Author: Cameron, James E
Date published: June 1, 2012
Do hockey players possess characteristics that correspond to the position they play? Some anecdotal evidence, such as the common observation that goaltenders are a "breed apart" (Plante, 1972, p. ii) suggests that this is the case. We examined this issue from two complementary perspectives: (a) the correlation of self-reported personality traits with hockey position, and (b) the characteristics perceived to be associated with position. The latter approach views hockey positions-goaltender, defense, and forward-as social categories, and therefore subject to biased perception such as stereotyping.
Hockey Position and Self-Reported Personality
Personality and Position. Between-groups comparisons on personality dimensionsinvolving athletes and non-athletes, more and less successful athletes, and different sportsrepresent an early theme in the sport psychology literature (Van den Auweele, Ny s, Rzewnicki, & Van Mele, 2001), but there are only a handful of findings regarding the correlation between personality and position. Kircaldy's (1982) analysis of university team sports indicated, for example, that offensive players were less emotionally stable, more toughminded, and (at least for male athletes) more extraverted than less defensive players. Other studies investigated personality-position associations in baseball (Greenwood & Simpson, 1994) and football (Cox & Yoo, 1995; Schurr, Ruble, Nisbet, & Wallace, 1984), but various idiosyncrasies-different positions across sports, and different measures across studies-mean that the generalizability of these findings is unclear. Moreover, many sports do not have a goaltender, which is arguably "the most conspicuous of all positions in hockey, if not in all sport" (Lonetto, Marshall, Moote, & Green, 1975; p. 8). It follows from this distinctiveness that goalies, in hockey as well as in soccer (e.g., Glanville, 1972), are the frequent subjects of observation and speculation.
Goalies are Different. According to former Montreal Canathens goaltender Ken Dryden (1983), "goalies are different": "Predictably, a goalie is more introverted than his teammates, more serious. ..more sensitive andmoody ('ghoulies'), more insecure" (p. 118). Such traits are broadly consistent with a theme of social dislocation that emerges from other anecdotal observations, but no systematic empirical evaluations. Former National Hockey League (NHL) player Johnny Gottselig, for example, said that "goalies are probably the loneliest guys in the world" (Fischler, 1994, p. 88), whereas goaltender Cesare Maniago recalled the pre-game feeling that members of his position "weren't just one of the boys.. ..The other guys talk, defensemen with defensemen, forwards with forwards. But the goalie wants to be by himself (Irvin, 1995, p. 122).
Hockey Positions as Social Categories
Although Dryden (1983) maintained that "the differences between 'players' and 'goalies' are manifest and real" (p. 118), there is the additional possibility that such beliefs are subject to various biases, including stereotyping and in-group bias.
Stereotypes of Positions. To the extent that positions provide a meaningful way to structure the perceptual fields of both players and observers-positions are distinguishable in terms of their roles, tasks and, in the case of goalies, their equipment-we can expect that perceptions of those categories are patterned by systematic cognitive and evaluative biases, including stereotyping (i.e., sets of beliefs and attributes that are consensually associated with a specific social category; Gardner, 1994).
One possibility is that stereotypes of hockey positions reflect the role-based structural distinctions between them (Eagly & Steffen, 1984). In the present case, the stereotype content might correspond to the prototypical images conveyed by the "division of labor" between players of forward, defense, and goal. For example, goaltenders occupy a relatively circumscribed space that makes their separateness highly salient. In terms of inferring dispositions, these role-based differences may be amplified by other social cognitive biases, so that the ascription of, say, introversion to goalies can be interpreted as an instance of the fundamental attribution error (Ross, 1977), where the positional feature of being alone on the ice is misjudged as a reflection of a dispositional preference for solitude. The distinctiveness of the goaltending position also means that the "goalies-are-different" stereotype could well be shaped and perpetuated by illusory correlation (Hamilton, 1981), whereby unusual qualities and behaviors are erroneously associated with minority status.
In-G roup Bias. An additional source of bias arises from the fact that one's own position assumes a special significance: it, like other in-groups, contributes to self-definition and selfevaluation. From this follows the notion, articulated in social identity theory (Tajfel & Turner, 1979), that intergroup perceptions have a self-esteem maintenance function: favoring the ingroup helps to sustain a positive self-image. On this basis we hypothesized that hockey players tend to evaluate members of their own position in relatively positive terms.
The purpose of this study is to systematically examine the association between hockey position and both self-reported and ascribed personality traits, using the Big-Five dimensions of personality as a framework for analysis (see John & Srivastava, 1999, for a review). With little scientific basis for predictions regarding self-reported personality differences in this context, we regarded the analyses as largely exploratory. Nevertheless, as outlined above, we had a specific interest in the characteristics of goalies, and we evaluate three specific hypotheses that reflect the anecdotal observations of Dryden (1983) and others: goalies are (on average) more introverted, more neurotic, and more psychologically peripheral team members than other hockey players.
We control for age, which is correlated with Big-Five scores (Helson, Kwan, John, & Jones, 2002; Srivastava, John, Gosling, and Potter, 2003): older people tend to be more agreeable and conscientious, and less extraverted, neurotic, and open to experience, than younger people. A second covariate is level of play, which is an important consideration for two reasons. First, if personality traits do correlate with position, then the defining qualities should be most evident at relatively high levels of play (e.g., Perlini & Halverson, 2006), since success is presumably facilitated by one's "fit" to the demands of a particular position. Second, Morgan (1985) hypothesized that successful athletes evince an "iceberg" profile of dispositional mood, scoring relatively high on positive dimensions of mood, and low on negative ones. Others (e.g., Van den Auweele et al., 2001) question the robustness of this relationship, with good reason: Rowley, Landers, Kyllo, and Etnier's (1995) meta-analysis found that the effect accounted for only 1% of the variance.
In terms of ascribed personality, we were generally interested in the patterns of intergroup perceptions across positions. We reasoned that in-group bias would be evident on the Big-Five dimensions that are most clearly evaluative and thus most relevant to self-esteem: agreeableness, conscientiousness, and neuroticism.
Participants and Procedure
We recruited 578 male participants (163 defensemen, 305 forwards, and 1 10 goaltenders) with the requirement that they be current hockey players. The primary mode of recruitment occurred via the distribution of questionnaires to contacts of the authors, who then distributed them to their teams, or to other individual players they knew. An additional 70 participants (including 54 goalies) completed an online version of the questionnaire.
The sample had a mean age of 35.61 (SD = 12.56), had played hockey for an average of 25 years (2 1 at their current position), and played an average of 2.2 1 times per week. Most lived in Ontario (n = 345) or Nova Scotia ( 1 86), with others from Alberta (34), Saskatchewan (5), New Brunswick (2), and 1 each from Québec, Prince Edward Island, Manitoba, Nunavut, Minnesota, and New York state. Respondents indicated the highest level of organized hockey they had played, which we coded into one of two categories: non-competitive (e.g., minor, recreational, or intramural hockey; n = 244), or competitive (e.g., major, junior, university, semiprofessional, or professional; n = 306).
Participants rated their own personality (I see myself as...) first, followed by their perceptions of hockey players of different positions-goalies (/ see goalies as...), forwards, and defensemen-on separate pages and in a counterbalanced order.
Personality. We adapted Gosling, Rentfrow, and Swann's (2003) Ten-Item Personality Inventory (TIPI), a very brief operationalization of the Big-Five personality domains. The TIPI has two items (one of which is reverse-scored) for each of the five domains, rated on a 7-point Likert-type scale. Gosling et al. demonstrated substantial convergent validity between the TIPI and standard measures of the Big-Five dimensions. In our sample, coefficient alphas for self-reports were somewhat lower than Gosling et al. 's, but in a comparable pattern and range of magnitude (.63, .37, .34, .60, and .38 for extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience respectively).
Self-concept. The questionnaire included two valid, single-item measures of self-concept: (a) Tropp and Wright's (2001) operationalization of group identification, in which the response options are seven increasingly overlapping circles representing the relationship between self and group ("Please circle the one picture... that you feel best represents your own level of identification with your hockey team," with the response scored from 1 to 7); and (b) Robins, Hendin, and Trzesniewski's (200 1) single-item measure of global self-esteem ("I have high self-esteem," 1 = disagree strongly to! = agree strongly).
We conducted 3 (Position: defense/forward/goaltender) × 2 (Highest Level of Play: noncompetitive/competitive) between-participants ANOVAs on the five personality variables, including age as a covariate. Considering first the main effect of position (Table 1), we found no significant differences for extraversion, F(2, 53I)=I .75, p = .18, agreeableness, F(2, 531) = 1.06, p = .35, conscientiousness, F(2, 530) = 0.50, p = .60, or neuroticism, F(2, 531) = 2.54, p = .08. The effect was just beyond the margin of significance for openness to experience, F(2, 530) = 3.00,p = .051, with defensemen tending to be less open than forwards and goaltenders. Neither comparison was significant (p = .08 and p = . 14, with a Bonferroni adjustment).
Level of play was associated with significant effects for four of the five personality dimensions. Participants who reported playing at a competitive level were more extraverted than those whose highest level was non-competitive (M= 4.87, SD = 1 .39; M = 4.57, SD = 1 .54, respectively), F(1 , 530) = 5.22, ? = .02, eta2 = .01; they were also less agreeable (M= 4.63, SD = 1.22;M=4.95,SD = 1.38),F(l,530) = 4.01,p = .046,eta2 = .01,andmoreneurotic(M=2.96,XD = 1.25; M = 2.67, SD = 1.31), F(I, 530) = 4.74, p = .03, eta^sup 2^ = .01. There was a significant interaction between position and highest level of play for conscientiousness, F(2, 529) = 6.09, p = .002, eta2 = .02. Post hoc comparisons indicated that whereas forwards did not differ across level of play (competitive: M = 5.72,S1D = 1.10; non-competitive: M = 5.83, SD = 1.08), t(280) = .76,p = .38, there were trends for defensemento be more conscientious if they had played at a higher level (competitive: M = 5.98, SD = 0.98; non-competitive: M = 5.64, SD = 1.00), t(148) = 2.07, p = .04, whereas goalies displayed the opposite tendency (competitive: M = 5.57, SD = 1 .06; non-competitive: M = 6.24, SD = 0.70), t( 1 04) = 3 .49, p = .00 1 . A Bonferroni adjustment (0= .0167) rendered only the comparison of goaltenders significant.
There were two effects involving age: compared to younger respondents, older players tended to be more agreeable, F(1, 530) = 62.76, p < .001 , eta^sup 2^ = . 1, and more conscientious, F(α , 529) = 32.00, p < .001, eta^sup 2^ = .06.
Identification and Self-Esteem
A3 (Position) × 2 (Highest Level of Play) between-participants ANOVA (controlling for age) indicated a significant relationship between position and team identification, F(2, 532) = 5.81,p = .003, eta^sup 2^ = .02. Pairwise comparisons showed that goalies had, on average, significantly lower levels of identification than defensemen and forwards (Table 1). Self-esteem did not differ by position, F(2, 527) = 1 .8 1 , p = . 17.
We conducted a series of 3 (Participant's Position) × 3 (Target Position) ANOVAs, with repeated measures on the second factor. We focus here on (a) the main effects of target, which indicate whether players are seen as different on the personality dimensions; and (b) the Participant's Position × Target Position interactions, which indicate whether intergroup perceptions are systematically biased by people's own position.
The effect of target was significant for all personality variables: extraversion, F(2, 551) = 1 92.62, p < .001, eta^sup 2^ = .4 1 , agreeableness, F(2, 550) = 77.0 1 ,p < .00 1 , eta^sup 2^= .22, conscientiousness, F(2, 550) = 156.35,p < .001, eta^sup 2^= .36, neuroticism, F(2, 550) = 78.92,p < .001, eta2 = .22, and openness to experience, F(2, 550) = 84.85, p < .001, eta^sup 2^= .24.
Bonferonni-adj usted pairwise comparisons (p < .05) indicated that overall, relative to both defensemen and goalies, forwards were seen as significantly more extraverted, less agreeable, less conscientious, more neurotic, and more open to experience (see Table 2). Goaltenders were judged to be the most conscientious (and more open than defensemen), whereas defensemen were rated as the most emotionally stable.
There were significant interactions between target position and own position for agreeableness F(4, 1102) = 10.31 ,p < .001 , eta^sup 2^ = .04, conscientiousness, F(4, 1 102) = 18.49,p < .001, eta^sup 2^ = .06, neuroticism, F(4, 1 102) = 13. 1 8,p < .001 , eta^sup 2^ = .05, and openness to experience, F(4, 1102) = 8.43, p<. 001, eta^sup 2^= .03, but not extraversion, F(4, 1104)= pA5,p = .21. To interpret these interactions, we conducted 18 planned comparisons for each personality variable: (a) contrasts of ratings by players of each position of their own and the other positions (9 withinparticipants comparisons, 3 for each in-group position); and (b) ratings of each target across the positions (9 between-participants comparisons, 3 for each target position). These two types of comparisons correspond to two ways to assess in-group bias; for example, if goalies tend to favor their own position they might (a) rate goalies as more agreeable than they rate defensemen and forwards, and (b) rate goalies as more agreeable than defensemen and forwards rate goalies. We used a Bonferroni-adjusted a = .002 (.05/1 8) to maintain a familywise Type I errorrate of .05.
Agreeableness. Whereas there was consensus that forwards are less agreeable than other players, the interaction reflects the fact that this discrepancy was least apparent in the judgment of forwards themselves, and most apparent in the eyes of goaltenders. Forwards judged members of their own position to be more agreeable than goaltenders saw them, /(407) = 5.75, p<. 001.
Conscientiousness. A similar but even more striking pattern was evident for conscientiousness (because it typifies the patterns of interest, we depict it in Figure 1). Forwards were generally regarded as less disciplined than players of other positions, but again this was relatively muted in forwards' own ratings. Forwards judged members of their own position to be more conscientious than did defensemen, /(45 1) = 3.5 1 ,p < .001, and goalies, r(407) = 5.01 , p < .00 1 . Similarly, goaltenders were seen as most conscientious by the netminders themselves, compared to the ratings made by defensemen, t(262) = 4.0 1 , p < .00 1 , and forwards, t(402) = 4.43, p < .001. Defensemen, on the other hand, ascribed more discipline to their own category than did goalies, t(264) = 3. 25, p < .001.
Neuroticism. Defensemen were seen as the most emotionally stable players by defensemen and forwards, but not goaltenders. Defensemen saw their own position as singularly stable, whereas goalies singled out forwards as particularly neurotic. Several betweenparticipants contrasts were also consistent with in-group bias: forwards rated members of their own position as less neurotic than defensemen (t(450) = -3 .43 , p < .00 1 ) and goaltenders (t(407) = -4.25, p < .001) saw them, defensemen judged themselves to be less neurotic than forwards did, r(451) = -3 .48, p < .00 1 , and goalies rated their own group as less neurotic than forwards judged them to be, t(402) = -3.25, p < .001.
Openness to experience. Only goaltenders did not view forwards as distinctively open to experience, judging them as no different from their in-group on this dimension. One between-participants contrast was significant: goalies rated their in-group as more open than defensemen rated them, t(262) = 4.65, p < .001 .
The game of hockey is a social world unto itself, with its own structure, traditions, myths, and characters. Although in the popular imagination certain traits tend to cluster around positions-the goalie prominent among them-these images might be perpetuated by biased social cognition. Our data suggest that personality differences between goalies, defensemen, and forwards are mostly in the eye of the beholder, and that such perceptions can be underStood from an intergroup perspective.
We found little evidence that self-reported personality traits correlated in any straightforward way with hockey players' position. The one clear difference to emerge, in terms of selfrated characteristics, was that goalies tended to report lower levels of team identification than other players. This does not appear to have any implications for self-evaluation, but it is consistent with anecdotal observations about the separateness of goalies in the team context.
Correlations between age and personality were consistent with previous research (Kelson et al., 2002; Srivastava et al., 2003), with older people describing themselves as more conscientious and agreeable than younger respondents. We also found some effects associated with level of play: respondents who had played competitively tended to be more extraverted (consistent with a positive orientation toward spending time with other team members), less agreeable (consistent, perhaps, with a competitive orientation), and more neurotic (contrary to the "iceberg" profile; Morgan, 1985). Consistent with Rowley et al. 's (1995) meta-analysis, these effects were very small (much smaller than those of age, for example); nevertheless, they may be of interest to sport psychologists interested in the dispositional factors that predict athletic success and commitment. Although the characteristics of the sample limited our ability to evaluate interactions involving position and level of competition, indications that personality types might be more evident for more skilled players (e.g., the conscientious defender) warrant further study of elite-level players.
Limitations: The Operationalization of Personality
Our operationalization of the Big Five did not allow us to distinguish between various facets of the five domains (e.g., Costa & McCrae, 1992), which would have provided a more fine-grained basis for comparing the positions. Moreover, as some critiques of the Big Five model have suggested, the dimensions are abstracted from context, and they do not explicitly invoke motivational concerns (e.g., McAdams, 1995). To use the example of the goaltender, a contextualized (game-specific) measure of neuroticism would be more sensitive to the special demands of the position. After all, even the most dispositionally anxious goalies must, in order to excel, cope effectively with their fear prior to, and during the game. For instance, NHL goalie Glenn Hall interpreted his legendary anxiety in game-specific and adaptive terms:
They talk about me throwing up before a game as a weakness. I've always looked at it as one of the great strengths I had. I didn't take things home with me after a game too much.. ..I was really quite relaxed (Irvin, 1995, pp. 55-56).
The results should also be interpreted in light of the psychometric limitations of the TIPI. The small number of items entails internal consistency estimates that tend to fall below normal expectations; for this reason Gosling et al. (2003) noted that alternative estimates, such as testretest reliability coefficients, are more appropriate for these scales. Whereas this limitation is offset by content validity and other evidence of construct validity, Gosling et al. also found that the TIPI tended to correlate less strongly with other variables than standard Big-Five measures. Thus, in the present context the TIPI may well be less sensitive to between-group differences than multi-item operationalizations of the personality domains. Finally, while Gosling et al. recommended that the TIPI be used when "personality is not the primary topic of interest" (p. 504), in the spirit of exploratory research we found the TIPI to provide a useful first step toward an understanding of personality traits in a hockey context. Nevertheless, in follow-up work we would encourage the use of instruments that allow more precise measurement, and we note the availability of other relatively brief operationalizations of the Big Five that might achieve both efficiency and reliability.
Perceiving Positions: Stereotypes and In-Group Bias
Despite the absence of main effects of position on self-reported personality, there was some convergence of opinion about what different players are like: compared to defensemen and goaltenders, forwards were seen as relatively outgoing, quarrelsome, undisciplined, and creative, whereas defensemen tended to be ascribed more emotional stability. As we anticipated, however, these patterns were qualified-for all dimensions but extraversion-by participants' own position, which in every case worked in the direction of enhancing in-group favorability. The results indicated a number of direct and indirect routes to achieving a positive in-group image, within the constraints of the intergroup context and the dimensions at hand (Tajfel & Turner, 1979): (a) each position had one dimension on which they viewed themselves as positively distinctive relative to both other positions (defensemen on emotional stability, forwards on openness to experience, and goaltenders on conscientiousness); (b) where the ingroup stereotype had negative implications, it was minimized; in particular, forwards expressed a muted version of the consensual opinion that they are relatively disagreeable and careless; (c) some intergroup ratings had the effect of denying positive distinctiveness to out-groups (e.g., goalies did not grant defensemen any special status with respect to emotional stability); and (d) bias was pervasive enough that between-group contrasts, particularly on conscientiousness and neuroticism, were also significant; thus, for example, goalies and forwards each attributed more conscientiousness and less neuroticism to their in-group than the other group ascribed to them.
Overall, then, these data provide strong evidence that hockey position represents a psychologically meaningful in-group for players. Two other patterns emerged: (a) forwards generally stood out as most distinctive, whereas defensemen and goalies were seen as more similar; and (b) goalies had the most biased intergroup ratings, which specifically targeted forwards. We suggest that underlying these effects is the fact that forwards are "offensive." In other words, goalies and defensemen are "on the same team" because they comprise a rolebased sub-category-defenders-whereas forwards' primary task is to score. Thus, forwards are "against" both kinds of defenders, and because scoring comes most visibly at the cost of goalies' failure, it is not surprising that those in the last line of defense are least inclined to view forwards favorably.
Given the absence of any compelling "kernel of truth" in participants' self-reported personality, these observations lead us to interpret these patterns in terms of the perspective of the different positions rather than their personalities. In addition to the tendency toward ingroup bias, the perceptions of positions are consistent with the role-based structural distinctions between them (Eagly & Steffen, 1984): scoring is aided by activity, aggressiveness, and creativity, in contrast to the stolidity and discipline of the defender. If social structure underlies stereotype content, then role-based qualities can easily be misattributed to individuals' dispositions. It is worth emphasizing, however, that because our data are limited to the judgments of players by other players, we cannot directly address whether the same patterns characterize the perceptions of hockey fans, or those of observers in other roles associated with the game (e.g., coaches, referees).
The image of the "crazy" goalie might be compelling for the same reason that it is generally untrue: it is a stereotype sustained by common perceptual biases. However, our data suggest that goalies are different from other hockey players in two ways: (a) they identify less with the team; and (b) they have sharply biased intergroup perceptions, which presumably reflect the perpetual threat of being scored on. Both, however, like the in-group-favoring tendency that all players shared, appear to be products of category-based perspective rather than personality.
Costa, P.T., Jr., & McCrae, R.R. (1992). Revised NEO Personality Inventory (NEO-Pl-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources.
Cox, R. H., & Yoo, H. S. (1995). Playing position and psychological skill in American football. Journal of Sport Behavior, 18, 183-194.
Dryden, K. (1983). TVz^ game. Toronto: Macmillan.
Eagly, A.M., & Steffen, V.J. (1984). Gender stereotypes stem from the distribution of women and men into social roles. Journal of Personality and Social Psychology, 46, 735-754.
Fischler, S. (1994). Goalies: Legends from the NHL' s toughest job. Toronto: McGraw-Hill Ryerson.
Gardner, R.C. (1994). Stereotypes as consensual beliefs. In M.P. Zanna & J.M. Olson (Eds.), The psychology of prejudice: The Ontario symposium (Vol. 7, pp. 1-31). Hillsdale, NJ:Erlbaum. ,
Glanville, B. (1972). Goalkeepers are different. New York: Crown Publishers.
Gosling, S.D., Rentfrow, PJ., & Swann, W.B., Jr. (2003). A very brief measure of the BigFive personality domains. Journal of Research in Personality, 37, 504-528.
Greenwood, M., & Simpson, W. K. (1994). Personality traits of intercollegiate baseball athletes playing central versus noncentral defensive positions at three levels of competition. Applied research in coaching and athletics annual (Boston), 1994, 1530.
Hamilton, D.L. (1981). Illusory correlation as a basis for stereotyping. In D.L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup behavior (pp. 115-144). Hillsdale, NJ: Erlbaum.
Kelson, R., Kwan, V.S.Y., John, O.P., & Jones, C. (2002). The growing evidence for personality change in adulthood: Findings from research with personality inventories. Journal of Research in Personality, 36, 287-306.
Irvin, D. (1995). In the crease: Goaltenders look at life in the NHL. Toronto: McClelland & Stewart.
John, O.P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In L.A. Pervin & O.P. John (Eds.), Handbook of personality: Theory and research (pp. 102-138). New York: Guilford.
Kirkcaldy, B. D. (1982). Personality and sex differences related to positions in team sports. International Journal of Sport Psychology, 13, 141-153.
Lonetto, R., Marshall, J., Moote, D., & Green, G. (1975). The goaltender and goaltending: The Canadian Hockey Library Volume 1. Guelph and Toronto, Ontario, Canada: The Research Institute for the Study of Sport and LR & Associates Ltd.
McAdams, O.P. (1995). What do we know when we know a person? Journal of Personality, 63, 365-396.
Morgan, W.P. (1985). Selected psychological factors limiting performance: A mental health model. In D.H. Clarke & H.M. Eckert (Eds.), Limits of human performance (pp. 70-80). Champaign, EL: Human Kinetics.
Perlini, A.H., & Halverson, T.R. (2006). Emotional intelligence in the National Hockey League. Canadian Journal of Behavioral Science, 38, 109-119.
Plante, J. (1972). Goaltending. New York: Collier Books.
Robins, R. W., Hendin, H.M., and Trzesniewski, K.H. (2001). Measuring global self-esteem: Construct validation of a single-item measure and the Rosenberg Self-Esteem scale. Personality and Social Psychology Bulletin, 27, 151-161.
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 10, pp. 173-220). New York: Academic Press.
Rowley, A. J., Landers, D. M., Kyllo, L. B., & Etnier, J. L. (1995). Does the iceberg profile discriminate between successful and less successful athletes? A meta-analysis. Journal of Sport and Exercise Psychology, 17, 185-199.
Srivastava, S., John, O.P., Gosling, S.D., & Potter, J. (2003). Development of personality in adulthood: Set like plaster or persistent change? Journal of Personality and Social Psychology, 84, 1041-1053.
Schurr, K. T., Ruble, V. E., Nisbet, J., & Wallace, D. (1984). Myers-Briggs Type Inventory characteristics of more and less successful players on an American football team. Journal of Sport Behavior, 7, 47-57.
Tajfel, H., & Turner, J.C. (1979). An integrative theory of intergroup conflict. In W.G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 3347). Monterey, CA: Brooks-Cole.
Tropp, L. R., & Wright, S.C. (2001). In-group identification as the inclusion of in-group in the self. Personality and Social Psychology Bulletin, 27, 585-600.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5, 207-232.
Van den Auweele, Y., Nys, K., Rzewnicki, R., & Van Mele, V. (2001). Personality and the athlete. In R.N. Singer, H.A. Hausenblas, & C.M. Janelle (Eds.), Handbook of Sport Psychology (2nd ed., pp. 239-268). New York: Wiley.
James E. Cameron
Saint Mary's University
James M. Cameron
Michigan State University
Richard N. Lalonde
Address correspondence to: Jim Cameron, Department of Psychology, Saint Mary's University, Halifax, Nova Scotia, Canada B3H 3C3. E-mail: email@example.com