-concepts in the leader/follower relationship. Organizational
Behavior and Human Decision Processes, 78(3), 167-203. doi:10.1006/obhd.1999.2832
Ma, Y., Mazumdar, M., & Memtsoudis, S. G. (2012). Beyond Repeated measures ANOVA:
Advanced statistical methods for the analysis of longitudinal data in anesthesia research.
Regional Anesthesia and Pain Medicine, 37(1), 99-105.
doi:10.1097/AAP.0b013e31823ebc74
Mageau, G.A., & Vallerand, R.J. (2003). The coach-athlete relationship: A motivational model.
Journal of Sports Science, 21(11), 883-904. doi:10.1080/0264041031000140374
Mageau, G. A., Vallerand, R. J., Charest, J., Salvy, S. J., Lacaille, N., Bouffard, T., & Koestner,
R. (2009). On the development of harmonious and obsessive passion: The role of
autonomy support, activity specialization, and identification with the activity. Journal of
Personality, 77(3), 601-646. doi:10.1111/j.1467-6494.2009.00559.x
77
Mallett, C. J., Trudel, P., Lyle, J., & Rynne, S. B. (2009). Formal vs. informal coach education.
International Journal of Sports Science & Coaching, 4(3), 325-334.
doi:10.1260/174795409789623883
Marsden, E., & Torgerson, C. J. (2012). Single group, pre- and post-test research designs: Some
methodological concerns. Oxford Review of Education, 38(5), 583-616.
doi:10.1080/03054985.2012.731208
Michie, S., & Johnston, M. (2012). Theories and techniques of behaviour change: Developing a
cumulative science of behaviour change. Health Psychology Review, 6(1), 1-6.
doi:10.1080/17437199.2012.654964
Michie, S., Richardson, M., Johnston, M., Abraham, C., Francis, J., Hardeman, W., … Wood,
C.E. (2013). The behaviour change technique taxonomy (v1) of 93 hierarchically
clustered techniques: Building an international consensus for the reporting of behaviour
change interventions. Annals of Behavioural Medicine, 46(1), 81-95.
doi:10.1007/s12160-013-9486-6
Michie, S., van Stralen, M.M., & West, R. (2011). The behaviour change wheel: A new method
for characterizing and designing behaviour change interventions. Implementation
Science, 6(42), doi:10.1186/1748-5908-6-42
More, K. G., & Franks, I. M. (1996). Analysis and modification of verbal coaching behaviour:
The usefulness of a data-driven intervention strategy. Journal of Sports Sciences, 14(6),
523-543. doi:10.1080/02640419608727739
Nicholls, J. G. (1989). The competitive ethos and democratic education. London, UK: Harvard
University Press.
78
Oberfield, D., & Franke, T. (2013). Evaluating the robustness of repeated measures analyses:
The case of small sample sizes and nonnormal data. Behavior Research Methods, 45(3),
792-812. doi:10.3758/s13428-012-0281-2
Partington, M., & Cushion, C. (2013). An investigation of the practice activities and coaching
behaviors of professional top-level youth soccer coaches. Scandinavian Journal of
Medicine & Science in Sports, 23(3), 374-382. doi:10.1111/j.1600-0838.2011.01383.x
Pelletier, K.L. (2010). Leader toxicity: An empirical investigation of toxic behaviour and
rhetoric. Leadership, 6(4), 373-389. doi:10.1177/1742715010379308
Petitpas, A., Cornelius, A.E., Van Raalte, J.L., & Jones, T. (2005). A framework for planning
youth sport programs that foster psychosocial development. The Sport Psychologist,
19(1), 63-80. doi:10.1123/tsp.19.1.63
Quested, E. J., Appleton, P. R., Wold, B., Balauger, I., Sarrazin, P., ... Cruz, J. (2015). The
impact of the Empowering Coaching training programme on the motivational climate
and player responses. 14th FEPSAC Congress, Bern, Switzerland.
Rowold, J. (2006). Transformational and transactional leadership in martial arts. Journal of
Applied Sport Psychology, 18(4), 312-325. doi:10.1080/10413200600944082
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental
designs for generalized causal inference. Boston, USA: Houghton Mifflin Company.
Skogstad, A., Einarsen, S., Torsheim, T., Aasland, M. S., & Hetland, H. (2007). The
destructiveness of laissez-faire leadership behavior. Journal of Occupational Health
Psychology, 12(1), 80-92. doi:10.1037/1076-8998.12.1.80
79
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(2), 249-257. doi:10.1016/j.psychsport.2012.10.002
Smith, N., Quested, E., Appleton, P. R., & Duda, J. L. (2017). Observing the coach-created
motivational environment across training and competition in youth sport. Journal of
Sports Sciences, 35(2), 149-158. doi:10.1080/02640414.2016.1159714
Smith, R. E., Smoll, F. L., & Cumming, S. P. (2007). Effects of a motivational climate
intervention for coaches on young athletes' sport performance anxiety. Journal of Sport
and Exercise Psychology, 29(1), 39-59. doi:10.1123/jsep.29.1.39
Smith, R. E., Smoll, F. L., & Curtis, B. (1979). Coach effectiveness training: A cognitive-
behavioural approach to enhancing relationship skills in youth sport coaches. Journal of
Sport Psychology, 1(1), 59-75. doi:10.1123/jsp.1.1.59
Smith, R. E., Smoll, F. L., & Hunt, E. (1977). A system for the behavioural assessment of
athletic coaches. Research Quarterly – American Alliance for Health, Physical
Education, and Recreation, 48(2), 401-407.
Smith, N., Tessier, D., Tzioumakis, Y., Fabra, P., Quested, E., Appleton, P., ... Duda, J. L.
(2016). The relationship between observed and perceived assessments of the coach-
created motivational environment and links to athlete motivation. Psychology of Sport
and Exercise, 23, 51-63. doi:10.1016/j.psychsport.2015.11.001
Smith, N., Tessier, D., Tzioumakis, Y., Quested, E., Appleton, P., Sarrazin, P., ... Duda, J. L.
(2015). Development and validation of the Multidimensional Motivational Climate
Observation System. Journal of Sport & Exercise Psychology, 37(1), 4-22.
doi:10.1123/jsep.2014-0059
80
Smith, N., Tzioumakis, Y., Tessier, D., Fabra, P., Sarrazin, P., ... Duda, J. L. (2015). The
objective motivational climate in grassroots football: Measurement development and
intervention effects. 14th FEPSAC Congress, Bern, Switzerland.
Smith, R. E., Zane, N. W. S., Smoll, F. L., & Coppel, D. B. (1983). Behavioral assessment in
youth sports: Coaching behaviors and children`s attitudes. Medicine and Science in
Sports and Exercise, 15(3), 208-214. doi:10.1249/00005768-198315030-00005
Smoll, F. L., & Smith, R. E. (1989). Leadership behaviors in sport: A theoretical model and
research paradigm. Journal of Applied Social Psychology, 19(18), 1522-1551.
doi:10.1111/j.1559-1816.1989.tb01462.x
Smoll, F. L., Smith, R. E., Barnett, N. P., & Everett, J. J. (1993). Enhancement of children’s self-
esteem through social support training for youth sport coaches. Journal of Applied Sport
Psychology, 78(4), 602-610. doi:10.1037/0021-9010.78.4.602
Smoll, F. L., Smith, R. E., & Cumming, S. P. (2007). Effects of coach and parent training on
performance anxiety in young athletes: A systematic approach. Journal of Youth
Development, 2(1), 20-38. doi:10.5195/jyd.2007.358
Smoll, F. L., Smith, R. E., Curtis, B., & Hunt, E. (1978). Toward a mediational model of coach-
player relationships. Research Quarterly – American Alliance for Health, Physical
Education and Recreation, 49(4), 528-541.
Stuntz, C. P., & Spearance, A. L. (2010). Cross-domain relationships in two sport populations:
Measurement validation including prediction of motivation-related variables. Psychology
of Sport and Exercise, 11(4), 267-274. doi:10.1016/j.psychsport.2010.02.007
81
Tejeda, M. J., Scandura, T. A., & Pillai, R. (2001). The MLQ revisited: Psychometric properties
and recommendations. The Leadership Quarterly, 12(1), 31-52. doi:10.1016/S1048-
9843(01)00063-7
Torgerson, C., & Torgerson, D. (2008). Designing and running randomised trials in health,
education and the social sciences: An introduction. Basingstoke, UK: Palgrave
Macmillan.
Turnnidge, J., & Côté, J. (2016a). Applying transformational leadership theory to coaching
research in youth sport: A systematic literature review. International Journal of Sport
and Exercise Psychology, 1-16. doi:10.1080/1612197X.2016.1189948
Turnnidge, J., & Côté, J. (2016b). Observing coaches’ leadership behaviours in sport: The
development of the Coach Leadership Assessment System (CLAS). Journal of Sport and
Exercise Psychology, 38, S265.
Turnnidge, J., & Côté, J. (2017a). Transformational coaching workshop: Applying a person-
centered approach to coach development programs. International Sport Coaching
Journal, 4(3), 314-325. doi:10.1123/iscj.2017-0046
Turnnidge, J., & Côté, J. (2017b). An exploration of coaches’ leadership behaviours in youth
sport (Unpublished doctoral dissertation). Queen’s University, Kingston, ON.
Turnnidge, J., Côté, J., Hollenstein, T., & Deakin, J. (2014). A direct observation of the dynamic
content and structure of coach-athlete interactions in a model sport program. Journal of
Applied Sport Psychology, 26(2), 225-240. doi:10.1080/10413200.2013.821637
Turnnidge, J., Evans, B.E., Vierimaa, M., Allan, V., Côté. (2016). Coaching for positive youth
development. In N.L. Holt (Ed.), Positive youth development through sport (p.137-150).
New York, NY: Routledge.
82
Turnnidge, J., Barling, J., Bruner, M. W., Erickson, K., Gainforth, H., Hollenstein, T., & Côté, J.
(2016, November). The development of a transformational coaching workshop:
Empirical findings and recommendations for practice. Poster presented at the Coaching
Association of Canada Sport Leadership Conference, Richmond, BC, Canada.
Vella, S. A., Oades, L. G., & Crowe, T. P. (2010). The application of coach leadership models to
coaching practice: Current state and future directions. International Journal of Sports
Science & Coaching, 5(3), 425-434. doi:10.1260/1747-9541.5.3.425
Vella, S. A., Oades, L. G., & Crowe, T. P. (2012). Validation of the Differentiated
Transformational Leadership Inventory as a measure of coach leadership in youth soccer.
The Sport Psychologist, 26(2), 207-223. doi:10.1123/tsp.26.2.207
Vella, S., Oades, L., & Crowe, T. (2013). A pilot test of transformational leadership training for
sports coaches: Impact on the developmental experiences of adolescent athletes.
International Journal of Sports Science and Coaching, 8(3), 513-530. doi:10.1260/1747-
9541.8.3.513
Visek, A. J., Achrati, S. M., Mannix, H. M., McDonell, K., Harris, B. S., & DiPietro, L. (2015).
The fun integration theory: Toward sustaining children and adolescents sport
participation. Journal of Physical Activity & Health, 12(3), 424-433.
doi:10.1123/jpah.2013-0180
Webster, C. A., Wellborn, B., Hunt, K., LaFleche, M., Cribbs, J., & Lineberger, B. (2013).
MPOWER: An observation system for assessing coach autonomy support in high school
varsity boys’ soccer practices. International Journal of Sports Science & Coaching, 8(4),
741-754. doi:10.1260/1747-9541.8.4.741
83
Wilson, A. J., Liu, Y., Keith, S. E., Wilson, A. H., Kermer, L. E., Zumbo, B. D., & Beauchamp,
M. R. (2012). Transformational teaching and child psychological needs satisfaction,
motivation, and engagement in elementary school physical education. Sport, Exercise,
and Performance Psychology, 1(4), 215-230. doi:10.1037/a0028635
84
Appendix A
General Coding Guidelines
Overview
The Coach Leadership Assessment System (CLAS) has been designed so that it can be used to
code videos that have been uploaded into a computer program (e.g., Noldus Observer). When
using the tool, coders review videos of coaches – pausing the video to code each coach
behaviour. This tool is intended for use with a continuous coding approach and thus every
possible type of coach behaviour must be classified into one of the categories of this instrument.
Each behaviour is then coded according to the individual initiating the behaviour (e.g., coach)
and the leadership tone of the behaviour (i.e., the primary code). In addition, coders complete
ratings to further describe the behaviour, and indicate the more specific content of the behaviour,
as well as with the recipient (i.e., athlete) of the behaviour and the context in which the
behaviour occurs (i.e., scrimmage).
This instrument is designed to be able to capture leadership behaviours in a variety of sports
(e.g., volleyball, soccer, swimming, synchronized swimming, beach volleyball, hockey,
basketball, etc.) and sport activities (e.g., competition, practices).
Coding sessions will involve coding the continuous behaviour of the coach during a 15-minute
video clip. The amount of time required to code each 15-minute segment in its entirety is highly
variable, and may depend on factors such as the experience of the coder or the content of the
particular segment.
The codes in the CLAS are based on classification of behaviours across two (2) dimensions:
1. Coach Behaviour, which includes:
a. The initiator subject (i.e., whose behaviour is being coded)
b. The leadership tone of a given interactive behaviour
c. The content modifier of a given interactive behaviour
d. The recipient of a given coach behaviour (i.e., to whom a coded behaviour is
directed).
2. Context (the training or competition activity context in which a given coach behaviour
takes place).
Rules
• 3-second rule
o Wait three (3) seconds before coding ‘neutral’ (leadership tone) when changing
from any actively communicative code. Code for this behaviour only if it
continues past the three (3) second waiting period, at which point you would
rewind the video three seconds and begin coding it at its true initiation point. If
within three (3) seconds a different actively communicative behaviour occurs, do
not wait to code that behaviour.
85
o Wait three (3) seconds before coding ‘observation’ (content modifier) when
changing from any actively communicative code. Code for this behaviour only if
it continues past the three (3) second waiting period, at which point you would
rewind the video three seconds and begin coding it at its true initiation point. If
within three (3) seconds a different actively communicative behaviour occurs, do
not wait to code that behaviour.
o Wait three (3) seconds before coding ‘uncodable’ when changing from any other
code. Code for this behaviour only if it continues past the three (3) second waiting
period, at which point you would rewind the video three seconds and begin
coding it at its true initiation point. If within three (3) seconds a different
behaviour visibly or audibly occurs, do not wait to code that behaviour.
• Default codes
o For the coach leadership tone dimension, specific behaviour codes are to be coded
by default if criteria for any other behaviour within the dimension are not met.
That is, use the default codes in the absence of any other codable behaviour:
• Leadership tone: 14 (Neutral)
o However, priority is given to more extreme codes in the leadership tone
dimension. If deciding between the default code and an active code, always select
the more extreme code
• E.g., if deciding between ‘neutral’ and ‘transactional’ leadership tone,
code ‘transactional’
o For the content modifier dimension, the default code to be used in the absence of
any other codable behaviour is:
• If the coach is not actively communicating with any of the athletes,
Leadership tone + content modifier: 14 (Neutral), 4 (Observation)
• If the coach is actively communicating with any of the athletes,
Leadership tone + content modifier: 14 (Neutral), 3 (General
Communication)
o For the recipient modifier, the default code is:
• Team (if behaviour is directed to two or more athletes)
o No default categories exist for the context dimension as this must be directly
observed.
Subject – Initiator or Recipient (letters)
As there are multiple participants in all videos (i.e., individual athletes and the coach), the coder
must specify which subject’s behaviour is being coded. Once assigned a subject ID, athletes must
be coded as same subject for all videos. Note: Athletes could be coded individual (i.e., athlete A,
athlete b, etc.) or could be coded as individual vs. team, depending on the research question or
design. The subject codes can be used as follows:
86
CODE
z – Coach
a – Athlete A
b – Athlete B
c – Athlete C
d – Athlete D
e – Athlete E
f – Athlete F
g – Athlete G
h – Athlete H
i – Athlete I
j – Athlete J
etc.
Dimension – Leadership Behaviour
Overview
• The leadership behaviour dimension is comprised of (a) leadership tone and (b) a class of
coach behaviour content modifiers.
• Each leadership behaviour code (i.e., all codes other than uncodable) is linked to the
content modifier codes. For every observed leadership behaviour, a leadership tone code
and a content modifier code MUST be scored. Thus, each observed behaviour is
categorized by the combination of two (2) codes – a leadership tone code followed by
modifier code(s) (e.g., “Discussing/modelling pro-social values or behaviours + General
Communication” or “Eliciting athlete input + Instruction/Feedback, etc.)
• If there is a change in any of these codes (leadership tone OR content modifier), begin a
new entry and code as new independent coach behaviour. Thus, if the coach begins with
“Discussing/modelling pro-social values or behaviours + General Communication” and
moves immediately to “Discussing/modelling pro-social values or behaviours +
Instruction/Feedback” in the same continuous interaction, code as two (2) separate
behaviours.
• Leadership tone behaviours and content modifier codes are intended to encompass
BOTH verbal and non-verbal behavioural indicators. For instance, the leadership tone
code “Recognizing athlete achievements/contributions” could include giving high-fives, a
thumbs up, etc., whereas the content modifier code “Organization” could include whistle-
blowing, etc. For non-verbal behaviours, they must be easily identifiable (i.e., there must
be a definite behavioural cue).
• For the leadership tone behaviour dimension, please use theoretical constructs (i.e., the 4
I’s, transactional, etc.) as a general guide for interpreting the “general message” of the
87
behaviour. For instance, when deciding whether a coach behaviour truly “fits” with a
particular behavioural code, such as eliciting athlete input, it may be useful to assess
whether this behaviour aligns with the general concept of intellectual stimulation.
LEADERSHIP TONE
Idealized Influence: Behaviours conveying the coach as (a) a positive role model, (b) an
individual of high moral/ethical standing, or (c) trustworthy and respected.
Categories:
▪ 01- Discussing/modelling pro-social values or behaviours
o Prosocial values/behaviours generally refer to values/behaviours that are
intended to benefit others and that are prompted by empathy, morality, or a
sense of social responsibility, rather than a desire for personal gain.
o Can include general social or moral topics (e.g., displaying respect,
supporting others, empathy/understanding, etc.).
o Can include deliberate attempts to foster pro-social attitudes or skills among
the athletes, etc. (e.g., teaching responsibility, highlighting the importance
of assisting teammates, etc.).
o Can include humour-based behaviours (e.g., humor as an initial ice-breaking
method, a stress-relieving method, or a means of motivation, team
communication, energy, or enjoyment promotion). Note: humour that is
mean-spirited, sarcastic, or at the expense of others should not be included
here, but in modelling anti-social behaviours; self-deprecating humour
should not be included here, but in showing vulnerability/humility.
o Can also include non-verbal behaviours (e.g., helping athletes gather their
equipment).
o E.g., “It is really important that we stay friendly and respectful on the
court.”
▪ 02-Showing vulnerability/humility
o Discussions where they recognize gaps in their knowledge, understanding,
and may involve asking an athlete for help.
o Can include admitting to, or apologizing, for mistakes.
o Can include discussing personal information with athletes (e.g., telling
stories where they felt discouraged, saying they’ve also had bad days,
sharing that they get nervous too, etc.).
o E.g., “Sorry guys, I messed up and gave you the wrong set.”
Inspirational Motivation: Behaviours through which a coach demonstrates that they hold (a)
high expectations for their athletes, or (b) a compelling vision of the future for either individual
athletes or the team as a whole. Also includes behaviours through which a coach promotes team
spirit, enthusiasm, and meaning/challenge.
88
Categories:
▪ 03-Discussing goals/expectations
o Expectations can be for a particular training session, a particular drill, or as
a part of a larger picture, such as an upcoming game or goals for the season.
o Can include discussion of goal(s), goal setting, etc. (can be for a particular
training session, a particular drill, or as a part of a larger picture, such as an
upcoming game or goals for the season). Can also involve asking athletes to
write down or vocalize their own goals.
o E.g., “For this drill, I want to see you guys giving 100 percent.
▪ 04-Expressing confidence in athlete potential
o Talking optimistically/enthusiastically about what the athlete(s) can achieve.
o Providing challenging task(s), etc. (E.g., “I think you guys can handle this
higher intensity, so we’re going to go for it today.”
o E.g., “I know you guys can do this.”
▪ 05-Promoting team concept
o Encouraging team spirit/attitude towards team members.
o E.g., team chants, discussing the importance of coming together as a team,
teamwork, etc.
o Can also include clarifying roles with team, discussing team issues
o Note: This category involves behaviours that involve a collective
goal/vision etc. If the behaviours involve a moral/ethical element (i.e., this is
how we should behave towards others), it should be coded as 01-
discussing/modelling pro-social values or behaviours.
o E.g., “Come on everyone, it’s ‘let’s go Vikings’ on three.”
▪ 06-Providing rationales/explanations
o Behaviours through which the coach highlights the value/meaning of certain
activities/drills (i.e., “This drill is important because. . . ”).
o Can include providing reasoning behind decisions (i.e., highlighting the
method behind the madness).
o Can include connecting activities to a larger picture (e.g., connecting
particular activities to athlete/team goals; “This will help us reach finals”).
o E.g., “It’s really important that we get this defensive drill right because you
know our opponents on Saturday are defensive all-stars.”
Intellectual stimulation: Behaviours that convey a view of the athlete(s) as capable decision
makers and contributing members of the situation. Also includes behaviours that encourage
athletes to think and act in novel and creative ways.
Categories:
▪ 07-Eliciting athlete input
89
o Questioning. Must allow an answer reflecting athlete input. These
questions should require a higher level of thinking. For instance, asking
critical questions regarding practice activities or social issues.
o Can involve encouraging athletes to (a) solve problems and to look for
alternative solutions, (b) have open discussions, and (c) contribute new and
alternative ideas.
o Note: This category relates to coach-initiated athlete input. If the athlete
offers input and the coach listens and/or incorporates their input, this should
be coded as 10-Showing interest in athletes’ feelings/needs/concerns.
Questions that do not require a higher level of thinking should not be
included (e.g., How many sets have you finished?), code as 14-Neutral
rather than 07- Eliciting athlete input.
o E.g., “How can we use what we have learned from this drill to make us more
successful in our games?” or “What can we learn from a drill/game when
things did not go as expected?”
▪ 08-Sharing decision making/leadership responsibilities
o Can involve providing athletes with choice(s) such as providing different
drill/task options.
o Can involve offering opportunities to show initiative, leadership, etc. (i.e.,
demonstrating skills for teammates, leading a warm-up, helping younger
athletes etc.).
o E.g., “Today, it’s Maddie’s turn to lead the warm-up set. She will decide the
stroke and distance.”
▪ 09-Emphasizing the learning process
o Encouraging or recognizing athlete(s) who seek or engage in challenging
tasks.
o Can include encouraging athlete(s) after mistakes or discussing the value of
mistakes.
o Can involve behaviours that emphasize effort.
o E.g., “That was a really great try Amy, mistakes like that only help us get
better.”
Individualized consideration: Behaviours through which a coach recognizes an athlete’s
individual needs, considers their unique abilities, and displays genuine care and concern.
Categories:
▪ 10-Showing interest in athlete feelings/needs/concerns
o Adapting activities to suit the needs of the athlete(s).
o Listening to athlete(s) and considering/incorporating their opinions.
o Can include discussing personal issues with the athlete(s).
o Can include referencing to past events, interactions, etc.
o E.g., “I know you weren’t feeling well yesterday, how are you today?”
90
▪ 11-Recognizing athlete achievements/contributions
o Can include thanking the athletes for their hard work, help, etc.
o Note: Recognition should have some level of specificity (i.e., a particular
performance or a particular athlete) and a higher degree of enthusiasm. For
example, a passive “Good job” would not fit with this category and would
be coded as 14-Neutral.
o E.g., “That’s excellent, Jamie! Fantastic job on the turn!”
Transactional: Behaviours that reinforce standards/expectations through rewards or
punishments.
Categories:
▪ 12-Discussing rewards/penalties
o Examples: If ___, then ___ statements.
o Clarifying, negotiating, and tying specific rewards/penalties to performance.
o E.g., “If you guys don’t complete this set properly, then everyone is going to
run laps”; “If you try one more time, then we can have a scrimmage.”
▪ 13-Searching for/responding to deviations from rules or standards
o Focusing on errors/mistakes (could include negative reactions to undesirable
athlete(s) behaviours).
o E.g., “Stop blowing bubbles and get back on task.”
14-Neutral: Absence of leadership related tone.
Notes
• Only code if no criteria from any other category is met.
• If the behaviour seems to meet any of the other criteria, choose the more active
category (i.e., categories other than neutral).
Laissez-Faire: Behaviours that convey coach’s disinterest in or ambivalence towards the
athletes or practice activities.
Categories:
▪ 15-Showing Disinterest
o Not paying attention to the athlete(s) or practice (e.g., seeming distracted).
o E.g., Playing with music, talking with others (assistant coaches, parents)
about non-relevant (i.e., not sport related) matters.
o Avoiding action.
o Note: Leaving the coaching area (and consequently being out
view/inaudible should be coded as x-Uncodable, NOT laissez-faire/showing
disinterest.
Toxic Leadership: Behaviours that convey that a coach holds negative attitudes/feelings
towards the athlete(s).
91
Categories:
▪ 16-Expressing anger/hostility
o Can include threats, intimidation (e.g., yelling).
o Can involve both verbal (e.g., threats) or non-verbal (body language,
shaking fists, etc.).
o E.g., “Stop what you’re doing right now or you’ll be sorry!”
▪ 17-Modelling anti-social behaviours
o Examples: Being rude, sarcasm, swearing.
o Can also include criticising, belittling, ridiculing, insulting, devaluing
athlete(s) input, making negative comments about athlete(s) to others.
o Can include verbal or non-verbal behaviours.
o Note: These behaviours may not necessarily be delivered in an angry/hostile
tone.
o Can include excluding athlete(s) from particular drills/activities.
o E.g., “Joey, that’s a terrible idea. No wonder you didn’t make the team last
year.”
MODIFIERS
1-Instruction/Feedback: Technical and/or tactical and/or teaching instruction or feedback from
coach, directed at athlete(s) motor performance or skill execution. Also includes
communication from coach related to individual mental/psychological skills,
characteristics, qualities, or aspects of performance
Notes
• Includes prescriptive/corrective technical information in reference to the quality of
the movement or skill execution (e.g., how it should be performed, what could be
improved, etc.)
• Can be directed at general psychological topics related to performance (e.g.,
confidence, focus, mental toughness, etc.)
2-Organization: Communication from coach related to organization of practice tasks and athlete
actions, NOT intended to directly influence performance.
Notes
• E.g., “Now we’re doing ___ drill”, “Go over there”, “Do 10 of these”, etc.
• Can include discipline, keeping control, etc.
• Can include timing or counting during skill execution/drills.
• Can include non-verbal behaviours (e.g.,
• CANNOT include any technical instruction related to movement quality (code 2).
Code for each separately, even if these behaviours occur in immediate sequence.
3-General communication: Communication from coach not directly related to task,
performance, or organization in the current team/training/performance context.
Notes
92
• Default code if coach is actively interacting with athlete(s) but criteria is not met for
other conversational categories (i.e., Instruction/feedback or Organization codes).
• E.g., joking with athletes, talking about school, etc.
4-Observation: Coach engaged in observing/watching athletes during training/performance
activities, though not directly communicating with athletes.
Notes
• Default code if coach is engaged in training/competition activities, but criteria is not
met for any actively communicative code.
• 3-second rule in effect before coding for ‘observation’ from an active
communication code.
MODIFIER – TARGET
• Individual athletes (ind or athlete a, b, c, etc.). Depending on the depth of coding,
individual athletes can be coded broadly as ind or individually identified as athlete, b, c,
d, etc.)
• Team: Default code if coach behaviours are not targeted towards anyone in particular or
if behaviours are targeted towards 2 or more athletes.
• Other: Only use for assistant coaches when discussing matters that are relevant to the
sport. If discussing non-sport related matters with an assistant coach or any other person
(e.g., ref, lifeguard, parents, friends, etc.), code as 15-Showing disinterest.
CONTEXT
• 1-Warm-up or cool-down
• 2-Structured drills/exercises
• 3-Instruction (if athlete(s) is stationary/listening to coach)
• 4-Scrimmage
• 5-Free play
• 6-Break – e.g., water break
93
Appendix B
Research Ethics Board Letter of Approval
94
Appendix C
Coach Letter of Information and Consent Form
COACH LETTER OF INFORMATION: WORKSHOP PARTICIPATION
Title of the study: Exploring Leadership in Youth Sport: A Transformational Perspective
I would like to invite you to take part in a research project. This research is being conducted by Sarah
Lawrason, MSc candidate, and Jennifer Turnnidge, PhD candidate, under the supervision of Dr. Jean Côté
in the School of Kinesiology and Health Studies, at Queen’s University in Kingston, Ontario. This study
has been granted clearance by the General Research Ethics Board according to Canadian research ethics
principles (http://www.ethics.gc.ca/default.aspx) and Queen's University policies
(http://www.queensu.ca/urs/research-ethics).
What is this study about? The purpose of this study is to develop and test a transformational coaching
workshop for coaches of athletes aged 12-17 in sport. This workshop is designed to enhance coaches’ use
of transformational leadership behaviours in youth sport.
What does participating in this study involve?
Pre/Post Workshop Component: Members of the research team will visit your practice facility twice
during the season: One time before the workshop and one time following your completion of the
workshop. During each of these visits, the practice session will be videotaped using two different video
cameras, one of which will be set up with a wide angle capturing the broad play area and the other of
which will be set up to capture coach-athlete interactions in greater detail. You will be asked wear a
wireless microphone, while a second microphone will be connected to one of the video cameras. The
videotapes will be watched by members of the research team to understand the dynamics of coaches’
leadership-based interactions. At each of the recorded practices, you will be asked to complete a
questionnaire about your leadership behaviours (before or after practice, based on your preference). The
athletes you coach will also be asked to complete questionnaires in a separate room that will ask them to
reflect on their experiences within the sport environment. We anticipate that ¼ of the athletes on your
team will complete the questionnaires. The questionnaires will be completed on-site and should take
approximately 30 minutes to complete. A member of the research team will be on-site to answer any
questions and to pick-up all questionnaires. There will be no deception used in this study.
Workshop Component: You will be asked to participate in a three-hour workshop relating to coaches’
use of transformational leadership behaviours in the youth sport environment. This workshop will consist
of lectures, discussions, and reflective activities designed to help you develop strategies for using
transformational leadership behaviours in your own coaching practice. Following the three-hour
workshop, you will be asked to complete a 30-minute questionnaire to provide feedback on the content
and delivery of the workshop material. The workshop will be administered free of charge and will take
place in the location of your organization’s choice either (a) on Queen’s campus or (b) at your practice
facility.
Optional Interview Component: One month following your completion of the workshop, you
will have the option of participating in a one-on-one, 1-hour interview to gain a deeper
understanding of your experiences in the workshop. The interview will take place in person at a
location of your choice, either at the interviewer’s office or in a private room at a practice facility
(while other co-workers are present in the building). As part of this research, an audio recording
device will be used to record your interview. It will allow the interview to be transcribed at a
later date. In the event that you participate in this interview, a follow-up email will be sent for
95
verification of the transcript of the interview. You will be asked for your permission to use an
audio recorder within the attached consent form. You are not obligated to participate in this
phase in order to attend the workshop. The total time commitment of the study will be 7.5 hours
for all three components (6.5 hours without the optional interview component).
Is my participation voluntary? Yes. Please answer questionnaire items and interview questions
as honestly as possible, but you are not obliged to answer any material that you find
objectionable or that makes you feel uncomfortable. You may withdraw at any time, for any
reason, without explanation from the study with no repercussions for yourself, your team, or
your league. If you choose to withdraw from the study you may choose to have your data
removed from analysis. Questionnaire data and transcripts will be destroyed and your video data
will be deleted. To withdraw from the study, please contact Sarah Lawrason or Jennifer
Turnnidge (contact information provided below).
What are the risks of participating? There are minor risks associated with participating this study as
you will be asked personal questions about your coaching style in the questionnaires and you will be
asked to reflect on your coaching experiences. Although there are minor risks, we will do everything
possible to minimize these risks. You are not required to answer any material that you do not feel
comfortable with. It is also important to highlight that should you agree to participate in this study, there
will be other coaches present at the workshop. Therefore, while we will make every effort to ensure that
your responses to the questionnaires and interview (if applicable) will remain confidential, we cannot
guarantee your confidentiality at the workshop. We will ask those involved in the workshops to keep the
personal information (e.g., names) of the coaches confidential.
What are the benefits of participating? We cannot guarantee or promise that you will receive
any direct benefits from being in the study. The results of the study will, however, contribute to a
better understanding of designing effective leadership-based coaching workshops. The study has
the potential to influence policy directions and services in relation to coach education and youth
development in sport.
What will happen to my responses? Your responses will be kept confidential to the extent
possible. With regards to the questionnaire data, the information you provide us will be de-
identified and given a code so that no one will be able to identify you from the information
unless they have the participant code key. The only individuals with access to identifying
information include Sarah Lawrason, Jennifer Turnnidge and Jean Côté. Members of the
research team will have access to the videos, de-identified questionnaire data, and interview
transcripts– all of whom are required to complete a confidentiality agreement. All the
information provided through the interviews, questionnaires, videotapes, and consent forms will
be confidential and will be stored in a controlled-access location (e.g., locked office, password
protected files) in the School of Kinesiology and Health Studies at Queen’s University for a
minimum of seven years after the completion of the study. While the information collected may
be presented at academic conferences and published in relevant academic journals,
confidentiality of all participants will be maintained. All images of athletes and coaches will be
viewed only by the research team during the analysis of the data. No images will be shown in
conference presentations or in any publications. All personal contact information and the
participant code key will be destroyed after it is no longer required to authenticate study results.
Will I be compensated for participating? We would like to show our gratitude by offering you
a clipboard (approximate value $25) for use in your practice sessions and a resource book of the
96
material covered in the workshop. Beverages/snacks will also be provided during the workshop.
We would also like to share with you our study results when they are ready for publication. If
you are interested in receiving a copy, please indicate so on the consent form. If you choose to
withdraw from the study, you are still entitled to the clipboard and to receive a copy of the
study’s findings.
If you decide that you would like to be a part of this study, please complete the attached Consent
Form. Please retain a copy of this Letter of Information and the Consent Form for your records.
Any questions about study participation may be directed to Sarah Lawrason and/or Jennifer
Turnnidge at 613-533-6000, Ext. 78207 or [email protected], [email protected] or Jean Côté
at [email protected] or 613-533-6000 Ext. 79049. If you have any ethical concerns or
complaints, you may contact the Chair of the Queen’s University General Research Ethics Board
at 1-844-535-2988 or [email protected].
PARTICIPANT CONSENT FORM – COACH
Title of the study: Exploring Leadership in Youth Sport: A Transformational Perspective
Name (please print clearly): ________________________________________
1. I have read the Letter of Information and have had any questions answered to my
satisfaction.
2. I understand that I will be participating in the study called Exploring Leadership in Youth
Sport: A Transformational Perspective. I understand that this means that I will be asked to (a)
complete a 30-minute questionnaire assessing my leadership behaviours before and after the
workshop, (b) have a practice session video and audio-recorded before and after the
workshop, and (c) take part in a 3-hour workshop, after which I will be asked to complete a
30-minute questionnaire assessing workshop content and provide feedback on the quality of
the workshop. I will also be asked to participate in an optional one-on-one, 1-hour interview
regarding my experiences in the workshop. I may indicate my desire to participate in the
optional interview component below.
3. I understand that my participation in this study is voluntary and I may withdraw at any time.
I understand that every effort will be made to maintain the confidentiality of the data now
and in the future. Only team members and supervisor(s) in this research project will have
access to the data. The data will be kept in a locked office. Electronic files of videos,
questionnaire data, and transcripts will be password protected. The data may also be
published in professional journals or presented at scientific conferences, but any such
presentations will never breach individual confidentiality. All data containing personal
identifiable information will be destroyed after seven years. Should I be interested, I am
entitled to a copy of the findings, which will be emailed to me.
4. I am aware that if I have any questions about study participation, they may be directed to
Jennifer Turnnidge at [email protected] or 613-533-6000 Ext. 78207 or Jean Côté at
[email protected] or 613-533-6000 Ext. 79049. Any ethical concerns about the study may be
97
directed to the Chair of the General Research Ethics Board at [email protected] or 1-
844-535-2988.
I have read the above statements and freely consent to participate in this research:
Signature: _____________________________________ Date: _______________________
1. I agree that the I can be audio-recorded and video-recorded.
… Yes.
… No.
2. …Yes, I would like to receive a summary of the study’s results.
Please send them to me at this email address _________________________________
3. I agree to be contacted about a follow-up interview, and understand that I can always decline
the request.
... Yes. Please contact me at: __________________________________________
98
Appendix D
Coach Information Sheet
Date (dd/mm/yyyy): _ _ /_ _ / _ _ _ _
Part A: Background information
First three letters of your FIRST NAME: __ __ __
First three letters of your LAST NAME: __ __ __
Date of birth (DD/MM/YY): ________
Place of birth: _____________________________________
Gender: ________________
How do you describe yourself in terms of ethnic origin? (Check all that apply)
• White
• Chinese
• Black
• Filipino
• Latin American
• Arab
• Japanese
• Russian
• African
• South Asian (e.g., East Indian, Pakistani, etc.)
• South East Asian (e.g., Vietnamese, Cambodian, etc.)
• West Asian (e.g., Iranian, Afghan, etc.)
• Korean
• Native/Aboriginal
• Other (please specify): ______________________
How would you describe yourself in terms of profession? (e.g., police officer, student, etc.)
______________________________________________________________________________
__________________________________________________________________
Part B: Coaching Information
Please list any and all of your coaching and/or education qualifications (e.g., NCCP level 1; BSc
Biology, etc.)
99
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________
Sport you coach: _________________ Level: ______________________________
Age of athletes: __________________ Gender of athletes: ____________________
How long have you been coaching this sport? __________________________________
Each week, roughly how many hours to do you spend coaching this sport in a formal practice?
_____________________
Each week, roughly how many hours to do you spend in coaching this sport in a competition?
__________________
Have you coached any other sports (please include sport, level and how many years you coached
for)?
________________________________________________________________________
________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
____________________________________________________________
100
Appendix E
Implementation Checklist and Workshop Fidelity
101
102
103
Appendix F
Table F1
Mean, Standard Deviation, and Pearson Correlation (r) for Leadership Behaviours for Pre 1 and Pre 2 Observations (n = 5).
Variable M (s) SD (s) II1 IM1 IS1 IC1 TC1 NEU1 LF1 TX1
1. II1 78.63 43.50
2. II2 69.12 18.25 -.478
3. IM1 315.07 170.48
4. IM2 468.53 93.74 .216
5. IS1 140.74 83.65
6. IS2 190.93 99.11 .868
7. IC1 343.54 153.73
8. IC2 446.37 187.09 .627
9. TC1 245.38 49.98
10. TC2 256.96 74.15 .599
11. NEU1 2197.14 310.42
12. NEU2 2039.36 143.96 .618
13. LF1 240.02 163.89
14. LF2 82.51 88.70 .659
15. TX1 27.28 39.32
16. TX2 31.33 23.04 .740
Note: II = Idealized Influence, IM = Inspirational Motivation, IS = Intellectual Stimulation, IC = Individualized Consideration, TC =
Transactional, NEU = Neutral, LF = Laissez-Faire, TX = Toxic
Appendix G
Table G1
Mean, Standard Deviation, and Pearson Correlation (r) for Content Modifiers for Pre 1 and Pre
2 Workshop Observations (n = 5).
Variable M (s) SD (s) GC 1 I/F 1 ORG 1
1. GC 1 267.44 176.65
2. GC 2 249.99 113.76 -.134
3. I/F 1 1279.92 290.23
4. I/F 2 1685.36 235.99 -.379
5. ORG 1 1786.76 204.13
6. ORG 2 1569.52 283.69 .393
Note: GC = General Commmunication, I/F = Instruction/Feedback, ORG = Organization.
105
Appendix H
Table H1
Mean, Standard Deviation, and Pearson Correlation (r) for Recipient Modifiers for Pre 1 and
Pre 2 Workshop Observations (n = 5).
Variable M (s) SD (s) Individual 1 Team 1
1. Individual 1 1148.95 307.80
2. Individual 2 1361.20 235.80 .539
3. Team 1 2198.83 225.22
4. Team 2 2143.67 297.70 .724
106
Appendix I
Table I1
Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Leadership
Behaviours for Pre 1 and Pre 2 Workshop Observations (n = 5).
Practice
Pre 1 Pre 2
Behaviour M SD M SD t 95% CI p
Idealized
Influence
78.63 43.52 69.12 18.25 0.39 [-58.338, 77.364] .717
Inspirational
Motivation
315.07 170.48 468.53 93.74 -1.95 [-371.894, 64.985] .123
Intellectual
Stimulation
140.74 83.65 190.93 99.11 -2.28 [-111.405, 11.025] .085
Individualized
Consideration
343.54 153.73 446.37 187.09 -1.53 [-289.432, 83.779] .201
Transactional 245.38 49.98 256.96 74.15 -0.43 [-85.779, 62.505] .687
Neutral 2197.14 310.42 2039.36 143.96 1.42 [-151.082, 466.642] .229
Laissez-faire 240.02 163.89 82.51 88.70 2.82 [2.546, 312.482] .048
Toxic 27.28 39.32 31.33 23.04 -0.33 [-37.761, 29.653] .755
Note: For all variables, df = 4. This table, split between pre 1 and pre 2 workshop practices,
displays the duration of coaches’ full-range of leadership behaviours when actively interacting
with athletes. Means are represented in rate per hour format (e.g., 78.63 s per hour).
107
Appendix J
Table J1
Descriptive Statistics According to Duration (s) and Paired Samples t-Tests for Content and
Recipient Modifiers for Pre 1 and Pre 2 Workshop Observations (n = 5).
Practice
Pre 1 Pre 2
Modifier M SD M SD t 95% CI p
Content
GC 267.44 176.65 249.99 113.76 0.18 [-258.917, 293.829] .869
I/F 1279.92 290.23 1685.36 235.99 -2.07 [-949.235, 138.342] .107
ORG 1786.76 204.13 1569.52 283.69 1.75 [-126.584, 561.078] .154
Recipient
IND 1148.95 307.80 1361.20 235.80 -1.77 [-545.663, 121.170] .152
Team 2198.83 225.22 2143.67 297.70 0.60 [-200.234, 310.561] .581
Note: GC = General Communication, I/F = Instruction/Feedback, ORG = Organization, IND =
Individual. For all variables, df = 4. This table, split between pre 1 and pre 2 workshop practices,
displays the duration of coaches’ content modifiers and recipient modifiers when actively
interacting with athletes. Means are represented in rate per hour format (e.g., 267.44 s per hour).