Abstract: It is important to examine the correlates of exercise dependence because of its negative physical and psychological effects. The purpose of this study was to examine if high and low exercise-dependent individuals different on exercise behavior and perfectionism. Participants were 79 university students who completed self-report measures of their exercise behavior, perfectionism, and exercise dependence symptoms. It was found that the high exercise-dependent group reported more perfectionism and exercise behavior than the low exercise-dependent group. Further research examining the relationship between perfectionism and exercise dependence symptoms may aid in identifying people who are at-risk for exercise dependence.

Exercise behavior can be viewed on a continuum, with sedentary individuals and excessive exercisers representing the two extremes. In the middle of this continuum are the individuals who moderately exercise, and therefore reap the health-related benefits of physical activity. At either end of the continuum, however, negative health consequences occur (Szabo, 1998; United States Department of Health and Human Services [USDHHS], 1996, 2000). For example, sedentary behavior places individuals at risk for physical ailments such as cardiovascular disease, obesity, and diabetes (USDHHS, 1996). In contrast, excessive exercise may manifest into exercise dependence, a condition where individuals feel compelled to exercise despite injuries, obligations, or attempts to reduce their activity (Hausenblas & Symons Downs, 2002a).

Exercise dependence is "a craving for leisure-time physical activity that results in uncontrollable excessive exercise behavior and that manifests in physiological and/or psychological symptoms" (Hausenblas & Symons Downs, 2002a, p. 90). There are two types of exercise dependence: primary and secondary (Veale, 1995). For primary exercise dependence, the physical activity is an end in itself. That is, people are intrinsically motivated to exercise. In contrast, for secondary exercise dependence, which is often evident in eating disordered populations, people are extrinsically motivated to exercise in an attempt to control or alter their body size/shape (American Psychological Association [APA], 1994; Veale).

Several explanations have been advanced in an attempt to account for the occurrence of exercise dependence (Carron, Hausenblas, & Estabrooks, 2003). These explanations are classified into the physiological (i.e., •-endorphin and sympathetic arousal hypothesis; Goldfarb & Jamurtas, 1997; Pierce, Eastman, Tripathi, Olson, & Dewey, 1993; Thompson & Blanton, 1987), psychobiological (i.e., general theory of addiction; Jacobs, 1986), and psychological (e.g., personality traits, anorexia analogue hypothesis; Chapman & DeCastro, 1990; Crossman, Jamieson, & Henderson, 1987; Yates, Leehey, & Shisslak, 1983) domains.

The personality traits explanation has received the most research attention (Carron et al., 2003). It is based on the belief that exercise dependent people have personality characteristics such as obsessive-compulsiveness, neuroticism, extroversion, low self-esteem, and high trait anxiety. For example, researchers have usually found a positive relationship between exercise dependence symptoms with obsessive-compulsiveness (Davis, Brewer, & Ratusny, 1993; Spano, 2001), trait anxiety (Coen & Ogles, 1993; Rudy & Estok, 1987; Spano, 2001), and extroversion (Yates, Leehey, & Shisslak, 1983). A negative association has been evidenced for exercise dependence symptoms with neuroticism and self-esteem (Rudy & Estok, 1987).

Another personality trait that may be related to exercise dependence is perfectionism. Perfectionistic people are driven to pursue superior accomplishments by setting high and demanding goals. Recently, researchers have noted that perfectionism can be characterized as either a positive or a negative force in a person's life (Enns & Cox, 2002). People who have positive perfectionism derive pleasure from their efforts, and they strive to excel while still being able to accept personal and situational limitations. In comparison, people who have negative perfectionism are driven by an intense need to avoid failure and they are unable to derive satisfaction from their accomplishments, which results in continually striving to achieve unrealistic goals (Blatt, 1995; Enns & Cox). It is plausible that individuals with exercise dependent symptoms may display traits of negative perfectionism.

The two located studies examining exercise dependence symptoms and perfectionism reported a positive relationship between these constructs (Coen & Ogles, 1993; Hausenblas & Symons Downs, 2002b). These researchers, however, have not distinguished between primary versus secondary exercise dependence (Coen & Ogles, 1993; Hausenblas & Symons Downs, 2002b). This is an important distinction because eating disordered individuals (who may also have secondary exercise dependence) tend to be perfectionist (APA, 1994). Furthermore, a positive relationship between physical activity and perfectionism with eating disordered populations exits (Davis, 1999; Narduzzi & Jackson, 2000). Thus, differentiating between primary and secondary exercise-dependent individuals when examining the relationship between exercise dependence and perfection is important because the exercise motives of these two groups are different (Veale, 1995).

The primary purpose of this study was to examine the relationship between primary exercise-dependence symptoms and perfectionism. The secondary purpose was to examine the relationship between primary exercise-dependence symptoms and exercise behavior. It was hypothesized that the high exercise-dependent individuals would score higher on perfectionism and exercise behavior than the low exercise-dependent group (Coen & Ogles, 1993; Hausenblas & Symons Downs, 2002b).

Participants were 79 male and female students from a large southeastern university in the United States (M age = 21.78, SD = 3.31); 66.3% women. Most of the students were Caucasian (82.6%). For academic standing, the participants were primarily juniors (36%), followed by seniors (33.7%), freshmen (11.6%), graduate students (9.3%), and sophomores (8.1%).

Drive for Thinness Subscale. The Drive for Thinness Subscale is a 7-item subscale from the Eating Disorder Inventory-2 (EDI-2; Garner, 1991) that measures the pursuit of thinness, which is the cardinal feature of an eating disorder. Participants respond to statements assessing excessive concern with dieting, preoccupation with weight, and the fear of gaining weight on a 6-point Liken Scale anchored at the extremes with 1 (never) and 6 (always). Examples of these items are "I feel extremely guilty after overeating" and "I am preoccupied with the desire to be thinner." A high score indicates a high level of drive for thinness. The Drive for Thinness Subscale has adequate reliability and validity (Garner, 1991), and its reliability in this study was excellent (alpha = .91).

Perfectionism Subscale. The Perfection Subscale is a 6-item measure from the EDI-2 (Garner, 1991) that assesses the extent and desire of the need to excel and be the best. The statements are anchored on a 6-point Likert Scale anchored at the extremes with 1 (never) and 6 (always). Although the Perfection Subscale does not separate positive and negative perfectionism questions, the Subscale offers examples of negative perfectionism. Example of negative perfectionism include "My parents have expected excellence of me" and "I feel that I must do things perfectly or not do them at all." Higher scores indicate higher levels of perfectionism. The psychometric properties of this subscale are good (Garner, 1991), and its reliability in this study was adequate (alpha = .75).

Exercise Dependence Scale. The Exercise Dependence Scale (Hausenblas & Symons Downs, 2002b) is a 30-item scale that assesses exercise dependence symptoms such as withdrawal and tolerance. Respondents indicate their agreement to the statements using a 6-point Liken Scale anchored at the extremes with 1 (never) and 6 (always). A high score indicates more exercise dependence symptoms. Questions refer to beliefs and behaviors that have occurred within the last three months. Examples of items include "I feel anxious if I cannot exercise," "I organize my life around exercise," and "I often exercise with more intensity than I intend." The scale has adequate reliability and validity (Hausenblas & Symons Downs), and in this study the scale's reliability was excellent (alpha = .95).

Leisure-Time Exercise Questionnaire (LTEQ). The LTEQ (Godin, Jobin, & Bouillon, 1986) is a self-report, three-item scale that assesses leisure-time exercise. Respondents are asked to recall the number and the intensity of exercise sessions undertaken within a typical week. The exercise intensities are mild, moderate, and strenuous. To determine the metabolic equivalents (METS), the frequency of exercise is multiplied by the activity intensity. The frequency of mild exercise is multiplied by three, moderate by five, and strenuous by nine. The values for mild, moderate, and strenuous exercise are added to determine the total exercise index. A high score represents a greater activity level. This questionnaire has adequate reliability and validity (Jacobs, Ainsworth, Hartman, & Leon, 1993).

Permission was received from the class instructors to administer the questionnaires to their students during class time. Once informed consent was received, 93 participants completed the questionnaires in about 15 min. The Drive For Thinness Subscale, which assesses the cardinal feature of an eating disorder, was used to screen for secondary exercise dependence (i.e., exercise dependence that is secondary to an eating disorder) that could confound the results. Participants scoring in the at-risk range (i.e., a score of 14 or greater;, Garner, 1991) on the Drive For Thinness subscale were omitted from further analysis (n = 9). Thus, the final sample was 79 university students.

The high (n = 40) and low (n = 39) exercise dependence symptoms (ED) groups were created using a median split from the Exercise Dependence Scale scores. A one-way analysis of variance revealed that the high ED group reported significantly more exercise dependence symptoms than the low ED group [F( 1,77) = 123.15, p < .001, W² = .60] (see Table 1). In regard to exercise behavior, it was found that the high ED group reported engaging in more strenuous [F( 1,76) = 16.40, p < .001, W² = .16], moderate [F( 1,75) = 4.93, p < .05, W² = .05], and mild exercise [F( 1,76) = 4.45, p < .05, W² = .04], compared to the low ED group. For perfectionism, it was found that the high ED group reported more perfectionism than the low ED group [F( 1,74) = 3.99, p = .05, W² = .04]. The effect sizes for the results were in the medium to large range (Cohen, 1988).

The study objectives were to examine if a relationship existed between primary exercise dependence symptoms, exercise behavior, and perfectionism. Consistent with the hypothesis and past researchers, it was found that the high ED group displayed more perfectionism than the low ED group (Coen & Ogles, 1993; Hausenblas & Symons Downs, 2002b). Because of the correlational design, however, the cause and effect of this relationship is unknown. Future research using a longitudinal design may assist in determining if perfectionism is an antecedent or a consequence of exercise-dependence symptoms. Also, further research is needed to comprehensively examine the nature of perfectionism with high exercise-dependent symptom individuals. That is, perfectionism is a striving for excellence that can manifest into either positive or negative perfectionism. In particular, and of relevance to exercise dependence, negative perfectionism is characterized by a motivation to avoid negative consequences, compulsive tendencies and doubting, inflexibly high standards, and an inability to experience pleasure from labors (Enns & Cox, 2002). Research is needed to examine which of these characteristics of negative perfectionism is evident in people who report high exercise-dependence symptoms. This information may aid in establishing effective prevention and treatment programs.

Consistent with the hypothesis, the high ED individuals participated in more exercise than the low ED group. Although this finding was expected, the group differences for mild, moderate, and strenuous exercise were larger as the intensity increased. That is, the effect size between the high and low ED groups for strenuous exercise was large, while the effect sizes for moderate and mild exercise were medium (Cohen, 1988). Thus, the high ED group not only performed more bouts of exercise, but they also exercised at a higher intensity than the low ED group. This increase in exercise behavior may lead to overuse injuries, which is a characteristic of exercise dependence (Hausenblas & Symons Downs, 2002a).

There are study limitations that must be mentioned. First, self-report measures have an inherent bias because participants may answer a question based on what they believe is socially desirable and acceptable (Krosnick, 1999). Second, a multidimensional assessment of perfectionism may provide more information about the nature of perfectionism and exercise-dependence symptoms (Enns & Cox, 2002; Frost, Marten, Labatt, & Rosenblate, 1990). Finally, a college sample was used which limits the generalizability of these results. Thus, future directions include using direct measures of exercise (e.g., accelerometers), multidimensional measures of perfectionism, and expanding the population assessed.

In conclusion, the results of this study supported the hypothesis that individuals who display primary exercise-dependent symptoms demonstrate more perfectionism than those exhibiting fewer exercise-dependent symptoms. The information gained about perfectionistic tendencies in this study can be used to detect individuals who may be at-risk for developing exercise dependence symptoms. Thus, perfectionism is not limited to secondary exercise dependence, but it is also evident in primary exercise dependence.

Table 1. Mean (M) and Standard Deviation (SD) Scores for the Exercise Dependence Scale (EDS), Perfectionism Subscale, and the Leisure-Time Exercise Questionnaire (LTEQ)
Legend for Chart:

B - Low Exercise Dependent M (SD) N = 39
C - High Exercise Dependent M (SD) N = 40
D - Total M (SD) N = 79


EDS(*) 47.46 (10.04) 86.98 (19.90) 67.47 (25.35)

Subscale(*) 5.82 (3.23) 7.65 (4.66) 6.71 (4.07)


Mild(*) 10.03 (6.95) 16.80 (18.60) 13.50 (14.49)

Moderate(*) 16.45 (10.26) 22.95 (14.94) 19.74 (13.18)

Strenuous(*) 14.68 (15.15) 32.85 (23.38) 24.00 (21.69)

(*) The high exercise dependent group scored significantly
higher than the low exercise dependent group.
American Psychological Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Blatt, S. J. (1995). The destructiveness of perfectionism: Implications for the treatment of depression. American Psychologist, 50, 1003-1020.

Carron, A. V., Hausenblas, H. A., & Estabrooks, P.A. (2003). The psychology of physical activity. New York: McGraw-Hill.

Chapman, C. L., & DeCastro, J. M. (1990). Running addiction: Measurement and associated psychological characteristics. Journal of Sports Medicine and Physical Fitness, 30, 283-290.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.

Coen, S., & Ogles, B. (1993). Psychological characteristics of the obligatory runner:. A critical examination of the anorexia analogue hypothesis. Journal of Sport & Exercise Psychology, 15, 338-354.

Crossman, J., Jamieson, J., & Henderson, L. (1987). Responses of competitive athletes to lay-offs in training: Exercise addiction or psychological relief? Journal of Sport Behavior, 10, 28-38.

Davis, C. (1999). Excessive exercise and anorexia nervosa: Addictive and compulsive behaviors. Psychiatric Annals, 29, 221-224.

Davis, C., Brewer, H., & Ratusny, D. (1993). Behavioral frequency and psychological commitment: Necessary concepts in the study of excessive exercising. Journal of Behavioral Medicine, 16, 611-628.

Enns, M. W., & Cox, B.J. (2002). The nature and assessment of perfectionism: A critical analysis. In G. L. Flett and P. L. Hewitt (Eds.). Perfectionism: Theory research, and treatment. Washington, DC: American Psychological Association.

Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449-468.

Garner, D. M. (1991). Eating Disorder Inventory - 2 manual. Odessa, FL: Psychological Assessment Resources.

Godin, G., Jobin, J., & Bouillon, J. (1986). Assessment of leisure time exercise behavior by self-report: A concurrent validity study. Canadian Journal of Public Health, 77, 359-361.

Goldfarb, A. H., & Jamurtas, A. Z. (1997). b-Endorphin response to exercise: An update. Sports Medicine, 24,8-16.

Hausenblas, H.A., & Symons Downs, D. (2002a). Exercise dependence: A systematic review. Psychology of Sport and Exercise, 3, 89-123.

Hausenblas, H.A., & Symons Downs, D. (2002b). How much is too much? The development and validation of the Exercise Dependence Scale. Psychology and Health, 17, 387-404.

Jacobs, D. (1986). A general theory of addictions. A new theoretical model. Journal of Gambling Behaviors, 2, 15-31.

Jacobs, D. R., Ainsworth, B. E., Hartman, T. J., & Leon, A. S. (1993). A simultaneous evaluation of 10 commonly used physical activity questionnaires. Medicine & Science in Sports & Exercise, 25, 81-91. Krosnick, J. A. (1999). Survey research. Annual Review of Psychology, 50, 537-567.

Narduzzi, K. J., & Jackson, T. (2000). Personality differences between eating-disordered women and a nonclinical comparison sample: A discriminant classification analysis. Journal of Clinical Psychology, 56, 699-710.

Pierce, E. F., Eastman, N. W., Tripathi, H. L., Olson, K. G., & Dewey, W. L. (1993). b-endorphin response to endurance exercise: Relationship to exercise dependence. Perceptual and Motor Skills, 77, 767-770.

Rudy, E. B., & Estok, P. J. (1989). Measurement and significance of negative addiction in runners. Western Journal of Nursing Research, 11, 548-558.

Spano, L. (2001). The relationship between exercise and anxiety, obsessive-compulsiveness, and narcissism. Personality and Individual Differences, 30, 87-93.

Szabo, A. (1998). Studying the psychological impact of exercise deprivation: Are experimental studies hopeless? Journal of Sport Behavior, 21, 139-147.

Thompson, J. K., & Blanton, P. (1987). Energy conservation and exercise dependence: A systematic arousal hypothesis. Medicine and Science in Sports and Exercise, 19, 91-99.

United States Department of Health and Human Services (1996). Physical activity and health: A report of the Surgeon General. Atlanta: U. S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention of Health Promotion.

United States Department of Health and Human Services (2000). Healthy People 2010 (Conference Edition, in Two Volumes). Washington, DC: January 2000.

Veale, D. (1995). In J. Annett, B. Cripps, & H. Steinberg (Eds.), Exercise Addiction: Motivation for Participation in sport and Exercise. (pp. 1-5). Leicester, UK: British Psychological Society.

Yates, A., Leehey, K., & Shisslak, C. M. (1983). Running -- An analogue an anorexia? New England Journal of Medicine, 308, 251-255.

Responsibility I - Assessing Individual and Community Needs for Health Education

Competency B - Distinguish between behaviors that foster and those that hinder well-being.

Sub-competency 1 - Investigate physical, social, emotional, and intellectual factors influencing health behavior


By Amy L. Hagan, M.S.E.S.S. and Heather A. Hausenblas, Ph.D.

Amy L. Hagen, M.S.E.S.S. is a Doctoral Student in the Department of Exercise and Sport Sciences at the University of Florida. Heather A. Hausenblas, Ph.D. is an Assistant Professor in the Department of Exercise and Sport Sciences at the University of Florida. Address all correspondence to Amy L. Hagen, M.S.E.S.S., ESS FLG 106i, P.O. Box 118205, University, of Florida, Gainesville, FL, 32611-8205, PHONE: 352.392.0580 ext. 1389, FAX: 352.392.5262, E-MAIL:

Share this with your friends