Taking the Next Step: Ways Forward for Coaching Science

Abstract

Coaching is no longer a subset of physical education or sport psychology but is rather an established vocation for research. In reaching such a position, we argue that a broad range of epistemologies have been used to investigate coaching such as sociology and cognitive psychology. However there is danger that, in the search for new ground, research becomes increasingly esoteric, having less and less impact on the domain that it is researching—namely coaching. As a step against this trend, we argue for and attempt to establish the commonalities across these research approaches suggesting that coaching is social, political, and pedagogical in nature. We accept that coaching is inherently complex but argue that coaches can be educated to cope with complexity through a professional judgment and nested decision making process. To facilitate this process, we offer a model for coaching that is inclusive of the commonalities across coaching research, summarizes our major theoretical points yet practical enough for application by coach educators and coaches.

Full text

Abraham and Collins Ways Forward for Coaching Science
Page 1 of 21
Taking the Next Step: Ways Forward for
Coaching Science
Andrew Abraham and Dave Collins
1Coaching is no longer a subset of physical education or sport psychology but is
rather an established vocation for research. In reaching such a position, we argue
that a broad range of epistemologies have been used to investigate coaching such
as sociology and cognitive psychology. However there is danger that, in the search
for new ground, research becomes increasingly esoteric, having less and less
impact on the domain that it is researching—namely coaching. As a step against
this trend, we argue for and attempt to establish the commonalities across these
research approaches suggesting that coaching is social, political , and pedagogical
in nature. We accept that coaching is inherently complex but argue that coaches
can be educated to cope with complexity through a professional judgment and
nested decision making process. To facilitate this process, we offer a model for
coaching that is inclusive of the commonalities across coaching research,
summarizes our major theoretical points yet practical enough for application by
coach educators and coaches.
Reflecting its evolutionary status, there have recently been a number of
attempts to provide a ‗state of the nation‘ overview for coaching science, together
with options for future development. The need for ‗tidying the field,‘ which such
state of the nation overviews represent, is arguably an inevitable feature of coaching
science as an emerging discipline. As any new applied human science evolves, it
undergoes an exponential explosion in theory: new ways of explaining, predicting,
and modifying behavior are presented and, hopefully, tested to see which offer the
most parsimonious and positive outcomes. Another important series of way-marks
must also take place, however. Unless a science is to repeatedly split and subsplit
into factions the theories need to be ‗culled,‘1 with strong, high explicative power
ideas refined to offer ever better service to the practitioners who should be the main
consumers of the research in an applied discipline such as coaching science. As a
key feature of this important action, theories can be combined to strengthen
relevance, boost the percentage variance for which they account, and offer improved
implications for practice, training, and further investigation.
Without this cull, two problems arise. Firstly, the applied science becomes
ever less applied, as practitioners (i.e., coaches and coach educators) increasingly
1Abraham is with the Carnegie Faculty, Leeds Metropolitan University, Leeds, United Kingdom. Collins
is with the Institute of Coaching and Performance, University of Central Lancashire, Preston, United
Kingdom.

Abraham and Collins Ways Forward for Coaching Science
Page 2 of 21
turn away from, or even ignore, the results as holding less and less relevance for the
real world (cf. parallel experiences in motor control—Christina & Bjork, 1991). As a
second consequence, academics become increasingly subtle and esoteric in their
work, shifting focus to ‗newer,‘ hereto unexplored territory to maintain and enhance
their publishing ‗reputation.‘ Unfortunately, both problems combine to make the
discipline less and less impactful in the domain for which, in many cases, it may
have been designed.
Reflecting these concerns, we believe that a cull in coaching science is
somewhat overdue. New coach training initiatives often show little or no evidence of
a research influence, while the subdivision of ideas using ‗new and discrete‘ topics
such as a social or political perspective (e.g., Potrac & Jones, 2009) seems to
challenge the inherent integration which should surely characterize real-life practice.
In short, what applied disciplines need to generate are theories which can strongly
influence professional practice in the real world, where coaching behavior, session
design, social environment, and playing politics are all part of the one essential
game.
Reflecting this need, our paper considers two interlinked features that may
offer such a new direction, or at the very least generate debate based on a more
integrated application. Firstly, we present a brief review of current ‗positions‘ in the
literature, seeking commonality rather than distinction against a benchmark of
practical implication. Secondly, we explore the process of planning and doing
coaching as a decision-making exercise. Our suggested integrative focus on the
Professional Judgment and Decision Making (PJDM) of coaches is compared with
other parallel professions (such as teaching), and also to the distinctions between
different types or styles of DM which are starting to emerge in the coaching
literature. As the final, third section, synthesis of these two theoretically based,
empirically supported, and clearly applied considerations leads to the suggestion of
an integrated model, termed ‗nested thinking‘ which can offer a stronger model for
testing and training professional practice.
Part 1: Real-World Coaching: Applying Politics, Social
Sciences, and Pedagogy
Perhaps the key issue that researchers in coaching have been trying to address for the
last three decades is defining coaching practice. Initially, practice was viewed very
much through a behavioral psychology lens that examined coach behaviors using
different contexts as the independent variable. The integration of findings to practice
was to identify behavioral profiles that could then be prescribed to more novice
coaches (e.g., Smith, Smoll, & Curtis, 1979). More recently, research has turned to
examining the cognitive processes (including what can/does influence these
processes) that coaches use to deliver behavioral (i.e., verbal and body language) and
physical (i.e., plans) outputs in relevant coaching contexts, be it training sessions,
games or planning time. Such a change of emphasis led Abraham, Collins, and
Martindale (2006) to conclude that there is a level of consensus that coaching (that
would be inclusive of all ‗practice‘) is a decision-making process. However, while
such consensus is building, there is far less of a consensus as to the types of
decisions that are made or what knowledge is required to make these decisions.
Furthermore, it is not always apparent whether research within the coaching
domain is working toward actually directing the coaching process. As such, a

Abraham and Collins Ways Forward for Coaching Science
Page 3 of 21
consensus often has to be inferred from research choosing to generate results of the
coaching process rather than for the process, per se. For example, approaches such as
Jowett and Cockerill (2003) have delivered vast amounts of data relating to coach-
athlete relationships and/or interactions, yet this work has become so reliant on
questionnaire data that the ideographic nature of coaching is missed. Consequently,
although there is implicit reference to coaches needing to change behavior (requiring
explicit reflections on current ways of thinking and behaving) the work is so de-
contextualized from the full scope of individual coach decision making that the
scope for drawing transferable development conclusions is limited.
This issue notwithstanding, we accept that the bulk of recent research has
progressed with a goal of improving our understanding of coaching as a complex
process. Unfortunately, even here research has not then gone on to explicitly
consider the subsequent required changes to development methods such that coaches
are able to learn how to cope and excel within this inherent complexity. Against this
background, we would argue that the extant research that does hold applications for
coaching practice or development can be grouped as coming from a socio political
stance (i.e., the strategic and political goals and problems that coaches face), a
sociological/social stance (i.e., the social setting within which coaching occurs) or a
pedagogical stance (i.e., how coaches create meaningful learning and development
opportunities for athletes and/or teams). We further suggest that all three offer ideas
that are ‗correct‘ but that all appear to lack a big picture outlook to really guide their
integration into coaching development and/or practice.
There is clearly a great deal of research within coaching that could be grouped
under these titles but it is not our intention to review all of it. Rather, we offer an
exemplary review of approaches that see coaching as drawing on these distinct
knowledge sources.
Exemplifying Knowledge Source Impacts on Coaching
Research and Application
Pedagogic Research
Based on the review conducted by Gilbert and Trudel (2004) we suggest that five
key themes emerge; coaching practice can be modeled and that knowledge is
required to perform the role of coaching, coaches use a range of strategies in
practice, coaches reflect to become better, and coach-athlete relations are linked to
efficacy and knowledge of self.
Coaching Models
Various coaching models have emerged over the last 20 years (e.g., Abraham et al.,
2006; Chelladurai, 1990; Côté, Salmela, Trudel, Baria, & Russell, 1995; Lyle, 2002),
typically with the goal of offering a structured account of a complex field by
acknowledging the broad issues that coaches need to consider in completing their
role. We suggest that these models can be classed as being either first or third person
models. The former are typically focused on viewing coaching through the coach‘s
eyes and seem therefore directly relevant for influencing coaches‘ practice (i.e.,
research primarily conducted for coaching). Conversely, the latter are generally
focused on identifying human and structural factors that can influence a coaching
environment (i.e., research of coaching). Typically psychometric in nature, this has

Abraham and Collins Ways Forward for Coaching Science
Page 4 of 21
produced lots of data but, we would suggest, mostly with limited impact for practice.
Consequently, it is the models for coaching that more directly offer structured ideas
for improving coaching practice, identifying a process to guide the decision making
of coaches. Contrast, for example, the defined knowledge categories suggested by
Abraham, Collins, Morgan, & Muir (2009) and by Côté and Gilbert (2009) with the
rather broad ideas of Chelladurai (1990). The bottom line from our perspective is
that broad and defined sources of requisite knowledge have been identified, but only
by comparatively few studies.
Coach-Athlete Relationships
Despite (in our view) being too reductionist in nature and reliant on psychometric
measures, the work of both Jowett (e.g., Lorimer & Jowett, 2009) and Myers (e.g.,
Myers, Payment, & Feltz, 2004) on coach-athlete relationships also reinforces the
need for coaches to make greater use of knowledge from the pedagogic domain.
However, there is additional recognition that making use of other domains of
knowledge, such as counseling or conflict management, may also enable coaches to
create effective working relationships. This is an interesting development as it links
with the requisite knowledge identified by both Abraham et al. (2009) and Côté &
Gilbert (2009) (i.e., knowing the athlete and self, inter and intra personal skills), and
other work on the social and political issues within coaching, a point we will return
to later.
In Situ Studies: Coaching Strategies
While models and psychometric measures exist, offering both a structure and context
to coaching practice, they have often been criticized for being too structured and
unable to explain the unpredictable nature of coaching (Cushion, 2007; Jones, 2007),
especially within coaching sessions and games. Much of this criticism is drawn from
the few in-depth investigations of practice (e.g., d‘Arripe-Longueville, Saury,
Fournier, & Durand, 2001; Saury & Durand, 1998) where qualitative, on -task data
have been collected. Such research has recognized the intentionality of coaching;
specifically, that some coaching is well planned and thought out, drawing on
pedagogic knowledge bases to inform practice, and providing implicit support for
the previous ideas of modeling. However, there is greater reference to the dynamic
and complex nature of coaching, questioning that required knowledge is only drawn
from formal disciplines such as skill acquisition or physiology. Rather, additional
requisite knowledge may be in the form of intervention ‗recipes,‘ encountered in the
form of drills or ‗preset‘ questions. Recognition of the formal and informal nature of
coaching by Saury and Durand (1998) led to a research focus that examined the
social complexities of coaching.
Social and Sociological Research
Over the last ten years there has been a significant amount of work that has
examined the social complexities of coaching (Bowes & Jones, 2006; Cushion,
Armour, & Jones, 2006; Potrac & Cassidy, 2006), much of which has been driven by
the note from Saury and Durand (1998) that coaching just isn‘t systematic. While the
social interactions examined have largely been in pedagogical settings, the argument
has been that the subtleties of the environment were being missed because of the
positivist approaches being used. The implication being that too much of this

Abraham and Collins Ways Forward for Coaching Science
Page 5 of 21
research was not recognizing the beliefs or assumptions of the various stakeholders
(coaches, athletes, parents, etc.) which drive much of the behavior observed in social
interactions in coaching.
There are obvious connections here with the reflection-based research we refer
to later. For example, the work of Strean, Senecal, Howlett, and Burgess (1997)
who, drawing on the work of Brookfield (1995), identify that reflective practice must
connect with paradigmatic assumptions (deeply held beliefs) if practice is really to
be influenced. Typical paradigmatic assumptions would include, self serving bias
when attributing success and failure, managing impressions of self in front of others,
and seeking power and/or control over self and others. Unfortunately all of these
assumptions are often tacit but have a great deal of influence over behavior which
means that, for social researchers, there will always be an element of second
guessing the intentions that underpin interactions; especially since people may not
actually be aware of what their intentions are! Consequently, from this perspective,
much of coaching often goes unsaid and unrecognized by researchers.
In addition to the issues of recognizing the role of beliefs on coaching, an
argument exists that so much is hard to predict that, even when findings from the
positivist sciences such as physiology or biomechanics do have applications to
coaching (and these are frequently already included in coach education), the chances
of real impact are slim without recognizing the complexity into which they are
expected to integrate (Jones, 2007). In summary, for such researchers formal
development in positivist disciplines is unlikely to have a significant impact of
coaching quality since it cannot allow for the inherent complexity.
While drawing attention to the ‗complexity‘ problem, Jones and Wallace
(2006) also offer a working concept, described as orchestration, for dealing with this
complexity. They described orchestration as a ―coordinated activity within set
parameters expressed by coaches to instigate, plan, organize, monitor, and respond
to evolving circumstances . . .‖ (Jones & Wallace, 2006, p. 61), going on to suggest
that
The detailed planning and coordination functions inherent in orchestration are
crucially characterized by flexibility . . . retaining short term flexibility through
incremental planning while attempting to retain some coherence through
longer term planning cycles. Plans are coordinated and frequently updated both
formally and informally based on detailed monitoring and evaluation of
practice. (p. 61–62)
We believe that these descriptions offer some guidance to influence the
development of coaches. The use of ―set parameters‖ to structure work and the use
of a hierarchy between short and long term objectives seem particularly relevant.
However, even on full reading of this definition and the associated explanations, it is
difficult to operationalize and teach actual mechanisms and practical skills that could
be deployed. Once again, the ideas seem to lack explicit direction that may be used
to drive coach development. Furthermore, we would argue that the criticism of
positivistic approaches runs the risk of throwing the baby out with the bath water.
Finally, we suggest that the descriptions and processes offered in this literature are
so complex that the inherent coaching complexity remains unaddressed.

Abraham and Collins Ways Forward for Coaching Science
Page 6 of 21
Applying Politics
A more recent development for examining and influencing coaching has come from
Potrac and Jones (2009), who have considered behavior through the micro politics of
relationships. Continuing and developing the theme developed by the second author,
these authors argue that coaching is not ―an unproblematic, progressive process but
as (sic) an arena for struggle‖ (p. 233). As such, they draw on a definition of micro
politics from Blase (1991, cited in Potrac and Jones 2009, p. 225) in that it
. . . refers to the use of formal and informal power by individuals and groups to
achieve their goals. In large part, political actions result from perceived
differences between individuals and groups, coupled with a motivation to use
power and influence and/or to protect. . . . Both cooperative and conflictive
actions and processes are part of the realm of micro-politics [while] the macro
and micro frequently interact. (p. 11)
Again, we believe this interpretation holds value for coaching, especially as it
recognizes that this form of behavior is not necessarily a bad thing—in fact it may
well be crucial to engaging effectively in coaching and performance environments
(cf. the managerial work of Butcher & Clarke, 2008). Yet again, however, there is
lack of specific structure or idea of how this work can be used to influence coaching
practice. So, while Potrac and Jones, (2009) argue that leverage points are often
searched for by the politically aware person, and that developing coaches‘ ability to
engage in this behavior proactively is probably crucial, (once again) no mechanism
for achieving this is offered.
Reflective Practice in Coaching: How Coaches (May or May
Not) Get Better
We have already made several references to the role of reflective practice in
coaching and this has been one of the major domains for coaching research in recent
years. This research has employed experiential learning theory (Brookfield, 1995;
Schön, 1987) to examine coaches‘ practice and coach development. Consequently,
this work has offered a potential dual impact within coaching. Firstly, it prescribes a
method for understanding how coaches may have developed through experience.
Secondly, it represents a meta-cognitive knowledge base that can be taught to
influence their efficacy in engaging in reflective practice to become better at
becoming better! Both have clear links with the notion of coaches knowing
themselves better as indentified by both Abraham et al. (2009) and Côté and Gilbert
(2009). It is also exactly what research from the social and political perspectives is
referring to when identifying how coaches must become more aware of the social
norms/assumptions that drive people‘s (coaches, athletes, parents, managers, etc.)
behavior. As such, reflective practice is often seen as being the answer to
understanding and developing coaching more effectively.
In an attempt to structure and optimize such return, Gilbert and Trudel (2001)
offer a model to guide reflective practice that encourages coaches to explore issues
that have arisen in their coaching in a rigorous, in-depth manner. Crucially from our
perspective, this encourages coaches to be reflective against the standards offered by
both other coaches and research to make critical and informed decisions about future
behavior (cf. the ‗judgment with standards‘ requirement proposed by Strean et al.,
1997). However, there is a lack of structure to guide best use of these external

Abraham and Collins Ways Forward for Coaching Science
Page 7 of 21
standards: specifically, how do coaches know that all relevant issues have been
explored and what research they need to seek out? Thus, while the model offered by
Gilbert and Trudel (2001) offers a structure to guide the mechanics of reflection, it
doesn‘t operationalize the actual reflective process of issue setting, unpacking, and
solving which are seemingly left to the biases of the people involved, an issue we
will return to later. At its core level therefore, this situation exemplifies the idea we
promote throughout this paper, namely that good (correct?) ideas are being offered
by research but also that there are bits of the full picture missing, such as how
exactly such ideas are best presented (by us as researchers) and critically applied (by
them as coaches).
Summary
Conflicts and Agreements
On reviewing the research presented, two broad approaches present themselves,
those which seek to provide structure to the coaching process through some level of
modeling (overt or scaffolded through broad conceptual ideas) or those which
suggest that the complexity in coaching presents too many issues for simplistic
models to adequately explain the coaching process. This issue is encapsulated by the
discursive paper by Cushion (2007) and respondent commentary by Lyle (2007),
with the former offering the complexity argument and critiquing models of coaching,
then the latter countering with a critique that the complexity argument was ‗over
egging the pudding.‘ Reflecting the same arguments we have made earlier, Lyle
(2007) suggests that structure is crucial to counter the complexity.
However, we suggest that their theoretical posturing was an (albeit required)
attempt at establishing some clear blue water between these philosophical positions
when, in fact, none (or at least very little) needs to exist, most notably from a
practical perspective. As identified earlier, we believe all of the research reviewed
presents arguments that are ‗correct‘ and, in fact, final conclusions from these
authors agree more than they disagree. If we compare the sociologically derived
ideas of Jones and Wallace (2006), the politically derived of Potrac and Jones
(2009), and the cognitive-behavioral perspective from Côté and Gilbert (2009), there
seems a clear and general agreement that coaching requires a high degree of
flexibility, knowledge and thinking to excel. In short, there is truth in all of these
interpretations and we argue that (good) coaching is and indeed must be systematic.
We just have to identify and develop systems to cope most effectively with the
‗swampy lowlands‘ of practice (Schön, 1987); the issues to be addressed if the
coaching process is to be optimized.
What Should Practical Guidelines Be Like?
Throughout this paper we have been somewhat critical about the lack of structure or
mechanisms offered: but what do we mean by structure or mechanisms? Vygotsky‘s
(1978) concept of scaffolding is probably the best analogy. In essence a scaffold
offers guidance on what elements of a problem need to be attended to, what
knowledge may be required, strategies that can be used to address the problem, and
encouragement for the ‗performer‘ to recognize and use the knowledge and skills
they do have and seek out the knowledge and skills they do not. Consequently, given
the suggested complexity in coaching, research needs to provide coaches (through
coach educators) with a structure to scaffold their approach.

Abraham and Collins Ways Forward for Coaching Science
Page 8 of 21
Typically, Vygotsky (1978) suggests that this needs to be done through
explicit guidance such as questions and instructions. Despite the criticism of models
in the coaching literature, evidence from teaching suggests that models of practice
(e.g., Entwistle & Peterson, 2004; Mosston & Ashworth, 1994) can be crucial in
developing these mechanisms (scaffolds) to guide relevant questions and
instructions; often enabling subsequently more self-directed growth. So, reflecting
this approach, any new model must ensure that the early criticisms of over-
positivism are accounted for, and that the pedagogical, social, and political factors of
coaching are both encompassed and integrated—one of the goals of this paper but
through the unifying focus of decision making.
Part 2: Professional Judgment and Decision Making (PJDM)
as a Unifying Focus
As is now hopefully obvious, we see more commonality than difference in what are
often considered somewhat orthogonal approaches to conceptualizing the coaching
process. Indeed, we also see the ways in which coaches decide what to do as a
common issue; decisions all researchers suggest are best made against a set of
external criteria critically internalized by the coach (be they generated by experience,
research, reflection or other coaches). This leads us to an expressed focus on
understanding and enhancing coach PJDM as the most integrative and parsimonious
pathway to improved coaching. This stance is already established (e.g., Abraham &
Collins, 2011) and parallels similar emphases in other applied fields such as teaching
(Entwistle & Peterson, 2004), sport psychology (Martindale & Collins, 2007), and
forensics (homeoffice.gov.uk, 2011).
In pursuing this approach, we first consider some of the underlying theoretical
and practical questions, several of which are already under consideration in the
coaching field. For the purposes of clarity and simplicity, we will delimit our
exemplars and applications to performance coaching with adults although, as we
occasionally demonstrate throughout the rest of this paper, the tenets of our
integrative approach work equally well in other environments, albeit with different
considerations and foci.
Classical and Naturalistic Decision Making: Thinking and
Recognizing
So if PJDM is crucial to coaching, examining research on decision making should
provide useful insight. Typically, two schools of thought are apparent; Classical
Decision Making (CDM) and Naturalistic Decision Making or NDM. Approaches
cited in the previous section refer strongly (with evidence) to the need for
practitioners to be able to make considered decisions. Effectively completed, indeed
as a characteristic of expertise in that field, these decisions will efficiently and
appropriately compare and contrast potential options for understanding and solving a
problem before a choice is made on which particular action to take—in short CDM
seems to match up well with coaching (and other applied domains) practice. This
approach would typically be applied during planning, implementation, and review
stages of practice to progress their athletes (or clients) toward set goals.
Recently, however, this approach has been criticized within the coaching
literature for being unable to explain how coaches operate in settings where there

Abraham and Collins Ways Forward for Coaching Science
Page 9 of 21
isn‘t sufficient time to make thought through and considered decisions (Cushion,
2007; Lyle, 2010). Indeed, the recently emerging literature examining the decision
making of professionals suggest that CDM cannot adequately explain decisions that
were made in a mix of: ―ill structured problems, uncertain dynamic environments,
shifting, ill defined or competing goals, action/feedback loops, time stress, high
stakes, multiple players, organizational goals, and norms‖ (Montgomery, Lipshitz, &
Brehmer, 2005, p. 2). This has led Lyle (2010) to focus strongly on the NDM work
of Klein and colleagues (e.g., Klein & Hoffman, 1993; Montgomery et al., 2005),
and its application to coaching.
The NDM model, originally developed based on observations of high pressure,
real life settings such as fire-fighter decision-making, is descriptive, providing a
frame within which characteristics of experts can be distinguished from those of
novices. One mechanism suggested to be involved in NDM is the ability to connect
recognized (i.e., seen before) cues from the environment to a method of action (this
is very similar to the notion of specific proceduralized knowledge by Anderson,
1982). Given the vast array of cues available in the coaching environment, learning
to recognize pertinent cues and to then develop relevant actions clearly takes a long
time, which is why NDM appears to differentiate experts from novices.
Certainly therefore, this approach seems to be relevant and is clearly
parsimonious for the decision making of coaches in time pressured, ill structured
domains such as those referred to by Saury and Durand (1998) in coaching sailing.
In fact, Lyle (2010) considers this approach to fit so well that he significantly (we
suggest overly) downplays (to the point of irrelevance) the role that CDM
approaches can fulfill (i.e., ―decision taking is not about choices . . . but about
coming to the most appropriate decision on an on-going basis‖ (Lyle, 2010, p. 29).
This would be supported by Klein‘s (1998) work which supports the notion that
experts do not routinely appear to directly compare multiple options in naturalistic
settings. However, we contend that a sole focus on NDM completely obviates the
complexity inherent in the coaching environment while also offering no tenable
explanation as to how coach behavior can evolve or be developed.
Rewarding Outcome not Process May Promote Biases,
Overconfidence, and Heuristics
So what is correct, CDM or NDM? It could be argued that this depends on the
desired end product. Despite a substantial knowledge base, experts often make snap
judgments. The good news is that these are often correct; the bad news is that they
aren‘t always. Furthermore, even though experts do successfully use NDM
approaches, this doesn‘t describe the methods used to develop up to the point of
expertise, which almost certainly involved a high degree of thought, problem
solving, and learning (i.e., CDM). Finally, such slower reflection and weighing up of
alternatives is also essential to drive the constant refinement and innovation of
practice which is so necessary in the rapidly evolving challenge of performance
sport. In keeping with the integrative theme of this paper therefore, it seems that both
CDM and NDM are correct, but in a proper balance and place. The relative roles of
NDM or CDM are perhaps best explored by examining the limitations of both, rather
than only focusing on their strengths. We have already noted how thoughtful and
considered problem solving (CDM) struggles to explain expert decision making in
time pressured environments but what of NDM?

Abraham and Collins Ways Forward for Coaching Science
Page 10 of 21
One of the biggest reasons why NDM approaches are so attractive is that
society (and sport) rewards experts who are quick to respond to ‗difficult‘ questions.
Experts are very quickly hung by their indecisiveness—just watch coaches/managers
and or politicians interviewed on TV. However, it is not just the press wh o reinforce
quick answers and punish slow (apparently) rambling responses; patients do it with
doctors, students do it with lecturers, (poor) lecturers do it with students,
interviewers do it with interviewees, and performers certainly often do it with their
coaches.
In this regard, Kahneman and Klein (2009) answer a question that coaching
researchers have been asking for some time, what makes an expert (coach)? These
authors suggest that the answer can best be found through peer review (i.e., the
expert is the person recognized as being so by his or her peers). As such, if the
emerging NDM expert receives sufficient peer support, then self-confidence is likely
to grow quickly as is their belief in their own expertise. This confidence will
continue to grow, or certainly will not be diminished, if decisions taken are viewed
as good ones. It is here that Kahneman and Klein (2009) suggest problems can
quickly arise, in that overconfidence can lead to mistakes being ignored or blamed
elsewhere, with biases in decision making becoming established and unchallenged
simply because they are the ‗expert.‘ This neglects the relativistic nature of expertise
and power. If people fail to recognize why they became expert, expertise may be
quickly lost and biases quickly established.
This issue is perhaps best exemplified in the domain of talent identification
and development. Recent research (e.g., Abbott, Button, Pepping, & Collins, 2005)
has displayed that past and present approaches are readily based around the decisions
of coaches toward applying high performance philosophies of selection and
deselection to attain more immediate and short term goals of having the best age
group team. This approach is heavily influenced by the coach‘s desire to generate an
impression of being a winning coach, which was socially reinforced by significant
others in the first place. As such, the coach continues to be drawn to select age group
players who fit a profile and coach them in a way to achieve results rather than work
to long-term development agendas; in short, apply a flawed decision system or
‗heuristic.‘
So are these naturalistic decisions ‗expert‘ or ‗biased‘? The answer to this lies
in the root cause of why experts stand still and go backward or consciously work
hard to become and remain expert. This important distinction is perhaps best
summarized by Tetlock, (2005):
One is the work of cognitive conservatism: the reluctance of human beings to
admit mistakes and update beliefs. The other is the ―self-serving‖ attribution
bias: the enthusiasm of human beings for attributing success to ‗internal‘
causes, such as the shrewdness of one‘s opinions, and failure to external ones
such as bad luck. (p. 128)
Note also the piquant observations of Galbraith (2002) ―when faced with the choice
between changing one‘s mind and proving that there is no need to do so, almost
everyone gets busy on the proof.‖ Both quotes point to ‗thinking through decisions‘
as being low on people‘s priorities if ready-made solutions exist. Of course, if no
solution exists, people should have to think things through especially if the answer is
important, given that they can resist socially generated pressures to provide quick
decisive answer so as to maintain their apparently ‗expert‘ status.

Abraham and Collins Ways Forward for Coaching Science
Page 11 of 21
If motivated to think carefully to design a new answer, people will reflect on
their experiences and should/may seek external input (we acknowledge knowing
where to look for this is probably crucial) to develop understanding and better
actions. It is when individuals stop (or never start) doing this that biases develop,
especially if important peers, perhaps unwittingly, recognize actions as being
‗expert.‘ There are obvious knock on effects here for how ‗communities of practice‘
operate. As one important example, social reinforcement of increasingly ‗routine-
ized‘ decisions will almost inevitably stunt the growth of the coach, even if s/he had
already achieved justifiably expert status. Without constant critical reflection and
appropriate innovation, such ‗decisive‘ experts will fall behind.
Of course, it could be argued that the problems of such ‗feels right‘ coaching
are as a result of these coaches learning their trade through informal methods and
that formal learning around talent development coaching would prevent these
problems. There are two immediate issues which challenge this assertion: firstly why
haven‘t coach development organizations already done something about this—are
there biases in the way coach development has in itself developed? Secondly, the
problems of bias have been observed in professions where formal learning does
occur such as clinical psychology (Kahneman & Klein, 2009). Formal learning may
prevent biases but only if it is developed to do so. Time locked licensing (i.e., where
coaches are licensed to practice for fixed periods of time before having to ‗relicense‘
against evidence of continuing professional development) is one method used
considered and used in other professions (e.g., medicine, GMC, 2010) to ensure that
expertise is reinvested, an approach also being implemented within some coach
education practice such as the Lawn Tennis Association (LTA, 2011) and is even
written into law in Bulgaria! However, we would question if this idea is being
considered with full recognition of what the role of licensing should be (i.e., is it an
opportunity to ensure continuing professional development or can it also be an
opportunity to reinforce the need for a more rigorous peer review system between
coaches and relevant stakeholders as well?).
A further related problem with an over reliance on NDM can occur from
expecting experts to always be expert in their responses to immediate events. While
NDM relies on the expert being able to recognize environmental cues and connect
them to relevant actions so a naturalistic decision can be made, we must also
acknowledge that coaches do come across situations that they do not recognize, with
subsequent actions being far from expert (Bowes & Jones, 2006). Consequently, the
coach will inevitably default to use of a heuristic that is biased by the moment they
are in: for example, the first time a coach comes across a performer who breaks
down in tears during training and needs to respond. This response will be
heuristically based and probably driven by the emotion the coach was feeling at the
time. Whatever the response, it may often not be driven by expert recognition.
As Kahneman and Klein (2009) would argue, if the coach recognizes this
response as being weak (Anderson, 1987) then the moment can be debriefed and
development can occur [i.e., the coach should then engage in critical incident
reflection (Gilbert & Trudel, 2001) but notably against the kind of external criteria
described earlier]. However, if there is an over-confidence and/or no
recognition/consideration that a suboptimum solution may have been used, then no
learning will occur and mistakes will inevitably occur again in the future.
Returning to the initial question therefore, of what are the limitations of NDM,
the answer lies in when coaches forget what made them expert in the first place.

Abraham and Collins Ways Forward for Coaching Science
Page 12 of 21
Subsequently, most situations are encountered with an (over) confidence that a quick
and correct response exists; consequently NDM can quickly turn to biased and/or
heuristic problem solving.
There is no doubt that NDM for a coach is crucial at key moments in training
and games, it is also extremely useful to the armed forces and firefighters. However,
the ability to make good calls at high pressure moments in all three of these domains
arrives from significant periods of training, well structured planning and reflection
against established principles, and a broad range of simulated and actual practice (cf.
Richards, Mascarenhas, Collins, 2009). Often missed and unrewarded by ‗outsiders‘
such as the media or badly informed researchers, are the hours of thinking,
rethinking, and ‗internally experimenting‘ (cf. Schön, 1987), executed through the
medium of CDM, that goes into the planning of this training on behalf of the trainee.
So, experts are expert as a result of hours (10,000?) of structured work and careful
thought (i.e., not just talent but talent exposed to meaningful learning opportunities)
(Hambrick, 2003).
Avoiding These Problems: Good Use of Mental Models
Given the complexity of CDM and NDM within performance domains such as
coaching, it is worthwhile noting that mental models have been suggested as a
means to make decision making generally, and NDM specifically, more efficient
within other domains such as fire-fighting and the military (Zimmerman & Harris-
Thompson, 2008). As such, developing models is seen as a useful training tool for
people who operate in NDM situations (cf. our earlier comments about providing a
scaffolding structure). However, a review of the mental model literature (Collins,
Brown, & Holum, 1991; Klein, 1998; Zimmerman & Harris-Thompson, 2008)
suggests that optimum employment of mental models requires attention to three key
factors. Firstly, the implementation of these models requires an initial high level of
cognitive engagement and thought, meaning that CDM is an essential component.
For example, Kahneman and Klein (2009) suggest ‗premortems‘ where plans are
assumed to have gone wrong and alternative solutions are sought even before the
plan has been put in place. A similar strategy of if-then planning has been employed
in teacher education (Tjeerdsma, 1995). Secondly, the initial development of models
requires a high level of critical analysis of the role of the NDM person, which
generally means a high level of CDM underpins any model (Militello & Hutton,
1998). Thirdly, elements of mental models may be more relevant in different
situations such that there are times where a whole model may be useful for CDM
opportunities and others when less is relevant to NDM opportunities due to
attentional demands on working memory.
So how effective are the currently available coaching models in offering
mental models to support coaching practice and development? Unfortunately, we do
not know the full answer to that question since, to our knowledge, none of these
models have been tested in intervention studies. Indeed, only one model, the
coaching schematic of Abraham et al. (2006), is reported in tandem with explicit
support from coaches as reflecting their role. The coaching model of Côté et al.
(1995) may also possess a similar level of validity in that it was built from the
responses of coaches but these elements were never presented back to participants
for validity checking. However, irrespective of these issues, there is probably now
sufficient data to suggest that neither of these models are doing enough to reflect the
multi dimensional nature of coaching as presented here.

Abraham and Collins Ways Forward for Coaching Science
Page 13 of 21
Since current models are not good enough (in the absence of evidence to the
contrary), an alternative solution is needed. Typically, mental models have been
developed by understanding ‗expert knowledge‘ through the use of Cognitive Task
Analysis (Militello & Hutton, 1998). The assumption being that if experts are
selected and their naturalistic (tacit) practice unpacked, an understanding of the
demands of the role and the perceptual and decision-making skills required to
become expert can be mapped. Given the breadth of research reviewed here we
would suggest that a broad task analysis has already been concluded in coaching and
that this research has sufficiently exposed the tacit and explicit elements of coaching
such that a new coaching model can be developed. Consequently we offer a new
model in Figure 1 as a basis to structure and guide research into coaching practice
and also to be used by coaches and coach educators as an initial mental model to
scaffold their practice. In doing so, we acknowledge that this new model is ripe for
adaptation or rejection through relevant testing, just as should have been applied
with others to date.
\ insert figure 1 \
Part 3: Answering These Challenges Through Nested Goals,
Plans, and DM Balance
If we consider the ‗evident need‘ for an NDM/CDM balance based on the ideas
presented here, four principles become evident:
1. Coaching environments encompass pedagogical, social, and sociopolitical
contexts that require decisions to be made, where possible against ‗external
criteria,‘ on how to interact with and influence (and be influenced by) various
stakeholders
2. NDM (or at least good NDM) grows out of off-line premortems, cognitive
experimentation (cf. Schön, 1987), critical planning, debriefs, and reflections
which consider critically what did and didn‘t work and which feeds into the
next NDM situation. In short, NDM is ‗grown‘ by offline debate, practice, and
thinking, with this thought being both situation-specific and generic
3. The environment must employ CDM and NDM at different times and under
different circumstances (e.g., in strategic long term versus annual versus session
planning). However, this ‗blend‘ is almost never 100% one or the other
4. This ‗blend‘ principle applies to all components of the coaching environment,
although other constraints will determine the balance for any particular
challenge.
The basis of our model in Figure 1 is that coaches engage in nested thinking,
(an idea originally offered but unexplored by Abraham et al., 2006) where decisions
taken at a micro level are embedded (nested) within medium term agendas which
themselves are linked to (nested within) longer term aims. As such, naturalistic
decisions are explicitly linked to decisions taken at a more classical level when time
is available to think through ideas. This acknowledges the following principles:

Abraham and Collins Ways Forward for Coaching Science
Page 14 of 21
• That the balance of classical to naturalistic decision making changes as a
result of the environment and the level of thinking time likely to (or at least
should) exist. It also acknowledges the political, social, and pedagogical
demands
• That the knowledge required to think and operate at each level changes in
subtle ways from more formal sources of declarative knowledge to more
interventionist procedural knowledge to reflect the demands of the situation
• That this approach reflects feed forward critical thinking process and a
critically reflective feedback process so that the whole process is dynamic and
flexible.
The Socio-Political-Strategic Level (Macro)
Reflecting these principles, coaches should critically consider the required strategic
sociopolitical goals of their work: for example, defining key educational and health
goals for the children‘s coach, defining retention, transition/progression skill
expectancies for the talent development coach, or managing upwards on
performance expectations for the performance coach. Once these are identified,
further parameters include who will need to buy into them, how they are
communicated and who they will need to be communicated to. In taking this stance,
coaches can be proactive in developing a sociopolitical environment that meets their
needs rather than just ‗tolerating constraints‘ that they have had no input to and/or
control of. Indeed, operating at this level of policy making is seen as being a vital
component of being a ‗professional‘ by Downie (1990). Given such planning is
fundamental to achieving long term goals, we would suggest that this should be a
predominantly CDM process to which a good deal of time should be allocated.
However, this does not mean that there won‘t be some element of NDM occurring;
this is inevitable in any form of planning process. Furthermore, as a r esult of critical
reflection, quick changes to long-term plans may be needed.
The Socio-Tactical-Motivational Level (Meso)
Once the Macro level of goal setting and planning is agreed, the coach can then
begin to work toward goal setting and planning for the sociomotivational and tactical
environment that will be required for macro goals to be achieved. While this level of
planning would likely focus on developing the environment required to support
athlete development and/or achievement, it can equally focus on creating the right
environment for parent and assistant coach buy in. If self-determination is so
important for intrinsic motivation (Deci & Ryan, 2000) then the self-determination
of all those with an active involvement in enabling the development of athlete needs
to be planned for (Pensgaard & Roberts, 2002; Weigand, Carr, Petherick, & Taylor,
2001). It is here that we believe annual planning for athletes fits since the major goal
of this approach is to develop self determination and ownership for th e athlete; it
also allows for coaches to consider optimal methods for engagement. The tactical
element of planning here is not necessarily just about approaches to game play but is
inclusive of hitting important performance and development markers that ‗evidence‘
progress to anyone with a vested interest. It is interesting to note that this ‗correct
environment‘ approach is a typical marker of excellence in teaching where specific
plans address the initial rigors of teaching new classes so that more meaningful

Abraham and Collins Ways Forward for Coaching Science
Page 15 of 21
relationships with students are established within agreed behavioral guides (Fink &
Siedentop, 1989). Once again, we would suggest that this process needs to be a
predominantly CDM process where ideas are challenged and thought through by
active agents (coaches and senior players for example). Since this level of planning
is closer to the realities of practice it will probably be more influenced by day to day
reflections, necessitating a NDM process that tweaks goals as progress is reviewed.
The Idio-Tactical Level (Micro)
Finally, the micro level of planning and delivery directly reflects an approach
focused on implementing meso targets (i.e., meta-cognitive, performance) with
individual athletes and/or groups of athletes focused on sport specific targets. At this
level, coaches are obviously expected to respond to situations as they arise, so NDM
becomes much more prevalent. However, as a result of taking a nested approach and
premortem-ing possible challenges, the coach is better prepared to both make
naturalistic decisions and more able to recognize when a decision is heuristic based
or biased and needs externally referenced critique. It is here that coaches really start
to learn how to cope with the swampy lowlands of practice because they are able to
learn the difference between expert NDM and heuristic guess work.
Conclusion: What Next?
In delivering this paper we have presented an overview of coaching research
drawing out the similarities and agreements identifying that coaching is a decision-
making endeavor (Abraham et al., 2006; Jones & Wallace, 2006; Lyle, 2010), is
nested in nature working to various levels of goals and involves working with
various stakeholders toward a variety of goals; social, political, performance,
individual etc (Abraham et al., 2006; Jones, 2007; Jowett & Cramer, 2009[AUQ1];
Potrac & Jones, 2009; Weigand et al., 2001), and which demands that coaches work
in both naturalistic (Lyle, 2010) and classical ways (Abraham et al., 2006; Gilbert &
Trudel, 2001). We have presented PJDM, through CDM and NDM, as a unifying
theory that, we argue, is sufficiently parsimonious to act as an umbrella theory for
coaching practice since it seems to allow for the application of all epistemological
positions. Finally, we have offered a model that summarizes the application of
differing perspectives within a nested approach accounting for the differing use of
CDM and NDM. We believe this model offers a scaffold to guide coach education
practice, coaching practice, and future research. We should highlight however, that
the application and understanding of this model will ultimately be limited by
practitioners‘ engagement with underpinning theory. In keeping with the theme of
this paper the model is not a black and white answer; rather, it is a structured entry to
navigate the coaching process—a mental model to guide rather than restrict.
Notably, however, despite the evidenced review of coaching and development
of a view and model on and of coaching, there is still a need to explicitly test the
explanatory power of this model and its underpinning theories to expert coaches‘
practice: a feature which we feel has been somewhat lacking with much of the
theoretical posturing and unempirical model making to date. Furthermore, if this
model is fully valid in reflecting coaching expertise, it should also represent an
accurate basis from which to implement a coach development intervention (Collins
et al., 1991). Given that this approach is intrinsically linked to long term
commitments to continued nested thinking and reflecting using NDM and CDM

Abraham and Collins Ways Forward for Coaching Science
Page 16 of 21
methodology, longitudinal studies tracking coaching practice must now be one the
goals for coaching researchers to test the introduction, application, and adherence of
these ideas by coaches and coach educators. This will form the basis of our future
work.
Notes
1 We use this term advisedly. We do not wish to remove the right to personal epistemologies
but there is a need to balance personal philosophies with providing research that makes sense
to practitioners and therefore impacts on practice in avowed applied disciplines.
References
Abbott, A., Button, C., Pepping, G-J., & Collins, D. (2005). Unnatural selection:
Talent identification and development in sport. Nonlinear Dynamics Psychology
and Life Sciences, 9(1), 61–88.
Abraham, A., & Collins, D. (2011). Effective Skill Development: How Should
Athletes‘ Skills Be Developed? In D. Collins, H. Richards, & A. Button (Eds.),
Performance psychology: A practitioner’s guide. Kidlington: Elsevier.
Abraham, A., Collins, D., & M artindale, R. (2006). The coaching schematic:
Validation through expert coach consensus. Journal of Sports Sciences, 24(6),
549–564.
Abraham, A., Collins, D., Morgan, G., & Muir, B. (2009). Developing expert coaches
requires expert coach development: Replacing serendipity with orchestration. In
A. Lorenzo, S.J. Ibanez, & E. Ortega (Eds.), Aportaciones Teoricas Y Practicas
Para El Baloncesto Del Futuro. Sevilla: Wanceulen Editorial Deportiva.
Anderson, J.R. (1982). Acquisition of a cognitive skill. Psychological Review, 89(4),
369–406.
Anderson, J.R. (1987). Skill acquisition: Compilation of weak-method problem
solutions. Psychological Review, 94(2), 192–210.
Bowes, I., & Jones, R. (2006). Working at the edge of chaos. Understanding coaching
as a complex interpersonal system. The Sport Psychologist, 20, 235–245.
Brookfield, S.D. (1995). Becoming a critically reflective teacher . San Francisco:
Jossey-Bass.
Butcher, D., & Clarke, M. (2008). Politics and worthy management agendas.
Management Online Review, August, 1–10.
Chelladurai, P. (1990). Leadership in sports: A review. International Journal of Sport
Psychology, 21, 328–354.
Christina, R.W., & Bjork, R.A. (1991). Optimizing long term retention and transfer. In
D. Druckman & R.A. Bjork (Eds.), In the mind’s eye: Enhancing human
performance (pp. 23–56). Washington, D. C.: National Academy Press.
Collins, A., Brown, J.S., & Holum, A. (1991). Cognitive apprenticeship: Making
thinking visible. American Educator, Winter, 6–11, 38–46.
Côté, J., & Gilbert, W. (2009). An integrative definition of coaching effectiveness and
expertise. International Journal of Sports Science & Coaching, 4(3), 307–323.
Côté, J., Salmela, J., Trudel, P., Baria, A., & Russell, S. (1995). The coaching model:
A grounded assessment of expert gymnastic coaches‘ knowledge. Journal of
Sport & Exercise Psychology, 17, 1–17.
Cushion, C.J. (2007). Modeling the complexity of the coaching process. International
Journal of Sports Science & Coaching, 2(4), 395–401.

Abraham and Collins Ways Forward for Coaching Science
Page 17 of 21
Cushion, C.J., Armour, K., & Jones, R. (2006). Locating the coaching process in
practice: models ‗for‘ and ‗of‘ coaching. Physical Education and Sport
Pedagogy, 11(1), 83–99.
d‘Arripe-Longueville, F., Saury, J., Fournier, J., & Durand, M. (2001). Coach -athlete
interaction during elite archery competitions: An application of methodological
frameworks used in ergonomics research to sport psychology. Journal of Applied
Sport Psychology, 13, 275–299.
Deci, E.L., & Ryan, R.M. (2000). The ―what‖ and ―why‖ of goal pursuits: Human
needs and the self dete rmination of behaviour. Psychological Inquiry, 11(4),
227–268.
Downie, R.S. (1990). Professions and professionalism. Journal of Philosophy of
Education, 24(2), 147–159.
Entwistle, N.J., & Peterson, E.R. (2004). Conceptions of learning and knowledge in
higher education: Relationships with study behaviour and influences of learning
environments. International Journal of Educational Research, 41, 407–428.
Fink, J., & Siedentop, D. (1989). The development of routines, rules, and expectations
at the start of the school year. Journal of Teaching in Physical Education, 8(3),
198–212.
Galbraith, J. (p.7) cited in Gilovich, T., Griffin, D. 2002. Introduction–heuristics and
biases: Then and now. In: T. Gilovich, D. Griffin, D. Kahneman (eds), Heuristics
and biases: The psychology of intuitive judgment . New York: Cambridge
University Press. pp. 1–18
Gilbert, W.D., & Trudel, P. (2001). Learning to coach through experience: Reflection
in model youth sport coaches. Journal of Teaching in Physical Education, 21,
16–34.
Gilbert, W., & Trudel, P. (2004). Analysis of coaching science research published from
1970-2001. Research Quarterly for Exercise and Sport, 75, 388–399.
Gilbert, W.D., & Trudel, P. (2005). Learning to coach through experience: Conditions
that influence reflection. Physical Educator, 62, 32–43.[AUQ2]
General Medical Council. (2010). Revalidation. Retrieved 29 April, 2011, from
http://www.gmc-uk.org/doctors/revalidation.asp
Hambrick, D.Z. (2003). Why are some people more knowledgeable than others? A
longitudinal study of knowledge acquisition. Memory & Cognition, 31(6), 902–
917.
Homeoffice.gov.uk. (2011) Retrieved 4 March, 2011, from
http://www.homeoffice.gov.uk/publications/police/forensic -science-
regulator1/codes-conduct-practice?view=Binary.
Jones, R. (2007). Coaching redefined: An everyday pedagogical endeavour. Sport
Education and Society, 12(2), 159–173.
Jones, R., & Wallace, M. (2006). The coach as ‗orchestrator‘: more realistically
managing the complex coaching context. In R. Jones (Ed.), The sports coach as
educator: Reconceptualising sports coaching (pp. 51–64). Abingdon: Routledge.
Jowett, S., & Cockerill, I.M. (2003). Olympic medalists‘ perspective of the athlete-
coach relationship. Psychology of Sport and Exercise, 4, 313–331.
Kahneman, D., & Klein, G.A. (2009). Conditions for intuitive expertise: A failure to
disagree. The American Psychologist, 64(6), 515–526.
Klein, G.A. (1998). Sources of power: How people make decisions. Cambridge: MIT
Press.

Abraham and Collins Ways Forward for Coaching Science
Page 18 of 21
Klein, G.A., & Hoffman, R.R. (1993). Seeing the invisible: Perceptual-cognitive
aspects of expertise. In M. Rabinowitz (Ed.), Cognitive science foundations of
instruction (pp. 203–215). Hillsdale, NJ: Lawrence Erlbaum Associates.
LTA. (2011). Coach licensing. Retrieved 2nd May, 2011, from
http://www.lta.org.uk/Coaches-coaching-assistants/Coach-accreditation/Coach-
licensing/
Lorimer, R., & Jowett, S. (2009). Empathic accuracy, meta-perspective, and
satisfaction in the coach-athlete relationship. Journal of Applied Sport
Psychology, 21(2), 201–212.
Lyle, J. (2002). Sports coaching concepts: A framework for coaches’ behaviour.
London: Routledge.
Lyle, J. (2007). Modelling the complexity of the coaching process: A commentary.
International Journal of Sports Science & Coaching, 2(4), 407–409.
Lyle, J. (2010). Coaches‘ decision making: A Naturalisitc Decision Making analysis.
In J. Lyle & C.J. Cushion (Eds.), Sport coaching: Professionalisation and
practice (pp. 27–41). London: Churchill Livingstone.
Martindale, A., & Collins, D. (2007). Enhancing the evaluation of effe ctiveness with
professional judgement and decision making. The Sport Psychologist, 21, 458–
474.
Militello, L.G., & Hutton, R.J.B. (1998). Applied cognitive task analysis (ACTA): A
practitioner‘s toolkit for understanding cognitive task demands. Ergonomics,
41(11), 1618–1641.
Montgomery, H., Lipshitz, R., & Brehmer, B. (2005). Introduction: From the first to
the fifth volume of naturalistic decision-making research. In H. Montgomery, R.
Lipshitz, & B. Brehmer (Eds.), How professionals make decisions. New Jersey:
Lawrence Erlbaum.
Mosston, M., & Ashworth, S. (1994). Teaching physical education. New York:
Macmillan.
Myers, N.D., Payment, C.A., & Feltz, D.L. (2004). Reciprocal relationships between
collective efficacy and team performance in women‘s ice hockey. Group
Dynamics, 8(3), 182–195.
Pensgaard, A.M., & Roberts, G.C. (2002). Elite athletes‘ experiences of the
motivational climate: The coach matters. Scandinavian Journal of Medicine &
Science in Sports, 12(1), 54–59.
Potrac, P., & Cassidy, T. (2006). The co ach as ‗a more capable other. In R. Jones (Ed.),
The sports coach as educator. Abingdon: Routledge.
Potrac, P., & Jones, R. (2009). Power, conflict and cooperation: Toward a
micropolitics of coaching. Quest, 61, 223–236.
Richards, P., Mascarenhas, D.R.D., & Collins, D. (2009). Implementing reflective
practice approaches with elite team athletes: parameters of success . Reflective
Practice, 10(3), 353–363.
Saury, J., & Durand, M. (1998). Practical knowledge in expert coaches: on site study
of coaching in sailing. Research Quarterly for Exercise and Sport, 69(3), 254–
266.
Schön, D. (1987). Educating the reflective practitioner. San Francisco: Jossey-Bass.
Smith, R.E., Smoll, F.L., & Curtis, B. (1979). Coach effectiveness training: A
cognitive-behavioural approach to enhancing relationship skills in youth sport
coaches. Journal of Sport Psychology, 1, 59–75.

Abraham and Collins Ways Forward for Coaching Science
Page 19 of 21
Strean, W.B., Senecal, K.L., Howlett, S.G., & Burgess, J.M. (1997). Xs and Os and
what the coach knows: Improving team strategy through critical thinking. The
Sport Psychologist, 11, 243–256.
Tetlock, P. (2005). Expert political judgment: How good is it? How can we know?
Princeton University Press.
Tjeerdsma, B.L. (1995). ―If-then‖ statements help novice teachers deal with the
unexpected. Journal of Physical Education, Recreation & Dance, November-
December, 22–24.
Vygotsky, L.S. (1978). Mind in society: The development of higher psychological
processes. Cambridge, MA: Harvard University Press.
Weigand, D.A., Carr, S., Petherick, C., & Taylor, A. (2001). Motivat ional climate in
sport and physical education: the role of significant others. European Journal of
Sport Science, 1(4), 1–13.
Zimmerman, L.A., & Harris-Thompson, D. (2008). Developing expertise. ARA
Technology Review, 4(1), 17–22.

Abraham and Collins Ways Forward for Coaching Science
Page 20 of 21
Figure 1 — The Nested Model: an approach to guiding classical and naturalistic decision making. In this instance high performance sport provides
the context; differing contexts would/could lead to differing objectives, timelines, and content.

Abraham and Collins Ways Forward for Coaching Science
Page 21 of 21
Author Queries
[AUQ1] The in-text citation "Jowett & Cramer, 2009" is not in the reference list. Please correct the citation, add
the reference to the list, or delete the citation.
[AUQ2] Reference "Gilbert, Trudel, 2005" is not cited in the text. Please add an in-text citation or delete the
reference.