REVIEW ARTICLE Sports Med 2001; 31 (12): 875-890

0112-1642/01/0012-0875/$22.00/0

© Adis International Limited. All rights reserved.

Psychological Factors in
Sport Performance
The Mental Health Model Revisited
John S. Raglin
Department of Kinesiology, Indiana University, Bloomington, Indiana, USA

Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 875
1. Mental Health Model (MHM) Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 878
1.1 Introversion-Extroversion and Athletic Success . . . . . . . . . . . . . . . . . . . . . . . . . . . 879
2. Dynamic Psychological Factors in the MHM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 880
3. State Anxiety and Sport Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 882
4. Assessment of Mood State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883
5. Reviews of the MHM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883
6. Recommendations for Assessment of Mood State . . . . . . . . . . . . . . . . . . . . . . . . . . . 885
7. Mental Health and Athletic Injury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 886
8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 887

Abstract The Mental Health Model (MHM) of sport performance purports that an in-
verse relationship exists between psychopathology and sport performance. The
model postulates that as an athlete’s mental health either worsens or improves
performance should fall or rise accordingly, and there is now considerable support
for this view. Studies have shown that between 70 and 85% of successful and
unsuccessful athletes can be identified using general psychological measures of
personality structure and mood state, a level superior to chance but insufficient
for the purpose of selecting athletes. Longitudinal MHM research indicates that
the mood state responses of athletes exhibit a dose-response relationship with
their training load, a finding that has shown potential for reducing the incidence
of the staleness syndrome in athletes who undergo intensive physical training.
The MHM also has implications for the general care of athletes as support services
have traditionally been limited to preventing or treating physical problems. De-
spite its simple premise and empirical support, the MHM has often been
mischaracterised in the sport psychology literature and recently some authors
have questioned its validity. This overview will summarise MHM research, in-
cluding the more recent work involving the model’s dynamic features in an effort
to resolve disputes surrounding the model.

From the inception of sports medicine, efforts cess have largely centred on biological variables,[1]
to identify the factors most crucial for athletic suc- but the contribution of psychological factors was
876 Raglin

recognised early on as well. As stated by Dudley[2] ures were too general in context, and contended it
in an article on the psychology of the athlete pub- would be necessary to develop sport-specific meas-
lished in 1888, ‘in all success in athletics the mental ures before the potential role of personality in sport
qualities of the athlete figure largely’ (page 43). could be adequately evaluated.[17] This antithesis
However, assertions such as this could not be em- to the credulous position has since been referred to
pirically tested until the development of valid psy- as the sceptical perspective.[10] However, it has been
chological measures. Early psychological research recognised that much of the evidence used to refute
in sport focused largely on the relationship between the position that an association exists between per-
personality traits with sport participation or suc- sonality and athleticism was the consequence of a
cess in athletics.[3,4] Personality traits have been variety of methodological flaws common to much
defined as: ‘relatively enduring differences among of the research, rather than a demonstration of the
people in specifiable tendencies to perceive the world inadequacy of trait theory or general measures of
in a certain way and in dispositions to react or be- personality.[9,10,18] Among the most serious prob-
have in a specified manner with predictable regu- lems noted was the consistent failure to test for the
larity’(page 1).[5] There continues to be debate over presence of response distortion, defined as the ten-
what are the fundamental factors of personality struc- dency for some individuals to respond falsely to
ture,[6] but most personality measures include di- psychological questionnaires. Several styles of re-
mensions related to emotionality (e.g. emotional
sponse distortion have been identified but the most
stability-neuroticism) and social adaptability (e.g.
prevalent is referred to as social desirability, whereby
introversion-extroversion).[7]
questionnaires are completed in a stereotypically
Reviews of the sport psychology literature pub-
positive manner.
lished during the 1960s generally indicated person-
Subsequent reviews of the personality and sport
ality factors such as extroversion or emotional sta-
literature have not supported either of the extreme
bility were moderately and positively associated
positions described by the credulous or sceptical
with participation in sport and athletic success.[3,4]
perspective.[9,10,18] Carefully conducted studies that
Some sport psychologists proposed that the associ-
accounted for potential confounds such as response
ation was far stronger. For example, Ogilvie and Tut-
ko[8] developed the Athletic Motivation Inventory, distortion consistently revealed several personality
a personality measure specifically for athletes and factors, particularly extroversion and emotional sta-
claimed: ‘outstanding athletes can be identified by a bility, to be moderately associated with sport par-
number of common, desirable personality charac- ticipation or performance. Unfortunately, even meth-
teristics’ (page 10). This extreme credulous perspec- odologically sound studies of personality and sport
tive on the efficacy of personality has since been have often been purely descriptive, and the use of
refuted,[9-11] yet a variety of psychological meas- a theoretical framework in this line of inquiry re-
ures, including assessments of personality, continue mains infrequent.[19] One exception to this trend has
to be widely used as a means to select athletes for been the Mental Health Model (MHM) of sport per-
professional teams in the US.[12,13] formance which was developed by Morgan,[10,11,18]
In contrast, overviews of this literature publish- who described the model as indicating ‘that suc-
ed during the 1970s were nearly unanimous in con- cess in sport is inversely correlated with psychopa-
cluding that personality was not a meaningful fac- thology’ (page 71).[11] The model predicts that ath-
tor in sport,[14-17] leading one reviewer to call for letes characterised by elevated scores on measures
the ‘trait psychology approach to be abandoned’ of psychological constructs such as neuroticism,
(page 35).[14] A related viewpoint expressed by other trait anxiety, depression, confusion and fatigue will
reviewers deemed that existing personality meas- tend to be unsuccessful compared with athletes

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 877

scoring in the low to normal range on these meas- ality other than the absence of psychopathology’
ures. (page 46).1 This description fails to acknowledge
Another way of presenting the basic premise of that Morgan’s conceptualisation of the model explic-
the model is that positive aspects of mental health itly assigned positive psychological characteris-
should be associated with greater success in sport. tics to successful athletes such as emotional stabil-
Hence, successful athletes would be predicted to ity. Other reviewers have correctly recognised that
exhibit desirable psychological traits such as emo- the MHM does, in fact, include several desirable
tional stability, low trait anxiety and high psychic psychological attributes, but have erroneously lim-
vigour. The MHM also includes a dynamic aspect ited the model to the relationship between mood
involving the impact that changes in mental health states and performance.[21,26,27] Finally, some re-
over time have on performance that, with rare ex- searchers have used the MHM to examine the im-
ception (e.g. Vanden Auweele et al.[20]), has not pact of precompetition mood state assessments on
been acknowledged in other reviews of the model. performance,[27] an application that is incongruent
Findings from research involving this aspect of the with Morgan’s own research on the model.
MHM indicate that intensive physical training rou- Unfortunately, misinterpretations of the MHM
tinely results in mood disturbances and decrements have often been reported at face value in sport psy-
in performance in athletes who exhibit positive men- chology textbooks,[28,29] and their pervasiveness in-
tal health profiles off-season or during periods of dicates it is timely to revisit the model. The remain-
easy training. der of this review will summarise the major findings
Despite the simple premise and conceptual parsi- of MHM research. Results from research on the dy-
mony of the MHM, it has often been misdescribed namic aspect of the model will also be included as
in sport psychology literature. In some cases, re- previous reviews have often not addressed this work.
viewers have overstated the effectiveness of the Findings regarding the influence of precompeti-
model in identifying successful athletes, contending tion anxiety on performance will be presented in
it should be capable of ‘fine grain distinctions’,[21] an attempt to distinguish the nomothetic perspec-
or misquoting Morgan as claiming that the results tive of the MHM from an idiographic model of
from MHM research support the credulous perspec- sport performance. Recommendations will be for-
tive.[22] Meyers et al.[23] have questioned the valid- warded regarding the use of mood state measures
ity of the MHM based on a similar misunderstand- for MHM research as will some suggestions for
ing, declaring that the model predicts elite athletes extending the model to incorporate more complex
should never exhibit poor mental health. Yet, Mor- interactional or hierarchical models of behaviour.
gan[10,11,18] has consistently stated that psycholog-
ical variables that the MHM was derived from are 1 The issue of psychopathology in athletes is of additional
only moderately associated with athletic perfor- importance because a prevailing model of sport psychology
focuses on education rather than treatment. In an influential
mance, emphasising that the accuracy of the model article, Danish and Hale[24] argued against a more conserva-
was both ‘far from perfect’ and ‘not acceptable for tive approach based on providing athletes standardised men-
selection purposes’ (page 79).[11] Moreover, the dy- tal health services. In fact, these authors contended that
namic aspect of the MHM predicts that elite ath- psychological help ‘could actually come at the expense of
the athletes’ self-esteem and welfare’ (page 98). This per-
letes undergoing intensive training should experi- spective was echoed by statements made in an earlier publi-
ence an elevation in negative mood measures. cation by Olgivie and Tutko[8] who also questioned the value
The psychological variables the MHM was of psychotherapy for athletes. Yet, various forms of mental
illness are far from uncommon in sports, and in a review of
founded on have also been misrepresented in the
this literature by Brewer and Petrie[25] it was concluded that
literature. For example, Vealey[22] has criticised the psychopathology occurs in athletes ‘at rates equal to and
model for ‘failing to focus on attributes of person- sometimes greater than the general population’ (page 269).

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
878 Raglin

1. Mental Health Model gan and Johnson[35]). For the clinical method, pre-
(MHM) Research dictions of success or failure were based on a cri-
terion of ±1 standard deviation of the sample mean.
The MHM was introduced by Morgan over 20 The overall accuracy of the statistical and clinical
years ago[10,18] and was described in detail in an methods designed to identify successful or unsuc-
article published in 1985[11] that summarised re- cessful athletes averaged ≈80%, and generally ranged
sults from 8 investigations conducted with athletes from 70 to 85%.[11] Statistical tests consistently in-
from several sports. Sample sizes ranged from 16 dicated that these prediction rates were more accu-
to 735 and the athletes ranged in skill from pre-elite rate than those achieved by chance alone, and re-
to elite or international calibre. The protocol typi- gression analyses also yielded results in the moderate
cally employed in this research consisted of having range (e.g. multiple regression coefficient = 0.73).
athletes complete standardised psychological meas- The results from these studies indicated that suc-
ures of personality and mood. Individuals were then cessful athletes possessed mean scores on psycho-
categorised in discrete groups (i.e. successful, un- logical measures that indicated better mental health
successful) based on pre-existing criteria such as compared with their less successful peers, includ-
number of varsity letters earned or starting team ing measures of mood state and emotional stability.
status. Psychological assessments in other studies In the case of mood, successful athletes scored low
(e.g. Nagle et al.[30]) were used to make predictions in undesirable factors of tension, depression, anger,
of team status before athletes were chosen for cham- fatigue and confusion, and above average in the
pionship teams.2 desirable factor of vigour. Morgan[11] referred to this
In some of these studies predictions of success characteristic mood response as the ‘iceberg pro-
and failure were conducted statistically by means file’ in reference to the contour the mood scores
of discriminant function analysis (e.g. Nagle et described when graphically plotted in T-score units.
al.[30]), whereas objective clinical evaluations of Early cross-sectional MHM research indicated that
the psychometric data were used to classify ath- differences in mood state between successful and
letes before selection in other research (e.g. Mor- unsuccessful athletes were generally more pro-
2 The strategy of comparing groups rather than intraindivid- nounced during periods of intense training,[30,35]
ual assessments in which performance is rated along a con- a finding which contributed to the development of
tinuum has since been criticised.[21,26] However, Morgan[11] the dynamic aspect of the model which will be de-
recognised that given constraints inherent in studying homo- scribed in the following section.
geneous samples of athletes ‘the use of extreme scores can
sometimes be quite effective’ (page 77). Such an approach Successful athletes were also less likely to be
circumvents the challenges that arise from attempting to introverted, and scored lower on various measures
compare performances across sports, or even between differ- of response distortion.[30,36] The importance of the
ent events in a given sport. Moreover, the theoretical view latter variable was demonstrated by Nagle et al.[30]
that intraindividual measures of sport performance are supe-
rior to between-group measures[31] has not been supported who observed that a measure of response distortion
by research[32,33] that has compared the efficacy of these proved to be the single most effective psychologi-
approaches and found them to be comparable. Similar con- cal variable in distinguishing successful and unsuc-
straints have also been noted when using physiological vari- cessful contenders for a US Olympic Wrestling team,
ables in the attempt to predict athletic performance. For
example, in studies of heterogeneous groups maximal oxy- with the unsuccessful athletes exhibiting profiles
gen uptake is consistently and highly correlated with endur- indicative of social desirability. Aside from the meth-
ance performance. Yet its utility is far less impressive in the odological value of including measures of response
case of highly trained athletes. As stated by Costill:[34] ‘Max-
distortion, recent research indicates that social de-
imal oxygen consumption frequently fails to predict the
winners when a group of similarly talented runners compete’ sirability has health consequences that could po-
(page 13). tentially affect performance. Brody et al.[37] found

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 879

a positive correlation between social desirability tive.[35] In one study,[30] it was found that wrestlers
scores and salivary cortisol levels in young adults, who qualified for the US Olympic team possessed
leading them to suggest that elevated social desir- higher Profile of Mood States (POMS) anger scores
ability could result in ‘poorer stress coping, greater than the nonqualifiers, implying that anger may
depression and greater risk for long term hormonal be a desirable factor in wrestling. However, other
and immunologic damage’ (page 228). POMS research has failed to find variations in mood
It is important to note that the mental health profiles among samples of successful athletes in
profiles of unsuccessful athletes typically did not different sport,[38,39] contradicting proposals that
differ significantly from the published norms, in- mood state profiles associated with success are sport-
dicating the lack of success was more often asso- specific.
ciated with average psychological health rather than
an unhealthy profile or the presence of clinically 1.1 Introversion-Extroversion and
documented psychopathology. Hence, differences Athletic Success
in mental health observed between successful and
unsuccessful groups were seldom large in magni- Based on the findings of these studies, Morgan[11]
tude nor were clinical conditions noted in unsuc- proposed that the absence of introversion is more
cessful competitors. It should also be pointed out important to sport success than possessing high ex-
that the psychological assessments employed were troversion scores as was hypothesised in earlier re-
not designed for the purpose of diagnosing mental views.[3,4] Hence, better performers would be more
illness. To some degree the failure to observe greater likely to score anywhere from average to high in
differences in psychological factors between suc- extroversion. Perceptual factors associated with in-
troversion may act to mediate success in some sports.
cessful and unsuccessful may also have been the
For example, there is evidence that introverts pro-
result of conservative tests of the model. It has been
vide higher ratings of perceived exertion than ex-
noted[21,26] that athletes defined as unsuccessful in
troverts at matched exercise workloads.[40] Research
several studies by Morgan and colleagues could
also indicates that extroversion is associated with
easily be regarded as elite as they involved candi-
desirable coping styles. Extroverts are more likely
dates for Olympic or World teams.
to choose active coping strategies that focus on the
Despite the observation that successful and source of stress and social support, whereas indi-
unsuccessful athletes could consistently be distin- viduals scoring high in neuroticism are more likely
guished at levels better than chance solely on the to use coping strategies that involve avoiding or
basis of psychological variables, the level of accu- ignoring the stressor.[41,42] Some reviewers have pro-
racy attained was deemed insufficient for the pur- posed that introverted athletes possess lower levels
pose of selecting individuals for competition. Not of pain tolerance than extroverts.[9,43] However, re-
unexpectedly given the modest efficacy of the model, search on the topic of pain perception and tolerance
Morgan[11] reported that the statistical and clinical in athletes is surprisingly absent, and recent work
prediction methods each consistently resulted in has failed to find an association between personal-
errors such as false negatives whereby individuals ity and pain perception.[44]
are successful despite the absence of positive men- More recently, the relevance of the personality
tal health. Conversely, there were cases of athletes dimension of introversion-extroversion to athleti-
with uniformly positive mental health who did not cism has been challenged. Based on a meta-analysis
succeed (i.e. false positives). Predictions using the of the sports personality literature, Vanden Auweele
clinical method were not always possible because and colleagues[20] concluded that above average ex-
in some instances the psychological profiles of ath- troversion is not characteristic of the elite athlete.
letes were neither uniformly positive nor nega- The review, however, was restricted to studies that

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
880 Raglin

measured extroversion with either the 16-Personality studies of elite athletes), and contradictory results
Factors Inventory (16-PF)[45] or the Eysenck Per- for some individuals were also noted. Neverthe-
sonality Inventory (EPI),[46] and it has been noted that less, positive results were replicated across sam-
these measures of extroversion are not highly in- ples of elite and non-elite athletes in several sports,
tercorrelated.[7] Results based on the Eysenck Per- and other researchers have since replicated these
sonality Questionnaire (EPQ)[47] were also omitted findings,[39,49] indicating the model possesses con-
as the authors contended it lacked appropriate norms siderable external validity. On the basis of these
for American athletes. Extroversion research based results, Morgan[11] concluded that there was sup-
on other scales (e.g. Minnesota Multiphasic Per- port for the MHM, although cause and effect could
sonality Inventory) was also excluded, yet the anal- not be established, and it was acknowledged that
ysis did include results based on foreign language participation in sport could have improved mood
translations of the EPI or 16-PF. No indication was state. Although the MHM was conceived on the
given whether the studies included in the analysis basis of research with athletes, there is also evi-
tested for response distortion, despite the fact that dence that success in other performance domains,
the EPI and 16-PF each include scales for this pur- particularly those requiring physical skill or condi-
tioning, is associated with superior mental health
pose. In cases where extroversion scores between
and mood states. Examples include military flight
successful and less successful athlete groups were
school trainees,[50,51] astronaut candidates,[52,53] arc-
provided, the data were collapsed to yield a single
tic expedition members,[54] scuba divers[55] and mili-
effect size. Moreover, it has been contended by Ey-
tary special forces personnel.[56] These findings sug-
senck et al.[9] and Hardman[43] that introversion may
gest that the tenets of the MHM are not limited to
be advantageous for particular sports such as pistol
sport performance, but also apply to success in a
shooting, and this is relevant as 15% of the athletes wide range of physical tasks.
in this meta-analysis were involved with shooting
sports.
Given the previous concerns, it is proposed that 2. Dynamic Psychological Factors in
the dimension of introversion-extroversion remains a the MHM
viable topic for study in sport. Future research may
In his review of the MHM, Morgan[11] proposed
also benefit from considering the association of dif-
that the model could be improved by assessing psy-
ferent aspects of extroversion (e.g. social boldness,
chological variables over time, stating that: ‘There
social warmth) and its interaction with other per- is evidence that an athlete’s psychologic response
sonality factors[44] with athleticism in various sports. to training, not his or her base-line characteristics,
Recent findings regarding the neurobiology of ex- represents the most important issue’ (page 79).[11]
troversion and its contribution to behaviour may Specifically, the accuracy by which athletic perfor-
also be relevant to personality and sport research.[48] mance can be predicted from mental health status
In summary, tests of the MHM based on person- should be enhanced beyond the level attained with
ality measures and stable aspects of mood state – trait assessments by incorporating assessments of se-
the static aspect of the model – revealed a consis- lected psychological variables (i.e. states and mood)
tent trend for more successful athletes to possess at regular intervals, particularly during training. Ini-
better mental health compared with less successful tially, work involving the dynamic aspect of the
athletes. The magnitude of the difference in mental MHM was limited,[11] but subsequent research find-
health measures between groups typically was found ings have provided strong support for this aspect
to be modest, as was the predictive efficacy of the of the model. Results from studies by Morgan and
model. Several of the studies were limited by the colleagues[57,58] of over 500 athletes in endurance
relatively small sample size (a problem inherent in sports such as swimming found that the mental

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 881

health of athletes, as measured by mood state, is diate this response, but the findings thus far have
closely tied to physical training. This research in- been inconclusive.[63]
dicates that as training load is increased there is a The results from monitoring research based on
corresponding elevation in mood disturbance and the dynamic aspect of the MHM have been applied
a plateau or even a decline in performance. as an intervention strategy intended to reduce the
During light training, scores on mood distur- incidence of staleness. Berglund and Säfström,[64]
bance measures are typically 1 to 2 standard devi- had world-class race canoeists complete the POMS
ations below published norms. With elevations in regularly throughout a training season in prepara-
training volume or intensity, mood disturbance val- tion for the Olympic games. POMS scores were
ues increase and at the peak training load (i.e. over- evaluated and athletes who possessed what were
training) will typically equal or exceed the published deemed as excessive total mood disturbance scores
norms with the positive variable of vigour exhibiting (i.e. >50% above baseline values) had their training
a corresponding decline. Reductions in training load loads reduced. Conversely, training was increased
result in corresponding improvements in mood for in athletes who displayed significantly lower mood
most athletes. Intensive anaerobic conditioning pro- disturbance scores than their teammates. This up-
grammes have also been found to result in eleva- ward titration of training is consistent with the MHM;
tions in negative mood factors.[59] Hence, changes athletes with ‘better than average’ mental health
during training should be capable of tolerating, and
in training load are associated with mood distur-
in theory, should benefit from even harder training
bance in the form of a dose-response function.[57-61]
regimens. Training loads were either increased or
Fluctuations in mood state that occur during ath-
decreased at some point during training in over half
letic training have also been found to be associated
the sample. No cases of staleness occurred and sev-
with biological measures relevant to training state
eral of the athletes earned Olympic medals, leading
including endocrine[60] and neuromuscular vari-
the authors to conclude that the intervention strat-
ables.[62]
egy was successful.
Most athletes can cope with intensive overtrain-
As proposed by Morgan,[11] the dynamic mood
ing, but ≈10% display excessive disturbances in
response to training can be integrated with the static
mood state and respond poorly to training tapers. aspect of the model to provide a more complete
These individuals experience the staleness syn- depiction of the association between psychological
drome, also referred to as overtraining syndrome; variables and athletic success. In a study that utilised
a condition associated with a chronic impairment this approach,[65] women athletes vying for a posi-
in physical performance, depression and various tion on a collegiate rowing team were assessed on
other physical and psychological symptoms.[57,58] mood state (POMS), the trait of self-motivation,
Athletes who develop staleness manifest compar- and rowing ergometer performance at the outset of
atively greater elevations in each of the POMS mood the season. Mood state was also monitored bimonthly
factors, with depression exhibiting the largest in- during the entire season. These variables were con-
crease.[58] In athletes who overtrain but do not de- trasted between women who discontinued training
velop staleness, the largest changes in POMS fac- with those who adhered to the entire season. Com-
tors occur in vigour and fatigue, whereas elevations parisons were also made between adherers selected
in confusion and depression often do not reach sta- to compete in a National competition (i.e. finalists)
tistical significance. Hence, mood disturbances in and rowers who were not selected (i.e. nonfinalists).
stale and healthy athletes who overtrain differ both It was found that self-motivation scores and row-
in magnitude and pattern. Some attempts have been ing ergometer performances were lower for indi-
conducted to identify psychological traits that me- viduals who dropped out of the team, but did not

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
882 Raglin

differ between finalists and nonfinalists. Ath- uniformly harmful to sport performance or benefi-
letes with higher self-motivation scores also had cial only for physically demanding gross motor tasks.
faster times in a rowing ergometer test. The baseline The findings of Hanin and others[68] indicate
mood disturbance scores did not differ across groups that models based on individual (i.e. idiographic)
and were positive compared with the published col- responses are more efficacious than group-based
lege norm. However, during heavy training the non- (i.e. nomothetic) theories for studying the effects of
finalists were found to be more likely to exhibit transient mood state changes resulting from compe-
elevations in mood disturbance which persisted fol- tition on the performance of psychologically healthy
lowing a training taper, whereas the mean mood dis- athletes. This distinction is important as the static
turbance value of the finalists returned to baseline. aspect of the MHM is nomothetic, and concerns the
These results indicate that employing both static association of personality factors and trait-like meas-
and dynamic psychological assessments can enhance ures of mood with performance in groups of ath-
the understanding of the association of psycholog- letes. However, the dynamic aspect of the MHM
ical factors with various aspects of athletic success, has both nomothetic and idiographic elements: mood
such as adherence to training and team status (i.e. disturbances associated with intensive training have
nonfinalists, finalists). MHM research would ben- been found to be consistently associated with im-
efit by conducting tests to determine if these and paired performance, but significant individual vari-
other psychological variables exhibit a hierarchical ability exists across athletes in their tolerance of
structure,[66] and by attempting to more explicitly hard training.[70] In turn, the IZOF model is entirely
integrate the static and dynamic aspects of the model. idiographic and indicates that considerable interin-
dividual differences should exist not only in the
3. State Anxiety and Sport Performance level of anxiety athletes experience before compe-
tition but the impact of anxiety (i.e. beneficial or
The stress of sport competition commonly re- detrimental) on performance as well.
sults in acute elevations in state anxiety that occur Some sport psychology research has utilised
hours or sometimes even days before the event. both the MHM and the IZOF to study athletic per-
Considerable interindividual variability in the in- formance. In studies of elite men and women dis-
tensity of state anxiety responses have been ob- tance runners, Morgan and colleagues[38,71] found
served,[67,68] and more importantly, its impact on uniformly positive mood state profiles in the sam-
performance also varies from athlete to athlete. Re- ples as would be predicted by the MHM. Yet the
search on several sports indicate that 30 to 45% of same athletes reported widely differing levels of
athletes report optimal sport performance while ex- optimal precompetition anxiety, leading the authors
periencing significant elevations in state anxiety.[68] to conclude that these results did ‘conform to the
Such findings serve as the basis for Hanin’s Indi- theoretical views of Hanin’ (page 129).[71] The find-
vidual Zones of Optimal Functioning (IZOF) model, ing that models based on nomothetic and idiographic
which posits that many athletes perform optimally levels of analysis can be used conjointly has poten-
only under high levels of state anxiety that arise tial applications in creating an integrative, state-
from stressors specific to sport competition. Ac- trait methodology to study the psychology of the
cording to the IZOF model, factors such as sport athlete in the context of different stressors such as
type, experience, or skill exert no systematic or group training and competition.
effect on the level of anxiety associated with op- However, the complementary application of the
timal performance,[67-69] and the model stands in IZOF and MHM has not always been recognised,
strong contrast to traditional theoretical explana- and instead these two approaches have been de-
tions in which elevated anxiety is seen as either picted as competing explanations of the influence

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 883

of precompetition mood state on sport performance. 65-item Likert questionnaire that measures the spe-
Prapavessis and Grove,[72] for example, used the cific mood states of tension, depression, anger, vig-
POMS to assess the acute mood state (i.e. ‘right our, fatigue and confusion. These factors are often
now’) of pistol shooters immediately before sev- combined to create a higher order factor referred
eral competitions in an attempt to contrast the rel- to as global or total mood disturbance. The use of
ative efficacy of IZOF with the MHM. Differences the term ‘state’ to describe the mood variables as-
in performance were found only for tests of the sessed by the POMS can be misconstrued as the
IZOF model, leading the authors to reject the MHM. test-retest reliabilities of the POMS factors (range
As previously noted, however, these models focus = 0.56 to 0.74)[77] are considerably higher than true
on the consequences of different stressors (i.e. train- state measures such as the state anxiety (range =
ing vs competition) on performance, examine the 0.16 to 0.33),[5] but fall short of measures of per-
association between psychological variables and sonality traits (range = 0.80 to 0.90).[46] Conse-
sport performance at different levels of analysis quently, the mood state factors assessed by the POMS
(i.e. group vs individual), and involve different psy- are actually somewhat trait-like as defined by test-
chological measures (traits vs states). More recently, retest reliability. This characteristic makes the POMS
Prapavessis[73] and other researchers[74] have ac- well suited for overtraining research as it yields
knowledged these distinctions and concluded that measures that depict typical mood, yet are also sen-
the IZOF model is more appropriate for tests of the sitive to ‘acute treatment effects’ (page 2)[77] or the
influence of precompetition mood on performance. influence of major stressors.
Hanin has recently extended the IZOF model A more transitory measure of mood comparable
beyond anxiety to incorporate idiographic measures with true state inventories can be assessed with the
of both positive and negative affect. This concept- POMS by administering it with the ‘right now’ or
ualisation is consistent with the emergent perspec- ‘today’ instructional set. This modification has been
tive that emotional regulation and coping involves found to be useful in studies examining mood re-
increased intensity of both positive and negative sponses to rapid increases in training load occur-
emotion that are interpreted differently depending ring over a span of days rather than weeks. Over-
upon the context,[66,75] whereas the traditional per- training research based on this version of the POMS
spective which remains popular throughout much indicates that as few as 2 days of intensified train-
of sport psychology is that negative affect should ing can result in significant increases in acute mood
be largely absent. Given the fact that many sporting disturbance,[78] with elevations most pronounced
events are long in duration, it has been proposed that for athletes showing signs of staleness.[79] Hence,
the study of secondary emotional responses that detrimental psychological consequences of intense
occur during the event would be useful.[76] Whether physical training can become manifest quite rap-
the MHM and IZOF model can be melded into a idly. Unfortunately, many studies of athletes using
truly interactional framework remains to be deter- the POMS have failed to consider the profound
mined from future studies; however, it is clear that effect training can have on mood state; elevations
employing these models conjointly can provide a in negative mood state factors resulting from inten-
greater understanding of how psychological traits sive training can be misinterpreted as repudiating
and states independently influence athletic perfor- the MHM rather than supporting the dynamic as-
mance. pect of the model.[80]

4. Assessment of Mood State 5. Reviews of the MHM

The psychological measure most commonly used Several recent reviews have examined the effi-
in overtraining research has been the POMS,[77] a cacy of the MHM.[21,26,27] It is significant that each

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
884 Raglin

review was confined to POMS research and MHM to that reported by Rowley et al.[21] do in fact have
studies that used other psychological measures in practical utility for homogeneous groups that ex-
concert with the POMS were not evaluated. While hibit considerable range restriction,[50] as is the case
being uniformly critical of the POMS, the specific for elite athletes. For example, in a review compar-
conclusions drawn from these reviews differed. In ing the efficacy of psychological testing to medical
a narrative review, Renger[26] charged that mood test validity, Meyer et al.[81] concluded that ‘psy-
state was ineffective in distinguishing successful chologists studying complex behaviour should be
from less successful athletes. Echoing criticisms of rather satisfied that they can identify independently
the sport psychology literature made over 20 years measured constructs that are of the magnitude ob-
earlier, Renger recommended sport psychology re- served for antihistamine medication (r = 0.11)’ (page
searchers interested in using personality measures 134). It also has been demonstrated by Rosenthal[82]
to identify successful athletes ‘abandon the POMS’ that even small effect sizes can have practical sig-
(page 83). However, Renger[26] did conclude that ath- nificance in terms of gains in accuracy.
letes generally possessed more favourable mood state The previous reviews concluded that the MHM
profiles than nonathletes. This conclusion differs was flawed or invalid, yet they each only focused
from that reached in a narrative review by Terry,[27] solely on POMS research. The reviews also included
who asserted that the POMS offered only ‘a limited studies that did not account for the influence of
capacity to distinguish the athlete from the nonath- physical training or competition on mood state. These
lete’ (page 322). Terry also stated that successful issues were recognised in a review of POMS re-
athletes generally possessed desirable mood pro- search by Vanden Auweele and colleagues[20] who
files immediately before performance by means of only considered studies in which mood state was
the ‘right now’ set, but it should be noted that none assessed during baseline conditions or during easy
of the MHM studies conducted by Morgan and col- training. The potentially confounding effects of sport
leagues[11,35,38] involved POMS assessments imme- competition on mood state were also recognised
diately before competition using these instructions. and the authors excluded studies that used the ‘right
Rowley et al.[21] conducted a meta-analysis of now’ instructional set of the POMS. On the basis
POMS research to test the efficacy of the MHM of the results it was concluded that ‘elite American
under the premise that ‘successful athletes possess athletes typically display the iceberg profile, espe-
more of an iceberg profile than less successful ath- cially during periods some time before competi-
letes’ (page 186). An analysis of 33 studies resulted tion’ (page 263).[20]
in an overall mean effect size of 0.15 (p < 0.05), Vanden Auweele et al.[20] defined positive mood
which increased to 0.20 when the most extreme in terms of the iceberg profile, as did the other re-
subscale scores were contrasted. However, no anal- viewers of POMS research with athletes. Employ-
ysis was conducted for results based on the widely ing the iceberg profile as a criterion measure or
used POMS total mood disturbance score. Several experimental variable is problematic because a gen-
methodological weaknesses, all of which have been erally accepted demarcation between a desirable (i.e.
previously noted by Morgan[10,18] and others,[19] were iceberg) and undesirable mood profile has not been
judged as reducing the magnitude of the effect sizes established. This has resulted in strikingly different
yielded. In large part because of the small effect definitions of the iceberg profile, involving criteria
size, the authors questioned the efficacy of the model as diverse as subjective visual inspection to quan-
and they proposed that future efforts be directed titative analysis. Terry,[27] for instance, has proposed
elsewhere rather than toward improving the meth- that extremely positive ‘Everest’ shaped POMS pro-
odology of MHM research. Other researchers have files are nearly always associated with successful
contended that effect sizes of a magnitude similar performance. Yet, exaggerated profiles may also be

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 885

indicative of response distortion and positive mood 6. Recommendations for Assessment

profiles have been observed in both average and of Mood State
unsuccessful performers, as well as nonathletes.
In a study widely cited as evidence against the The iceberg profile is best regarded simply as a
MHM, Durtschi and Weiss[80] employed a more visual representation of desirable emotional health
quantitative approach in a comparison of elite and as indicated by mood state. It can be useful as a
non-elite marathoners. The runners completed the clinical tool,[35] but its utility as an experimental
‘right now’ version of the POMS on each of the 10 variable is questionable. A preferable alternative
days preceding a race, and the total number of ‘ice- would be to contrast the T-scores of POMS subscales
bergs’ was compared between groups. The authors or, as encouraged by the authors of the instrument,
defined an iceberg profile as one in which scores to use total mood disturbance scores.[77] Reliance
of the negative POMS factors were each below a on the POMS alone to establish mental health pro-
T-score of 50 and the positive factor of vigour was files in athletes is not recommended. Other psycho-
above 50. However, by using this method, mood logical measures of personality structure should be
profiles only slightly better or worse than the norm included to yield a more comprehensive psycho-
values would be classified as desirable (i.e. ice- logical profile of the athlete, and measures to de-
berg) or undesirable (i.e. non-iceberg). A prefera- tect response distortion should routinely be em-
ble alternative would be to employ criterion cut- ployed as research has found responses indicative
offs of 1 standard deviation above and below the of social desirability in some athletes.[30,39] Future
published norms. In addition, it would be neces- MHM research may benefit by assessing addition-
sary to control for the influence of daily training al psychological variables associated with mental
load on mood. health such as coping skills[44,83,84] or social sup-
Despite the call to ‘abandon the POMS’[26] for port. Investigations employing measures based on
the study of personality in the athlete, when used more recent personality theories such as the 5-factor
appropriately it remains perhaps the most useful of model, may also prove efficacious.[7]
all psychological instruments in MHM research, in In tests of either the static or dynamic aspect of
large part because of its utility in both static and the MHM, baseline assessments should be conducted
dynamic contexts. Unfortunately, researchers often during periods of rest or easy training out of the
have not distinguished these aspects of the MHM, competitive season. If this is not possible, having
leading them to erroneously conclude that it is in- athletes complete the POMS on the basis of how
valid. The POMS has also been criticised for not they recall feeling during periods of easy training
having been expressly developed for athletes, and has been found to be a sufficiently accurate alter-
it has been argued that sport-specific mood instru- native.[84] In overtraining research, a common strat-
ments are needed.[28] However, this perspective lacks egy is to compare an individual athlete’s current
both a compelling theoretical and empirical basis. mood state score with the mean of other athletes
Research has shown the POMS to be responsive to undergoing comparable training, as opposed to the
both general and sport-specific stressors such as published norms. This method has been success-
training and competition. In fact, the nonspecific na- fully used in overtraining research as a means to
ture of the POMS makes it responsive to stressors out- monitor athletes showing signs of the staleness syn-
side the realm of sport, factors which sport-specific drome.[64] The presence of general stressors related
measures are often insensitive to but which are sa- to social and interpersonal events, as well as health
lient to the athlete. However, even when the POMS status and injury, should also be considered as there
is used properly, its association with sport perfor- is considerable evidence that these factors also in-
mance is likely to remain modest. fluence mood state.[77]

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
886 Raglin

When using the ‘right now’ version of the POMS the combination of poor coping skills and low so-
as a means to assess the responses to short term cial support accounted for 30% of variance in the
training regimens, it is important to account for the length of time missed because of injury. Explained
potential influence of impending competitions on variance dropped considerably when the influence
mood state. Researchers[27] have developed POMS of these variables was considered singly. Conversely,
athlete norms based on the ‘right now’ instructional resiliency against injury was best explained by a
set but do not recommend they be used in the con- disjunctive model. Athletes least likely to become
text of overtraining.[74] injured possessed either high coping skills or had
adequate social support, but the presence of both
7. Mental Health and Athletic Injury factors did not further enhance resiliency.
These results, along with the findings of Hanin,[68]
Research on the psychology of sports injury has have potential bearing on research on the psychol-
not been conceptualised with the MHM in mind, ogy of sport injury. Current theory[87] contends that
but findings involving the influence of mental health negative emotions such as anxiety and psycholog-
are consistent with the model. For example, in a ical stress increase the risk of injury because of
prospective study of US Alpine skiers, May et al.[85] elevated muscle tension or reductions in visual span.
found that depression and other psychological vari- Yet the impact of such changes would clearly de-
ables were associated with illness and injury expe- pend on the psychobiological requirements of the
rienced later during the season. Athletes with ele- sport task. For example, sport psychologists com-
vated levels of depression at the outset of the season monly teach athletes relaxation techniques to avoid
were more than twice as likely (50 vs 21%) to be injury. However, research by Mote[88] has found
dropped from the team. High life change scores and
that low muscle tension can be counterproductive
low well-being were also significantly correlated
in the sport of skiing and stated that ‘contraction of
with team status (r = 0.25 and 0.29, respectively),
the musculature is of the utmost importance to skier
and skiers possessing positive well-being were more
safety. Skiers should not relax when they fall’ (page
likely to make the elite team. The results indicate
323). The well established consequences of stress-
that positive mental health is related to athletic suc-
related emotions on physical health factors such as
cess defined either as the absence of injury or ill-
ness, or as elite designation. immune function, cognition and physiological en-
durance and strength would be more likely to result
Smith et al.[86] examined the contribution of
psychological factors and sport injury in 451 high in injury under general circumstances. In addition,
school athletes who completed assessments of life research has shown that depression diminishes fine
experiences, social support and psychological cop- motor skill and increases movement time,[89] fac-
ing skills before the athletic season. Rather than tors critical to performance in many sports.
merely determining the composite contribution of In summary, there is evidence that psychologi-
these factors in subsequent injury, the predictive cal variables relevant to the MHM can have a sig-
efficacy of two interactional models was contrasted. nificant impact on athletic performance. The like-
Based on a conjunctive model, optimal prediction lihood of experiencing emotional disorders such as
should result from a specified combination of mod- depression often can be reduced by possessing good
erator variables, with deviations from this combi- coping skills or being part of a strong social net-
nation reducing explained variance. In a disjunc- work, and these attributes may benefit the recovery
tive model, sufficient scores of any of the moderator from injury as well. Moreover, there is evidence
variables should optimise prediction. The conjunc- that extroversion is positively correlated with seek-
tive model best explained vulnerability to injury. In ing social support and use of coping strategies,[90]
athletes who experienced major negative life events, and this presents another means by which the per-

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 887

sonality dimension of introversion-extroversion im- health of athletes, as would be the case for mood
pacts upon sport performance. Future MHM research state monitoring to prevent the development of the
would benefit from examining the interactive ef- staleness syndrome in athletes who overtrain.[64]
fects of extroversion, social support and coping skills However, it is important that mood state monitor-
on athletic success. These findings also suggest that ing protocols be supervised by individuals with ap-
athletic success should be conceptualised as a pro- propriate training in psychology. In tests of the ef-
cess rather than solely in terms of outcomes such ficacy of any overtraining monitoring strategy, it
as team status (starter/nonstarter; successful/unsuc- would be ethically indefensible to experimentally
cessful) or meet performance. Factors such as prone- manipulate training load to deliberately create fully
ness or resistance to injury, illness and the staleness developed cases of the staleness syndrome in ath-
syndrome each contribute to athletic success and letes.
in turn, are associated with psychological variables. It has been contended that the MHM over-
emphasises the importance of psychopathology in
8. Conclusion athletic performance at the expense of positive fac-
tors such as psychological skills,[22] and some sport
The MHM proposes that athletes with poor to psychologists have claimed that psychotherapy may
average mental health should on the whole, per- actually be counterproductive for athletes.[8,24] How-
form worse than athletes with good mental health, ever, the finding by Meyers et al.[23] that 85% of
a perspective that has come under criticism in sev- elite athletes who requested performance enhance-
eral reviews.[21,26,27] Significantly, each of these re- ment services at the US Olympic training centre
views defined the model in terms of the iceberg actually experienced more serious psychological
profile and focused solely on mood state, rather
issues effectively argues against this perspective,
than considering the entire MHM literature. More-
and supports the relevance of the MHM and the
over, with one exception,[20] these reviews failed
efficacy of providing clinical services to athletes.
to adequately consider the influence of athletic train-
Prospective research with athletes training for the
ing on mood state, described by Morgan as the dy-
Olympic skiing team indicates that psychological
namic aspect of the MHM.[11] In contrast, research
measures, including depression, are associated with
in which the effects of physical training and athletic
competition are either controlled for[39,49] or ex- reduced performance and a greater incidence of
plicitly incorporated into the research design,[57,59-64] injury.[85] Such findings underscore what may be
has consistently found that desirable emotional the most important implication of the MHM.
health is associated with sport performance. Traditionally, in sports medicine there has been
The modest accuracy of the MHM in identify- an almost exclusive focus on maintaining the phys-
ing successful and unsuccessful competitors con- ical health of the athlete. Considerable resources
traindicates its use as a means to select athletes for have long been devoted toward this end with little
sport teams, and evidence that other psychological consideration of mental health concerns. The MHM
evaluation strategies could provide greater accu- model indicates that it is equally important to focus
racy has been lacking. Ethical concerns also miti- on psychological health, both in terms of its impact
gate against the use of psychological measures to on sport performance and for the general welfare
select athletes. Athletes may feel coerced to partic- of the athlete.
ipate in psychological testing and give up their rights Findings from research on idiographic perspec-
to privacy, thereby allowing coaches access to poten- tives such as the IZOF model can be integrated into
tially sensitive information. This application dif- MHM research to provide a theoretically comple-
fers from using psychological assessments as a mentary state-trait framework for examining the
means to help maintain the mental and physical association of psychological variables with sport

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
888 Raglin

performance. Morgan[10,11] also emphasised that the 19. Vealey RS. Sport personology: a paradigmatic and methodolog-
ical analysis. J Sport Exerc Psychol 1989; 11: 216-35
efficacy of a state-trait approach could be further 20. Vanden Auweele Y, De Cuyper B, Van Mele V, et al. Elite per-
improved by the inclusion of relevant physiological formance and personality: from description and prediction to
information to create a more representative psycho- diagnosis and intervention. In: Singer RN, Murphey M, Ten-
nant LK, editors. Handbook of research in sport psychology.
biological model of athletic performance.[11] Fu-
New York (NY): Macmillan, 1993: 257-89
ture MHM research could benefit by implementing 21. Rowley AJ, Landers DM, Kyllo LB, et al. Does the iceberg
both of these strategies as well as exploring more profile discriminate between successful and less successful
complex interactional relationships that involve ad- athletes? A meta-analysis. J Sport Exerc Psychol 1994; 17:
185-99
ditional psychological variables. 22. Vealey RS. Personality and sport: a comprehensive view. In:
Horn TS, editor. Advances in sport psychology. Champaign
(IL): Human Kinetics, 1992: 23-59
References 23. Meyers AW, Whelan JP, Murphy SM. Cognitive behavioral
1. Costill DL. Inside running: basics of sports physiology. Carmel strategies in athletic performance enhancement. In: Hersen
(IN): Cooper, 1986 M, Eisler RM, Miller PM, editors. Progress in behavior mod-
2. Dudley AT. The mental qualities of an athlete. Harvard Alumni ification. Vol 30. Pacific Grove (CA): Brooks/Cole, 1996:
Mag 1888; 6: 43-51 137-64
3. Cooper L. Athletics, activity and personality: a review of the 24. Danish SJ, Hale BD. Toward an understanding of the practice
literature. Res Q 1969; 40: 17-22 of sport psychology. J Sport Psychol 1981; 3: 90-9
4. Warburton RW, Kane JE. Personality related to sport and phys- 25. Brewer BW, Petrie TA. Psychopathology in sport and exercise.
ical activity. In: Kane JE, editor. Readings in physical educa-
In: Van Raalte JL, Brewer B, editors. Exploring sport and
tion. London: P.E. Association, 1966: 61-89
exercise psychology. Washington, DC: American Psycholog-
5. Spielberger CD, Gorsuch RL, Lushene PE, et al. In: Manual for
ical Association, 1996: 257-74
the state-trait anxiety inventory (Form Y). Palo Alto (CA):
26. Renger R. A review of the Profile of Mood States (POMS) in
Consulting Psychologist Press, 1983
the prediction of athletic success. J Appl Sport Psychol 1993;
6. Eysenck HJ. Four ways five factors are not basic. Perspect Ind
5: 78-84
Differ 1992; 13: 667-73
27. Terry P. The efficacy of mood state profiling with elite perform-
7. Digman JM. Personality structure: emergence of the five-factor
ers: a review and synthesis. Sport Psychol 1995; 9: 309-24
model. Ann Rev Psychol 1990; 41: 417-40
28. Hardy L, Jones G, Gould D. Understanding psychological prep-
8. Ogilvie BC, Tutko TA. Problem athletes and how to handle
aration in sport. New York (NY): John Wiley & Sons, 1997
them. London: Pelham, 1966
29. Weinberg RS, Gould D. Foundations of sport and exercise psy-
9. Eysenck HJ, Nias KBD, Cox DN. Personality and sport. Adv
chology. 2nd ed. Champaign (IL): Human Kinetics, 1999
Behav Res Ther 1982; 1: 1-56
30. Nagle FJ, Morgan WP, Hellickson RO, et al. Spotting success
10. Morgan WP. The credulous-sceptical argument in perspective.
traits in Olympic contenders. Phys Sports Med 1975; 3: 31-4
In: Straub WF, editor. An analysis of athlete behavior. Ithaca
(NY): Movement Publications, 1978: 218-27 31. Krane V. Conceptual and methodological considerations In
11. Morgan WP. Selected psychological factors limiting perfor- sport anxiety research: from the inverted-U to catastrophe
mance: a mental health model. In: Clarke DH, Eckert HM, theory. Quest 1992; 44: 72-87
editors. Limits of human performance. Champaign (IL): Hu- 32. Raglin JS, Turner PE. Anxiety and performance in track and
man Kinetics, 1985: 70-80 field athletes: a comparison of the inverted-U hypothesis with
12. Smith TW. Punt, pass, and ponder the questions. The New York ZOF theory. Pers Individ Diff 1993; 14: 163-72
Times 1997 Apr 20; Sect F: 11 33. Turner PE, Raglin JS. Variability in precompetition anxiety and
13. Hruby, P. Tests of character. The Washington Times 2000 Feb performance in college track and field athletes. Med Sci
25; Sect B: 2 Sports Exerc 1996; 28: 378-85
14. Kroll W. Reaction to Morgan’s paper: psychological conse- 34. Costill DL. Applied exercise physiology. American College of
quences of vigorous physical activity and sport. In: Scott MG, Sports Medicine: 40th Anniversary Lectures. Indianapolis
editor. The academy papers. Iowa City (IA): American Acad- (IN): American College of Sports Medicine, 1994: 69-79
emy of Physical Education, 1976: 35 35. Morgan WP, Johnson RW. Personality characteristics of suc-
15. Martens R. The paradigmatic crisis in American sport psychol- cessful and unsuccessful oarsmen. Int J Sport Psychol 1978;
ogy. Sportwissenschaft 1975; 5: 9-24 9: 119-33
16. Rushall BS. An evaluation of the relationship between person- 36. Johnson RW, Morgan WP. Personality characteristics of college
ality and physical performance categories. In: Kenyon GS, athletes in different sports. Scand J Sport Sci 1981; 3: 41-9
editor. Contemporary psychology of sport. Chicago (IL): Ath- 37. Brody S, Wagner D, Heinrichs M, et al. Social desirability
letic Institute, 1970: 157-65 scores are associated with higher morning cortisol levels in
17. Singer RN, Harris D, Kroll W, et al. Psychological testing of firefighters. J Psychosom Res 49: 227-8
athletes. J Phys Educ Rec 1977; 48: 30-2 38. Morgan WP, O’Connor PJ, Ellickson KA, et al. Personality
18. Morgan WP. The trait psychology controversy. Res Q Exerc structure, mood states, and performance in elite male distance
Sport 1980; 51: 50-76 runners. Int J Sport Psychol 1988; 19: 247-63

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
Psychological Factors in Sport Performance 889

39. Newcombe PA, Boyle GJ. High school students’ sports person- 59. Raglin JS, Eksten F, Garl FT. Mood state responses to a pre-season
alities: variations across participation level, gender, type of conditioning program in male collegiate basketball players.
sport, and success. Int J Sport Psychol 1995; 26: 277-94 Int J Sport Psychol 1995; 26: 214-25
40. Morgan WP. Psychophysiology of self-awareness during vig- 60. O’Connor PJ, Morgan WP, Raglin JS, et al. Mood state and
orous physical activity. Res Q Exerc Sport 1981; 52: 385-427 salivary cortisol changes following overtraining in female
41. McCrae R, Costa P. Personality, coping and coping effective- swimmers. Psychoneuroendocrinology 1989; 14: 303-10
ness in an adult sample. J Person 1986; 4: 587-601 61. Raglin JS, Morgan WP, O’Connor PJ. Changes in mood states
42. Gomez R, Holmberg K, Bounds J, et al. Neuroticism and extra- during training in female and male college swimmers. Int J
version as predictors of coping styles during early adoles- Sports Med 1991; 12: 585-9
cence. Pers Individ Diff 1999; 27: 3-17 62. Raglin JS, Koceja DM, Stager JM, et al. Mood, neuromuscular
43. Hardman KA. Dual approach to the study of personality and function, and performance during training in female swim-
sport in performance. In: Whiting HTA, Hardman KA, Hendry mers. Med Sci Sport Exerc; 28: 372-7
LB, et al., editors. Personality and performance in physical 63. Raglin JS, Wilson GS. Overtraining in athletes In: Hanin, YL,
education and sport. Lafayette (IN): Balt Publishers, 1973: editor. Emotions in sport. Champaign (IL): Human Kinetics,
77-122 1999: 191-207
44. Cook DB, O’Connor PJ, Eubanks SA, et al. Naturally occurring 64. Berglund B, Säfström H. Psychological monitoring and modu-
muscle pain during exercise: assessment and experimental lation of training load of world-class canoeists. Med Sci Sports
evidence. Med Sci Sports Exerc 1997; 29: 999-1012 Exerc 1994; 26: 1036-40
45. Cattell RB. Manual for the 16-PF. Champaign (IL): Institute for 65. Raglin JS, Morgan WP, Luchsinger AE. Mood state and self-
Personality and Ability Testing, 1972 motivation in successful and unsuccessful women rowers.
46. Eysenck HJ, Eysenck SBG. Manual for the Eysenck personality Med Sci Sports Exerc 1990; 22: 849-53
inventory. San Diego (CA): Educational and Industrial Test- 66. Hirt ER, Levine GM, McDonald HE, et al. The role of mood in
ing Service, 1968 quantitative and qualitative aspects of performance: single or
multiple mechanisms. J Exp Soc Psychol 1997; 33: 602-9
47. Eysenck HJ, Eysenck SBG. Manual for the Eysenck personality
67. Hanin YL. State-trait anxiety research in the USSR. In: Spiel-
questionnaire. San Diego (CA): Educational and Industrial
berger CD, Diaz-Guerrero R, editors. Cross cultural anxiety.
Testing Service, 1975
Vol 3. Washington, DC: Hemisphere, 1986: 45-64
48. Depue RA, Collins PF. Neurobiology of the structure of person-
68. Hanin YL. Emotions and athletic performance: individual zones
ality: dopamine, facilitation of incentive motivation and ex-
of optimal functioning model. In: Seiler R, editor. European
traversion. Behav Brain Sci 1999; 22: 491-569
yearbook on sport psychology. Vol 1. Sank Augustin: FEB-
49. Mahoney MJ. Psychological predictors of elite and non-elite
SAC, 1997: 30-70
performance in olympic weight lifting. Int J Sport Psychol
69. Raglin JS. Anxiety and sport performance. Exerc Sports Sci
1989; 20: 1-12
Rev 1992; 20: 243-74
50. Bartram D. The predictive validity of the EPI and 16PF for
70. Costill DL, Flynn MG, Kirwan JP, et al. Effects of repeated days
military flight training. J Occup Organ Psychol 1995; 68:
of intensified training on muscle glycogen and swimming
219-36 performance. Med Sci Sports Exerc 1988; 20: 249-54
51. Jessup GH. Validity of the Eysenck personality inventory in 71. Morgan WP, O’Connor PJ, Sparling PB, et al. Psychological
pilot selection. Occup Psychol 1971; 21: 158-69 characterization of the elite female distance runner. Int J
52. Fassbender C, Goetters K. Psychological evaluation of Eu- Sports Med 1987; 8 Suppl.: 124-31
ropean astronaut applicants. Aviat Space Environ Med 1994; 72. Prapavessis H, Grove JR. Precompetitive emotions and shoot-
65: 925-9 ing performance: the mental health and zone of optimal func-
53. Rose RM, Fogg LF, Helmreich RL, et al. Psychological predic- tion models. Sport Psychol 1991; 5: 223-34
tors of astronaut effectiveness. Aviat Space Environ Med 73. Prapavessis H. The POMS and sports performance: a review. J
1994; 65: 910-5 Appl Sport Psychol 2000; 12: 34-48
54. Leon GR, McNally C, Ben-Porath YS. Personality charac- 74. Terry PC, Lane AM. Normative values for the profile of mood
teristics, mood, and coping strategies in a successful North state for use with athletic samples. J Appl Sport Psychol 2000;
Pole expedition team. J Res Pers 1989; 23: 162-79 12: 93-109
55. Raglin JS, O’Connor PJ, Carlson N, et al. Responses to under- 75. Walden TA, Smith MA. Emotion regulation. Motiv Emot 1997;
water exercise in scuba divers differing in trait anxiety. Under- 21: 7-24
sea Hyperb Med 1996; 77-82 76. Cerin E, Szabo A, Hunt C, et al. Temporal patterning of com-
56. Johnson RF, Merullo DJ. Psychological mood profiles of army, petitive emotions: a critical view. J Sport Sci 2000; 18: 605-26
marine corps, and special operations forces personnel. Pro- 77. McNair DM, Lorr M, Dropplemann LF. Profile of mood states
ceedings of the Human Factors and Ergonomics Society 41st manual. San Diego (CA): Educational and Testing Service,
Meeting; 1997 Sep 22-26; Albuquerque (NM), 594-8 1992
57. Morgan WP, Brown DL, Raglin JS, et al. Psychological moni- 78. O’Connor PJ, Morgan WP, Raglin JS. Psychobiologic effects
toring of overtraining and staleness. Br J Sports Med 1987; of three days of increased training in female and male swim-
21: 107-14 mers. Med Sci Sports Exerc 1991; 23: 1055-61
58. O’Connor PJ. Overtraining and staleness. In: Morgan WP, edi- 79. Morgan WP, Costill DL, Flynn MG, et al. Mood disturbance
tor. Physical activity and mental health. New York (NY): following increased training in swimmers. Med Sci Sports
Hemisphere, 1997: 145-60 Exerc 1988; 20: 408-14

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)
890 Raglin

80. Durtschi SK, Weiss MR. Psychological characteristics of elite cial support, and coping skills, and adolescent sports injuries.
and nonelite marathon runners. In: Landers DM, editor. Sport J Pers Soc Psychol 1990; 58: 360-70
and elite performers. Champaign (IL): Human Kinetics, 1984: 87. Williams J, Roepke N. Psychology of injury and rehabilitation.
74-80 In: Singer RN, Murphey M, Tennent LK, editors. Handbook
81. Meyer GJ, Finn SE, Eyde LD, et al. Psychological testing and of research on sport psychology. New York (NY): Macmillan,
psychological assessment. Am Psychol 2001; 56: 128-65 1993: 815-39
82. Rosenthal R. Evaluation of procedures and meta-analytic re- 88. Mote Jr CD. The forces in skiing and their implication to injury.
sults. In: Meta-analytic procedures. Applied Social Research Int J Sport Biomech 1987; 3: 309-25
Methods Series. Vol 6. London: Sage, 1991: 127-36 89. Sabbe B, Hulstijn, W, Van Hoof J, et al. Fine motor retardation
83. Haney CJ, Long BC. Coping effectiveness: a path analysis of and depression. J Psychiat Res 1996; 30: 295-306
self-efficacy, control, coping, and performance in sport com- 90. Amirkhan JH, Risinger RT, Swickert RJ. Extraversion: a ‘hid-
petitions. J Appl Soc Psychol 1995; 25: 1726-46 den’ personality factor in coping? J Pers 1995; 63: 189-212
84. Kenttä G, Hassmén P, Raglin J. Training practices and over-
training syndrome in Swedish age-group athletes. Int J Sport
Med 2001; 22: 460-5
85. May JR, Veach TL, Reed MW, et al. A psychological study of
health, injury, and performance in athletes on the US alpine Correspondence and offprints: John S. Raglin, Ph.D., De-
ski team. Phys Sports Med 1985; 13: 111-5 partment of Kinesiology, HPER 112, Bloomington, IN 47405,
86. Smith RE, Smoll FL. Ptacek JT. Conjunctive moderator vari- USA.
ables in vulnerability and resiliency research: life stress, so- E-mail: raglinj@indiana.edu

© Adis International Limited. All rights reserved. Sports Med 2001; 31 (12)