Printer Friendly

Superstitious behavior in sport: levels of effectiveness and determinants of use in three collegiate sports.

From professional athletic organizations to the local little league team, the use of superstitious behavior in sport is evident from numerous TV and newspaper reports covering athletics. These rituals can come in the form of ice baths before a football game to listening to a certain song before a gymnastics event. Athletes may have lucky charms that inspire them during the contest or they may repeat a secret phrase known only to them before shooting a free throw. Other superstitious behaviors related to food, pre and post-game activity, clothing, and behavior during competition are prevalent in all major sports (Buhrmann, Brown, and Zaugg, 1982). Superstitious behaviors in sport can be defined as actions which are repetitive, formal, sequential, distinct from technical performance, and which the athletes believe to be powerful in controlling luck or other external factors (Womack, 1992). The repetitive nature of such events allows for the term "ritual" to be used to describe these superstitious behaviors. Buhrmann et al. (1982) and Becker (1975) discuss how use of superstition in sport can serve other purposes for the athlete as well, such as the lowering of anxiety levels and enhancing the outcomes of performance. In addition, these behaviors or rituals can be either personalized for each individual athlete, or they may be team-generated and agreed upon by the larger group.

Superstitious behavior or ritual can be distinguished from preperformance routines. Preperformance routines are learned, behavioral and cognitive strategies which are intentionally used by athletes in order to facilitate physical performance (Cohn, 1990). Examples of techniques used in cognitive preperformance routines include imagery, relaxation techniques, focusing strategies and coping strategies. These strategies are then supplemented with behaviors, such as physical practice of the desired movement. While preperformance routines have been shown to be effectively used by athletes to modify their performance (Eklund, Gould and Jackson, 1993; Ravizza and Osborne, 1991), they differ substantially from superstitious behavior and ritual in sport. First, preperformance routines are developed for an individual or team by an expert, such as a sport psychologist, after individual or team functioning has been assessed. Thus, where rituals are created, sometimes spontaneously and seemingly at random, by the athlete, preperformance routines are taught by an outside and expert source. Second, preperformance routines focus on cognitive self-control as a means by which an athlete can personally, directly and literally affect individual performance outcomes. In contrast, superstitious rituals include a wide range of behaviors, which correlate with actual performance only in the sense that they lower anxiety levels of athletes and are perceived to effect some measure of control over luck. The power afforded superstitious rituals is given only insofar as the athlete believes these rituals to be effective.

The role of superstitious ritual in sport is not new. As far back as the beginning of this century, Gardiner (1925) had already observed the use of superstitious behavior in athletes. Malinowski (1927) suggested that these rituals occur primarily when conditions of uncertainty or chance are present and occur throughout cultures. In an athletic environment where one can sustain a career ending injury one minute or rise to the heights of individual or team performance the next, uncertainty is the rule. As alluded to above, superstitious behaviors or rituals may be used for a variety of reasons in athletics. Ritualistic behaviors are hypothesized to be used to decrease an athlete's anxiety and increase perceived chance of success (Womack, 1992). The inherent competitiveness of athletes and the societal pressure to succeed in sport can influence an athlete to resort to external means, such as superstitious behavior, to control the outcome of an athletic contest. Douglas (1966), as well as Venturi (1986) have implied that the ambiguity inherent in sport may cause athletes to seek control and certainty through the use of superstitious rituals, as well as use of magical charms or talismans. Thus, an athlete with a high need to succeed seeks to develop control over outcomes through use of superstitious behaviors, or uses ritual to gain control over chance elements or events under others' control.

Superstitious ritual use has often been associated with high-risk activities (Womack, 1992), where either physical danger to the individual or possibility of failure is at stake. Womack studied various professional athletic teams and discovered a variety of rituals used to respond to different situations. She postulated that superstition was used by professional athletes as a means of maintaining emotional stability in order to perform optimally and also as a means of dealing with conditions of stress, anxiety and danger. Especially in highly stressful situations, athletes have resorted to superstitious ritual to seek stability and control over the sport situation or outcome.

This feeling of control or stability can help calm an athlete before a contest, thus allaying excitement and anxiety, while also increasing perceived confidence (Becker, 1975). Due to the perceived effectiveness of ritual in sport, superstitions are used during all aspects of competition, pregame as well as during actual competition (Womack, 1992).

Despite the believed prevalence of superstitious behavior in sport, there has been very little empirical research in this area. Recent studies have examined the superstitious behaviors athletes use before games and how use of such rituals relate to variables, such as religiosity (Buhrmann and Zaugg, 1983; Buhrmann, Brown, and Zaugg, 1982). Research concerning use of superstitious ritual in sport has also indicated that female athletes are more likely to use rituals than male athletes and that age of the athlete is positively correlated with amount of ritual use (Buhrmann and Zaugg, 1981). However, this research was done primarily with basketball players, and no cross-sport comparisons of superstitious behavior use were made. To date, no research was found that examined the relationship between anxiety and use of superstitious sport rituals, or how superstitious behavior relates to importance of success for an athlete. Nor was research found that addressed the possible link between use of superstitious rituals and attributional variables in sport, such as locus of control. Last, no studies were found that replicated early work which indicated a relationship between religiosity and use of superstitious rituals.

The present study addresses all of these areas using a survey methodology. First, this study provides information about use of superstitious behavior and perceived effectiveness of those behaviors in three, Division I, collegiate sport teams. Second, this study predicts a positive relationship between use of superstitious behaviors, level of anxiety and perceived importance of success. As more self-perceived importance is placed on winning, sport anxiety levels may increase and the athlete may try to lower that anxiety through use of superstitious rituals. Third, it is hypothesized that sport attribution is related to use of superstitious behaviors and rituals. It could be reasoned that if one's locus of control or causality for success is external (e.g. luck or powerful others) then an effort may be made to alter one's luck or somehow gain control over powerful others. Of course, just the opposite hypothesis could also hold true. Those with external locus of control will not feel they have the ability to alter fate, so their use of superstitious behaviors will be lessened, while those with an internal locus of control will believe in their ability to alter fate, thus using more superstitious behaviors. The relationship between internal and external locus of control and use of superstitious behaviors will be examined in the present study. Last, this study predicts a positive relationship between use of superstitious ritual and religiosity. It is hypothesized that the belief that a higher power controls one's life is reflective of an external locus of control. If this is so, those individuals with higher religiosity will also report greater use of ritual in order to win the favor of, or gain control over, this higher power.



Participants in this research were athletes from three, NCAA, Division I, college athletic teams at Southern Utah University. In total, the sample was comprised of 87 men and 20 women, with a mean age of 20.5 years. As a whole, their average number of years on an athletic team at the University was 2.4 years. Within this sample were athletes from the football, gymnastics and track teams. For the football team, 77 players completed the survey. These men ranged in age from 18 to 25 with a mean age of 21 years and an average of 2.5 years on the football team. The gymnasts were 12 women, ranging in age from 18 to 21 with a mean age of 19 and, on average, 1.9 years of collegiate sport participation. The track team consisted of 10 men and 8 women, ranging in age from 18 to 21 with a mean age of 19.6 years and an average of 1.9 years of collegiate sport participation. Each group participated voluntarily receiving no compensation, monetary or academic, for their assistance with this project.


Demographics. Each athlete answered a set of standard demographic questions designed for use in the present study. Information was collected about participants' sex, age, years on their team and importance of success in their sport. Level of importance of success was measured using a 5 point Likert scale response with 1 (not at all important) to 5 (very important).

Religiosity Measure. Religiosity was assessed using a standard measure of religiosity. Subjects were asked to provide information about how many times they had attended church services in the past month, as part of their religion. Tittle (1980) and Tittle and Welch (1983) indicate that frequency of church attendance is the most widely used and valid measure of religiosity within the social research domain.

Locus of Control. Locus of control was measured using the Fitness Locus of Control Scale (FITLOC :Whitehead and Corbin, 1988). The FITLOC is an instrument designed to measure three dimensions of locus of control in fitness domains: internal locus of control, powerful others and chance. The FITLOC has 11 items responded to using a 6 point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). Internal LOC is measured using three items; powerful others LOC is measured using four items, and chance LOC is measured using four items. In order to obtain a total score for each subscale, the responses for each are summed, giving a possible range of summed score of 3 to 18 for the internal subscale, and a range from 4 to 24 for the powerful others and chance subscales. Reported test-retest reliability scores for the FITLOC ranged from .59 to .75 and inter-item reliability coefficients were reported to range between .62 and .84. For use in the present study, the FITLOC was modified slightly to reflect each sport surveyed. In each of the items, the term physical fitness was substituted with either football participation, gymnastics participation or track participation, corresponding to the team surveyed. Because the FITLOC was modified for the present study, inter-item reliability and construct validity were tested using the present sample. Cronbach alphas for the modified FITLOC were .85 for the internal locus of control scale, .78 for the powerful others subscale and .92 for the chance subscale. The FITLOC subscales were then correlated with the Exercise Locus of Control Scale (EOLOC: Mc Cready and Long, 1985). The EOLOC also measures internal, chance and powerful other dimensions of control in exercise settings. The Pearson correlation between the FITLOC internal subscale and the EOLOC internal subscale was. 58, p [less than] .01; the powerful others subscales correlated at .48, p [less than] .01 and the chance subscales correlated at .66, p [less than] .01. These correlations indicate construct validity for the modified FITLOC.

Sport Anxiety Scale. (Smith, Small and Schutz, 1990). The Sport Anxiety Scale is a self-report measure of three components of anxiety in competitive sport situations. The scale measures the dimensions of somatic anxiety, worry and concentration disruption. Each item is responded to using a Likert scale response set with 1 (not at all) to 4 (very much so). Subscores are created by summing the responses for all items in the subscale. The somatic anxiety subscale contains nine items with total scores ranging from 9 to 46. The worry subscale contains seven items with possible scores ranging from 7 to 28, and the concentration disruption subscale contains five items with scores ranging from 5 to 20. Inter-item reliability scores for all three factors ranged from .81 to .92. Test-retest reliability of the scale was reported to be .77 after 18 days.

Superstitious Beliefs Measure. A scale measuring superstitious beliefs, behaviors and rituals was created for use in this study, based on an earlier scale used by Buhrmann, Brown and Zaugg (1982). Buhrmann et al. categorized superstitious behavior into seven categories: clothing and preparation, fetishes, pregame, game, team ritual, prayer and coach. This early measure had a total of 40 items. For the present study, the Buhrmann et al. measure was modified slightly. Most of the original items in the scale were kept except for seven items specific to basketball teams (e.g. free throw rituals) and, after speaking with athletes from all the sports surveyed in the present study, explaining to them what superstitious behaviors were and asking them to generate examples within their sports, 12 more items were added for a total of 45 superstitious rituals. In the present study, the original grouping categories were maintained and extra items placed into the appropriate category of behavior. The present scale asked each athlete to indicate whether or not he/she used each superstitious behavior listed and also asked the respondent to indicate the degree of effectiveness of each ritual used for him/her. Effectiveness was measured using a five point Likert scale with 1 (not at all effective) to 5 (very effective). A total superstitious behavior score was created by summing the number of rituals used per participant. A copy of the superstitious behavior survey used in the present study is contained in Appendix A.


Survey packets were administered to each team in a group setting by the primary researchers. Subjects were informed that the survey they were being asked to complete was a questionnaire measuring their use of superstitious behaviors and rituals in their sport, as well as related personality characteristics. Information in each packet was coded by social security number as to maintain subject confidentiality and not to allow individual identification by name. The researchers were available during survey administration to clarify any questions which arose from the participants.


Demographic Information

Demographic information about the entire sample and broken down by team is presented in Table 1. In these variables of interest, team differences were present only in the total number of superstitious behaviors used, and in all three levels of anxiety. More specifically, gymnasts used a significantly higher number of superstitious rituals than football players or track athletes, F (2, 103) = 5.68, p [less than] .01. Gymnasts also showed higher levels of somatic anxiety than football players but were no different than track athletes in somatic anxiety, F(2,102) = 4.28. p [less than] .05. Track athletes, however, showed higher levels of worry, F(2,101) = 5.61, p [less than] .01, and concentration disruption F(2,102) = 12.64, p [less than] .01, than did football players.


Top Superstitious Rituals and Behaviors Used by Team and Ratings of Effectiveness for Top Behaviors

The top ten superstitious behaviors used by each team and their corresponding levels of effectiveness are presented in Table 2. For football players, the most used superstitions were prayer and clothing oriented. It is interesting to note, while the effectiveness ratings of prayer are correspondingly high for these players, the effectiveness ratings of widely-used clothing rituals are among the lowest for football players' most used superstitious rituals. For gymnasts, appearance-related superstitions were important, as were superstitious rituals having to do with eating, prayer and team cohesion (e.g. team prayer and team cheer). For track athletes, clothing rituals and luck-related superstitious behaviors (e.g. having a lucky charm or making lucky marks on shoes) were important.

Correlations Between Total Superstitious Behavior Use, Religiosity, Locus of Control and Anxiety

In order to attempt to address why superstitious behaviors or rituals are used in sport, total use was correlated with other variables of interest. The results of these Pearson correlations are presented in Table 3. The correlations are presented for the entire sample, then for each sport individually. It can be noted that there is little support for religiosity, locus of control or anxiety relating to total superstitious behavior use among any of the three groups of athletes surveyed.

Religiosity, Locus of Control and Anxiety as Predictors of Specific Superstitious Behaviors

Although none of the predictor variables in the study contributed significantly to total superstitious ritual use, it can be hypothesized that personal beliefs or aspects of personality (e.g. religiosity, LOC or anxiety) may be more predictive of use of a specific ritual. This hypothesis was tested for three superstitious rituals. First, it was predicted that higher religiosity was indicative of higher use of prayer ritual (ritual F1 in Appendix A). Second, it was predicted that higher monitoring of appearance was related to locus of control beliefs (Ritual A1 in Appendix A). Checking one's appearance is a behavior which gives the individual athlete internal control over one small aspect of his/her sport presentation. Social psychologists discuss impression management as a means by which individuals try to exert control over social situations and the people in them (Aronson, Wilson and Akert, 1995). Last, it was predicted that eating rituals would be related to anxiety (Ritual A14 in Appendix 1). Anxiety has been related to eating behaviors in normal as well as eating disordered populations (Fisher, Schneider, Pegler and Napolitano, 1991; Mizes, 1988). T-tests were done to test these hypotheses. In the first of these analyses, it was found that higher religiosity was related to [TABULAR DATA FOR TABLE 2 OMITTED] [TABULAR DATA FOR TABLE 3 OMITTED] greater use of prayer as a sport superstition, t (104) = -3.748, p [less than] .01. In the second analysis, lower belief in chance was related to higher usage of appearance monitoring ritual, t(101) = 2.176, p [less than] .05. Belief in internal LOC or powerful others LOC was not related to this appearance ritual. Last, only one aspect of sport anxiety, concentration disruption, was significantly related to use of eating rituals, t(103) = -2.183, p [less than] .05). This analysis indicated that higher levels of concentration disruption occurred with higher usage of an eating ritual.


Why athletes use superstitious behaviors or rituals in the competitive milieu and what types of behaviors are used pose interesting questions for investigation by sport psychologists. The purpose of the present study was to examine usage of superstitious behaviors in three different sports: football, gymnastics and track. Each of these three sports is different in terms of sex of participant, and team versus individual performance. Due to these differences, it was hypothesized that superstitious behaviors would vary among athletes within the groups studied. Indeed some differences were found to exist. Overall, gymnasts reported more superstitious ritual use than football players or track athletes. It is suspected that due to the focus and importance placed on each individual performance in gymnastics that more rituals are used in order to increase one's chance of success. This finding is also consistent with earlier research by Buhrmann and Zaugg (1981), which found higher ritual use in female basketball players than male basketball players.

The present study further indicated that although commonalties were found in superstitious ritual usage across sport, each sport also had rituals specific to it. For example, football rituals were centered around clothing and prayer. Although gymnasts also emphasized clothing rituals, team rituals and pregame food rituals were also very important to these athletes. Still different were track athletes. Track athletes focused again on clothing, but they were the only group that mentioned lucky items of clothing or lucky markings on shoes. What rituals athletes use does, indeed, seem dependent upon their particular sport.

Unlike earlier studies, the present study addressed perceptions of effectiveness associated with each superstitious behavior used by the athletes. It would seem obvious to believe that rituals, which are most used, are engaged in because they are also believed to be highly effective. Judging from mean effectiveness ratings of most -used superstitions, this does not appear to be the case. In the present study, ritual effectiveness was rated on a five point scale with 3 being neutral, 1 being low effectiveness and 5 being high effectiveness. As the mean effectiveness ratings indicated in Table 2, even the most widely used rituals vary in their perceived effectiveness from a low score of 2.64 to a high score of 4.43. Examination of trends in this data indicated that prayer behaviors seemed to garner high effectiveness ratings across groups of athletes, while clothing rituals (although widely used) garnered some of the lowest effectiveness ratings across sports. The conclusion can be drawn that it is not necessarily effectiveness of a superstitious sport behavior which determines its popularity among members of a sport team.

Last, this study made a preliminary attempt to determine why superstitious rituals were used. Previous research and theory has indicated that religiosity (Buhrmann and Zaugg, 1983), locus of control beliefs, importance of success and alleviation of anxiety (Womack, 1992) are all factors which may facilitate the use of superstitious behaviors in sport contexts. The present study found little support for a relationship between ritual use and any of the above mentioned factors. Judging from these results, or lack thereof, it can be concluded that religiosity, locus of control, importance of success and anxiety do not contribute meaningfully to total level of superstitious ritual usage.

It may be that use of superstitious behavior is much more personalized in nature and the actual ritual used depends upon the athlete using the ritual, as well as other factors, such as the sport the athlete is in, or the effectiveness of the ritual. Initial support for this assumption was shown in the present study. It was found that religiosity, locus of control and anxiety were related to higher usage of specific superstitious rituals. Thus, it seems that athletes may develop rituals that correspond with their personalities and personal belief systems. It, then, is not total use of rituals which is important, but which specific rituals these athletes choose to employ.

Overall, this study expands upon the existing literature concerning use of superstitious behavior or ritual in sport in a number of ways. First, the study sampled three different sports teams. Second, this study examined, not only which superstitious behaviors these athletes used, but how effective the perceived rituals were to the athletes. Last, the study made an attempt to test some of the underlying assumptions about why athletes use rituals in their sport environment. The use of superstitious behavior in sport is an intriguing area for future study. Further understanding of the personal mechanisms within athletes that drive use of specific rituals is needed. In addition, follow-up research needs to be done to relate ritual usage and perceived ritual effectiveness to actual sport performance.


Acknowledgements: The authors would like to thank James G. Bleak and Craig S. Morrison for their contribution to this research.


Aronson, E., Wilson, T. D., & Akert, R. M. (1994). Social psychology: The heart and the mind. New York: HarperCollins College Publishers.

Becker, J. (1975). Superstition in sport. International Journal of Sport Psychology, 6(3), 148-152.

Buhrmann, H. G., Brown, B., & Zaugg, M. K. (1982). Superstitious beliefs and behavior: A comparison of male and female basketball players. Journal of Sport Behavior, 5, 175185.

Buhrmann, H. G. & Zaugg, M.K. (1981). Superstitions among basketball players: An investigation of various forms of superstitious beliefs and behavior among competitive basketballers at the junior high to university level. Journal of Sport Behavior, 4, 163-174.

Buhrmann, H. G., & Zaugg, M. K. (1983). Religion and superstition in the sport of basketball. Journal of Sport Behavior, 6, 146-157.

Cohn, P. J. (1990). Preperformance routines in sport: Theoretical support and practical applications. Sport Psychologist, 4, 301-312.

Douglas, M. (1966). Purity and danger. London: Routledge and Kegan Paul.

Eklund, R.C., Gould, D. & Jackson, S.A. (1993). Psychological foundations of Olympic wrestling excellence: Reconciling individual differences and nomothetic characterization. Journal of Applied Sport Psychology, 5, 35-47.

Fisher, M., Schneider, M., Pegler, C., & Napolitano, B. (1991). Eating attitudes, health-risk behaviors, self-esteem, and anxiety among adolescent females in a suburban high school. Journal of Adolescent Health, 12, 377-384.

Gardiner, E. N. (1925). Olympia: Its history and remains. Oxford University Press.

Malinowski, B. (1927). Coral gardens and their magic. London: Routledge and Kegan Paul.

McCready, M. L. & Long, B.C. (1985). Locus of control, attitudes toward physical activity, and exercise adherence. Journal of Sport and Exercise Psychology, 7, 346-359.

Mizes, J.S. (1988). Personality characteristics of bulimic and non-eating disordered female controls: A cognitive behavioral perspective. International Journal of Eating Disorders, 4, 541-550.

Ravizza, K. & Osborne, T. (1991). Nebraska's 3-R's: One-play-at-a-time preperformance routine for collegiate football. Sport Psychologist, 5, 256-265.

Smith, R. E., Smoll, F. L., & Schutz, R.W. (1990). Measurement and correlates of sport-specific cognitive and somatic trait anxiety: The Sport Anxiety Scale. Anxiety Research, 2, 263-280.

Tittle, C. (1980). Sanctions and social deviance: The question of deterrence. New York: Praeger.

Tittle, C. & Welch, C. (1983). Religiosity and deviance: Toward a contingency theory of constraining effects. Social Forces, 61, 653-682.

Venturi, I. (1986). Superstition in sports. Movimento, 2, 23-27.

Whitehead, J.R. & Corbin, C. B. (1988). Multidimensional scales for the measurement of locus of control of reinforcements for physical fitness behaviors. Research Quarterly for Exercise and Sport, 59, 108-117.

Womack, M. (1992). Why athletes need ritual: A study of magic among professional athletes. In Shirl Hoffman (Ed.), Sport and Religion (pp. 191-202), Champaign, IL: Human Kinetics.


In the June, 1997 issue (Volume 20, No. 2), the title for Wiechman and Williams' article, "Relation of Athletic Identity to Injury and Mood Disturbance," was incorrectly printed. The title should have been, "Factors Affecting Athletic Identity and Expectations in High School Student Athletes". Additionally, the factor analysis for the Athletic Identity Measurement Scale was not included in the publication. Please contact Shelley Wiechman if you would like a copy of that analysis.
COPYRIGHT 1998 University of South Alabama
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bleak, Jared L.; Frederick, Christina M.
Publication:Journal of Sport Behavior
Date:Mar 1, 1998
Previous Article:Actor-observer bias and perceived sensitivity to internal and external factors in competitive swimmers.
Next Article:Phenomenological meanings of martial arts participation.

Related Articles
Psychological skills assessment and athletic performance in collegiate rodeo athletes.
Effects of self-administered visuo-motor behavioral rehearsal on sport performance of collegiate athletes.
Collegiate coaches: an examination of motivational style and its relationship to decision making and personality.
Superstitious rule generation is affected by probability and type of outcome.
Characteristics associated with superstitious behavior in track and field athletes: are there NCAA divisional level differences?
Use of social ecology model to address alcohol use among college athletes.
Sport superstition as a function of skill level and task difficulty.
The impact of leadership behavior on satisfaction of college tennis players: a test of the leadership behavior congruency hypothesis of the...

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |