Ffooff

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/234092858
Validation of a Multidimensional Measure of Fear of Failure in a British Sample: The

Performance Failure Appraisal Inventory (PFAI)
Article in International Journal of Sports Science & Coaching · January 2010
CITATIONS READS
16 8,062
2 authors:
Sam Sagar Sophia Jowett

Leeds Beckett University Loughborough University
13 PUBLICATIONS 815 CITATIONS 153 PUBLICATIONS 8,334 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Sophia Jowett on 28 May 2014.
The user has requested enhancement of the downloaded file.

International Journal
of Coaching Science
Volume 4 / Number 1 / January 2010
CONTENTS
Coach Communication of Non-Selection in Youth Competitive Sport/ 3

A. Lauren Capstick and Pierre Trudel University of Ottawa, Canada
Should coaches use personality assessments in the talent identification process? A 15 year
predictive study on professional hockey players/ 25
Chris J. Gee University of Toronto, Canada
John C. Marshall & Jared F. King Self Management Group, Canada
The learning formats of coach education materials/ 35

John Lyle University of Queensland, Australia
Sue Jolly & Julian North SportsCoach UK, UK
Validation of a Multidimensional Measure of Fear of Failure in a British Sample:

The Performance Failure Appraisal Inventory (PFAI)/ 49
Sam S. Sagar Leeds Metropolitan University, UK
Sophia Jowett Loughborough University, UK
International Journal of Coaching Science (ISSN 1975-8286) is published biannually in January and July
by 21C Education Publishing Company, #301 Wonjin Building, 157-1, Ssanglimdong, Joonggu, Seoul
100-400, Republic of Korea. Tel 82-2-2269-6561, Fax 82-2-2269-6568
Co-Publishers: Dr. Jeong-Keun Park (jkpark@hoseo.edu), John Bales (jbales@coach.ca)
Subscription and Business Office: Co-Publisher of IJCS, Dr. Jeong-Keun Park

Department of Sport & Exercise Science, Hoseo University, Asan city, Chungnam 336-795, Republic of
Korea. Tel 82-41-540-5872, Fax 82-41-540-5876, Cell: 82-19-306-4917
E-mail: jkpark@hoseo.edu
One year subscription fee: U.S.$30 for individual; U.S.$120 for institution
International Journal of
Coaching Science
Vol. 4 No. 1
January 2010. pp.3-23
Coach Communication of
Non-Selection in Youth Competitive
Sport
A. Lauren Capstick* and Pierre Trudel University of Ottawa, Canada
Abstract
Previous research in youth sport has overlooked the experience of non-selection from the
collective perspectives of all those involved. The purpose of this descriptive research was to gain
understanding about the communication of non-selection within competitive youth sport, while
also seeking to investigate the methods used by coaches to inform youth of their non-selection.
This qualitative research involved interviews employed within an interpretive phenomenological
analysis (IPA). Participants (n=15) were comprised of coaches, athletes, and parents. Findings
indicate that coaches, athletes, and parents all consider the task of communicating non-selection
to be troublesome. An inventory of communication methods used by coaches to inform athletes
of their tryout results was created by noting the experiences of all participants. Lastly, coach
profiles revealed that coaches make communication choices based on factors such as their past
experience, willingness to provide feedback, as well as the age of their athletes. Findings
highlight the importance of experiential learning, as coaches learn primarily about the process of
communicating non-selection through their experiences as coaches and as parents.
Keywords: tryouts, elimination, coaching, interpretive phenomenological analysis
* Please direct all correspondence regarding this article to A. Lauren Capstick, School of Human Kinetics, University of Ottawa,
125 University, Ottawa, Ontario, Canada, K1N 6N5, or via e-mail at acaps073@uottawa.ca
3
A. Lauren Capstick and Pierre Trudel
Introduction
The potential benefits of participation in physical activity and sport are well documented
(e.g., Fraser-Thomas, Côté, & Deakin, 2005; Thompson, Humbert, & Mirwald, 2003). Currently,
the escalating competitive nature of youth sport (Côté & Fraser-Thomas, 2007) requires an
increasing number of youth to tryout for membership on sport teams. Lipsyte (1979) offers a
particularly vivid perspective of the process of tryouts within youth sport and attributes the (mal)
practice of eliminating youth to what he calls the ‘Varsity Syndrome’. The ‘Varsity Syndrome’
“refers to a selection process operating within sports which systematically denies opportunities for
participation to the larger portion of the population. Only those with exceptional talent…. can
reap the rewards of sport participation that should belong to all” (p. 15).
Interestingly, the research available on the topic of tryouts deals primarily with the examination
of those who are chosen for membership on sport teams. That is, in seeking more information about
who deserves to be selected, researchers have tended to focus on talent identification (Abbott &
Collins, 2004; Durand-Bush & Salmela, 2001), and the development of expertise (Côté, Baker, &
Abernethy, 2007; Wall & Côté, 2007). Within research examining expertise, many researchers are
interested in the relative age effect and how children born in the earlier portions of the selection
year tend to be chosen for teams and are thus over-represented within the highest levels of sport
(e.g., Côté et al., 2007; Musch & Grondin, 2001; Sherar, Baxter-Jones, Faulkner, & Russell, 2007).
Other work examines intersections between relative age and selection effects related to athlete
birthplace (Baker & Logan, 2007; Côté, MacDonald, Baker, & Abernethy, 2006).
When investigating athletes’ removal from sport, existing literature tends to focus on dropout
(e.g., Fraser-Thomas, Côté, & Deakin, 2008; Sarrazin, Vallerand, Guillet, Pelletier & Cury, 2002),
burnout (e.g., Cresswell, 2008; Goodger, Gorely, Lavallee, & Harwood, 2007), and retirement (e.g.,
Lally, 2007; Taylor & Ogilvie, 1994; Warriner & Lavallee, 2008). Issues associated with the loss
of athletic identity following unsuccessful tryouts are discussed (Alfermann, 2000; Brown & Potrac,
2009; Grove, Fish, & Eklund, 2004) yet the studies to date deal with research participants who
were in their late teens or college/ university students. The communication of non-selection within
the youth sport context is presently lacking and there is absolutely no research addressing the
perspectives of all actors involved: coaches, athletes, and parents.
The little scholarly research that is available spans many decades and calls into question the
practice of eliminating youth from sport through selections due to the negative psychological
outcomes (Barnett, 2006; 2007; Blinde & Stratta, 1992; Lebovitz, 1978; Orlick & Botterill, 1975).
Although these articles bring attention to the negative consequences associated with being cut
from a team, they offer a somewhat utopian conclusion when suggesting that the practice of
sport tryouts be abolished within youth sport. Even though being cut from a sports team as a
4
International Journal of Coaching Science Vol. 4 No. 1 January 2010
youngster may likely “represent the first encounter by the child with a socially sanctioned system
of exclusion” (Pease, Locke, & Burlingame, 1971, p. 46) we argue that tryouts and team
selection processes will remain pervasive amongst youth sport. With that in mind, it is surprising
that the question posed by Pease et al., nearly 40 years ago, remains essentially ignored. They
stated that one of the “crucial questions remaining” within coaching research is that of “how
does the coach and/ or athlete perceive the exclusion process?” (1971, p. 46), and to that we
could add, ‘how do parents perceive the exclusion process?’.
If it is accepted that tryouts amongst young athletes continue to occur with each new
sporting season in many community and school settings, and that this practice is likely to persist,
it is important to get the perspective from all actors involved with the process in order to
understand exclusion from youth sports following tryouts. The critical role that coaches play in
shaping the experiences of youth cannot be taken lightly, and coaches are cautioned not to
“underestimate their role in influencing athletes’ development” (Fraser-Thomas, & Côté, 2009, p.
20). Lastly the fact that coaches are required to have strong communication skills (Fraser-Thomas,
& Côté, 2009; Gilbert, Gilbert, & Trudel, 2001; Turman, 2008), means that an investigation into
how coaches communicate their tryout results is not only long overdue, but it is essential when
considering the process of non-selection in youth sport.
The perspectives of athletes who are eliminated, and those from parents whose children are
eliminated, are important to consider alongside the views of coaches in order to create holistic
understandings of non-selection. The purpose of this descriptive research was to gain
understanding about the communication of non-selection within competitive youth sport by
appreciating the perspectives and experiences of all actors involved. The purpose of this article is
to discuss the methods used to inform youth of their non-selection and to convey the experiences
that lead coaches to choose a certain method to communicate their tryout results. The broad
questions directing this article are:
1. What methods are being employed in order to communicate non-selection within youth
team sports?
2. What experiences contribute to the choices coaches make when choosing a method to
communicate tryout results?
Method
This qualitative research study employed interpretive phenomenological analysis (IPA) which
assumes an inherent interpretive worldview in which the researchers acknowledged their place
within the interpretation of the information presented by participants. Ellis (1998) notes that
5
interpretivists “do not seek a uniquely correct or ‘accurate’ interpretation, but rather the most
adequate one that can be developed at that time” (p. 27). As such, when considering the process
of communicating team sport tryout results to eliminated youth, we have attempted to derive the
most suitable interpretations within the particular social context of our participants. Furthermore,
when exploring novel areas, such as the inquiry into methods to communicate non-selection, IPA is
the ideal methodological choice as it allows for a detailed examination of a process which ensures
that the subjective reality is maintained (Smith & Osborn, 2003). That is to say, each participant’s
account of team elimination was honoured for both its uniqueness, and similarity to the others’.
Interviews were used to collect data because they are particularly relevant for IPA, and
allow for meaningful perspectives of others to become known (Patton, 2002; Potrac, Brewer,
Jones, Armour, & Hoff, 2000; Smith & Osborne 2003). While the interview guide contained key
questions to cover during the course of the discussion, the interviews were informally conducted
and allowed the “participants to engage in the process more freely” (de Marris, 2004, p. 53), as
they were the ones who primarily guided the direction of the discussion. A pilot interview was
conducted, transcribed, and reviewed in order to determine the suitability and order of the key
interview questions, and to gain analysis experience with IPA. Following this, fifteen participants
were recruited using purposeful sampling (Patton, 2002). The primary investigator used word of
mouth within her coaching and sport community, and ‘snowball sampling’ in which potential and
actual participants were involved in targeting other potential and actual participants (Patton, 2002).
Participants provided informed consent, and pseudonyms were assigned to protect anonymity.
Each participant was individually interviewed and conversations were audio recorded. None of the
participants’ experiences were related to one another, and some participants were able to speak
about more than one team elimination occasion. The fifteen participants were comprised of three
groups of people: coaches, athletes, and parents. The researchers sought participants who could
speak to elimination experiences of youth aged 13 to 15 in order to appreciate the non-selection
of youth trying to ‘specialize’. In order to contextualize the 13-15 year old age-group dealing
with non-selection, Côté’s (1999) Development Model of Sport Participation was used. This
model describes a three-stage journey within elite youth sport through sampling, specializing, and
investment which maintains “positive physical, psychological, and social development patterns”
(Fraser-Thomas et al., 2005, p. 31). Youth aged 13-15 are likely in their specializing years, and
are thus attempting to reduce the number of sports they sample in favour of becoming more
proficient within one or two sporting domains (Côté & Hay 2002; Soberlak & Côté, 2003).
The primary investigator spoke to five coaches and it was stipulated that they perform team
selection processes regularly (at least annually), of youth aged 13 to 15, and had at least five
team selection coaching experiences. Three male and two female middle-aged coaches were
interviewed. Five athletes were interviewed who were cut from a sports team when they were
6
aged 13 to 15, and to gain retrospective information it was stipulated that the elimination should
have happened at least five years prior. Three female and two male athletes all in their late
twenties were interviewed. Five unrelated parents, who consisted of four mothers and one father,
were interviewed and all parents were middle-aged with grown children. Criteria for their
inclusion maintained that at least one of their children was cut from competitive sport when aged
13 to 15 years, and that the elimination happened at least five years prior. Athletes and parent
participants were required to speak about non-selection from at least five years prior in order to
be able to contextualize the experience and to be able to speak to any consequences of the
non-selection in the years that followed.
Interviews were transcribed verbatim resulting in over 275 pages of double-spaced text, and
transcripts were analyzed in accordance with the guidelines for IPA research by Smith and Osborn
(2003). Under IPA the objective is to interpret the meanings, rather than measure frequency of any
given ‘type’ of response. Analysis began immediately following the first session of data collection
and was collaboratively done by two researchers in order to enhance the consistency of the
interpretations. After several readings of the transcripts, notes were made in the margins in order to
annotate interesting or significant areas mentioned by participants (Smith & Osborn, 2003). Under
IPA methodology, there are no rules or requirements to divide a certain number of experiences into
a certain number of meaning units or categories. Smith and Osborn are explicit in this point, in
noting how “some parts of the interview will be richer than others and so warrant more
commentary” (2003, p. 67). Consequently, there was no need to make equal amounts of
interpretation for each section of the interviews, and it was the richest portions of the interviews
that received the most analysis. Analysis proceeded from here as emerging themes were expressed
(either on the other margin or on a separate piece of paper) all the while attempting to “capture
the essential quality of what was found in the text” (Smith & Osborn, 2003, p. 68). Themes were
then examined for connectedness to each other; within and across different interviews.
Rigour and trustworthiness were assessed and maintained according to literature regarding
qualitative work (Patton, 2002). For example, researcher subjectivity was acknowledged and
assessed by use of a conceptual baggage diary by the primary investigator (Kirby &
McKenna,1989). Furthermore particularity was honoured by maintaining the integrity of different
or oppositional points of view, and interpretations of participant quotes were reviewed and
verified by each participant (Patton, 2002).
Results
The findings relevant to the purposes of this article are presented in three parts. First, we
7
present an unanticipated finding about the shared perspective of the participants regarding the task
of communicating non-selection. Next, we provide an inventory of the methods used to
communicate tryout results as expressed by the participants. We conclude by profiling each coach
in order to examine how their experiences led them to make certain choices about their methods
to communicate tryout results. In line with IPA and as recommended by Smith and Osborn
(2003), the findings will include numerous direct quotes from the participants in order to best
illustrate their original and nuanced responses.
Views about Communicating Non-Selection

Each participant had strong convictions about the communication of non-selection, and about
the implications of telling youth that they were unsuccessful in making a team. During interviews,
and while telling their own story of non-selection, every participant made note of the troublesome
nature of communicating non-selection. These perspectives were shared when participants expressed
the non-selection process in their own words and were not the result of direct questioning about
the challenges of non-selection. Their language was vivid; as coaches, athletes, and parents used
strong words such as ‘hate’, ‘dread’, and ‘the hardest part’, to express the actual and perceived
difficulty and anxiety experienced by coaches when communicating non-selection.
• “I think it’s one of the more unpleasant things about sport”. (Coach Will)
• “You have to remember that a lot of these coaches, they hate cutting, for one thing.
They absolutely hate it”. (Parent Heather)
• “Nobody likes informing someone of bad news”. (Athlete Erin)
• “It’s stressful for everybody. We call it, and you might know it too: the hardest part of
the season. It’s the most trying part, it’s the most stressful part; you lose sleep over it. I
don’t sleep well pre-season”. (Coach Dave)
• “I think no matter how you do it – it’s hard because it’s about not getting what you
want”. (Parent Mary)
• “I know there’s almost no good way to do it, but there are better ways and the coaches
need to find those ways”. (Athlete Brett)
• “It’s probably the hardest thing to do - to cut a child and whether 11 or 17, it’s so
hard”. (Coach Leigh)
• “I always hated the calls. I always dreaded the call, even if the people were nice about
it, and said ‘thanks for the opportunity’ and hung up. You still felt bad. You know
they’re sad”. (Coach Leigh)
• “That’s hard to do. I know it’s hard to do”. (Parent Adrienne)
8
Notably, not one participant indicated that communicating tryout results is easy or
straightforward. As Sarah, an athlete stated, “It’s a difficult task to do but you just have to do
it”. From the perspective of these participants, the idea of delivering information about
non-selection is wholeheartedly undesirable and arguably difficult. With this in mind, it is
important to know just how coaches actually communicate tryout results.
How Coaches Communicate Tryout Results

The first research question aimed to illuminate the different methods used by coaches to
inform athletes of their tryout results. In order to create an inventory, information was gathered
from each of the three participant groups. Athletes shared their experiences with tryouts and told
us how their coaches had informed athletes of their tryout result. Parents told us about how their
children had been informed of tryout results and coaches named all the ways they had
communicated tryout results. Analysis of their interviews suggests that coaches employ a variety
of communication methods as presented in Figure 1.
Method Variation 1 Variation 2
All attending tryouts

Speak only to athlete
Phone calls Eliminated Speak to athlete and parent
Selected for membership
Formal Speak only to athlete

Face-to-face Speak to athlete and parent
Informal Speak only to athlete
Posted in hallway
Selected for membership Read out loud
List Posted in local newspaper
Posed on website
Eliminated Posted in hallway
Written personal communication To all athletes attending tryouts Hard-copy letters

E-mail
Figure 1. Methods and variations to communicate tryout results in youth sport
9
The methods used to inform youth of their tryout results include phone calls, face-to-face
interactions, lists, and written personal communications. It is important to note that within the
same method, several variations exist. It was explained that phone calls can be made to everyone
who attended the tryout, only to those who are selected for membership, or only to those who are
eliminated. The participants described how sometimes coaches do not involve the athletes’ parents
during these phone conversations, and in some cases, such as this one, parents are included,
“During the phone call I will actually talk to the parents first and then we would let the athlete
know if they made the team or not” (Coach Leigh). Face-to-face interactions were explained to be
either formal, where the coach or group of coaches speak to the athlete in a private setting, or
informal, consisting of a short conversation off to the side of the athletic venue. In accordance
with the experiences of the participants, formal conversations can sometimes involve parents, but
in some cases they do not, as was the case for Parent Julia, “It was individually done, so the
boys went into the dressing room on their own and came out on their own”.
The participants informed us that lists are commonly used to communicate tryout results,
and can either name those selected for membership, or they can name those who are excluded.
Furthermore, when lists name those who are selected for team membership, it was revealed that
these lists are either read aloud or are posted in hallways, websites, or newspapers. Athlete Sarah
explained a typical situation in which, “The coach would say ‘okay the list is going to be up
tomorrow morning on the bulletin board’”. Lastly, it was explained that personalized written
communications can be given to each athlete who attends the tryout, and can be a
hand-delivered, mailed, or sent via e-mail. Coach Alyson recounted how she “would hand out
sheets at the end of the last try-out indicating who made the team, or who made the team as an
alternate, or who should find another program”.
Not included in the figure, yet related to the various options facing coaches about methods
to communicate tryout results are the options for follow-up. That is, after an athlete learns that
he or she has been unsuccessful in earning membership on a given team, whether by face-to-face
interactions, lists or otherwise, there are sometimes circumstances when coaches create
opportunities for follow-up with the eliminated athletes. Occasionally, the eliminated youth is
allowed to have a follow-up meeting with the coach, and at times the eliminated athlete with
his/her parents can call or e-mail the coach or league director to ask further questions. It was
revealed that in many circumstances, there is neither action nor further communication with
eliminated athletes following the original communication of the tryout result.
How Coaches Choose a Method to Communicate Tryout Results

Each of the five interviewed coaches is profiled here in order to understand the factors that
10
influenced their choices when selecting a method to communicate tryout results. Coaches in this
study explained that they made choices about the delivery of tryout results based on their
experiences as coaches, and sometimes as parents. In addition to reflecting upon their own
experiences and upon the examples of others, some considered the potential impact and tone of
their message of non-selection on youth.
Will’s profile. Will was a volunteer football coach at the community, school, provincial, and
university levels. Most of his coaching experiences were with boys under 15 years of age, and
he also worked with the Provincial Football Association in his province. He had been coaching
football for over 20 years.
Will’s coaching and administrative experiences with the announcement of tryout results led
him to believe that coaches ought to be cautious about inviting potentially abusive confrontations
with parents and athletes. Despite noting, “We’ve tried all sorts of ways to try to lessen the
blow”, some encounters with eliminated athletes and their parents taught him to select a method
of communication that reduces the chances of confrontations.
They just want to trip you up, they want to ask you all sorts of questions and if you come
back with an answer that they feel you’re inconsistent with or it doesn’t apply to other athletes
and stuff like that, then they can go raise holy hell with whoever is above you – a principal, a
president, something like that.
In order to decrease the occurrence of emotional conflicts with eliminated athletes or their
parents, Will explained how he needs to leave the sports fields after posting a list:
Sometimes you have to be like a secret agent trying to get out of there, you know, trying to
get to your car before some parent tracks you down in the parking lot. They’re emotional, and
in certain situations you become a punching bag, a verbal punching bag.
About posting lists, Will also acknowledged that this method allows for the communication
of tryout results to happen very quickly and to a large number of people all at the same time.
He recognized that eliminated athletes may want to have interactions with coaches but stated, “It
[posting a list] may seem to be a cop-out but there’s just too many kids”. When reflecting as a
parent, Will noted that he would want to know the reason why his child was cut, but
immediately added that as a coach, “logistically it’s a nightmare [to speak with each athlete]
with the number of kids that are there, but yeah, that is what I too would want as a parent”.
Along the same lines, Will appreciated the fact that the youth themselves usually have
unanswered questions when they find out of their elimination by a posted list:
11
They’ve been brought up to question things, to want to know why, and not to take things at
face value and not to give people in a position of authority a free pass. So that’s kind of made
it really interesting for coaches and I don’t know if there’s any easy solution out of that…. To
be honest with you I think it’s not necessarily a bad thing to allow kids to question things, but
at the end of the day, it’s very wearing on coaches because what they’re doing is questioning
the coach’s integrity in that situation. And there are some coaches that do need to have their
integrity questioned, but the vast majority of them don’t, and it’s tough on us.
Will considered the wishes of athletes and parents to have interactions with coaches
following unsuccessful tryouts, but weighed these against the outcomes of his own negative
experiences in these types of interactions. As a result of these considerations, and due to the
logistical challenges he faces, Will chose to employ lists to announce the results of his tryouts.
Consequently, Will did not willingly have interactions with parents, nor did his method allow for
the provision of feedback to the eliminated youth.
Dave’s profile. Dave was a volunteer hockey coach, and had been coaching boys’ hockey
for over 15 years. He most recently had focused his coaching energies on the Bantam Triple A
level which is the highest level of hockey for youth aged 13-15.
Dave explained how he considered the examples offered by his sons’ coaches when
choosing his own team selection delivery method:
I went through it as a hockey parent. My boys were released from hockey teams and they’ve
been released in different ways and I’ve just taken that information and think there’s a wrong
way of doing it and that’s a good way of doing it. Through my years as a parent and coach
I’ve decided that there’s no better way than being honest, up-front, and realistic.
He recounted a specific example where he observed the way in which his son’s hockey
coach delivered tryout results. He remembered that the coaches “were out in the parking lot
giving them results letters, with no feedback. Absolutely none”. He decided that this appeared
cowardly, and questioned the coaches’ need to “drive away in such a hurry” instead of staying
to provide answers to the athletes’ questions. He summarized how he assessed both the “nasty”
and “good” examples from other coaches, in order to learn that for him it was important to be
honest and to avoid being “harsh” with the young players.
Dave noted how he scrutinized and learned from the examples of others, and as a result,
chose to use face-to-face release interactions when communicating non-selection. He explained
how during the face-to-face interactions, he was able to provide feedback to the athletes, and
how this was something he regarded as important because it allowed him to communicate to the
athletes why they were not selected for membership on the team. “From my experiences, they
12
take it well, as long as you give them the proper feedback. They have to know. You give them
a real honest reason why”.
Dave also appreciated the role of parents when dealing with athletes in the 13-15 age range
and he welcomed questions from parents after he had spoken to non-selected youth and said,
“They are most definitely welcome to talk to us afterwards”. Dave communicated his value for
the important role that parents play in youth hockey, and how he aimed to include them in the
communication of tryout results. He did this because he believed that parents are likely able to
help their child navigate through the consequences of elimination. In addition to the face-to-face
interaction where he provided feedback to the athletes, he provided a letter detailing alternative
choices for each eliminated athlete as well as his “phone number for their parents so that if they
wish they can give me a call to ask follow-up questions”.
Unique to Dave’s situation was the fact that his hockey league provided recommendations
about how tryout results should be announced. Dave explained that face-to-face interactions were
discouraged from the league because it was believed that this method was “too hard on coaches”.
Coaches were, as a result, encouraged to use lists to communicate their results yet Dave said
“I’d like to talk to every player”. Dave considered the advice from his league against his own
experiences and surmised, “you want to do it within the guidelines that you’re technically taught
to do and what you think is morally right”. Dave consequently made choices after considering
his experiences with the communication of tryout results both as a father and as a coach and
weighed these against the information received from his league.
Rick’s profile. Rick was a volunteer coach in the sport of girls’ ringette, and had been
coaching for nearly 10 years. He had coached club, community, and provincial level teams, many
of which overlapping with the 13-15 year-old age range.
Rick referred to consequences he had witnessed when coaches were, in his opinion, too
removed with their team elimination procedures. He felt very strongly about making deliberate
choices regarding the manner in which non-selection is communicated:
The consequences of doing it the wrong way can be devastating. It can be. They could drop out
of the sport, they could take it personally; lose their self-esteem which can manifest itself in
many other complications in life. There’s all kinds of things that could trigger young people to
not feel good about themselves. There are then all kinds of consequences of that. The bottom
line is that I wouldn’t care if I never won a game – but I certainly would care if I was
responsible for setting a young person in a path where they self-destruct because I made them
feel so bad about themselves. If I didn’t take this seriously, I wouldn’t coach.
At one point, he considered using phone calls to communicate his tryout results but decided
13
against it because “you don’t see the expression on their face and how the 12 year old is taking
the news”. In order to do what he can to ensure that the elimination process can be a positive
learning experience for his athletes, he has rejected all methods he had observed, or was familiar
with, and instead used a face-to-face method preceded by continual discussions with all the
athletes throughout the tryout process.
Rick was also sensitive to the fact that the athletes he worked with were young teenagers,
and not adults:
You’re not making cuts for the Toronto Maple Leafs...you’re making cuts for a group of kids
who are 13-14 years old in a sport where they are not professionals. And remember we’re
talking 13-14 year olds; a very vulnerable age, especially if you are female. That’s what you are
dealing with – young people who are vulnerable and they have a million other things to worry
about and so the object wasn’t to completely destroy their morale and stuff – the object was to
say ‘yes you didn’t make the team’, but send them away feeling good about themselves.
By providing a large amount of both positive and constructive feedback to his athletes
throughout the tryout process, and again during the face-to-face release meeting, and by helping
his athletes to keep a balanced perspective on the tryout, Rick believed that he was being
sensitive to the impact this experience had on young people. He concluded:
We should be doing things a little bit better so that even though not everybody gets to play
with the A team, everybody should feel good about the fact that they did their best, and are not
coming away from a tryout with a negative feeling about the sport, and more particularly about
themselves.
Rick ensured that even though the parents were knowledgeable about their athlete’s progress
throughout the tryout camp, he made sure that they were present when he delivered his final
decision. That is, when he informed youth of their non-selection during a formal face-to-face
meeting, he provided additional feedback, coaching suggestions, and ensured that a parent was
present in order to be able to support their child. Rick embraced his method due to his
evaluation of other examples he had witnessed, and because of his appreciation of the impact the
decision and its delivery can have on young people.
Leigh’s profile. Leigh was a volunteer coach in the sport of soccer, and coached girls’
teams. She began coaching when her daughter’s team needed coaches, and had coached at
multiple age groups; the majority of which including the 13-15 year-old cohort studied here.
Leigh coached community competitive soccer, and had been doing so for over 10 years.
Leigh recalled “one horrific experience” she had when phoning athletes and their parents to
14
communicate tryout results. A parent of an eliminated athlete swore at her and used offensive
labels to describe her coaching practices. “To be verbally abused was shocking. I wasn’t going
to change my mind nor did I think I’d made the wrong decision, I was just hurt. But that was
an extreme example. Most parents were really quite okay”. She later added, “The phone calls,
they take time, but I don’t know, I think people are worth it”. Even though Leigh experienced
one terrible encounter with a parent, the sum of her other experiences when using phone calls
with athletes and parents to inform youth of their tryout results, resulted in a decision to
continue with this method.
For Leigh, being aware of the athletes’ age and likely impact of the undesirable news, led
to decisions to communicate non-selection in a way that involves parents. She noted that her
rationale is not based on the assumption that parents will be able to coach their children in the
areas requiring improvement. Leigh speculated, “A lot of parents aren’t sport-minded. They don’t
know that sport and so what their child needs to work on might not make any sense to them”.
For Leigh, the decision to include parents had more to do with the emotional support she hoped
they were able to offer:
We told the parents first, because I thought the child might break down, so I wanted to let one
of the parents know so they could be there to comfort and to help remind the child that we
found some positives in their tryout too.
The message that Leigh provided to eliminated youth was also guided by her understanding
about the sadness the youth experience upon learning their fate. She said, “I am most concerned
with making sure the kids are okay with this, or as okay as they could be”. In order to
accomplish this, she spoke with each child who attended the tryout and for those who are not
selected for membership she tried:
To put it delicately, I think it’s really important – there’s no easy way but I always try to
sandwich in the good stuff and get in that you didn’t make the team, but we’re glad you came
out. We want you to come out again.
Leigh was deliberate about offering positive and constructive feedback and she
communicated her team selection decisions by phoning each athlete who attended the tryout. She
ensured that parents were involved with these phone calls when dealing with non-selected youth,
and she chose this method after considering her experiences delivering tryout results, and by
considering the impact and needs of the youth upon learning of their non-selection.
Alyson’s profile. Alyson coached full-time in the sport of synchronized skating from the
developmental levels up to and including National and International teams. She had been
15
coaching for roughly 20 years and has held team tryouts and eliminations well over 50 times –
most of which covering the 13-15 age grouping studied here. The athletes she coached with are
almost always female.
Alyson noted how she has been able to learn from her experiences and feels as though she
has chosen a method that best reflects her learning. When she first started coaching, she posted a
list of the skaters who had made the team but was quick to say, “It’s not a nice thing to do to
the skaters! It gives you no opportunity to talk to them and let them know why they were
chosen as an alternate or why they were released from the team”. Alyson elaborated on why she
no longer uses list-posting:
When you choose your alternates, and if one of the girls decides that she doesn’t want to skate,
you don’t really have much recourse in calling back the girls who were cut. They feel like you
didn’t want them. They feel hurt. And if you call them back, they say ‘no thanks you, you’ve
hurt me’ or ‘you didn’t want me – you think I’m crap’. There are other ways of doing it that
work much better.
In her quote above, Alyson explained that when her skaters learned of their elimination via a
list, they were less likely to take her up on a later offer to join the team.
After deciding that list-posting was not the right method for her, Alyson used to hand out
sheets of paper which indicated “who made the team, or who made the team as an alternate, or
who should find another program”. She subsequently learned that this method also resulted in
negative consequences, “you would lose skaters generally. It hurts people’s feelings on the
bottom end”. She explained her most recent use of two methods and described how her choice
depends upon how many youth are involved and thus how many people need to be informed of
their tryout result. If there was a manageable number of athletes at a tryout, she noted that she
hosted face-to-face meetings where she can speak to each youth about their tryout. She explained
that the order of the meetings is important because she needed to know if skaters wanted to
decline an offer to be on the team, and if so, she offered their spot to another skater lower on
her list. The reason she would use this method when the group is small was due to the fact that
families had to wait around the arena for their time to meet with her. She explained:
The only problem with that is you have kids waiting out in the hall for quite a while
wondering why you haven’t called on them. They have a tendency to know where they stand on
the list. So it’s a little more obvious to the skaters who is doing what on the team. But is does
give you answers more promptly and you can talk face-to-face with the skaters and face-to-face
with the parents.
Alyson explained that when faced with a significant number of youth to inform she used
16
phone calls because the families did not have to wait at the arena following the tryout. In these
cases, she called skaters the night of the final tryout and during the next day. She explained that
families were aware and expecting the phone calls and she had asked that they be prepared to
accept or decline her offer at the time of the call so that she could take other skaters should
positions become available. Again, the order of the phone calls mattered, and she called the
skaters being offered core positions first, then skaters being offered alternate positions, and finally
skaters who she could not place on that particular team. She mentioned, “Sometimes I will have
an assistant coach help out by calling the core skaters, but like I said, I would much rather do
it all myself because then I know they are all getting the good feedback”. Alyson acknowledged
that some of the methods she used in the past resulted in outcomes incongruent with her
coaching values. She was clear in her desire to want to motivate skaters to pursue the sport of
synchronized skating, and she understood that the way in which she communicated tryout results
demonstrated and supported this.
In addition to evaluating her own previous tryout communication experiences, Alyson
provided details about how she was very aware of how her message of non-selection was
framed.
I have to put myself in their shoes. I need to remind myself of their age. What are they going
to feel? Of course they are going to be completely crushed if they don’t make the team. I am
absolutely making sure I know how I would want to be treated at that age in that situation…
We try to do it as nicely as possible, and especially because we are dealing with kids.
By being cognizant of the age of the athletes, and of the likely disappointment they will
feel upon learning they were not successful in earning membership on her team, Alyson talked
about how she needed to provide positive feedback to the skaters, and ensured they understood
that they “are not a total failure”.
When it came to the announcement of the tryout results, Alyson illustrated examples of
how parents are involved. She chose to place phone calls that involved both the athlete and their
parent. Alyson elaborated:
I think it’s extremely important that the parents know what is going on, what their skater needs
to work on, why they didn’t make it as a core position, otherwise they are not going to know
what to work on to improve to make it into the line even that year or the next year.
As a seasoned coach, Alyson was able to experience several methods to communicate tryout
results first-hand. She described how she learned from her experiences, and modified her choice
to reflect her learning in appreciation for the age of the athletes, and their likely reaction to
17
receiving undesirable news. As a result, she either met with athletes and their parents in
face-to-face meetings, or used phone calls involving the athletes and parents in order to
communicate non-selection following her team tryouts.
Discussion
All participants recognized the troublesome nature of informing youth of their non-selection.
While former literature about sport tryouts focused on the selection of the best athletes, this
research reveals a shared empathy by all those involved for the difficulty involved with telling
youth about their non-selection following tryouts. The fact that coaches find delivering undesirable
news a difficult task is not surprising and is supported by related research within organizational
behaviour and business noting that managers have a hard time discussing subordinates’ poor
performance with them (Larson, 1984; Levy, Albright, Cawley, & Williams, 1995; Manzoni,
2002). Fortunately for coaches however, results from this study indicate that parents and athletes
sympathize with coaches having to deliver tryout results. Armed with this knowledge, all parties
involved can reconsider the fact that conceptually speaking, they are starting from the same
place. That is, coaches should take comfort in knowing that parents have compassion for them
when they need to communicate tryout results to eliminate youth.
By learning from the experiences of athletes, parents, and coaches it was revealed that there
are five main methods used by coaches to communicate tryout results and when considering all
the different variations of these methods, there were 16 unique ways of communicating tryout
results. When taking into account the amount of feedback that is offered to the eliminated athlete
these methods differ tremendously from one another. One approach is to create an environment
in which the smallest amount of communication occurs after the tryout. That is, when using a
list, a coach can potentially avoid contentious conversations with athletes and parents because
when posting a list, one does not have to answer questions about why some children made the
team and others did not. Generally, the list contains the names of those selected for membership
(although we did learn of situations where a list conversely listed those who were not selected),
but it does not contain information behind the coach’s choices. Coach Will, despite
acknowledging that athletes prefer to receive feedback and answers to their questions about their
exclusion, noted that his logistical concerns outweighed the value he placed on providing
feedback. Will is not alone in recognizing that his time dedicated for coaching is divided by
many responsibilities that span very broad and often vague boundaries. The responsibilities of
coaches require a balancing act between many roles as educator, motivator, administrator,
fundraiser, manager and so on (e.g., Abraham, Collins, & Martindale, 2006; Becker, 2009; Gilbert
18
& Trudel, 2001; Jones & Turner, 2006; Lyle, 2002).

The other coaches in this study did however select methods to communicate non-selection
with more direct contact with the non-selected athletes, and in some cases, with their parents too.
These four coaches described how the method they used to communicate non-selection allowed
them to provide feedback to the athlete about why he or she was unsuccessful in making the
team. It is no surprise that coaches see value in providing feedback, as research linking feedback
to motivation and learning within academic settings (e.g., Butler & Winne, 1995; Tuckey,
Brewer, & Williamson, 2002) and sport settings (e.g., Mauch, 2007; Mouratidis, Vansteenkiste,
Lens, & Sideridis, 2008; Roemmich, Gurgol, & Epstein, 2004; Williams & Hodges, 2005) is well
established. Research investigating feedback in school settings reveals that students experience
negative outcomes when they are unable to determine the cause of their poor performance.
Thompson and Richardson (2001) contend that “unclear evaluative feedback – feedback which
fails to clearly specify the grounds on which students have met with achievement success or
otherwise – is likely to exacerbate negative outcomes, engender uncertain self-images and lead to
poor performance” (p. 153). If we liken this to situations in which young athletes who, upon
exclusion are unaware of the reasons behind their elimination, we can appreciate why coaches
might want to provide feedback. Four of the coaches in this study chose methods to
communicate non-selection that allowed the youth to learn about their performance and reasons
for not making the team.
It is also clear that some coaches find it important to include parents in the process of
communicating non-selection to youth aged 13-15, and it is argued that coaches are well advised
to include parents under such circumstances. Even though young teenagers may socialize more
frequently with their friends and depend upon their peers for social acceptance, it is parents who
remain the strongest figures of influence throughout adolescence (Atwater, 1996; Boyce, 2004;
Ryan, 2001). It is noted that teenagers are likely to depend on their parents for guidance on
important issues (Centers for Disease Control and Prevention, 2000), and it is suggested that
athletic exclusion may be one such issue (Wylleman, De Knop, Ewing, & Cumming, 2000).
Furthermore, the relationships between coaches and parents are an important consideration within
youth sport. Findings from this research are consistent with former research suggesting that
coaches and parents ought to keep open lines of communication in order to foster “healthy
reciprocal relationships” that will benefit the youth participants (Fraser-Thomas et al. 2008, p.
660).
When examining the coaches’ explanations of how they chose a method to communicate
non-selection it is worthwhile noting that none of the coaches said that they learned about this
through formal coach education. Coaches learn through different situations featuring a mix of
informal and formal experiences (e.g., Erickson, Bruner, MacDonald, & Côté, 2008; Mallett,
19
Trudel, Lyle, & Rynne, 2009; Nelson, Cushion, & Potrac, 2006; Rodgers, Reade, & Hall, 2007)
and it is suggested that their “path to learning how to coach is often idiosyncratic” (Werthner &
Trudel, 2006, p. 209). It is thus important to acknowledge that coaches will learn how to coach
in many different ways and through many different contexts. One of the first actions required of
many coaches is team tryouts, and this research illustrates that coaches learn about this process
through their experiences as parents, and as coaches. It is suggested that developers of formal
coaching education programs consider raising discussions about the communication of tryout
results, and this research reveals that there are many different ways to choose from.
Conclusion
This research offers perspectives about the communication of non-selection in youth sport,
and provides considerations for coaches to make when faced with decisions about the
announcement of tryout results. While the stories of the coach participants represent novel and
rich informative data about the communication of tryout results, and about factors contributing to
coach decisions to use one method of communication over the others, research measuring the
actual impact of various communication methods on drop-out, motivation, and athlete satisfaction
is warranted. That is, while the purpose of this exploratory research was not to measure best
practices in communicating non-selection, nor was the aim to probe athletes about the negative
psychosocial repercussions involved with certain methods of communication, future research could
examine these relationships in order to determine the ‘appropriateness’ of each method. While the
elimination of athletes is arguably a very difficult task to perform, this study represents the
foundation for future studies concerned with consequences surrounding tryouts within competitive
youth sport.
References
Abbott, A., & Collins, D. (2004). Eliminating the dichotomy between theory and practice in talent identification and
development: Considering the role of psychology. Journal of Sports Sciences, 22, 395-408.
Abraham, A., Collins, D., & Martindale, R. (2006). The coaching schematic: Validation through expert coach
consensus. Journal of Sports Sciences, 24(6), 549-564.
Alfermann, D. (2000). Causes and consequences of sport career termination. In D. Lavallee, & P. Wylleman (Eds.),
Career transitions in sport: International perspectives (pp. 45-58). Morgantown, WV: Fitness Information
Technology, Inc.
Atwater, E. (1996). Adolescence (4th ed.). Upper Saddle River, NJ: Prentice-Hall.
Baker, J., & Logan, A. J. (2007). Developmental contexts and sporting success: Birth date and birthplace effects in
20
national hockey league draftees 2000-2005. British Journal of Sports Medicine, 41(8), 515-517.
Barnett, L. (2006). Flying high or crashing down: Girls' accounts of trying out for cheerleading and dance. Journal of
Adolescent Research, 21(5), 514-541.
Barnett, L. (2007). "Winners" and" losers": The effects of being allowed or denied entry into competitive
extracurricular activities. Journal of Leisure Research, 39(2), 316-344.
Becker, B. (2009). It’s not what they do, it’s how they do it: Athlete experiences of great coaching. International
Journal of Sports Sciences & Coaching, 4(1), 93-119.
Blinde, E., & Stratta, T. (1992). The 'sport career death' of college athletes: Involuntary and unanticipated sport exits.
Journal of Sport Behavior, 15(1), 3-20.
Boyce, W. (2004). Young people in Canada: Their health and well-being No. H39 498/2004E. Ottawa: Health Canada.
Brown, G., & Potrac, P. (2009). You've not made the grade son: De-selection and identity disruption in elite level
youth football. Soccer & Society, 10(2), 143-159.
Butler, D., & Winne, P. (1995). Feedback and self-regulated learning: A theoretical synthesis. Review of Educational
Research, 65(3), 245-281.
Centers for Disease Control and Prevention. (2000). Life's first great crossroad, tweens make choices that affect their
lives forever. Atlanta, GA.
Côté, J. (1999). The influence of the family in the development of talent in sport. The Sport Psychologist, 13,
395-417.
Côté, J., Baker, J., & Abernethy, B. (2007). Practice and play in the development of sport expertise. In R. Eklund, &
G. Tenenbaum (Eds.), Handbook of sport psychology (3rd ed., pp. 184-202). Hoboken, NJ: Wiley.
Côté, J., & Hay. (2002) Children's involvement in sport: A developmental perspective. In Silva, J. and Stevens, D.
Boston (Eds.). Psychological foundations of sport (pp. 484-502). MA: Merrill.
Côté, J., & Fraser-Thomas, J. (2007). Youth involvement in sport. In P. R. E. Crocker (Ed.), Sport psychology: A
Canadian perspective (pp. 266-294). Toronto: Pearson Prentice Hall.
Côté, J., MacDonald, D., Baker, J., & Abernethy, B. (2006). When where is more important than when: Birthplace
and birthdate effects on the achievement of sporting expertise. Journal of Sports Sciences, 24(10), 1065-1073.
Cresswell, S. L. (2008). Possible early signs of athlete burnout: A prospective study. Journal of Science and Medicine
in Sport, 12, 393-398.
deMarris, K. (2004). Chapter 4: Qualitative Interview Studies: Learning Through Experience. In deMarris, K.B. &
Lapan, S. (Eds.). Foundations for research: Methods of inquiry in education and the social sciences. (pp.
51-68). Mahwah. N.J.: L. Erlbaum Associates.
Durand-Bush, N., & Salmela, J. (2001). The development of talent in sport. In R. Singer, H. Hausenblas & C. Janelle
(Eds.), Handbook of sport psychology (pp. 269-289). New York: Wiley.
Ellis, J. (1998). Interpretive inquiry as a formal research process. In Ellis, J. (Ed.). Teaching from understanding:
Teacher as interpretive inquirer. (pp. 15-32). New York: Garland Publishing, Inc.
Erickson, K., Bruner, M., MacDonald, D., & Côté, J. (2008). Gaining insight into actual and preferred sources of
coaching knowledge. International Journal of Sports Science and Coaching, 3(4), 527-538.
Fraser-Thomas, J., Côté, J. (2009). Understanding adolescents’ positive and negative developmental experiences in sport.
The Sport Psychologist, 29(3), 3-23.
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.
Fraser-Thomas, J., Côté, J., & Deakin, J. (2008). Understanding dropout and prolonged engagement in adolescent
competitive sport. Psychology of Sport and Exercise, 9(5), 645-662.
Gilbert, W., Gilbert, J., & Trudel, P. (2001). Coaching strategies for youth sports: Part 2: Personal characteristics,
parental influence, and team organization. Journal of Physical Education, Recreation & Dance, 72(5), 41-46.
21
Gilbert, W., & Trudel, P. (2001). Learning to coach through experience: Reflection in model youth sport coaches.
Journal of Teaching in Physical Education, 21(1), 16-34.
Goodger, K., Gorely, T., Lavallee, D., & Harwood, C. (2007). Burnout in sport: A systematic review. The Sport
Psychologist, 21(2), 127-151.
Grove, J., Fish, M., & Eklund, R. (2004). Changes in athletic identity following team selection: Self-protection versus
self-enhancement. Journal of Applied Sport Psychology, 16(1), 75-81.
Jones, R. & Turner, P. (2006). Teaching coaches to coach holistically: Can problem-based learning (PBL) help?
Physical Education and Sport Pedagogy, 11(2), 181-202.
Kirby, S. & McKenna, K. (1989). Experience, research, social change: Methods from the margins. Toronto: Garamond.
Lally, P. (2007). Identity and athletic retirement: A prospective study. Psychology of Sport & Exercise, 8(1), 85-99.
Larson, J. (1984). The performance feedback process: A preliminary model. Organizational Behavior and Human
Performance, 33(1), 42-76.
Lebovitz, H. (1978). Did you ever cut a boy? In R. Martens (Ed.), Joy and sadness in children's sports (pp.
142-144). Champaign, IL: Human Kinetics.
Levy, P. E., Albright, M. D., Cawley, B. D., & Williams, J. R. (1995). Situational and individual determinants of
feedback seeking: A closer look at the process. Organizational Behavior and Human Decision Processes,
62(1), 23-37.
Lipsyte, L. (1979). Varsity syndrome: The unkindest cut. Annals of the American Academy of Political and Social
Science, 445(1), 15-23.
Lyle, J. (2002). Sports coaching concepts: A framework for coaches' behaviour. London: Routledge.
Mallett, C., Trudel, P., Lyle, J., & Rynne, S. (2009). Formal vs. informal coach education. International Journal of
Sports Science & Coaching 4(3), 325-334.
Manzoni, J. (2002). A better way to deliver bad news. Harvard Business Review, 80(9), 114-119.
Mauch, L. (2007). Motivating students through formative feedback. Strategies: A Journal for Physical and Sport
Educators, 20(4), 23-27.
Mouratidis, A., Vansteenkiste, M., Lens, W., & Sideridis, G. (2008). The motivating role of positive feedback in sport
and physical education: Evidence for a motivational model. Journal of Sport Exercise Psychology, 30(2),
240-268.
Musch, J., & Grondin, S. (2001). Unequal competition as an impediment to personal development: A review of the
relative age effect in sport. Developmental Review, 21(2), 147-167.
Nelson, L., Cushion, C., & Potrac, P. (2006). Formal, nonformal and informal coach learning: A holistic
conceptualisation. International Journal of Sports Science & Coaching, 1(3), 247-258.
Orlick, T., & Botterill, C. (1975). Every kid can win. Chicago: Nelson-Hall.
Patton. M.Q. (2002). Qualitative Research and Evaluation Methods (3rd Ed.). Thousand Oaks: CA: Sage.
Pease, D., Locke, L., & Burlingame, M. (1971). Athletic exclusion: A complex phenomenon. Quest, 16, 1.
Potrac, P., Brewer, C., Jones, R., Armour, K., & Hoff, J. (2000). Toward an holistic understanding of the coaching
process. Quest, 52(2), 186-199.
Rodgers, W., Reade, I., & Hall, C. (2007). Factors that influence coaches' use of sound coaching practices.
International Journal of Sports Science & Coaching, 2(2), 155-170.
Roemmich, J., Gurgol, C., & Epstein, L. (2004). Open-loop feedback increases physical activity of youth. Medicine
and Science in Sports and Exercise, 36(4), 668-673.
Ryan, A. (2001). The peer group as a context for the development of young adolescent motivation and achievement.
Child Development, 74(4), 1135-1150.
Sarrazin, P., Vallerand, R., Guillet, E., Pelletier, L., & Cury, F. (2002). Motivation and dropout in female handballers:
A 21-month prospective study. European Journal of Social Psychology, 32(3), 395-418.
22
Sherar, L., Baxter-Jones, A., Faulkner, R., & Russell, K. (2007). Do physical maturity and birth date predict talent in
male youth ice hockey players? Journal of Sports Sciences, 25(8), 879-886.
Smith, J., & Osborn, M. (2003). Interpretive phenomenological analysis. In J. Smith (Ed.), Qualitative psychology: A
practical guide to research methods (pp. 51-80). London: Sage.
Soberlak, P., & Côté, J. (2003). The developmental activities of elite ice hockey players. Journal of Applied Sport
Psychology. 15, 41-49.
Taylor, J., & Ogilvie, B. (1994). A conceptual model of adaptation to retirement among athletes. Journal of Applied
Sport Psychology, 6(1), 1-20.
Thompson, A., Humbert, M., & Mirwald, R. (2003). A longitudinal study of the impact of childhood and adolescent
physical activity experiences on adult physical activity perceptions and behaviors. Qualitative Health Research,
13(3), 358-377.
Thompson, T., & Richardson, A. (2001). Self-handicapping status, claimed self-handicaps and reduced practice effort
following success and failure feedback. The British Journal of Educational Psychology, 71(1), 151-170.
Tuckey, M., Brewer, N., & Williamson, P. (2002). The influence of motives and goal orientation on feedback seeking.
Journal of Occupational and Organizational Psychology, 75(2), 195-216.
Turman, P. (2008). Coaches' immediacy behaviors as predictors of athletes' perceptions of satisfaction and team
cohesion. Western Journal of Communication, 72(2), 162-179.
Wall, M., & Côté, J. (2007). Developmental activities that lead to dropout and investment in sport. Physical Education
and Sport Pedagogy, 12(1), 77-87.
Warriner, K., & Lavallee, D. (2008). The retirement experiences of elite female gymnasts: Self identity and the
physical self. Journal of Applied Sport Psychology, 20(3), 301-317.
Werthner, P., & Trudel, P. (2006). A new theoretical perspective for understanding how coaches learn to coach. The
Sport Psychologist, 20(2), 198-212.
Williams, A. M., & Hodges, N. J. (2005). Practice, instruction and skill acquisition in soccer: Challenging tradition.
Journal of Sports Sciences, 23(6), 637-650.
Wylleman, P., De Knop, P., Ewing, M., & Cumming, S. (2000). Transitions in youth sport: A developmental
perspective on parental involvement. In D. Lavallee, & P. Wylleman (Eds.), Career transitions in sport (pp.
143-160). Morgantown, WV: Fitness Information Technology Inc.
23
Coaching Science
Vol. 4 No. 1
Should coaches use personality

assessments in the talent identification
process? A 15 year predictive study on
professional hockey players
Chris J. Gee* University of Toronto, Canada
John C. Marshall & Jared F. King Self Management Group, Canada
Abstract
Making an accurate and valid prediction about an athlete’s long term success in professional
sport is likely a difficult aspect of a professional coach’s role. Therefore, to aid them in this
evaluative process coaches routinely employ a battery of tests, all of which are intended to
inform their eventual selection decision. To date however, personality inventories have yet to
become common place within this evaluative process; and thus, their predictive utility within the
talent identification process has not yet been adequately tested (Aidman, 2007). Those research
efforts that have been concerned with personality’s role in predicting athletic success have been
overwhelmingly cross-sectional and descriptive in nature, and therefore do not mirror the applied
use (e.g., longitudinal prediction) of these instruments by coaches. Consequently, the purpose of
the current investigation was to address these previous limitations by employing a normative
measure of personality (SportsPro™; Marshall, 1979) and assessing its relationship to athletic
performance over a 15 year time period. Potential draft choices of a Canadian National Hockey
League team (N=124) were profiled prior to the 1991-92 entry draft and were followed until the
end of the 2005-06 NHL season. The proposed selection model was found to be a significant
* Please direct all correspondence to Chris Gee, Department of Exercise Sciences, University of Toronto, Toronto, Ontario,
Canada, M5S 2W6 or via e-mail to chris.gee@utoronto.ca, Telephone: (416)746-0444 ext246, Fax: (416)7466757
25
Chris J. Gee, John C. Marshall & Jared F. King
predictor of a player’s total NHL goals, NHL assists, and their overall NHL points. Overall,
when performance is assessed longitudinally within a relatively homogenous sample of athletes,
personality measures appear to add to a coach’s ability to predict an athlete’s longitudinal athletic
attainment.
Keywords: SportsPro, personality, predictive, performance, hockey
Introduction
It is a widely held belief within competitive sport circles that successful teams, especially
those that become legacies, are built through the draft (Sabino, 2009). Teams that are not only
able to identify the stars of today, but also the stars of the future, appear to build a solid
foundation of players upon which a winning team can be constructed. As such, talent
identification appears to be a paramount ingredient in a team’s long term success.
Current scouting and talent identification procedures are becoming increasingly more
multidimensional, comprised of a variety of anthropometric (e.g., height, weight, VO2max, lactic
acid threshold), psychological and interpersonal assessments. This movement towards a more
robust and comprehensive evaluative process within the professional ranks can likely be attributed
to the increased importance being placed on the draft process as a result of league imposed
salary cap restrictions. Moreover, this trend reflects the importance and breadth of knowledge
being generated within the sport sciences, and our current conceptualization of athleticism as a
multifactoral construct. Consequently, teams are looking for pieces of information that can add
predictive value to their selection considerations, while simultaneously helping them to avoid draft
blunders [e.g., Ryan Leaf (NFL), Todd Van Popple (MLB), Jason Bonsignore (NHL), Darco
Milicic (NBA)].
The influence of personality on athletic performance has been a widely researched and
heavily contested topic since the inception of sport psychology as an academic discipline (Deaner
& Silva, 2002). Previous research endeavors have been concerned with examining personality
differences between athletes and non-athletes (Eagleton, McKelvie, & de Man, 2007; Reiss, Wiltz,
& Sherman, 2001), athletes who occupy various positions within a sport (Greenwood & Simpson,
1994; Newcombe & Boyle, 1995; Singer, 1969), athletes from different sports (Eagleton et al.,
2007; McKelvie, Lemieux, Stout, 2003), and most importantly, top performing athletes and the
rest of the field (Beedie, Terry, & Lane, 2000; Gat & McWhirter, 1998; Piedmont, Hill, &
Blanco, 1999). Overwhelmingly, the applied angle of these studies has been to support the use of
personality testing in the talent identification process, and thus, highlight the utility of
26
understanding an athlete’s personality composition when attempting to predict their future athletic
attainment. Unfortunately, due to a variety of methodological shortcomings (e.g., heterogeneous
sample, cross-sectional analyses, ipsative tools) these studies have fallen significantly short in this
pursuit, which has led many sport scientists and coaches to view the hypothesized relationship
between personality and athletic performance in a cynical and pessimistic manner (Eysenck, Nias,
& Cox, 1982; Vealey, 1992).
One of the most frequently levied criticisms against academic studies of personality and
sport performance has been the pervasive use of ipsative inventories. Ipsative tests, as described
by Baron (1996), “are force-choice questionnaire formats, where respondents order sets of items
loading on different scales” (p.1). Due to this structural design, “ipsative scores not only fail to
meet the assumptions of classical psychometric analysis (especially the assumption of error
independence), they also constitute an essentially ordinal level of measurement” (Baron, 1996;
p.1). In an applied sense, ipsative personality tests (e.g., MBTI, DISC, TAP) provide a very
descriptive overview of a particular individual, and thus can be used for coaching and
intra-individual purposes, but due to their design properties cannot be used to make valid and
reliable interpersonal comparisons (Mead, 1994). On the other hand, normatively scored
personality tests are those instruments that employ a Likert scale format (e.g., 1 = strongly
disagree; 5 = strongly agree) and use the unrestricted sum of each scale to represent a subject’s
score on a given construct. This questionnaire format allows for the establishment of norms and
interpersonal comparisons, minimizes scale and item inter-correlations, and has been shown to
have much stronger psychometric properties (Meade, 1994). As such, it has been the
recommendation of several behavioral scientists that normative personality inventories be
considered “best practices” in the prediction of an individual’s future success (Cattell & Brennan,
1994; Heggestad, Morrison, Reeve, & McCloy, 2006; Johnson, Wood, & Blinkhorn, 1989;
Meade, 1994). Unfortunately, these suggestions and recommendations have not yet been adopted
by talent identifiers or academics.
Another major limitation associated with the personality literature in sport, as well as a
major disconnect from the applied use of these tools, has been its descriptive and cross sectional
nature (Aidman, 2007; Vealey, 1989). Previous studies have routinely employed research designs
where athletes are grouped by gender, position, sport, and sometimes current performance ranking,
and then tested and compared according to a personality inventory. As Aidman (2007; p.3) points
out, “sport psychologists’ brave but often simplistic pursuit of personality seems to have largely
ignored the fundamental developments in mainstream personality research.” Consequently, these
early studies concerned with personality and sport performance provided such varied results, much
of which was simply descriptive, that making external generalizations was deemed almost
impossible (Aidman, 2007; Aidman & Schofield, 2004; Van Auweele, Nys, Rzewnicki, & van
27
Mele, 2001). As such, these results have fuelled the skeptical argument against the utility of
personality tests in the prediction of athletic performance and are least partially responsible for
the limited adoption of these instruments in the talent identification process (Deaner & Silva,
2002).
One of the “fundamental developments” in mainstream personality research that Aidman
(2007) was referring to is the longitudinal manner in which personality is able to predict desired
outcome variables (e.g., athletic performance). Therefore, rather than being a strong predictor of
within competition or short-term performance metrics (e.g., daily or weekly performance),
personality’s predictive contributions are believed to be much more valid and reliable when
assessed long-term (Hogan, 1998; Hogan & Shelton, 1998). For example, Aidman (2007) found
that personality constructs were not effective predictors of junior level performance among a
sample of Australian Rules Football players; however, these same constructs were found to be
extremely predictive of who ultimately succeeded at the senior level of competition when
assessed seven years later. This operational shift towards a life-span approach to the study of
personality and athletic performance requires the use of longitudinal research designs, which
although called for in the area of sport psychology (Morris, 1995), have yet to become common
practice. Therefore, the purpose of the current investigation was to test the predictive contribution
of a normative personality profile on athletic success over a 15 year time period. Not only does
this address a current methodological shortcoming in the sport sciences, but also tests the utility
of personality assessments in a way that mirrors their current application in professional sport.
Methods
Subjects were 124 amateur hockey players attending an evaluation camp for a National
Hockey League team prior to the 1991-92 entry draft. Players were asked to complete a
normative personality profile (i.e., SportsPro™) by the team psychologist prior to arriving at
camp. Administration of the profile was done before the talent identification camp to minimize
social desirability bias and interpersonal comparisons, as these are both criticisms that have been
levied against self report measures of personality in the past (Vealey, 1992).
Only players who competed in at least 82 regular season games (N=49) were included in
the regression analysis [of the 124 players who completed the assessment, 110 were drafted into
the NHL, 27 of these draftees never played a game, and 34 played less than 82 games]. The 82
game criterion was established in an attempt to isolate “true” NHL players, and thus omit “call
ups” whose performance would reflect a lack of opportunity and not necessarily their inherent
potential.
28
The SportsPro™
The SportsPro™ (Marshall, 1979) is a self-report normative personality inventory that is
comprised of 168 adjective and 96 attitudinal items, all of which are scored on a 5-point Likert
scale. The SportsPro™ measures athletes on the following attributes: competitiveness, need for
achievement, independence, people orientation, analytical disposition, comfort with conflict,
coachability, self-confidence, pre-competitive anxiety, athletic identity, sportspersonship and finally,
their attitudes towards athletics.
The items and constructs that comprise the SportsPro™ have been extensively validated in
a variety of achievement contexts; most notably competitive sales cultures (King & Gee, 2009;
Marshall, 1979; Marshall & McHardy, 1997; Marshall & McHardy, 1999; McHardy & Marshall,
2003). For instance, using a sample of 30,393 commission-based sales representatives the internal
reliability of the scales ranged from .76 to .91. Moreover, test-retest reliability co-efficients
ranged from .90 to .97 for a subset of these individuals (N=25) who were tested again after a
month (Marshall, 1979). These test-retest results highlight the stable and dispositional nature of
the SportsPro’s personality constructs.
Previous research employing the SportsPro’s items and constructs in non-sporting
achievement contexts (e.g., business, academics) has provided a fairly stable personality profile
associated with “top performers” (Marshall, 1979; Marshall & McHardy, 1997; Marshall &
McHardy, 1999; McHardy & Marshall, 2003). Moreover, this “top performer” profile also closely
resembles many of the findings already present within the broader psychological literature on
personality and performance (Lamont & Lundstrom, 1977; Nicholson, 1998; Warr, Bartram, &
Martin, 1995) The stability and consistency of these attributes across different achievement
contexts appears to support their hypothesized relationship with performance. As such, the five
independent characteristics (i.e., competitiveness, need for achievement, independence potential, self
confidence, and coachability) associated with this “top performer” profile were synthesized to
create a composite personality score (0=not ideal, 5=top performer). Therefore, players who
possessed all five of these personality traits were given a score of 5, with those possessing none
of these attributes receiving a score of zero. This composite personality score was subsequently
entered as the predictor variable in the current regression design.
Performance
Players’ cumulative statistics were compiled after the 2005-2006 NHL season. Games
played, goals, assists, penalty minutes, and the number of NHL teams they played for were
included in the analyses. Goaltenders were omitted from the investigation because of the
performance metrics being used.
29
Information pertaining to the draft (e.g., whether or not the players were drafted, which
round, which pick) was not included in this study, because the personality information was not
used by teams during the selection process. As such, selection decisions during the 1991 – 1992
entry draft were based on internal scouting metrics, and therefore the personality data could only
be validly used in this study to predict future athletic performance.
Data Cleaning
As per Tabachinick and Fidell’s (1996) recommendations for multiple regression analyses,
univariate statistics were computed for all variables being entered into the models with no
extreme skewness or kurtosis observed. As only one predictor variable was being entered into the
regression designs, multicollinearity and multivariate outliers were a non-issue.
Results
In order to control for inflated Type I statistical error, regressions were only computed on
those variables that displayed a significant correlation with the composite personality score
(Tabachinick & Fidell, 1996). A total of three simultaneous regression models were computed.
Players’ composite personality scores were found to significantly predict the number of goals
2 2
[R =.084; F(1, 47)=4.31, p<.05], assists [R =.087; F(1, 47)=4.67, p<.05], and total points
[R2=.087; F(1, 47)=4.65, p<.05] that players’ accumulated over this 15 time period. Interestingly,
when height and weight (both commonly cited anthropometric indices used when scouting
amateur hockey players) were entered into each of the previously mentioned regression models,
they failed to significantly increase the amount of variance accounted for. This is likely due to a
lack of variability in these metrics among this homogeneous sample of draft-age hockey players.
Additional exploratory analyses uncovered a potential relationship between the comfort with
conflict scale and total penalty minutes (r=.22, p<.05), whereas the coachability attribute and the
number of times a player was traded were found to negatively correlate, but failed to reach
statistical significance (r=-.25, p<.08).
Discussion
The purpose of the current investigation was to assess the predictive potential of a
normative personality inventory on the long term success of a sample of junior hockey players.
In doing so, this study addresses a fundamental criticism levied against earlier personality
30
research (i.e., cross-sectional), while also assessing the utility of these instruments in an
ecologically valid manner.
The prediction of athletic achievement from personality traits was one of the earliest, and
continues to be one of the most attractive, applications of applied sport psychology (Aidman,
2007; Cooper, 1969; Griffith, 1928). However, the academic literature supporting the utility of
this practice has been equivocal at best, leading many coaches and sport teams to be skeptical of
its predictive potential (Deaner & Silva, 2002). Nevertheless, the use of personality-based
assessment tools is currently becoming more common among professional sports teams, as they
search for any advantage in the identification and prediction of an athlete’s likelihood for long
term success.
The results of the current study provide preliminary support for the use of normative
personality profiles in the prediction of athletic success when measured longitudinally in the form
of a career. Athletes who possessed the “top performer” profile on the SportsPro™ significantly
outperformed those athletes who lacked in one or more of these attributes over the 15 year time
period under evaluation. When entered into a regression model this composite personality profile
explained roughly 10% of the variance comprising the criterion performance metrics (i.e., goals,
assists, total points). As such, personality profiles do appear to significantly contribute to the
talent identification process; however, should not likely be used in an independent manner.
Therefore, when employed in conjunction with other commonly cited scouting (e.g., junior-level
performance, height, weight, strength, aerobic capacity) and psychological metrics (e.g., Mental
Skills Inventory, CSAI), personality profiles may help coaches, scouts and psychologists more
accurately separate the “best” from the “rest” (Humara, 2000; Spieler, Czech, Joyner, Munkasy,
Getner & Long, 2007). Future studies should look to regress personality traits simultaneously
with these other commonly cited scouting metrics in order to quantify their independent
contribution to the prediction of long term performance. Doing so will not only identify which
constructs are the most important predictors of athletic performance, but will also allow teams to
put forth the most parsimonious and predictive selection model possible.
The current study also reinforced the “top performer” profile that has emerged consistently
when the items and constructs comprising the SportsPro™ have been examined in achievement
contexts (Marshall, 1979; Marshall & McHardy, 1997; Marshall & McHardy, 1999; McHardy &
Marshall, 2003). Top performers possess an above average competitive disposition, are motivated
by challenge and reward, confident in their ability to succeed, are open to coaching and
feedback, and can operate both independently and as part of a group. Interestingly, it appears to
be the interaction of these traits, and thus the entire package that ultimately influences
performance as the removal of any one of these characteristics was shown to have a negative
impact on a player’s likelihood to succeed. This in turn reinforces the idea of assessing each
31
player in the most holistic manner possible in order to maximize the predictive validity of the
selection decision. This differs from previous studies, and in some cases the applied use of
personality and psychological assessments, where single constructs (e.g., mental toughness) have
been the sole focus of the evaluative process (e.g., mental toughness, emotional stability).
Applied Contribution
The results of this research endeavor have significant implications in the coaching sciences
and overall talent identification process. Most, if not all, professional sport teams place a great
deal of importance on talent identification and draft selection. As such, coaches and sporting
administrators now have empirical support for the utility of employing normative personality tests
in the prediction of athletic success and a better understanding of how these instruments should
be employed and ultimately interpreted (i.e., long term, not short-term predictions).
Personality inventories such as the SportsPro™ can also be used effectively as coaching and
development tools. These profiles provide a very thorough top-to-bottom description of each
athlete, which can inform the unique strategies that coaches and psychologists employ as part of
the athlete’s development. Moreover, as personality is a strong driver of behavior (i.e., typical
responses), understanding an athlete’s inherent disposition will allow coaches to more accurately
predict how athletes will respond in various competitive situations (e.g., failure, big game, clutch
situation, fame). As a result, coaches can proactively work on developing athletes in these
various areas and providing them with strategies and/or techniques to address these potential
growth opportunities.
Normative assessments like the SportsPro™ may also be useful in understanding, managing,
and ultimately modifying team dynamics. In situations where there is coach-athlete conflict,
helping the coach better understand how their natural style may be in contrast to the athlete’s
inherent preference for leadership may help the coach customize and tailor their style to suit this
particular athlete’s needs. This individualized approach to coaching should facilitate stronger
coach-athlete relationships while also having an indirect impact on athlete performance
(Chelladurai, 1978). Consequently, by using the personality inventories to gain additional insight
into their players, coaches can increase their likelihood of achieving both their social and
task-oriented objectives.
Overall, personality profiles should help coaches not only select higher probability for
success players in the future, but also maximize their ability to understand and relate to each
individual athlete.
32
References
Aidman, E.V. (2007). Attribute-based selection for success: The role of personality attributes in long term predictions
of achievement in sport. The Journal of the American Board of Sport Psychology, 3, 1-18.
Aidman, E.V., & Schofield, G. (2004). Personality and individual differences in sport. In T. Morris & J. Summers
(Eds.) Sport Psychology: Theory, Applications and Issues (2nd Ed.) (pp. 22-47). Milton: Wiley.
Baron, H. (1996). Strengths and limitations of ipsative measurement. Journal of Occupational and Organizational
Psychology, 69, 49-56.
Beedie, C.J., Terry, P.C., & Lane, L.M. (2000). The profile of mood states and athletic performance: Two
eta-analyses. Journal of Applied Sport Psychology, 12, 49-68.
Cattell, R.B., & Brennan, J. (1994). Finding personality structure when ipsative measurements are the unavoidable basis
of the variables. The American Journal of Psychology, 107, 261-274.
Chelladurai, P. (1978). A contingency model of leadership in athletics. Unpublished doctoral dissertation, University of
Waterloo, Canada.
Cooper, L. (1969). Athletics, activity, and personality: A review of literature. Research Quarterly, 40, 17-22.
Deaner, H., & Silva, J.M. (2002). Personality and sport performance. In J.M Silva & D.E. Stevens (Eds.)
Psychological Foundations of Sport (pp. 48-65). Boston, MA: Allyn and Bacon.
Eagleton, J.R., McKelvie, S.J., & De Man, A. (2007). Extraversion and neuroticism in team sport participants,
individual sport participants, and nonparticipants. Perceptual & Motor Skills, 105, 265-275.
Eysenck, H.J., Nias, D.K.B., & Cox, D.N. (1982). Sport and personality. Advances in Behavior Research and Therapy,
4, 1-55.
Gat, I., & Mcwhirter, B.T. (1998). Personality characteristics of competitive and recreational cyclists. Journal of Sport
Behavior, 21, 408-420.
Greenwood, M., & Simpson, W.K. (1994). Personality traits of intercollegiate baseball athletes playing central versus
noncentral defensive positions at three levels of competition. Applied Research in Coaching and Athletics
International, 1, 15-30.
Griffith, C.R. (1928). Psychology and Athletics. New York: Scribners.
Heggestad, E.D., Morrison, M., Reeve, C.W., & McCloy, R.A. (2006). Forced-choice assessments of personality for
selection: Evaluating issues of normative assessment and faking resistance. Journal of Applied Psychology, 91,
9-24.
Hogan, R. (1998). Reinventing personality. Journal of Social and Clinical Psychology, 17, 1-10.
Hogan, R., & Shelton, D. (1998). A socioanalytic perspective on job performance. Human Performance, 11, 129-144.
Humara, M. (2000). Personnel selection in athletic programs. Athletic Insight, 2,
http://www.athleticinsight.com/Vol2Iss2/Personnel.htm
Johnson, C.E., Wood, R., & Blinkhorn, S.F. (1989). Spuriouser and spuriouser: The use of ipsative personality tests.
Journal of Occupational Psychology, 61, 153-162.
King, J., & Gee, C.J. (2009). “Using personality assessments to improve the recruitment and selection of high
performing athletes”. Paper presented at the annual meeting of the Eastern Canada Sport and Exercise
Psychology Symposium, Toronto, ON, March.
Lamont, L.M., & Lundstrom, W.J. (1977). Identifying successful industrial salesmen by personality and personal
characteristics. Journal of Marketing Research, 14, 517-529.
Meade, A.W. (1994). Psychometric problems and issues involved with creating and using ipsative measures for
selection. Journal of Occupational and Organizational Psychology, 77, 531-552.
Marshall, J. (1979). SportsPro Validation Manual. Toronto, Canada: Selection Testing Consultants International.
Marshall, J., & McHardy, B. (1997). Selecting Sales Professionals. Toronto, Canada: Selection Testing Consultants
33
International.
Marshall, J., & McHardy, B. (1999). Principles of Self Management. Toronto, Canada:Selection Testing Consultants
International.
McHardy, B., & Marshall, J. (2003). Managing Effort Getting Results. Toronto, Canada: JM & RM Holdings.
McKelvie, S.J., Lemieux, P., & Stout, D. (2003). Extraversion and neuroticism in contact athletes, non contact athletes,
and non-athletes: A research note. Athletic Insight, 5,
http://www.athleticinsight.com/Vol5Iss3/ExtraversionNeuroticism.htm
Morris, T. (1995). Psychological characteristics and sports behaviour. In T. Morris & J. Summers (Eds.), Sport
psychology: theory, applications, and issues (pp. 3-27), Milton: John Wiley and Sons.
Newcombe, P.A., & Boyle, G. J. (1995). High school students’ sport personalities: Variations across participation level,
gender, type of sport, and success. International Journal of Sport Psychology, 19, 247-263.
Nicholson, N. (1998). Personality and entrepreneurial leadership: A study of the heads of the UK’s most successful
independent companies. European Management Journal, 16, 529-539.
Piedmont, R.L., Hill, D.C., & Blanco, S. (1999). Predicting athletic performance using the five factor model of
personality. Personality and Individual Differences, 27, 769-777.
Reiss, S., Wiltz, J., & Sherman, M. (2001). Trait motivational correlates of athleticism. Personality and Individual
Differences, 30, 1139-1145.
Sabino, D. (2009, June 9). MLB draft rankings: How has each team fared over the past 10years. Sports Illustrated.
Retrieved August 31, 2009 from http://sportsillustrated.cnn.com/2009/writers/david_sabino/06/07/draft.rankings/
Singer, R.N. (1969). Personality differences within and between baseball and tennis players. Research Quarterly, 40,
582-588.
Spieler, M., Czech, D.R., Joyner, A.B., Munkasy, B., Gentner, N., & Long, J. (2007).Predicting athletic success:
factors contributing to the success of NCAA division IAA collegiate football players. Athletic Insight, 9,
http://www.athleticinsight.com/Vol9Iss2/PredictingSuccess.htm
Tabachnick, B.G., & Fidell, L.S. (1996). Using multivariate statistics (3rd Ed.). New York, NY: Harper Collins
College Publishers.
Van Auweele, Y. Nys, K., Rzewnicki, R., van Mele, V. (2001). Personality and the athlete. In Singer, R.N.,
Hausenblas, H.A., & Janelle, C.M. (Eds.). Handbook of Sport Psychology (2nd ed.) (pp. 239-268). New York:
Wiley.
Vealey, R.S. (1989). Sport personology: A paradigmatic and methodological analysis. Journal of Sport and Exercise
Psychology, 11, 216-235.
Vealey, R.S. (1992). Personality and sport: a comprehensive review: In T.S. Horn (Ed.), Advances in Sport Psychology
(pp. 25-59). Champaign, IL: Human Kinetics.
Warr, P., Bartram, D., & Martin, T. (1995). Personality and sales performance: situational variation and interactions
between traits. International Journal of Selection and Assessment, 13, 87-91.
34
Coaching Science
Vol. 4 No. 1
The learning formats of coach

education materials
John Lyle* University of Queensland, Australia
Sue Jolly & Julian North SportsCoach UK, UK
Abstract
This paper reports on an investigation into the design of coach education materials. At a time of
significant change in coach education in the UK, it was appropriate to examine the extent to which
coach education materials conform to good practice in terms of adult learning design principles and
the formatting of the content. Coach education materials from Levels 1-3 of coaching awards in 5
sports were content analysed. These data were accompanied by opinions from 60 coaches and
tutors, a focus group, and number of interviews. In the absence of a substantial literature on this
aspect of coach education, a framework of analysis criteria was derived from the adult learning
literature and prescriptions of good practice from those who had substantial experience in materials
design. The framework elements (each with sub-groups of 5 to 18 items) were progression,
feedback, interactivity, reflection, operation, contextualisation, summarising, attractiveness, layout &
structure, and accessibility. The coach education materials were mixed in quality, and the analyses
were usefully focused on two issues - presentational formatting and learning principles design
(structural formatting). Overall, interactivity, reflection and feedback in the learning materials were
less positive than might have been expected. The project was able to provide a valuable set of
operational criteria for guiding the future production of learning materials in coach education.
Keywords: coach education, coaching resources, resource design, learning materials design
* Prof John Lyle, John Lyle Consulting, 0044 - (0)1670 511853, 0044 - (0)7590 108098, E-mail: jlyle.consulting@btinternet.com
35
John Lyle, Sue Jolly & Julian North
Introduction
This short paper provides a summary of a project to evaluate the learning design qualities
of coach education materials. The focus is on the construction of an analysis framework with
which to evaluate learning materials, and the issues and implications for coach education
managers. There is no doubt that there are many perceived weaknesses in the design and
delivery of coach education. These were acknowledged in earlier writings (Campbell 1993,
Coaching Review Panel 1991), and in more recent sources (Abraham & Collins 1998, Cassidy,
Jones & Potrac 2004, Cushion, Armour & Jones 2003, Jones, Armour & Potrac 2004,
Liukkonen, Laako & Temala 1996, Lyle 2002, Saury & Durand 1998). The academic literature
on coach education is very sparse (Jones 2006, Gilbert 2006, Lyle 2002, Lyle 2007, Trudel &
Gilbert 2006) and demonstrates that rigorous evidence for the effectiveness of coach education is
very rare. The design of coach education learning programmes and learning resources has not, to
date, been treated as problematic, although perhaps recognised to be so by practitioners. However,
the evident need to develop and improve the resources to accompany coach education provided
an opportunity for further investigation of these issues.
The investigation was designed to examine the existing state of knowledge on learning
material design, review current practice, ascertain and analyse coaches’ and tutors’ opinions, and
make recommendations for the development of effective learning material formats. In particular
we were concerned to establish the extent to which materials used to support coaches’ learning
had been designed in such a way that the basic principles of adult learning were demonstrated.
More specifically, the main aims of the research were (a) to undertake a critical analysis of the
learning material formats and design used by a sample of five sports, (b) to establish the
opinions and preferences of a sample of coaches and tutors; and (c) to provide recommendations
for the formatting and design of resource and learning materials.
It is important to describe the purpose of the work that was carried out. Although the
investigation was conducted with rigour, it did not satisfy many of the basic tenets of the
research procedures associated with systematic content analysis or qualitative analysis. Our
purpose was to create a usable, working tool for analysis and to use that tool to provide an
initial position statement, or insight, into coach education learning materials. We ask the reader to
accept that our devising of the analytical framework involved considerable interpretation, and our
simple scoring system produced trends rather than rigorously discriminating data. Nevertheless, the
paper provides a research agenda. Our purpose was to inform policy makers and to stimulate
consideration of an issue that had received insufficient attention to date. We present the paper in
that context. We conducted a number of workshops with coach education managers to accompany
36
the investigation and we are confident that it provoked significant questions about existing
practice in the design of materials in coach education.
Review of literature
Although there is an emerging academic literature in coach education, there is little tradition
of impact or evaluation studies, and no existing evidence on the effect of different learning
material formats or designs on the coach’s learning. Similarly, a preliminary search for more
general research evidence on the presentational design of learning materials was unsuccessful.
This body of literature is characterised by prescriptions of good practice based on the authors’
experience of producing such materials, albeit with an underpinning of sound learning principles.
There was no substantial body of empirical evidence. Further, there was a good deal of repetition
in prescriptions and in authors’ published outputs, and considerable overlap between prescriptions.
Nevertheless, there appeared to be a significant level of consensus on what constituted good
practice in formatting and design.
The correspondence between adult learning (also termed post-compulsory learning) and
coach education provision made this literature an obvious starting point. Rowntree (2000) profiles
adult learners, identifying traits that will impact their learning. These include being rich in
experience (or attitudes) relevant to the subject they are learning, goal-orientated, but with their
own agendas to pursue, and informed by previous learning experiences. Adult learners are also
likely to be balancing other life interests and pressures, and expecting value for time and money
invested. Race (2005) draws together the literature on learning and examines the implications for
design and delivery of often-cited authors in learning theory (Knowles 1990, Kolb 1984, and
McCarthy 1987). He then uses this literature and his considerable experience to identify and
emphasise five factors that underpin effective learning. These are: wanting to learn, taking
ownership of the need to learn, learning by doing, learning through feedback, and making sense
of what has been learned.
Race also provides a checklist of design principles (2004, 2005). Within this he suggests a
structured approach to facilitating a positive learning environment. This is achieved as the learner
moves from wanting/needing (which is initiated, enhanced and clarified through relevance and
attractiveness), to doing (for which ‘engagement’ is provided), then to digesting (through
structured application and reflection), and finally to evaluation (by giving the learner feedback).
The writing contains a number of statements that draw attention to key messages, particularly
about ensuring that information is ‘turned into’ knowledge, there is a structured learning
environment, and the role and value of feedback is emphasised.
37
Brown and Gibbs (1996) provide a useful reminder in the context of this project that the
emphasis is on the learning process created by the learning resource, and not merely the quality
and appropriateness of the content itself. Brown and Gibbs suggest that, without appropriate
learning activities, assessment demands, and feedback, the learning materials may fail to stimulate
and reinforce the learning that is assumed to take place. Race (1994) provides guidelines for
facilitating effective learning that can be interpreted as a set of quality indicators for learning
materials. This set of guidelines provides a challenging list of requirements for the design of
learning materials. They provoke consideration of presentational/formatting issues, and structured
learning engagement issues.
Rowntree has written extensively on matters relating to resource-based learning, based on
his very considerable experience of producing materials for the Open University (1990, 1992,
1994, 1997). In Making Materials-Based Learning Work (Rowntree 1997), he highlights key
design features, which he lists as access, interactivity, feedback, layout and structure, and
effectiveness. The contribution of interactivity is emphasised and this highlights the challenge for
the learning materials designer in producing the opportunities for levels of interaction that might
be present in face-to-face learning programmes. Rowntree (1999a) charts the development of
distance education through a number of phases. He points to the emergence of the Open
University as a turning point away from a reliance on textbooks to more learner-centred
workbooks. These newer distance-learning materials were learning-centred and emphasised
relevance, understanding and engagement. These good practice criteria are grouped into what he
calls ‘success factors in materials-based learning’ - activities, integration, interactivity, objectives,
structure and user-friendliness (1999b).
Brown and Smith (1996) acknowledge design and layout skills alongside other key resource
production features such as subject knowledge, understanding learning theory, and effective
evaluation skills. They highlight four design principles, (a) student engagement through interaction,
(b) developing a range of learner activities, (c) comprehensive feedback loops, and (d) regular
updating/review. They also identify simple design features to avoid (for example, word overload),
and those to incorporate, such as diversity and creativity, and self-progress checks.
Further design guidance comes from Rust and Wilson (1996). Within a series of
recommendations, they emphasise the need for progression, breaking the material into manageable
chunks, and making the material personal to the reader. The authors also have some specific
advice on layout; attractiveness, simplicity, signposting, and evidence of progress are their
recommendations. There is a subset of the literature that provides useful but relatively simplistic
guidelines for page formatting (see Parker 1998, Powell 1991). These tend to have been produced
for study guides and workbooks with a less ambitious scope and complexity of learning
ambitions/outcomes. Within the literature there is also a reminder to acknowledge cultural
38
diversity, and ‘be careful of references which include assumptions about gender, race, language,
history, and so on’ (Endean 2003).
There was very little direct evidence of effective coach education design. Recommendations
were based on sound theoretical positions, but had rarely been converted into ‘materials’.
McCullick et al (2005) reported that the coaches/instructors in their study favoured logical,
sequential and comfortable formats, wanted pedagogical knowledge taught and modelled, their
educators to be knowledgeable, and the introduction of pertinent pedagogical research to be
apparent. In many ways this is at odds with the more flexible and learner-centred approaches.
The authors suggest that the ‘golf coach/instructor’ context may have influenced the findings. One
answer to how coach education might allow coaches to develop the necessary learning is
described by Jones et al (2004) who suggest that coach education should be characterised by
contextualisation, problematisation, activity and application. Cassidy et al (2004) suggest
conceiving of coaching as an intellectual exercise, focusing on integration and synthesis, and
presenting an holistic picture of coaching practice. These authors focus upon the use of critical
tasks, problem-based learning, narratives, and mentoring. Trudel & Gilbert (2006) acknowledge
that programmes are traditionally ‘classroom based’, but identify alternative methods such as
videos, CD-ROM and distance learning.
Constructing an analytical framework
The investigation required a set of evaluation criteria that could be used to analyse existing
learning materials in coach education. Although there were fairly extensive prescriptions from a
number of authors, the information had not been sufficiently aggregated and no comprehensive
framework existed that ranged across both learning structure and matters of presentation and
layout. Therefore, the approach adopted was a pragmatic one. Literature sources were used to
construct a usable tool for analysis. It was considered more important to have a workable
framework than to validate the ‘instrument’ and its items.
A simple procedure was adopted for constructing a set of evaluation criteria. The literature
was reviewed for recommendations/prescriptions on good practice, and for principles underlying
good design in distance and adult education. A ‘long list’ of all relevant
themes/features/principles/recommendations was created. This resulted in a very extensive list of
items, although there was some considerable duplication. The list was refined in two ways. First,
overarching principles of adult learning, attractive design, and user-friendliness were used to ‘sort’
the materials. Second, a simple but more detailed process of refining (organising and
synthesising) the items into manageable and discrete categories took place. This involved a
39
process of trial and error; checking terminology overlaps, allowing a degree of overlap between
categories; and refining through discussion with colleagues.
This is a description of how we created the evaluation criteria, rather than a defence of its
reliability and validity. We were guided by the emphases that we had found in the literature, our
experience, and a desire for criteria that were comprehensive but usable. Identifying the
‘sub-items’ or elements within each category ensured sufficient coverage and minimum overlap.
Although we felt that discussion with colleagues, reinforcement within the literature, and
subsequent workshops gave the framework a certain validity, the key question was whether it
would be a useful tool. Our experiences suggest that this was the case. As a result of this
process, the criteria were finally ordered into 11 categories. The following table describes those
category headings and the number of sub-items or elements associated with each one. Three
examples of this list of elements are provided.
Category Brief Descriptor Elements

Progression Progressive steps from information to application in obvious incremental 12
steps; content structured from simple to complex, links to other resources;
a clear sequencing rationale
Feedback Promotion of independent study; comments on progress; provides 18
self-assessment questions or tasks; responses are provided that encourage
reflection
Interactivity Materials and their formatting encourage communication with others; 12
engages the coach in learning tasks; invites the coach to apply the material
Reflection Application to practice; effective self-evaluation skills are developed; 8
problem-solving scenarios are used; alternative interpretations are
encouraged
Operation The relationship between the materials and the remainder of the 6
course/programme is evident; evident instructions for use; adequate
signposting
Contextualisation No cultural, racial, gender or ethnic bias; the materials are related to the 7
coach’s role within the sport; obvious links to assessment and the learning
objectives
Handling The resources are easy to manipulate; store; can be amended or expanded 5
as required
Summarising Key points are emphasised; there is signposting of summaries; the 13
summaries are concise, simple and related to learning outcomes
Attractiveness Learning materials are pleasing to view, varied, and stimulating; a mix of 10
illustrations, diagrams and fonts; uses colour and boldness to provide
emphasis
Layout and Format of the materials is visually appealing; ‘chunked’ into appropriate 16
Structure sections and topics; provides feedback loops; uncluttered in presentation
Accessibility Learning materials are formatted to enable understanding of the language 12
and content; user-friendly; stimulating; appropriate indexing; language is
appropriate to the ‘level’ of the resource
40
Summarising
Layout is simple and illustrates summary points in a learner-friendly format
Structure is concise, simple, and un-confusing
Short versions of key points/findings are provided
Main points are given at the beginning of chapters/sections
Immediate summaries follow the end of a section
Bullet points are used to highlight summary sections
Signposting summaries are widely used
Key statements use ‘bold’
Colour is used to promote key points of learning
Simple formatting is used to write the summaries
Summaries highlight/emphasise the key learning points
Summaries were ‘chunked’ and related to learning outcomes
Any synopses are concise, and do not merely provide repetition
Attractiveness
The material pleasing/stimulating in appearance
The material was interesting (photographs, diagrams, line drawings)
The resource promoted questioning
Strong colour schemes were used to highlight key areas of learning
Font used was modern and readable
Text size varied
The resource had large amounts of white space, illustrations and diagrams
Material related in presentation to the role or sport
The material has an immediate attraction
Visual interest is stimulated by coherent design
Interactivity
The resource promotes (encourages, demands) communication within course cohort groups
The material design promotes ‘engaging’ (tasks, activities, questions) in order to prompt reaction and
reflection
The interactive theme of the resource is promoted by frequent group or individual study tasks which help
make sense of what has been learnt
The resource helps candidates focus on and enhance the desire to learn (by relating to practice and
improvement)
The resource is clear on ‘what has been learnt’ and ‘what still needs to be learnt’
The feedback and guidance is directly or indirectly ‘led’ by the tutor or mentor
Learner interaction is helped by clear instructions on how to engage or not (i.e. complete the written task
or simply read it)
The resource promotes free-ranging discussions, role plays and interviews
The resource promotes intra-group communication and generates discussion
The material promotes ‘acting’ on the information in own setting, being independent
The resource invites cooperation and encouraged personal input into tutorials
The resource material relates key findings to learner practice
Analysis of sport specific coach education learning materials
The purpose of this part of the investigation was to provide a snapshot of existing
coaching education learning materials in terms of their design and formatting and to highlight
strengths and weaknesses. This would allow us to identify potential issues.
41
Methodology
Coach education materials were obtained for three levels of award across a sample of 5
sports. Each of the resources was evaluated using a simple scoring system based on the
categories and criteria identified in the analytical framework. Each product was graded against the
assessment criteria contained in the analytical framework. Each element was given a score out of
three: 3 being an excellent example of that feature; 2 indicated a good/satisfactory example; 1
represented a poor example; and 0 was used if there was no evidence of that criterion. The
scores were then calculated as a percentage of the total possible scores to allow for comparison
with other resources, and to enable patterns of scoring between categories to be evaluated.
Efforts were made to ensure that scoring was carried out in a consistent and reliable
fashion. First, it was necessary to resolve the question of the standard against which the elements
should be evaluated. It was not considered feasible to identify empirically-based evaluation criteria
for each item. However, it was agreed with the project client that there would be some value in
identifying exemplars of good practice that could be used to initiate a discussion on relative
quality. Two resource products were identified that were considered to have appropriate quality
across a range of the categories and elements (one textbook, and one coach education resource
from another sport). These resources were used to provide some reference points for evaluation.
Second, the exemplar materials were used to provide a pilot set of evaluations. These were
carried out by a researcher and a team of external colleague from the client organisation. The
results of this blind testing were compared, and this led to further discussion about interpretation
of criteria. The evaluation was carried out by a researcher, and repeated by the lead author.
Results
The percentage mean and median scores for each category, across the 15 sets of resources,
are shown in the following table. The significant range of scores within each category suggests
that there were significant variations across elements and/or that further work is required to
identify key criteria. Resources were strong in handling, contextualisation, accessibility and
attractiveness. These relate more to the design of the resources and their obvious relevance to
coaching. The lowest scoring items were summarising, operation, and layout and structure. These
relate more to the structure of the content. Overall it was noticeable that some important features
(in relation to adult learning principles) did not score as highly as might have been expected:
interactivity, reflection, feedback and summarising. Given that these categories scored less well
than would be expected, and taking into account where the scoring could be improved, the
following elements would be recommended: introducing signposting, creating remedial feedback
loops, course maps and indexes; integrating study tasks into resources; providing summary
42
sections; contextualising learning into specific coaching roles; and providing uncluttered, visually
appealing material.
Summarised results of the analysis by category
Category Mean Rank Median Rank Range
Handling 88.0 1 73.3 1 53.3

Contextualisation 78.9 2 71.4 3 56.3
Accessibility 69.8 3 72.7 2 55.5
Attractiveness 66.9 4 63.0 6 56.7
Interactivity 64.7 5 66.7 4 50.0
Reflection 60.8 6 58.3 8 45.8
Feedback 60.5 7 64.8 5 38.9
Progression 59.9 8 53.8 9 35.7
Layout & Structure 56.8 9 52.1 10 50.0
Operation 56.2 10 61.1 7 63.9
Summarising 45.0 11 43.6 11 41.0
It became evident that there is some ‘linkage’ between categories. A theme throughout the
analysis is the extent to which the coherence of the presentation is fundamental in having an
overall strong learning framework. Signposting, remedial loops (reinforcement of key content),
reference to practice, and indexing are key to more than one category. For example, if there are
no signposts and instructions, this has an impact on progression, operation, layout and structure.
As a consequence, it was evident that materials that failed to provide an integrated structure
tended to be weak over more than one formatting category. Study tasks and summaries are key
to the management of the coach’s learning, and an absence of these features was an issue
throughout the assessed materials.
The results also suggested that it would be useful to distinguish between ‘presentational
format’ and ‘structural format’. The former may encourage attentiveness and engagement; the
latter is essential for stimulating the learning process. It is conceivable that learning materials
could exhibit positive features in one of these aspects but not achieve an overall ‘quality’ in
relation to the learning framework. There is also an issue related to the use of the resource. For
example, whether a resource is used for face-to-face or more independent learning (or as a
distant support material) may have implications for categories reflecting the structure of the
learning. Support materials may score highly for content but less well for learning structure
(because the learning is mediated by the face-to-face tutor), and stand-alone materials might be
expected to depend on a more sophisticated learning structure.
43
Survey and analysis of coaches’ and tutors’ opinions
This element of the research was designed to establish the opinions of a sample of coaches
and tutors about their preferences in relation to learning materials formats. This was based on
their recent experience of using these resources within a coach education award course. It was
clear from the responses received that the coaches and tutors were able to engage with the issues
and gave (often very lengthy) views on the resources they had experienced.
Methodology
There were a number of elements to the survey: a questionnaire, interviews and a focus
group. The methodology had been designed to elicit responses/perceptions about coach education
resources used in award courses in the sample sports. Questionnaires were distributed to a sample
of coaches and tutors who had recently (within 6 months) completed a coaching award course.
Interviews took place with four coach education tutors and two coach education managers. The
focus group consisted of an opportunity sample of coaches who had completed an award course
within the previous 6 months.
The questionnaire was designed to elicit responses aligned with those established in the
learning materials evaluation framework. It was designed as a series of statements reflecting the
quality/utility of an aspect of learning materials design and format, about which respondents could
provide an indication of the strength of their agreement (or otherwise) with the statement
[strongly agree, agree, disagree, strongly disagree, not applicable]. Item statements were designed
selectively to reflect the major analysis framework criteria: progression, accessibility, interactivity,
feedback, attractiveness, and a miscellany of ‘practical use’ items. The questionnaire went through
a number of iterations between researcher and client, and changes were made to the language
and design in order to make it as user-friendly as possible.
Overall, 60 responses were received (an approximate 20% response rate). Although initially
intended to provide quantitative evidence, the questionnaire also provided a useful source of
qualitative evidence since they elicited comment from respondents about their experience of using
the learning materials. The questionnaire invited comment on (a) what was most useful about the
learning materials, (b) what was the most difficult or annoying thing, (c) how the learning
materials could be improved, and (d) whether the delivery format of the materials was appropriate.
Results
Overall, the coaches and tutors in the sample gave a very positive response to the learning
materials. The statements/items were presented in a ‘positive’ way, that is, a ‘strongly agree’
44
score would imply support for the utility of the materials. In relation to the complete range of
30 items from each respondent, 31% elicited a ‘strongly agree’ response, 54% an ‘agree’, 14% a
‘disagree’, and only 1% ‘strongly disagree’. The coaches/tutors in the sample gave very positive
responses in terms of the learning framework design – emphasising activities, accessibility, and
relationship to practice. The activities were relevant and enjoyable. On the other hand, there were
less positive aspects of the learning structure: identifying the absence of model answers, lack of
information on self-pacing the learning, and few examples and case studies. Perhaps surprisingly,
a lack of motivation to study further was identified. Negative comments were also received about
the presentation; absence of variety, and lack of appeal were mentioned.
The lessons to be learned from this part of the investigation can be summarised as:
• Tasks and activities that are engaging and relevant are motivating
• Value is placed on examples and case studies
• Feedback is very important
• Coaches wanted the learning materials to be structured in such a way that it became a
distance learning resource and encouraged own-paced learning
• Respondents were able to distinguish what were considered to be visually appealing (and
by inference stimulating) formats from those that were less so.
Most respondents also gave very full comments, and these proved to be a very valuable
source of evidence about the learning materials. These were not analysed in a rigorous fashions,
but merely provided illustrative comment. Coaches and tutors identified a range of features as
‘most useful’. These features could be divided into structural and presentational, and, in general,
reinforced earlier design prescriptions. There was support for well-structured content, appealing
presentation, clarity, and feedback. The coaches did not like repetition or less challenging
activities. Suggested improvements related to model answers, examples, and good signposting.
Interviews were conducted with four coach education tutors and two coach education
managers. Once again the intention was to elicit generic opinions on learning resources rather
than to evaluate specific learning materials. In addition, a focus group discussion was held with
coaches (n=9) who had taken part in coach education award programmes within the past 6
months. Discussion was directed by the researcher to the learning materials provided to
complement their courses. To a remarkable degree, there was significant complementarity between
the findings resulting from the various forms of enquiry. Coaches’, tutors’ and focus group
members’ opinions were very similar. The strengths and weaknesses identified in interview could
readily be aggregated with the findings of the analyses of sport-specific project resources.
45
Discussion
The purpose of the research was to examine the formatting and design of coach education
learning materials. The substantive element in the project was the extent to which materials used
to support the coaches’ learning are designed in such a way that the basic principles of
successful adult learning are achieved. This necessitated the aggregation of two sets of principles
of practice; the first was adult learning prescriptions (relevant because of their similarity of
context and learner characteristics to coach education), and the second was the prescriptions for
good practice accumulated over time by those with considerable experience of producing open
and distance learning materials in other spheres.
It is clear that the analyses of learning materials carried out for the project and the
experiences of those surveyed and interviewed confirmed there were examples of good (in terms
of the framework criteria) and the less good. No hypothesis or expectations were established at
the outset since the research was entirely exploratory in nature. However, the evaluative
comments are inevitably couched in terms of a relationship to an ideal independent learning
model. Throughout the research a series of key words emerged. These are a useful reminder of
the themes that underpin the formatting and design issues, and we offer these as a stimulus to
further research and the development of an evaluation tool.
The key words to emerge on learning principles are learning by doing, feedback,
motivation, application, and understanding. The key words to emerge in the design of learning
materials are the introduction, clear objectives, signposting, activities/questions, feedback, a course
map, progression, remedial loops, summarising and consistency of approach. The key words to
emerge in presentational layout are white space, headings and sub-headings, lack of clutter,
diagrams and illustrations, colour, bolding and emphasis. These few sentences provide a valuable
shorthand guide.
In any interpretation of the quality of learning materials it is important to appreciate the
assumptions and intentions around which the learning materials have been designed and formatted.
In relation to coach education programmes, the accompanying materials have to be understood in
relation to the extent of open or distance learning intentions, the contextualisation of materials,
relationship to award courses, and cultural and ethnic diversity. Resources are not produced in
isolation. It is necessary to reiterate a number of related issues that arose during the research.
These included the extent to which the resource was part of a ‘package’ of materials (and may
not have been shown to advantage in isolation), the balance of tutor facilitation/independence, the
cost implications of production (also a function of generic versus course-specific learning
materials), and the ‘level’ of content in relation to candidates’ previous education.
In comparison to an ideal model of a positive learning framework some key features in the
46
learning materials in the project were less well represented than might have been expected. The
implementation quality of some aspects of good practice (for example, study tasks) may be less
challenging than expected. In addition, where materials were used as independent learning
products, the characteristics (clear outcomes, self-paced learning, progress checks, relationship to
role/practice, self-evaluation) were less good than might be expected. Poorer learning materials
tended to be ‘informational’ with little interactivity and uninspiring presentation (albeit some
specific products had good presentational aspects). On the other hand, the better learning
materials were attractive resources with well-structured sections/chunking and challenging
activities/study tasks (although with limited feedback in most cases).
The evaluation framework did not attempt to weight the criteria equally (reinforced by an
unequal number of elements). Our view is that there is a ‘core’ of learning principles to be
applied to learning materials, and that these should be given attention in evaluation and design.
This conclusion has emerged from being engaged in the project and to some extent has its
rationale in the literature review. Although the application of these principles to specific examples
of coach education materials requires further discussion, each resource might usefully be designed
and evaluated against a core set of principles. Such a list might contain generic criteria of
interactivity, reflection, feedback, summarising, attractiveness, accessibility, and progression.
To some extent the investigation into coach education materials began from a realisation
that, despite increased scrutiny of coach education, insufficient attention had been given to the
learning materials themselves. In particular, the extent to which they contributed positively to the
learning environment was being questioned. Anecdotal experience suggested that materials
designers and authors had no design agenda from which to operate. Most of the resources were
viewed as technical materials, but were not thought to be constructed in a way that promoted
learning, particularly if used as independent study materials. This investigation has demonstrated
that there are variations in the quality of existing materials, but that it is possible to identify
useful guidelines for both evaluation and design. The purpose of the paper will be achieved if
greater attention is paid to the extent to which learning materials adhere to sound presentational
and learning design guidelines.
References
Abraham, A., & Collins, D. (1998). Examining and extending research in coach development. Quest. 50, 59-79.
Brown, S., & Gibbs. (1996). Reasons for employing RBL, http://www.city.londonmet.ac.uk/deliberations/rbl/brown.html.
Brown, S., & Smith, B. (1996). Resource-Based Learning. London: Kogan Page.
Campbell, S. (1993). Coaching education around the world. Sport Science Review. 2, 62-74.
Cassidy, T., Jones, R., & Potrac, P. (2004). Understanding Sports Coaching. London: Routledge.
47
Coaching Review Panel (1991). Coaching Matters: A Review of Coaching and Coach Education in the UK. London:
Sports Council.
Cushion, C., Armour, K.M., & Jones R.L. (2003). Coach education and continuing professional development:
experience and learning to coach. Quest, 55, 215-230.
Endean, M. (2003). Learning materials at a distance. UK Centre for Materials Education, Higher Education Authority.
http://www.materials.ac.uk/guides/distance.asp.
Gilbert, W.D. (2006). Introduction to special issue: Coach education. The Sport Psychologist, 20, 123-125.
Jones, R.L., Armour, K.M., & Potrac, P. (2004) Sports Coaching Cultures: From Practice to Theory. London:
Routledge.
Jones, R.L. (2006). The sports coach as educator: Re-conceptualising sports coaching. London: Routledge
Knowles, M.S. (1990). The Adult Learner: A Neglected Species (4th Edition). Houston: Gulf Publishing.
Kolb, D.A. (1984). Experiential Learning. Englewood Cliffs, NJ: Prentice Hall.
Liukkonen, J., Laakso, L., & Telama, R. (1996) Educational perspectives of youth sport coaches: analysis of observed
coaching behaviours. International Journal of Sport Psychology. 27, 439-53.
Lyle, J. (2002). Sports Coaching Concepts: A Framework for Coaches Behaviour. London: Routledge.
Lyle, J. (2007). A review of the research evidence for the impact of coach education. International Journal of
Coaching Science. 1(1), 17-34.
McCarthy, B. (1987). The 4MAT System. Teaching to Learning Styles with Right/Left Mode techniques. Barrington, IL:
EXCEL.
McCullick, B., Belcher, D., & Schempp, P. (2005). What works in coaching and sport instructor certification
programmes? The participants’ view. Physical Education and Sport Pedagogy. 10, 121-137.
Parker, R.C. (1994). Looking Good in Print (3rd Edition). Chapel Hill, NC: Ventana Press.
Powell, R. (1991). Resources for Flexible Learning. Stafford: Network Educational Press.
Race, P. (1994). The Open Learning Handbook (2nd Edition). London: Kogan Page.
Race, P. (2004). 500 Tips on Open and Online Learning (2nd Edition). London: Routledge Falmer.
Race, P. (2005). Making Learning Happen: A Guide for Post-Compulsory Education. London: Sage.
Rowntree, D. (1990). Teaching Through Self-Instruction. London: Kogan Page.
Rowntree, D. (1992). Exploring Open and Distance Learning. London: Kogan Page.
Rowntree, D. (1994). Preparing Materials for Open, Distance and Flexible Learning. London: Kogan Page.
Rowntree, D. (1997). Making Materials-Based Learning Work. London: Kogan Page.
Rowntree, D. (1999a). A new way with words on distance education.
http://iet.open.ac.uk/pp/D.G.F.Rowntree/words_in_de.htm.
Rowntree, D. (1999b). Success factors in materials-based learning.
http://www.iet.open.ac.uk/pp/D.G.F.Rowntree/MBL.htm/introdex.html
Rowntree, D. (2000). Who are your distance learners? http://iet.open.ac.uk/pp/D.G.F.Ropwntree/distance_learners.htm.
Rust, C., & Wilson, J. (1996). Helping individual staff to develop and implement resource-based learning. In S. Brown
& B. Smith (Eds.) Resource-Based Learning (pp. 38-48). London: Kogan Page.
Saury, J., & Durand, M. (1998). Practical knowledge in expert coaches: on-site study of coaches in sailing. Research
Quarterly in Exercise and Sport. 69(3), 254-266.
Trudel, P., & Gilbert, W.D. (2006). Coaching and coach education. In D. Kirk, M. O’Sullivan & D. McDonald (Eds.)
Handbook of Physical Education (pp. 516-539). London: Sage.
48
Coaching Science
Vol. 4 No. 1
Validation of a Multidimensional
Measure of Fear of Failure in a British
Sample: The Performance Failure
Appraisal Inventory (PFAI)
Sam S. Sagar* Leeds Metropolitan University, UK
Sophia Jowett Loughborough University, UK
Abstract
The Performance Failure Appraisal Inventory (PFAI; Conroy et al., 2002) is a multidimensional
measure of threat appraisals associated with fear of failure. Whilst emerging research has supported the
validity and reliability of the PFAI with North American sport participants, its psychometric properties
remain untested with non-American samples. The present study, therefore, examined the psychometric
properties of the PFAI with a sample of British sport participants (N= 519) in order to ensure the
findings it generates can be relied on cross-culturally. Confirmatory factor analysis was employed to
test whether the proposed multi-factorial structure of the PFAI fits the British data well. The PFAI
appears to be a psychometrically sound measure that is valuable for use both in research and practice
to assess fear of failure also with British participants. Coaches can employ the PFAI in order to
understand what their athletes fear about failing in sport, while psychologists can employ it as a
diagnostic tool in assessment, diagnosis and treatment of fear of failure, and in assessing the efficacy
of their treatment programmes. As such, the findings of the present study are valuable for sport
coaches, coaching scientists, and psychologists who work with athletes and their parents and coaches.
* Correspondence concerning this article should be addressed to Sam S. Sagar, Ph.D, Department of Psychology, Faculty of
Health, Leeds Metropolitan University, Leeds LS1 3HE, UK., Phone: +44-113-812 0000, Fax: +44-113-812 3440, Email:
S.Sagar@leedsmet.ac.uk
49
Sam S. Sagar & Sophia Jowett
Keywords: Cultural Differences, Fear of Failure; PFAI, Shame; Validity, Reliability,

Confirmatory Factor Analysis
Introduction
Fear of failure (FF) is prominent in both sexes and across levels of actual and perceived
ability (Covington, 1992; Elliot & Church, 1997). It involves cognitive, emotional, and behavioral
experiences related to failure (i.e., non-attainment of one’s goal) in evaluative achievement
contexts (Heckhausen, 1991). Fear of failure has been conceptualized as a motive to avoid failure
in evaluative achievement contexts associated with anticipatory shame (Birney, Burdick, &
Teevan, 1969; McGregor & Elliot, 2005). This avoidance tendency typically prompts the pursuit
of avoidance based goals and strategies (e.g., self-handicapping, performance- and
mastery-avoidance goals, low achievement) that, in turn, exert adverse effects on individuals
(Conroy & Elliot, 2004; McGregor & Elliot, 2005; Urdan & Midgley, 2003). In the education
and sport contexts, where evaluation forms an important part of these contexts, high FF has been
associated with a prevalence of negative physical and psychological effects (e.g., depression,
eating disorders, anxiety, drug abuse; see Conroy, 2001a; Sagar, Lavallee, & Spray, 2007). For
example, sport-related research has associated FF with high levels of worry, stress, anxiety
(cognitive and somatic; Conroy, Willow, & Metzler, 2002), perfectionistic tendencies (Sagar &
Stoeber, 2009), and interpersonal antisocial behavior (Sagar, Boardley, Kavussanu, 2009). Research
with adolescent elite athletes’ reported FF to affect their interpersonal behavior, schoolwork,
sporting performance, and general well-being (Sagar, Lavallee, & Spray, 2009). Accordingly, FF
can have important implications for athletes. It is, therefore, important that sport practitioners
(e.g., coaches, coaching scientists, performance directors, psychologists) have knowledge and
understanding of the FF phenomenon in the sport context and, thus, are able to recognize
behavior/symptoms of FF in the athletes they work with.
Early and contemporary achievement motivation theorists (e.g., Atkinson, 1957; Elliot &
Thrash, 2004; McGregor & Elliot, 2005; McClelland, Atkinson, Clark, & Lowell, et al., 1953)
positioned shame at the core of FF. They conceptualized FF as a unidimensional motive
disposition that orients individuals to seek to avoid failure in achievement settings because one
feels shame upon failure. More recently, however, Conroy et al. (2002) conceptualized FF as
both a multidimensional and hierarchical construct. Their model is grounded in
Cognitive-Motivational-Relational Theory of Emotions (Lazarus, 1991) that posits that FF involves
appraising threat (in evaluative situations) to achieve personally meaningful goal(s) when one fails
in the performance and when one expects to fail. Such appraisal activates cognitive beliefs
50
(schemas) associated with the aversive consequences of failing and leads to experiencing fear.
Conroy’s et al.’s model underlines the multidimensionality of FF and suggests five beliefs about
the consequences of failure that are associated with threat appraisals. These are: experiencing
shame and embarrassment; devaluing one’s self-estimate; having an uncertain future; important
others losing interest; and, upsetting important others. Individual differences in FF can be
determined by measuring the strength of one’s belief about the likelihood of these aversive
consequences of failure to occur to them after they fail (Conroy et al., 2002) because individuals
have learnt to associate failure with aversive consequences and, therefore, experience fear and
apprehension in evaluative situations (Conroy & Elliot, 2004). Thus, FF represents a dispositional
tendency.
The Performance Failure Appraisal Inventory (PFAI; Conroy, 2001b; Conroy et al., 2002;
Conroy, Metzler, & Hoffer, 2003) measures threat appraisals associated with FF. It measures the
strength of individuals’ beliefs that failure is associated with the aversive consequences as
proposed by Conroy et al.’s (2002) model. The PFAI is the first FF measure developed from
meta-theory of emotions; it therefore examines FF as a function of person-environment interaction
(rather than a trait or state) and acknowledges the individual nature of perceptions of failure,
rather than assumes it to be the same for all performers.
The original PFAI (Conroy, 2001b) revealed some conceptual, empirical, and practical flaws
and, subsequently, Conroy and colleagues (2002) presented a revised version. This version
comprises 25 items that measure five dimensions of threat appraisals associated with FF: (1) fear
of shame and embarrassment; (2) fear of devaluing one’s self-estimate; (3) fear of having an
uncertain future; (4) fear of important others losing interest; and (5) fear of upsetting important
others. Confirmatory factor analysis revealed satisfactory goodness of fit indexes for the five
dimensions when tested as a five correlated factor model and as a higher-order factor model
when the five dimensions were subsumed. Across the analyses conducted, the five correlated
factor model yielded the following indexes: .92 for Comparative Fit Index (CFI), .91 Non-
Normed Fit Index (NNFI), .06 Standardised Root Mean Residual (SRMR), and .05 Root Mean
Square Error of Approximation (RMSEA). The hierarchical model recorded very similar scores:
.91 for CFI, .90 for NNFI, .07 for SRMR, and .06 for RMSEA. Although the five dimensions
of the PFAI do not assess all possible FF appraisals (e.g., beliefs that failure is associated with
wasting one’s effort, losing opportunity, tangible losses), it appears to assess key threat appraisals
that are associated with FF in a conceptually coherent and empirically validated manner.
Based on the 25-item PFAI, Conroy and colleagues (2002) validated a short version of the
instrument that comprises five items. These items were the most representative items from each
of the five dimensions of threat appraisals across all the validation analyses conducted for the
long version. Although the short version revealed a relatively good fit to the data (.94 for CFI,
51
.88 for NNFI, .05 for SRMR, and .11 for RMSEA), its internal consistency was slightly lower
(.72) than that of the long version (.91 for all 25 items). Conroy and colleagues suggested that
the short version of PFAI shows promise as an alternate form for researchers who are interested
in the general FF score, especially, as it requires less time to complete.
The short and long versions of the PFAI were developed in the United States of America
(USA) by American researchers for American sport- and exercise- specific populations. The PFAI
versions have been used in American research that continues to demonstrate their sound
psychometric properties including internal consistency, factorial validity as well as temporal
stability (e.g., Conroy et al., 2002, 2003; Conroy & Metzler, 2004; Conroy, Kaye, & Fifer, 2007;
Kaye, Conroy, & Fifer, 2008). To date, however, the PFAI has been employed in research in the
USA predominantly by Conroy and his colleagues to examine FF in sport and exercise settings
(e.g., Conroy & Metzler, 2004; Conroy et al., 2007; Kaye et al., 2008). To our knowledge, only
one study has been conducted using the long version of the PFAI outside the USA. In an
educational setting in Greece, Sideridis and Kafetsion (2008) examined FF in high school and
college students. They reported unacceptable internal consistency scores for two dimensions of the
PFAI: fear of upsetting important others and fear of experiencing shame and embarrassment. No
other psychometric properties were reported. As measures developed in one country are not
necessarily culturally equivalent to other countries (Sue & Sue, 1987; see also Duda & Hayashi,
1998), the present study aimed to examine the psychometric properties of the PFAI with a
British sample of sport participants. It sought to establish whether the USA-based PFAI is a
psychometrically sound measure in the United Kingdom (UK).
The UK and the USA can be considered as reflecting a socio-economically profile that is
similar. Moreover, English is spoken both in the UK and the USA and they both reflect an
individualistic orientation (Triandis, 1995). Such an orientation implies that individuals are
independent as opposed to belonging in collectives and are motivated by their own preferences
and needs, and not by the goals of the collectives. Within countries such as UK and USA,
striving for success and avoiding failure becomes a priority because success is promoted and
rewarded (e.g., praise, glory, money) while failure is disapproved and punished.
Triandis (1995) has explained that a multiplicity of factors such as historical period,
geographical location, the specific situation, individual characteristics and the degree to which a
culture reflects a specific orientation (i.e., individualism versus collectivism) define the culture.
Thus, although UK and USA may be similar in many ways, they are also distinct in others
including different educational systems, family values, and work attitudes, all of which are
aspects that can influence how one appraises the threat of failure in achievement or evaluative
situations. Therefore, this study aims to establish the degree to which threat appraisals in sport
settings as measured by the PFAI are capturing the same constructs and at the same levels in
52
UK with a British sample of sport participants. The findings of this study should provide
additional information of the psychometric properties of the PFAI to researchers and practitioners
who wish to employ this instrument and also increase our confidence that the findings it
generates can be trusted across cultures.
Method
Participants
A sample of 519 British participants (289 male, 230 female) volunteered for the present
study. Their ages ranged from 16 to 27 years (M = 20.26 years, SD = 2.38). They competed in
a variety of sports (e.g., athletics, badminton, basketball, football, gymnastics, tennis) at club level
and their sporting experience ranged from 1 to 19 years (M = 8.09 years, SD = 4.45 years).
Procedures
Participants were recruited from a number of sporting clubs in the UK (mainly in the
East-Midlands and Southern regions). The aims of the study, its voluntary nature, and the criteria
for participation (to be training and competing in their sport at the time of the study) were
explained (personally by the first author) to the participants. After signing an informed consent
form, the participants were requested to complete the questionnaire before the commencement of
their training session. The study procedures were approved by the institutional ethical committee.
Instrument
Performance Failure Appraisal Inventory (PFAI). The PFAI (Conroy et al., 2002) was
administered to measure the participants’ FF. It comprises 25 items measuring beliefs associated
with aversive consequences of failure. Its five subscales capture fear of experiencing shame and
embarrassment (7 items; e.g., “when I am failing, it is embarrassing if others are there to see
it”), fear of devaluing one’s self-estimate (4 items; e.g., “when I am failing, I blame my lack of
talent”), fear of having an uncertain future (4 items; e.g., “when I am failing, it upsets my
“plan” for the future”), fear of important others losing interest (5 items; e.g., “when I am not
succeeding, people are less interested in me”), and fear of upsetting important others (5 items;
e.g., “when I am failing, people who are important to me are disappointed”). Participants rated
how strongly they believed that each aversive consequence of failure was likely to occur to them
after failing or after not succeeding in a sporting competition. For this study, we altered the
response scale of the instrument from that designed by its developers. The response scale used in
53
the present study offered a five-point scale that included response options of 0 (“do not believe
it at all”), 2 (“believe it 50% of the time”), and 4 (“believe it 100% of the time”). This is
slightly different from the original five-point scale that offers response options of -2 (“do not
believe it at all”), 0 (“believe it 50% of the time”), and +2 (“believe it 100% of the time”). We
made this alteration to the response scale in order to prevent participants from omitting the
significance of the ‘+’ or ‘-’ sign. In an unpublished study we conducted, it was showed that
participants often (unintentionally) did not consider the ‘+’ and ‘-’ signs in their responses,
rendering these responses invalid and unusable. The original validation of the PFAI with U.S
sports participants (Conroy et al., 2002) recorded internal consistency values as follows: fear of
experiencing shame and embarrassment .80; fear of devaluing one’s self-estimate .74; fear of
having an uncertain future .80; fear of important others losing interest .81; and fear of upsetting
important others .78.
Data Analysis
Confirmatory factor analyses were conducted utilizing robust maximum likelihood estimation
with EQS 6 for windows (Bentler & Wu, 2002). A number of alternative a priori models were
tested and their fit was compared using the same data in an attempt to assess the structural
validity of the measure (see Hoyle & Smith, 1994). In this case, the aim was to identify the
model structure that best captures the dimensions of the FF. This approach of comparing
alternative models was utilized by Conroy et al. (2002) in the validation of the PFAI with a
sample of U.S sports participants and has become a common practice in the questionnaire
validation process for psychological measures in sport (e.g., Jackson & Marsh, 1996; Jowett &
Ntoumanis, 2004). Corresponding with the models tested in Conroy et al.’s validation study, we
tested four models in this present study. The first three models specified were first-order factor
models. Model 1 (M1) hypothesized a single-factor structure that reflects a general FF construct.
This model assumes that the construct is unidimensional. Model 2 (M2) hypothesized an
uncorrelated five-factor structure representing the five dimensions of FF: (1) fear of experiencing
shame and embarrassment, (2) fear of devaluing one’s self-estimate, (3) fear of having an
uncertain future, (4) fear of important others losing interest, (5) and, fear of upsetting important
others. This model, although assumes that the FF construct is multidimensional, its five
dimensions are not associated. Model 3 (M3) is similar to M2 with the only difference that the
hypothesized five-factor structure is correlated. Subsequently, M3 assumes that the five dimensions
that make up FF are associated. Finally, Model 4 (M4) hypothesised a higher order factor model
whereby the five dimensions of FF were subsumed under a general factor. This higher order
factor model assumes that the general factor, namely, FF can account for the correlations among
54
the five first-order factors. We also tested a final model (M5). M5 concerned the short version
of the PFAI (Conroy et al., 2002). Thus, this model contained five items all of which load onto
one first-order factor.
Various goodness-of-fit indexes were utilised to assess the adequacy of the factorial
structure of the four competing models as well as M5 (see Hair, Anderson, & Tatham, 1998).
2
The Satorra-Bentler scaled x statistic was used to evaluate the absolute fit between the models
tested and the data. The x2 statistic is known to be biased in large samples so it was
interpreted with caution. The robust Comparative Fit Index (CFI) and the Non-Normed Fit
Index (NNFI) were utilised to compare the various hypothesised models with the independence
model. For CFI and NNFI, we adopted values of greater than .90 (Bentler & Bonett, 1980;
Kline, 1998); although Hu and Bentler (1999) have recommended values greater than .95 as
being satisfactory. The Standardized Root Mean Square Residual (SRMR) was also utilised
because it represents the average of the standardised residuals between the specified and
obtained variance-covariance matrixes. Finally, the Root Mean Error of Approximation
(RMSEA) was also utilized to assess the degree to which the hypothesised model fit to the
population covariance matrix. For SRMR values of less than .09 and RMSEA values of less
than .08 were considered acceptable (Browne & Cudeck, 1993; Hu & Bentler, 1999).
Additionally, when the 90% confidence interval of the RMSEA contains 0.05, it indicates a
possibility of close fit (Browne & Cudeck, 1993).
Akaike’s (1987) Information Criterion (AIC) and Bozdogan’s (1987) sample-sized adjusted
statistic (CAIC) were used to examine whether a good model fit can be achieved with fewer
estimate parameters. Neither of these fit indexes has specified acceptable values but, amongst the
competing models, the one with the lowest AIC and CAIC would be the most parsimonious.
Results
Tables 1 and 2 show the means (M’s), standard deviations (SD’s), intercorrelations (r’s),
and internal consistency scores (α’s) of the 25 items of the PFAI. All mean scores were low
to moderate relatively to the response scale (0-4). The intercorrelations amongst the five
dimensions of the PFAI were also moderate ranging from .37 to .62. The internal consistency
values recorded with the samples of the present study were .81, .70, .71, .81, and .77,
respectively, all of which were either equal or above the recommended cut off point of .70
(Nunnally, 1978).
55
Table 1. Mean, standard deviations factor loadings error variances and squared multiple
correlationscores(SMC)
Std. Loading/
Item Mean SMC
Deviation Error
Fear of experiencing shame and embarrassment
10. When I am not succeeding, I am less valuable than when 1.93 1.30 .55/ .83 .31
I am succeeding
15. When I am failing, I get down on myself easily 2.27 1.22 .55/ .84 .30
18. When I am failing, it is embarrassing if others are there 2.20 1.15 .62/ .79 .38
to see it
20. When I am failing, I believe that everybody knows I am 2.03 1.13 .69/ .73 .47
failing
22. When I am failing, I feel that my doubters feel that they 1.73 1.19 .67/ .74 .45
were right about me
24. When I am failing, I worry about what others may think 2.10 1.17 .68/ .73 .47
about me .63/ .77 .40
25. When I am failing, I worry that others may think I am 2.11 1.16 .63/ .78 .39
not trying
Fear of devaluing one’s self-estimate
1. When I am failing, it is often because I am not smart 1.04 1.02 .46/ .89 .21
enough to perform successfully
4. When I am failing, I blame my lack of talent 1.56 1.19 .77/ .63 .60
7. When I am failing, I am afraid that I might not have 1.81 1.18 .84/ .54 .71
enough talent .62/ .78 .38
16. When I am failing, I hate the fact that I am not in control 2.33 1.16 .45/ .89 .20
of the outcome
Fear of having an uncertain future
2. When I am failing, my future seems uncertain 1.35 1.07 .70/ .71 .45
5. When I am failing, I believe that my future plans will change 1.53 1.13 .79/ .61 .62
8. When I am failing, it upsets my “plan” for the future 1.52 1.11 .79/ .61 .61
.55/ .82 .31
12. When I am failing, I am not worried about it affecting my 2.30 1.11 .21/ .98 .50
future plans
Fear of important others losing interest
11. When I am not succeeding, people are less interested in me 1.54 1.19 .68/ .73 .47
.51/ .85 .26
13. When I am not succeeding, people seem to want to help 1.24 1.05 .65/ .76 .43
me less
17. When I am not succeeding, people tend to leave me alone 1.31 1.05 .57/ .82 .33
21. WhenIamnotsucceeding,somepeople are not interested in me 1.33 1.06 .76/ .65 .58
anymore
23. When I am not succeeding, my value decreases for some people 1.49 1.13 .75/ .66 .56
Fear of upsetting important others
3. When I am failing, it upsets important others 1.45 1.01 .63/ .78 .39
6. When I am failing, I expect to be criticised by important others 2.18 1.15 .49/ .87 .24
9. When I am failing, I lose the trust of people who are .95 1.01 .57/ .82 .33
important to me
14. When I am failing, important others are not happy 1.56 1.07 .75/ .66 .57
19. When I am failing, important others are disappointed 1.70 1.10 .76/ .65 .58
.42/ .90 .18
Note. Values in the second row of items 24, 7, 8, 11 and 19 relate to the Short PFAI version.
56
The normalized estimate of Mardia’s coefficient was 31.01 for the competing models
(M1-M4), indicating multivariate non-normality and for that reason the robust Maximum
Likelihood estimation procedure was employed. This procedure provides more accurate standard
errors when the data are not normally distributed (Byrne, 1994). Subsequently, model evaluation
and comparison were carried out. Table 3 presents the fit indexes of the models tested. Both M1
2
and M2 had high scaled x /df ratios, low CFI and NNFI, and high RMSEA and SRMR. M3 had
better fit indexes compared to the first two models and M4. Specifically, the x2/df ratio was
below 3 (although a value of 2 or below is preferred), the CFI and the NNFI approached .90
whilst SRMR and RMSEA recorded values of .06 were meeting the proposed criteria.
Additionally, the 90% confidence interval of RMSEA included the 0.05 value. The AIC and
CAIC were useful fit indices for allowing comparison between the models tested (Dunbar, Ford,
Hunt, & Der, 2000) indicating the superiority of M3 relative to M1, M2, and M4.
Table 1 provides information of the factor loadings of M3 and unexplained unique error
variance of the items, as well as squared multiple correlation values (i.e., the proportion of
variance of the factors shared by their designated items). Overall the factor loadings were
relatively high (M factor loading = .65), ranging from .45 to .84, and were statistically
significant (p < 0.01). However, item 12 (see Table 1) recorded very low factor loading and
very high error variance rendering its quality suspect. Moreover, as Table 1 highlights, the SMC
values of certain items were particularly low (e.g., items 1, 16, and 6).
Finally M5, which represented the short version of the PFAI, recorded satisfactory CFI,
NNFI and SRMR (see Table 3). The factor loadings of the items comprising this scale were
moderate (M factor loading = .54), ranging from .42 to .63, and were statistically significant (p
< 0.01), though the error values (M error variance = .68) were high and SMC values were low
(M sq. multiple correlation = .31) (see Table 1).
Table 2. Intercorrelations (r’s) and internal consistency scores (α’s) of the PFAI’s dimensions
1 2 3 4 5 α
1. Fear of experiencing shame and embarrassment 1 .81
2. Fear of devaluing one’s self-estimate .62 1 .70
3. Fear of having an uncertain future .41 .44 1 .71
4. Fear of important others losing interest .61 .37 .40 1 .81
5. Fear of upsetting important others .55 .36 .59 .52 1 .77
57
Table 3. Fit indices for the five models tested
Scaled Robust RMSEA

Models tested df 2 NNFI SRMR AIC CAIC
x /df CFI (90% CI)
M1 275 5.58 .72 .69 .08 .09 (.09 – .10) 986.097 -458.177
M2 275 5.89 .70 .67 .24 .10 (.09 – .10) 1071.369 -372.905
M3 265 2.96 .89 .89 .06 .06 (.05 – .06) 254.490 -1137.264
M4 269 3.09 .87 .86 .07 .06 (.05 – .06) 294.224 -1118.538
M5 short version 5 7.68 .90 .90 .06 .11 (.08-.14) 28.442 2.182
Note. CFI = Comparative Fit Index, NNFI = Non-Normed Fit Index, SRMR = Standardized Root Mean
Square Residual, RMSEA = Root Mean Square Error of Approximation, 90% CI = 90% confidence
interval of RMSEA, AIC and its adjusted CAIC = Akaike Information Criterion, M1= one-factor model,
M2 = five-uncorrelated-factor model, M3 = five-correlated-factor model, M4 = one-factor second-order
model, M5 = one-factor model (5 items; short version).
Discussion
The present study examined the psychometric properties of both the short and long versions
of the PFAI in order to establish the degree to which these assessments are capturing the same
dimensions of FF at the same levels in a sample of British sport participants. Findings revealed
that the model (M3) where the five dimensions of the PFAI were correlated, generated the best
fit to the data than all the other models tested. Although these findings support Conroy et al.’s
(2002) findings, the goodness of fit indexes they reported were slightly better for both the five
correlated factors and the higher order factor. Furthermore, factor loadings, error variances and
squared multiple correlations for some of the items were especially poor (see Table 1, no.’s 1,
16, and 6). The only reverse-scored item (see Table 1, no. 12) was also empirically poor
recording the lowest factor loading and the highest error variance. Subsequently, these items may
need to be reworded or replaced. Especially, the negatively-worded item may be either replaced
with a positively-worded item (i.e., “When I am failing, I worry about it affecting my future
plans”) or develop an alternative item in its place (e.g., “When I am failing, I worry about it
hindering my future plans) (see also Conroy et al., 2002). Overall, the poor values recorded at
the item level suggest a degree of model misspecification that may be due to the PFAI being
less sensitive to the respondents’ cultural background.
The short version of the PFAI contained the five items utilized in its original validation
(see Conroy et al., 2002). Analysis revealed that, although, such fit indexes as CFI and NNFI
and SRMR were acceptable, the factor loadings, error variances and squared multiple correlations
of each one of the five items were relatively poor. These results, like those in the study
conducted by Conroy et al. should be interpreted with caution especially as the short version was
58
administered to the participants as part of the long version of the PFAI. Thus, future research
needs to examine the short version separately from its long version.
Finally, we used a modified response scale, which appears to have worked well. We,
therefore, suggest that researchers consider replacing the numerical aspect of the PFAI’s original
response scale (which ranges from -2 to +2), with a response scale that ranges from 0 (“do not
believe it at all”) to 4 (“believe it 100% of the time”). Our previous (unpublished) work has
indicated that participants unintentionally omitted the ‘+’ and ‘-’ signs leading to data that were
invalid and unusable. This problem can be prevented by simply modifying the response scale.
The findings from the present study are of important theoretical, psychometric and practical
value. From a theoretical point view, the findings support the conceptualization of FF as a
multidimensional but not as a hierarchical construct. On reflection, PFAI as an assessment of FF
has started to gain evidence of its psychometric equivalence across cultural groups (cf. Duda &
Hayashi, 1998). From an empirical point view, the findings suggest that future cross-validation
studies in the UK and elsewhere may need to consider ways to improve further the psychometric
properties of the PFAI. These changes may reflect the cultural backgrounds of the participants
studied. As a result, we encourage researchers to continue investigating the psychometric
properties of the PFAI in other cultures and in other languages. Moreover, as our results are
limited to sample of individuals aged 16-27 years, further research is needed to assess the
appropriateness of the PFAI in British children under the age of 16 years. Although it is possible
that younger athletes’ comprehension and interpretation correspond with the PFAI, more research
is required to establish its applicability across the different ages; especially with young children.
The PFAI should continue to be an influential and valuable instrument for researchers and
practitioners (e.g., sport coaches, coaching scientists, psychologists) who want to assess FF in
such diverse settings as sport, exercise and education. The PFAI has paved the way as a tool to
study and understand the FF construct and its origins. Future research should consider employing
it to investigate, for example, the link between FF and personality factors (e.g., optimism-
pessimism, self-handicapping, self-evaluative processes). Moreover, as PFAI is grounded in
meta-theory of emotions that emphasises the importance of person-environment interaction, the
scope for theoretical, empirical and practical research is immense. For example, the PFAI can
help researchers examine the role of parents, teachers, and sport coaches in the development of
FF in children (e.g., examining coach-athlete relationships, parent-athlete relationships). It can also
help them investigate different treatment frameworks (e.g., cognitive behavioral therapy,
rational-emotive therapy; see Sagar et al., 2009) of FF in settings such as family, education, and
sport and exercise. Such research will enhance knowledge and understanding of the FF construct,
which, in turn, will inform practitioners’ work in assessing and treating FF and other performance
anxiety problems in various evaluative settings. As such, research on FF can bridge scientific,
59
practitioner, and clinical perspectives.

From a practical point of view, the PFAI potentially can be employed as a diagnostic tool
in assessment, treatment and prevention of FF. It may be a useful tool in helping sport
psychologists to identify individuals’ maladaptive appraisals (or threat beliefs) associated with
failure, and to design treatment programmes accordingly. Subsequently, the PFAI can help them
to assess the efficacy of their treatment programme in modifying, reducing, or restructuring
individuals’ maladaptive appraisals associated with failure and, thereby, in reducing their FF levels
and alleviating associated problems. However, further research is needed to establish the PFAI
suitability as a diagnostic tool. Accordingly, we encourage sport psychologists to employ the
PFAI alongside other assessment tools, including interviews and observations, and to sample the
entire domain of FF appraisals, which are not sampled by the PFAI.
Finally, sport coaches and coaching scientists can employ the PFAI to ascertain what
athletes fear about failing in sport, and to measure their FF levels. It is important that they have
knowledge and understanding of the FF phenomenon in the sport domain and when working with
athletes as it can help them recognize behavior/symptoms of FF in their athletes and,
subsequently, refer such athletes to psychologists who can help to regulate or reduce these fears
over time. Moreover, knowledge of FF is also valuable for coaching practices (particularly for
coaches and performance directors) and can assist them when seeking to alter the environment
and the way in which their athletes work. Accordingly, the PFAI can contribute to the
enhancement of coaches’ work with their athletes, as well as to their athletes’ performance and
overall well-being.
To date, FF has been studied primarily in achievement domains such as business (e.g.,
Fried- Buchalter, 1997) and education (e.g., Martin & Marsh, 2003), but has been greatly
neglected in the sport context. This is unfortunate as sport is a highly evaluative situation and
competitive domain and is one of the most popular achievement contexts for youth (Treasure,
2001). In sporting competitions, athletes are evaluated by the public, parents, peers, coaches and
judges and, as such, failure often occurs publically. Indeed, concerns about performance failure
and negative social evaluation are the most prevalent sources of competitive stress and worry for
youth (Passer, 1988). The prevalence of problems attributed to high FF (Conroy, 2001a; Sagar et
al., 2007) and the importance of sport in youths’ lives, suggests that researchers need to pay
attention to the FF phenomenon in this context. Fear of failure induces distress that can create a
barrier to future participation (Conroy, 2001a, 2003). Ignoring FF and the problems associated
with it can have detrimental effects on individuals. Therefore, by reducing FF, practitioners can
help enhance individuals’ experiences in achievement activities, performance, and general
well-being.
60
Conclusions
The PFAI appears to be a psychometrically sound measure that comprises five correlated
dimensions and has adequate factor structure in a UK population. Whilst the PFAI is a valuable
instrument that can be used in practice and research to assess FF in the UK with British sports
participants, our findings also point to subtle changes that may need to be undertaken to improve
further its psychometric properties. Additionally, researchers may wish to examine the instrument’s
criterion validity (concurrent and predictive). Such research endeavors will, in turn, enhance our
confidence when interpreting the results the PFAI generates. The PFAI may be the impetus
needed to conduct more research on FF in sport and exercise settings as well as other
achievement contexts. Finally, the findings of the present study are valuable both for researchers
and practitioners (e.g., sport coaches, coaching scientists, psychologists) who work with athletes
and their parents and coaches.
References
Akaike, H. (1987). Factor analysis and the AIC. Psychometrika, 52, 317-332.
Atkinson, J. W. (1957). Motivational determinant of risk-taking behaviour. Psychological Review, 64, 359-372.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures.
Psychological Bulletin, 88, 588-606.
Bentler, P. M., & Wu, E. J. C. (2002). EQS 6 for Windows User’s Guide. Encino, CA: Multivariate Software.
Birney, R., Burdick, H., & Teevan, R. (1969). Fear of failure. New York: Van Nostrand.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen, & S. J. Long (Eds.).
Testing structural equation models (136-162). Newbury Park: CA: Sage.
Byrne, B. M. (1994). Structural equation modelling with EQS and EQS/Windows. London: Sage.
Conroy, D. E. (2001a). Fear of failure: An exemplar for social development research in sport. Quest, 53, 165-183.
Conroy, D. E. (2001b). Progress in the development of a multidimensional fear of failure measurement: the
Performance Failure Appraisal Inventory (PFAI). Anxiety, Stress & Coping, 14, 431-452.
Conroy, D. E. (2003). Representational models associated with fear of failure in adolescents and young. Journal of
Personality, 71, 757-783.
Conroy, D. E., & Elliot, A. J. (2004). Fear of failure and achievement goals in sport: Addressing the issue of the
chicken and the egg. Anxiety, Stress, and Coping, 17, 271-285.
Conroy, D. E., Kaye, M. P., & Fifer, A. M. (2007). Cognitive links between fear of failure and perfectionism.
Journal of Rational-Emotive & Cognitive-Behavior Therapy, 25, 237-253.
Conroy, D. E. & Metzler, J. N. (2004). Patterns of self-talk associated with different forms of competitive anxiety.
Journal of Sport & Exercise Psychology, 26, 69-89.
Conroy, D. E., Metzler, J. N., & Hofer, S. M. (2003). Factorial invariance and latent mean stability of performance
failure appraisals. Structural Equation Modeling, 10, 401-422.
Conroy, D. E., Willow, J. P., & Metzler, J. N. (2002). Multidimensional fear of failure measurement: The Performance
Failure Appraisal Inventory. Journal of Applied Sport Psychology, 14, 76-90.
61
Covington, M. V. (1992). Making the grade: A self-worth perspective on motivation and school reform. New York:
Cambridge University Press.
Duda, L. J., & Hayashi, C. T (1998). Measurements issues in cross-cultural research within sport and exercise
psychology. In L. J. Duda (Ed.), Advances in sport and exercise psychology measurements (pp. 471-484).
Morgantown, WV: Fitness Information.
Dunbar, M., Ford, G., Hunt, K., & Der, G. (2000). A confirmatory factor analysis of the Hospital Anxiety and
Depression Scale: Comparing empirically and theoretically derived structures. British Journal of Clinical
Psychology, 39, 79–94.
Elliot, A. J., & Church, M. A. (1997). A hierarchical model of approach and avoidance achievement motivation.
Journal of Personality and Social Psychology, 72, 218-232.
Elliot, A. J., & Thrash, T. M. (2004). The intergenerational transmission of fear of failure. Personality and Social
Psychology Bulletin, 30, 957-971.
Fried-Buchalter, S. (1997). Fear of success, fear of failure and the impostor phenomenon among male and female
marketing managers. Sex Roles, 37, 847-859.
Heckhausen, H. (1991). Motivation and action. New York: Springer.
Hoyle, R. H., & Smith, G. T. (1994). Formulating clinical research hypotheses as structural equation models: A
conceptual overview. Journal of Consulting and Clinical Psychology, 62, 429-440.
Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to under parameterized
model misspecification. Psychological Methods, 3, 424-453.
Jackson, S. A, & Marsh, H. W. (1996) Development and validation of a scale to measure optimal experience: The
Flow State Scale. Journal of Sport & Exercise Psychology, 18, 17-35.
Jowett, S., & Ntoumanis, N. (2004). The Coach - Athlete Relationship Questionnaire (CART – Q): Development and
initial validation. Scandinavian Journal of Medicine and Science in Sports, 14, 245-257.
Kaye, M. P., Conroy, D. E., & Fifer, A. M. (2008). Individual differences in incompetence avoidance. Journal of
Sport & Exercise Psychology, 30, 110-132.
Kline, R. B. (1998). Principles and Practice of Structural Equation Modeling, Guilford: New York.
Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University.
Martin, A. J., & Marsh, H. W. (2003). Fear of failure: Friend or foe? Australian Psychologist, 38, 31-38.
McClelland, D. C., Atkinson, J. W., Clark, R. A., & Lowell, E. L. (1953). The achievement motive. New York:
Appleton-Century-Croft.
McGregor, H. A., & Elliot, A. J. (2005). The shame of failure: Examining the link between fear of failure and
shame. Personality and Social Psychology Bulletin, 31, 218-231.
Nunnally, J. (1978). Psychometric theory (2nd ed.). New York: McGraw Hill.
Passer, M. W. (1988). Determinants and consequences of children’s competitive stress. In F. Smoll, R. Magill, & M.
J. Ash (Eds.), Children in sport. Champaign IL: Human Kinetics.
Sagar, S. S., Boardley, I.D., Kavussanu, M. (2009). Fear of failure and student-athletes’ interpersonal antisocial
behavior in sport and education. Manuscript under peer review.
Sagar, S. S., Lavallee, D., & Spray, C. M. (2007). Why young elite athletes fear failure: Consequences of failure.
Journal of Sports Sciences, 25, 1171-1184.
Sagar, S. S., Lavallee, D., & Spray, C. M. (2009). Coping with the effects of fear of failure: A preliminary
investigation of young elite athletes. Journal of Clinical Sports Psychology, 3, 73-98.
62
Sagar, S. S., & Stoeber, J. (2009). Perfectionism, fear of failure and affective responses to success and failure: The
central role of fear of experiencing shame and embarrassment. Journal of Sport and Exercise Psychology, 31,
602-627.
Sideridis, G., & Kafetsios, K. (2008). Parental bonding, fear of failure and stress during class presentations.
International Journal of Behavioural Development, 32, 119-130.
Sue, D., & Sue, S. (1987). Cultural factors in the clinical assessment of Asian Americans. Journal of Consulting and
Clinical Psychology, 55, 479-487.
Treasure, D. C. (2001). Enhancing young people’s motivation in youth sport: An achievement goal approach. In G. C.
Roberts (Ed.), Advances in motivation in sport and exercise (pp. 177-198). Champaign, IL: Human Kinetics.
Triandis, H. C. (1995). Individualism and collectivism. Oxford: West View Press.
Urdan, T., & Midgley, C. (2003). Changes in the perceived classroom goal structure and pattern of adaptive learning
during early adolescence. Contemporary Educational Psychology, 28, 524-551.
63
View publication stats

Ffooff

Uploaded by

Copyright:

Available Formats

Ffooff

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ffooff

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Validation of a Multidimensional Measure of Fear of Failure in a British Sample: The

Article in International Journal of Sports Science & Coaching · January 2010

Sam Sagar Sophia Jowett

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Coach Communication of Non-Selection in Youth Competitive Sport/ 3

The learning formats of coach education materials/ 35

Validation of a Multidimensional Measure of Fear of Failure in a British Sample:

Subscription and Business Office: Co-Publisher of IJCS, Dr. Jeong-Keun Park

A. Lauren Capstick* and Pierre Trudel University of Ottawa, Canada

Keywords: tryouts, elimination, coaching, interpretive phenomenological analysis

Views about Communicating Non-Selection

How Coaches Communicate Tryout Results

Method Variation 1 Variation 2

All attending tryouts

Selected for membership

Formal Speak only to athlete

Informal Speak only to athlete

Eliminated Posted in hallway

Written personal communication To all athletes attending tryouts Hard-copy letters

Figure 1. Methods and variations to communicate tryout results in youth sport

How Coaches Choose a Method to Communicate Tryout Results

& Trudel, 2001; Jones & Turner, 2006; Lyle, 2002).

Should coaches use personality

Keywords: SportsPro, personality, predictive, performance, hockey

The learning formats of coach

Constructing an analytical framework

Category Brief Descriptor Elements

Analysis of sport specific coach education learning materials

Summarised results of the analysis by category

Category Mean Rank Median Rank Range

Handling 88.0 1 73.3 1 53.3

Survey and analysis of coaches’ and tutors’ opinions

Keywords: Cultural Differences, Fear of Failure; PFAI, Shame; Validity, Reliability,

Table 3. Fit indices for the five models tested

Scaled Robust RMSEA

practitioner, and clinical perspectives.

You might also like