Personality characteristics of competitive and recreational cyclists.
Certain personality characteristics have been consistently found in athletes, such as introversion (Hagberg, Mullin, Bahrke, & Limburg, 1979), lower levels of cooperation (Harder, 1992) and narcissistic personality characteristics (Carroll, 1989). Among elite cyclists in particular, high self-confidence has been associated with strong performance (McCann, Murphy, & Raedeke, 1992). Athletes have also been found to posses the "iceberg" personality profile, as measured by the Profile of Mood States (POMS; Morgan, O'Connor, Ellickson, & Bradley, 1988). Athletes with the "iceberg" profile on the POMS scored lower than the population average on tension, depression, anger, fatigue, and confusion, and above the population average on vigor. This profile has been found among successful wrestlers (Silva, Shultz, Haslam, Martin, & Murray, 1985), crew team members (Morgan & Johnson, 1978) and distance runners (Morgan, O'Connor, Ellickson, & Bradley, 1988; Morgan & Pollock, 1977).
Exploring the psychological characteristics of athletes in sports psychology may help identify attitudes and personality traits which are helpful to performance. Information from psychological profiles may have implications for coaches and trainers in working with athletes on sport-specific psychological techniques designed to enhance performance.
Although research on the psychological characteristics or profiles of successful athletes may have practical relevance, its application has been limited due to methodological weaknesses in prior research (Davis & Mogk, 1994). Some studies fail to differentiate within their samples among athletes of different sports (e.g., Mahoney, Gabriel, & Perkins, 1987). Others use a plethora of diverse and often times poorly standardized measures (e.g., Mahoney, et al., 1987; Aamodt, Alexander, & Kimbrough, 1982). Still others employ statistical tests inappropriately (Hagberg, Mullin, Bahrke, & Limburg, 1979; Mahoney, et al., 1987). For instance, many studies failed to use appropriate statistical methods in determining the presence of the iceberg profile on the POMS, reducing it to only a descriptive measure. This calls into question the subjectivity associated with deciding what constitutes "above and below average" on each of the six mood states measured on this instrument. Even the few studies that have analyzed the POMS with statistical methods are suspect. For example, Fuchs and Zaichkowsky (1983) used multiple t tests to assess the differences in POMS profiles between bodybuilders and nonathletes, greatly enhancing the chance of Type I errors.
The present study rectifies some of the methodological weaknesses found in earlier studies by (a) studying athletes engaged in one sport, (b) using standardized and well-normed measures, and (c) using appropriate statistical tests. In this study, personality characteristics of competitive caliber (elite) cyclists, recreational cyclists, and nonathletes were investigated using the Personality Adjective Checklist (PACL; Strack, 1987), the Coolidge Axis Two Inventory (CATI; Coolidge & Merwin, 1992), and the Profile of Mood States (POMS; McNair, Loft, & Droppleman, 1971). In this study we investigated personality characteristics that have been identified in previous research as important among athletes, but that have not been examined among the specific population of elite and recreational cyclists.
The hypotheses of this study were the following: (a) competitive cyclists would be significantly more confident (as measured by the Confident scale of the PACL and Narcissistic scale of the CATI) and introverted (as measured by the lntroversive scale of the PACL) in comparison to nonathletes; (b) competitive cyclists would reflect lower levels of cooperation (as measured by the Cooperative scale on the PACL) than recreational cyclists and nonathletes, and finally, (c) both competitive and recreational cyclists, but not non-athletes, will reflect the classic "iceberg" profile on the POMS that is generally found in athletes.
Participants consisted of three separate groups of 17 participants (12 males, 5 females per group who were matched for age, education, and income; N = 51). The mean age of the competitive cyclists was 26.9 years (range = 18-46 years) and their mean education totaled 15.5 years (range = 12-19 years). Recreational cyclists averaged 27.0 years of age (range = 19-45 years) and they achieved an average education of 14.5 years (range = 12-19 years). Finally, the nonathletes mean age was 24.3 years (range = 19-43 years) and their mean education was 14.5 years (range = 12-19 years). Preliminary one-way analyses of variance revealed no significant differences between groups on either age, F(2,50) = .763, p = .47, or education level, F(2,50) = 1.31, p = .28. Participant categorical responses to level of income were subjected to a chi-square test of significant differences. No significant differences were found based on self-reported income between the three groups, [[Chi].sup.2] (2, N = 51) = 3.606, p = .16.
The competitive cyclists had been competing in Category I or II level as defined by the United States Cycling Federation for an average of 6.2 years, with an average of 47.6 races per year (range = 24-80 races per year).
Recreational cyclists were defined as individuals who cycle on their own, with friends, or in organized bicycle rides, but who had not competed and were not participating in organized cycling competitions at the time of this study. Recreational cyclists consisted of members of a local cycling club or were university undergraduate students who met this definition of a recreational cyclist.
Nonathletes were defined as individuals who were not participating in any competitive sports activities and not involved in organized cycling at the time of the study. These participants may have exercised at a health club or participated in a recreational activity, such as jogging or swimming, or may have played occasional sports with friends, but for no more than four times per week in any given activity. The nonathlete group also consisted of volunteer undergraduate students.
The psychological tests administered included measures based on Axis 11 personality definitions of the DSM classification system (Diagnostic and Statistical Manual, 3rd Edition-Revised, American Psychiatric Association, 1987). These measures were included to assess the personality constructs of interest, for cross study comparisons, and to replicate previous personality research on athletes. It is important to use measures that look at global personality characteristics based on common criteria (i.e., DSM-III-R) to allow for comparisons and the formulation of global theories in this field. Although the DSM-III-R defines abnormal behavior, it can be used as a basis for broad categories of differing personality styles and to compare studies investigating personality among athletes.
The Personality Adjective Check List (PACL; Strack, 1987) is comprised of a 105-item adjective checklist that measures personality traits reflective of DSM-III-R classifications. The PACL is a measure utilized with populations scoring in the normal range but is broadly based upon the DSM-III-R Axis II personality constructs. Therefore, each measure of the PACL was constructed based upon an Axis II personality category which are listed in parentheses after each of the PACL personality categories below. The PACL has been used in other research to help clarify how personality disorders are related to dimensions of normal personality traits (Pincus & Wiggins, 1990; Wiggins & Pincus, 1989) and in other personality research studies (Brokken & Smid, 1984; Hofstee, Brokken, & Land, 1981; Lewicka, 1983; Sanderman, Koeter, Ormel, & Raats, 1988).
The PACL measures the following: introversion (akin to DSM-III-R schizoid), inhibition (avoidant), cooperativeness (dependent), sociability (histrionic), confidence (narcissistic), forcefulness (antisocial and aggressive), respect (compulsive), and sensitivity (passive-aggressive and self-defeating). The PACL includes a problem indicator (PI) to assess mal-adaptive or problematic levels of the traits measured. The PACL has demonstrated acceptable test-retest reliability (r = .65; six months) and strong internal consistency (Cronbach's alpha = .82) (Strack, 1987).
The Coolidge Axis H Inventory (CATI; Coolidge & Merwin, 1992) is an inventory based on 13 personality disorders in the DSM-III-R and is used for the assessment of personality disorders. The CATI is a 200 item, self-report, four-point Likert scale questionnaire. The CATI was designed to represent all 117 criteria for personality disorders in the DSM-III-R and allows for a greater variability in responding in comparison with many of the previous measures used in assessing personality characteristics and athletes. The 13 personality disorder scales of the CATI have good test-retest reliability (r = .90; one week interval) and internal consistency (Cronbach's alpha = .76) in a sample of 891 participants (Coolidge & Merwin, 1992).
The Profile of Mood States (POMS; McNair, Lorr, & Droppleman, 1971) measures six mood states: tension, depression, anger, vigor, fatigue, confusion, and overall mood. The POMS consists of a list of 65 words or phrases that describe moods or feelings. The subject is asked how each word fits them according to how they have been feeling during the past week, including the day of testing. For each word they choose one of five choices ranging from not at all, to extremely. The POMS has strong test-retest reliability (r = .65 - .74; 20 days). It has been reported as a good measure of the personality characteristics that are highly predictive of success in athletes of several different sports (Morgan, 1980).
The three groups of participants were recruited in the following ways. A pool of elite cyclists were randomly selected from the United States Cycling Federation directory (N = 41). From this original group 17 indicated willingness to participate in the study. Demographic information was gathered on this group of elite cyclists, including age, education level, and income. In order to find matched comparison groups of recreational cyclists and nonathletes, the following procedures were used. Recreational cyclists were contacted through a local cycling club and invited to participate in the study if they met certain criteria; specifically, if they fell within the same age range and education level as the elite cyclist group and consented to participate. Additionally, since there were several elite cyclists in the younger age brackets, students were recruited from a university undergraduate psychology class and were asked to participate in the study if they met the definition of a recreational cyclist. Nonathletes were invited to participate from large undergraduate psychology courses if they met the target age range and education level of elite cyclists.
Participants were given a questionnaire packet containing specific written instructions, an informed consent form, the personality measures, and surveys to collect background and demographic information. Participants were asked to fill out the questionnaires at their convenience, but in one sitting. A debriefing statement was also provided to all participants.
For preliminary analyses, outliers, and the assumptions of ANOVA including normality and equal variances were checked using various SPSSx procedures for all three groups on each scale of the CATI, PACL, and POMS. There was one outlier (greater than three standard deviations from the group mean) in the nonathlete group on the Passive Aggressive scale of the CATI and one on the Confident scale of the PACL. Analyses including the outliers resulted in significant differences among groups on these scales; subsequent reanalysis after removal of both outliers eliminated significant differences on the Passive Aggressive scale, but results related to the Confident scale remained significant. The latter analyses are reported here.
One-way analysis of variance was used for each dependent variable. Demographic variables of age, education, and income were not taken into account when group differences were analyzed since there were no significant differences between groups on these variables. Fisher's Least Significant Difference (LSD) procedure was used for making post hoc planned comparisons. Univariate ANOVA results with group means for the scales of the CATI and PACL used in this study appear in Table 1.
Multivariate analysis of variance was not used for two reasons. First, Russell (1990) argues that MANOVA does not guard against inflation of the overall error rate. Second, differences between groups were hypothesized at the univariate level for only a few of the several dependent measures. Therefore, true differences may not have been reflected and may have been obscured in a multivariate analysis. Results are reported under the headings of the personality variable that was explored in analysis.
As hypothesized, a significant difference between the three groups on the Confident scale of the PACL emerged, F(2,49) = 3.81, p = .03. Fisher's LSD test indicated that the elite cyclists (M = 54.76) scored significantly higher than the nonathlete group (M = 47.47), suggesting higher levels of narcissistic personality traits within this group. No statistically significant differences emerged between groups, however, on the Narcissism scale of the CATI, F(2,50) = 1.40, p = .26. A power analysis of the Narcissism scale with this sample size indicated only a 20% chance of rejecting the null hypothesis given the current sample size [TABULAR DATA FOR TABLE 1 OMITTED] (Cohen, 1988). This power analysis was performed to assess the power of the current data set to find significant differences.
The hypothesis that competitive cyclists would score significantly higher on the Introversive scale of the PACL was not corroborated, F(2,50) = .272, p = .76. Calculations of power indicate less than 17% chance of rejecting the null hypothesis with the current sample size (Cohen, 1988).
The hypothesis that competitive cyclists would be less cooperative than the other two groups was confirmed. An ANOVA on the Cooperative scale of the PACL revealed a significant difference between groups, F(2,50) = 4.51, p = .02. Fisher's LSD test indicated that the elite cyclists (M = 41.59) scored significantly lower than both the nonathletes (M = 53.82) and the recreational cyclists (M = 52.65) on this scale of the PACL.
"Iceberg" Profile on the Profile of Mood States
Profile analysis (Tabachnick & Fidell, 1989) was the initial analysis used for examining the results on the POMS. Profile analysis distinguishes between group profiles and explores how they differ on a set of measures that have the same range of possible scores. Profile analysis involves a "test for parallelism," which was used to examine whether the profiles for the three groups exhibited the same pattern of highs and lows on the mood states measured by the POMS. The profile analysis test for parallelism was significant, F(10,240) = 2.40, p = .014. Therefore, the slopes between each measure of the POMS were significantly different between groups, indicating that the profiles were not parallel. More specifically, the elite cyclists and recreational cyclists profiles were significantly non-parallel, F(5,160) = 2.30, p = .047.
After utilizing the test for parallelism on the POMS, the presence of the "iceberg" profile was explored. Visually, both the elite cyclists and recreational cyclists reveal the classic "iceberg" profile (see Table 2 for scale means). Profile analysis results indicated that both the competitive cyclists profiles, F(5,80) = 14.03, p = .000, and the recreational cyclists profiles, F(5,80) = 6.49, p = .000, were not flat. Since all pairwise comparisons were not of interest, the Vigor scale was contrasted to the remaining five variables on the POMS for each group using t tests. A Bonferroni adjustment was included in these analyses to control for possible family-wise error. In the competitive cyclist group, Vigor (M = 56.94) was significantly greater than Tension (M = 44.47), Depression (M = 45.41), Fatigue (M = 46.24), and Confusion (M = 39.18), but not significantly different from Anger (M = 51.00). In the recreational cyclist group, Vigor (M = 55.41) was significantly greater than Tension (M = 44.29), Depression (M = 45.71), Anger (M = 47.41), and Confusion (M = 45.24), but not significantly different from Fatigue (M = 49.65). Therefore, both the elite cyclists and recreational cyclists revealed the classic "iceberg" profile on the POMS (Fuchs & Zaichkowsky, 1983; Morgan, 1980; see Table 2).
The nonathlete profile on the POMS was flat, F(5,80) = 1.70, p = .145, indicating no significant difference in the scale means for this group. Therefore, consistent with hypothesis 3, the nonathlete group did not conform to the "iceberg" profile that was found in both the elite and recreational cyclists groups.
The hypothesis that competitive cyclists would score significantly higher on the Confident scale of the PACL was confirmed, suggesting greater narcissistic traits among this group. The hypothesis that elite and recreational cyclists reflected the "iceberg" personality profile was also confirmed since Vigor was significantly higher than the other variables measured by the Profile of Mood States. This means that this sample of cyclists shows greater energy and less negative traits than the nonathletes sampled. This profile was determined through objective statistical means rather than subjective criteria used in previous studies.
Table 2 Summary Data (M and SD) for the Profile of Mood States Group Elite Recreational Nonathletes Cyclists Cyclists (n = 17) (n = 17) (n = 17) M SD M SD M SD Tension 44.47 6.98 44.29 5.28 49.29 10.83 Depression 45.41 6.94 45.71 7.94 46.82 9.92 Anger 51.00 9.70 47.41 6.96 51.12 10.52 Vigor 56.94 7.62 55.41 7.88 50.06 9.65 Fatigue 46.24 8.47 49.65 6.40 51.47 10.31 Confusion 39.18 5.04 45.24 8.30 44.65 10.92
The hypothesis that the elite cyclists would be significantly more Narcissistic on the CATI than the other groups, however, was not supported. While this may seem inconsistent with the earlier finding on the PACL, this finding may in fact simply be due to the lack of sensitivity of the CATI. That is, the CATI is more sensitive to pathological extremes of narcissism but less sensitive to its more moderate levels. Thus, it may not have registered differences with the current sample.
Attempts to replicate previous findings that elite cyclists are more introverted (Hagberg, Mullin, Bahrke, & Limburg, 1979) was not confirmed. This may be due to differences in the study participants. The Hagberg, et al. (1979) study which found a significant introversive style among elite cyclists involved a small team of highly competitive and successful cyclists. These individuals were members of the Pan-American and Olympic cycling teams, included two American road champions, and included many cyclists who were eventually selected for the World Team. These cyclists were full time competitors at a highly competitive point in their careers. In contrast, many of the elite cyclists in the present study held full time jobs or were students. Although all of the elite cyclists in this study were qualified to race on a national and/or international level, many were not doing so at the time of this study. Races for the elite cyclists in the current study consisted mostly of local competitions. Many of these cyclists knew each another. In fact, part of the motivation for these elite cyclists to compete in the local events may have been the rewards of the social atmosphere and contact that the races offered. Therefore, introversion may be a characteristic found more frequently in highly competitive world-class cyclists and less so in elite cyclists of lower caliber.
The significant finding that competitive cyclists were less cooperative than recreational cyclists and nonathletes is consistent with previous studies on competitive athletes and with the findings of cooperation-competition studies in applied social psychology in general. Cycling is a solitary sport, and even though each cyclist needs to support the team, most cyclists are motivated by their own individual achievements. Cooperativeness may be perceived to be detrimental to this motivational factor in cycling. Further, Eysenck, Nias, and Cox (1982) posit that traits such as aggressiveness and impersonal attitudes (characteristics of high psychoticism) contribute to success in many sports.
Any implications for coaching must be considered in light of the relatively small and focused sample of the current study. Nevertheless, there are certain coaching implications that might be of interest. For example, it is commonly believed that when the gun goes off at the starting line the cyclists with the strongest mental attitude will most likely come out ahead (Jacobs, 1986). Understanding personality characteristics of athletes could potentially be useful to coaches as they prepare a mental "game plan" and work with athletes on a daily basis. With a complete understanding of an elite cyclist's specific personality dynamics, the coach will have more information about how to help motivate, encourage, and psychologically prepare an athlete for successful competition.
The results of the elite cyclists in this study tend to have a narcissistic personality style. Since narcissistic personality styles possess a specific sensitivity toward criticism, coaches may need to be sensitive to the way they point out problems and to the manner in which they work with cyclists to correct such problems. Coaches can learn to "couch" criticisms in a way that will not increase defensiveness or anger in athletes. This sensitivity is not only important for building rapport but also for helping athletes positively progress in their training. Further replication utilizing random sampling procedures will allow for more global generalizations.
The fact that elite cyclists were shown to be less cooperative may also have important coaching implications. For example, it may be best to allow elite cyclists to train on an individual basis because problems may occur if these individual are required to make group decisions about training. Lower cooperativeness seems to be correlated with higher self-confidence. An athlete with higher self-confidence and lower cooperativeness will probably not respond well to a domineering coach. Therefore, coaches may want to allow individual cyclists to devise their own training program and then act as consultants in pointing out potential problems in the plan and in helping athletes make modifications to the original plan.
Continuing research on the personality characteristics of athletes is important in the development of improved mental and motivational components of training. The present study should be replicated, but with a more discrete group of elite cyclists (i.e., cyclists competing at higher levels). This would help determine if the personality characteristics found here also hold for more successful cyclists. This is especially important because the increased focus and popularity on fitness and health in society today has spurred more people to become active in exercise and organized sports. It is difficult to find individuals who do not participate in some type of sport or "friendly" competition, and although substantial differences exist, the gap between recreational athletes, the "weekend competitor," and highly competitive athletes is becoming increasingly more narrow.
Future research could also explore more specifically how empirical findings in personality traits of athletes can help to inform and improve coaching strategies as well as contribute to increased athletic performance. The present study represents one attempt to explore the personality characteristics of competitive and recreational cyclists, but greater research in this area is warranted. Well designed studies that use appropriate statistical methods and that help to correlate personality characteristics specific to competitive athletes with athletic performance would be helpful for advancing this area of research in sports psychology.
Aamodt, M. G., Alexander, C. J., & Kimbrough, W. W. (1982). Personality characteristics of college and nonathletes and baseball, football, and track team members. Perceptual and Motor Skills, 55, 327-330.
American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders, 3rd ed.-revised. Washington, DC: APA.
Brokken, F. B, & Smid, N. G. (1984). Measurement research: An extension of the standard personality adjective checklist. Nederlands Tijdschrift voor de Psychologie en haar Grensgebieden, 39, 348-352.
Carroll, L. (1989). A comparative study of narcissism, gender, and sex-role orientation among bodybuilders, athletes, and psychology students. Psychological Reports, 64, 999-1006.
Cratty, B. J. (1989). Psychology in contemporary sport (3rd. ed.). Englewood Cliffs, NJ: Prentice Hall.
Cohen, J. (1988). Statistical power analysis.for the behavioral sciences (2nd. ed.). Hillsdale, NJ: Lawrence Erlbaum.
Coolidge, F. L., & Merwin, M. M. (1992). Reliability and validity of the Coolidge Axis Two Inventory: A new inventory for the assessment of personality disorders. Journal of Personality Assessment, 59, 223-238.
Davis, C., & Mogk, J. P. (1994). Some personality correlates of interest and excellence in sport. International Journal of Sport Psychology, 25, 131 - 143.
Eysenck, H. J., Nias, D. K. B., & Cox, D. N. (1982). Sport and personality. Advances in Behavior Research and Therapy, 4, 1-56.
Fuchs, C. Z., & Zaichkowsky, L. D. (1983). Psychological characteristics of male and female bodybuilders: The iceberg profile. Journal of Sport Behavior, 6, 136-145.
Hagberg, J. M., Mullin, J. P., Bahrke, M., & Limburg, J. (1979). Physiological profiles and selected psychological characteristics of national class American cyclists. Journal of Sports Medicine, 19, 341-346.
Harder, J. W. (1992). Play for pay: Effects of inequity in a pay-for-performance contexts. Administrative Science Quarterly, 37, 321-335.
Hofstee, W. K., Brokken, F. B., & Land, H. (1981). Construction of a standard personality adjective checklist. Nederlands Tijdschrift voor de Psychologie en haar Grensgebieden, 36, 443-452.
Jacobs, A. (1986). Sport psychology and cycling. in E.R. Burke (Ed.), Science of Cycling (pp. 203 - 212). Champaign, IL: Human Kinetics Books.
Lewicka, M. (1983). Personality-trait adjective check-list. Przeglad Psychologiczny, 26, 703-713.
Mahoney, M. J., Gabriel, T. J, & Perkins, T. S. (1987). Psychological Skills and Exceptional Athletic Performance. The Sport Psychologist, 1, 181-199.
McCann, S.C., Murphy, S. M., & Raedeke, T. D. (1992). The effect of performance setting and the individual differences on the anxiety-performance relationship for elite cyclists. Anxiety, Stress and Coping: An International Journal, 5, 177-187.
McNair, D. M., Lorr, M, & Droppleman, L. F. (1971). Profile of mood states manual. San Diego: Educational and Industrial Testing Service.
Morgan, W. P. (1980, July). Test of champions. Psychology Today, pp. 92-108.
Morgan. W. P., & Johnson, R. W. (1978). Personality characteristics of successful and unsuccessful oarsmen. International Journal of Sport Psychology, 9, 119-133.
Morgan, W. P., O'Connor, P. J., Ellickson, K. A., & Bradley, P. W. (1988). Personality structure, mood states, and performance in elite male distance runners. International Journal of Sport Psychology, 19, 247-263.
Morgan, W. P., & Pollock, M. L. (1977). Psychological characterization of the elite distance runner. Annals of the New York Academy of Sciences, 301, 382-403.
Pincus, A. L., & Wiggins, J. S. (1990). Interpersonal problems and conceptions of personality disorders. Journal of Personality Disorders, 4, 342-352.
Russell, D. W. (1990). The analysis of psychophysiological data: Multivariate approaches. In J. T. Cacioppo & L. G. Tassinary (Eds.), Principles of psychophysiology (pp. 775-801). New York: Cambridge University Press.
Sanderman, R., Koeter, M, Ormel, H., & Raats, G. (1988). Generalizability of the short standard personality adjective checklist. Nederlands Tijdschrift voor de Psychologie en haar Grensgebieden, 43, 82-85.
Silva, J. M., Shultz, B. B., Haslam, R. W., Martin, T. P., & Murray, D. F. (1985). Discriminating characteristics of contestants at the United States Olympic Wrestling Trials. International Journal of Sport Psychology, 16, 79-102.
Strack, S. (1987). Development and validation of an Adjective Check List to assess the Millon Personality Types in a normal population. Journal of Personality Assessment, 51, 572-587.
Tabachnick B. G., & Fidell, L. S. (1989). Using multivariate statistics (2nd ed.). New York: Harper & Row.
Wiggins, J. S., & Pincus, A. L. (1989). Conceptions of personality disorders and dimensions of personality. Psychological Assessment, 1, 305-316.
|Printer friendly Cite/link Email Feedback|
|Author:||Gat, Irit; McWhirter, Benedict T.|
|Publication:||Journal of Sport Behavior|
|Date:||Dec 1, 1998|
|Previous Article:||Gender bias: a study of high school track & field athletes' perceptions of hypothetical male and female head coaches.|
|Next Article:||The influence of physical activity and theoretically relevant variables in the use of drugs: the deterrence hypothesis revisited.|