The aim of the present study was to identify the reasons that drive individuals towards participation in organized fitness programs in Greece. Three hundred and twenty (n = 320) individuals participated in the research and completed the inventory of fitness motives. The principal component analysis of the fifty-two items revealed the existence of nine factors which were defined as follows: challenge, relaxation, intellectual, enjoyment, addiction, social, health, escape, external influence. The total scale and the sub-scales were shown to have highly positive psychometric properties.

What motivates individuals to take part in organized fitness programs? The answer to this question has both theoretical and practical implications. Examining the perceived reasons that make exercise an attractive behavioral choice is one way of gaining insight into such motivational processes (McCullagh and Noble, 1998). Furthermore, once the reasons to exercise are fully understood, they can be applied to predictive models of exercise participation, which is of great importance to the practitioners.

Graef, Csikzsentmihalyi and Gianinno (1983) argued that intrinsic rather than extrinsic motivation was more dominant regarding leisure behavior, as it facilitated attempts to meet optimal levels of arousal. Intrinsic motivation has been reported to be influenced by a number of factors such as awareness, leisure repertoire, leisure ethic (Iso-Ahola & Weissinger, 1987) and perception of constraints (Ellis & Witt, 1984).

Freedom of choice is probably the crucial factor for enhancing an individual’s intrinsic motivation (Weissinger, 1986). The cognitive evaluation theory (Deci & Ryan, 1985) emphasizes the importance of intrinsic motivation in relation to personal goals and objectives. It is one of the main theoretical approaches to the study of exercise participation, motivation and adherence to participation. According to Deci & Ryan (1985), individual goals might be informational or controlling in nature. Intrinsic motivation can be enhanced when someone can feel the experience of choice and there is freedom from pressure. On the other hand, tension and pressure to perform undermine intrinsic motivation for the activity, and therefore create unfavorable conditions for long–term participation. McAuley, Wraith and Duncan (1991) argued that external reasons, such as health- and personal appearance-related ones, might initially direct individuals towards participation in exercise. However, as skills and physical conditioning improve, intrinsic reasons may become more salient.

The personal investment theory (Maehr and Braskamp, 1986) is another theoretical approach to the study of exercise motivation and adherence to participation. It is an attempt to employ an interactive framework to motivation (McCullagh and Noble, 1998). This approach emphasizes the importance of both personal and situational factors. The personal factor includes individual traits such as self-motivation (Dishman and Ickes, 1981), locus of control (McCready and Long, 1985) and goal orientation (Pemberton, 1986), whereas situational factors include mainly environmental factors. Personal investment theory suggests that a person’s individual characteristics and situational factors influence the individual to determine the meaning of the presented activity. This interpretation of meaning influences personal investment. There are three main components of this theory: sense of self, perceived options, and personal incentives. The sense of self includes perceptions of competence, self-reliance and goal directedness; perceived options refer to the behavioral alternatives available in a given situation; and personal incentives are the motivational focus of behavior (Markland and Hardy, 1993). As Markland and Hardy (1993) argue, both the cognitive evaluation and the personal investment theories suggest that a consideration of an individual’s reasons for exercising may be profitable.

Measures of Reasons for Exercise Participation

A variety of scales has been developed to measure sport and exercise motives. Various studies have been conducted in competitive sports, physical education classes (Gould, Feitz and Weiss, 1985Buanamano et al., 1995), leisure and recreation settings (e.g., Beard and Ragheb, 1983; Crandal, 1980; Clough et al., 1990), and physical activities (e.g., Markland and Hardy, 1993; Silberstein et al., 1988; Frederick and Ryan, 1993; Alexandris & Carroll, 1997).

A review of the published studies has indicated that Markland and Hardy's (1993) scale is one of the most detailed one. It was developed with the aim to measure reasons for participating in sport and exercise programs. The scale consisted of forty-four items, grouped into twelve dimensions. The scale was shown by the authors in order to obtain satisfactory psychometric properties. The validity of the scale was tested against gender differences. Previous findings (Mathes and Battista, 1985; Biddle and Bailey, 1985; Duda and Tappe, 1989; Finkenberg, 1991), as well as the personal investment theory (Maehr and Braskamp, 1986) suggested that affiliation and weight management-related motives were perceived by women as more important one than men, who on the other hand perceived competition-related motives as more important. Furthermore, the studies have shown (Duda and Tape, 1986; Beran, 1986) that older adults tend to endorse health- and fitness-related reasons for exercising more than younger adults.

Silberstein et al.'s (1988), Frederick and Ryan’s (1993) and Alexandris and Carroll’s scales are shorter and include twenty-four, twenty-three and twenty-seven items respectively. The first two scales emphasized mainly health/fitness and body-related motives (e.g., sexual attractiveness, general appearance, weight control, fitness/health). It should be pointed out that both these scales failed to recognise social motives. Furthermore, Frederick and Ryan's (1993) scale failed to measure relaxation-related motives. Research in the field of recreational motivation (e.g., Crandal, 1980; Beard and Ragheb, 1983; Clough et al., 1990) has repeatedly shown that relaxation and social motives are major incentives for participation. Furthermore, Silberstein et al. (1988) developed their scale utilizing a student sample, whose motives were different from those of the adults (Frederick and Ryan, 1993). Alexandris and Carroll’s (1997) scale has been developed for recreational sporting activities in general, and it is not fully applicable to fitness activities. Clough et al.'s (1990) scale was specifically developed for recreational running. The forty-five items were analyzed and grouped into six dimensions. The lack of discussion concerning aspects of reliability of the scale is worth noticing. Finally, Beard and Ragheb's (1983) scale was developed for leisure activities in general. The authors reported very good psychometric properties. Once again this was developed for leisure activities, and it is not fully applicable to fitness activities. The lack of health motives supports this argument. Health-related motives have been widely suggested (e.g., Alexandris & Carroll, 1997; Frederick and Ryan, 1993; Markland and Hardy, 1993; Davis et al., 1995) to be amongst the most distinct and important motivational dimensions for sport participation.

Rationality of the Study

Participation in organized fitness sessions has recently increased rapidly in Greece (Alexandris and Carroll, 1999). Organized fitness programs are offered by both private and public sports clubs. It is of great importance for the practitioners to know exactly the reasons that individuals take part in these programs. This will help them to fit the programs to their needs and wants. Considering the theoretical value of research on the reasons for fitness participation as discussed above and the practical implications of such research, the aim of the present study was to develop a scale measuring the reasons for taking part in organized fitness sessions.

Item generation

The items of the Fitness Motivation Inventory were generated by two main sources:

a) literature on sport, exercise and leisure motivation. Special references should be made to Markland and Hardy's (1993), Silberstein et al.'s (1988), Frederick and Ryan’s (1993), Alexandris and Carroll’s (1997), and Gould, Feitz and Weiss (1985) instruments. These scales were subjected to a critical analysis by the authors and three expert judges in the field. The aim was to cover all possible motivational dimensions which have been suggested in the literature: health/fitness, appearance, challenge/achievement, social, intellectual, relaxation/stress reduction, enjoyment, and addiction. The items were selected on the basis of their applicability in aerobics classes in sport centers;

b) twenty informal semi-structured interviews conducted with aerobics participants. The aim of these interviews was to generate items specifically related to aerobics motivation. A pool of sixty-five items was initially generated. This item pool was scrutinized by the authors and three expert judges. This procedure led to a final pool of sixty-two items, which then was reviewed by a teacher of Greek literature. The final step was to conduct a pilot study with one hundred and fifty fitness participants. The respondents in the pilot study were asked to comment on the clarity of expression and the layout of the instrument. Slight modifications were made, based on their suggestions. Three items were dropped based on the item and reliability analysis of the pilot study. These items had low item-total correlations (<0.35; Devellis, 1991). The sixty-item instrument was once again reviewed by the Greek literature teacher in order to ensure clarity and comprehension. A question asking participants to report the time of their club membership (in months) was included in the scale.

Sampling

Three hundred and one (n=301) individuals, who took part in organized fitness sessions provided by four public fitness clubs participated in the research. Respondents were asked to evaluate on a four-point Likert-style ranking scale (4 – “very important” to 1 – “unimportant”) the importance of a series of statements as reasons for their participation in the programs. The respondents filled the questionnaire before commencement of exercises. Women (67%), middle aged (33%) and married individuals (63%) were the majority in the sample.

Principal Component Analysis

A principal component analysis was performed in order to establish the factorial validity of the scale. Only those components with the eigenvalue greater than 1.0 were retained and rotated with both orthogonal and oblique rotations. Both methods gave similar results. Orthogonal rotation was retained because of conceptual simplicity and

Table 1. Principal Component Analysis for the Motivation Scale

ease of description (Tabachinick and Fidell, 1989). The nine factors that emerged accounted for 65% of the variance. The loading matrix indicated that five items did not load satisfactorily (>0.35) on the factors and were dropped. The nine factors were then defined as follows: 1) Challenge – seven items; 2) Relaxation – seven items; 3) Intellectual – six items; 4) Enjoyment – eight items; 5) Addiction – five items; 6) Social – seven items; 7) Health – six items; 8) Escape – three items; and 9) Social influence – three items. The nine factors and the loading matrix are presented in Table 1.

Internal Consistency Reliability

The Cronbach-alpha coefficient was calculated for the entire scale and for each of the sub-scales. The values of alpha showed very good internal consistency reliabilities. The alpha for the whole scale was found to be 0.95, while all the sub-scales reliabilities ranged from 0.78 to 0.90. All the items contributed positively to the internal consistency reliabilities of the sub-scales, and therefore no further modifications were required. The descriptive statistics and the reliabilities of the whole scale and the sub-sales are presented in Table 2.

Table 2. Descriptive Statistics and Reliability for the Sub-scales and the Total Scale

TEST-RETEST RELIABILITY

A test-retest reliability analysis was conducted in order to determine the stability of the inventory’s scales. A subset of 60 individuals from the original sample was re-administered on the scale, 7-9 weeks after the first administration. The correlations between the sub-scale scores on the two occasions ranged from 0.57 to 0.89, which indicated that the scores were relatively stable for this period of time (Table 2).

Validity of the Scale

Content Validity was probed by: a) critical selection and examination of the items comprising the scale by the authors and expert judges; b) item-total correlations which were shown to be satisfactory.

Face Validity was probed by: a) careful examination of the scale by the two teachers of Greek literature; b) respondents’ positive comments about the scale. They found it comprehensive.

Table 3. Scale means, SD and direct discriminant analysis for males and females

Construct Validity: As discussed above, the previous studies (Biddle and Bailey, 1985; Duda and Tappe, 1989; Markland and Hardy, 1993) as well as the personal investment theory (Maehr and Braskamp, 1986) suggested that affiliation- and weight management-related motives were perceived by women as more important; whereas men perceived competition- related motives as more important. In order to test the construct validity of the scale an examination of gender differences was conducted aiming to determine if the scale could discriminate between men and women. A MANOVA indicated statistically significant differences in the two groups’ scores (Wilk’s lambda = 0.85, F = 3.9, p<0.001). A follow-up discriminant function analysis indicated a statistically significant separation (canonical r = 0.39). The loading matrix of the correlations between the predictors and the discriminant function indicated that the health/appearance and challenge/status dimensions were the best predictors of group membership. Males scored significantly higher than females in the status/challenge dimension (F = 4.2, p<0.05), whereas females scored significantly higher in the health/appearance dimension (F = 8.7, p<0.005).

One hundred from the original sample completed also a Greek version of the Intrinsic Motivation Inventory (IMI, Ryan, 1982; Tsorbatzoudis et al., 1999). The IMI determines the level of intrinsic motivation for a task as an additive function of the four underlying dimensions of interest/enjoyment, perceived competence, effort/importance and pressure/tension. As McAuley (1989) and Markland and Hardy (1993) suggested, the wording of the IMI could be easily modified to reflect intrinsic motivation for any particular activity. A correlation analysis was conducted to examine the concurrent validity of the scale. The interest/enjoyment of the IMI scale had the highest correlations with enjoyment (r = 0.50, p<0.001), relaxation (r = 0.52, p<0.001) and intellectual (r = 0.56, p<0.001) scales. Furthermore, the effort/importance scale correlated significantly with the challenge (r = 0.50, p<0.001) scale. Unexpectedly, the effort/importance scale was also correlated with the social (r = 0.43, p<0.001) and influence (r = 0.46, p<0.001) scales. Finally, the tension/pressure scale had the highest correlation with the challenge scale (r = 0.55, p<0.001).

The aim of the present study was to identify the reasons that drove individuals towards participation in organized fitness programs in Greece. A new scale was developed in order to achieve the objectives of the study. The principal component analysis revealed nine motives dimensions. These dimensions are conceptually clear and can be compared with factor structures revealed in other similar studies (e.g., Markland and Hardy, 1993; Silberstein et al., 1988; Frederick and Ryan, 1993; Alexandris & Carroll, 1997; and Gould, Feitz and Weiss, 1985) despite the different nature of the activities used (e.g., leisure, recreation, and competitive activities). In contrasting the studies, it should be pointed out that the enjoyment, social, health and relaxation factors are common in most of the above studies. On the other hand, the addiction dimension indicates a difficulty in comparing studies with different samples. “Addiction” has been reported in studies which investigated motives for sport participation (e.g., Clough et al., 1990), but it has not been reported in studies which investigated motivation for leisure and recreation participation (e.g., Ragheb, 1983). The examination of the mean scores of the nine dimensions indicates that the relaxation- and enjoyment-related motives are the most important incentives to fitness participation. These findings are in line with Markland and Hardy’s (1993) findings. In their study they found that recreation and enjoyment were the most salient exercise incentives. The high scores in the relaxation, enjoyment and also escape dimensions support the recreational nature of fitness participation, and may suggest that sport providers and leaders should work very closely on the provision of interesting, attractive and fun programs, which should take place in relaxing and well-designed fitness facilities (Alexandris & Carroll, 1999). The aim should be to promote enjoyment and to facilitate participants’ relaxation. In contrast, the relatively low score of the health dimension (mean 2.63) is one of the surprising findings of the study, and it needs further investigation. Further development and extension of the health scale with the inclusion of more items related to weight management and fitness might clarify this issue. Finally, the very low score of the influence dimension (mean 1.50) is also worth pointing out. It indicates that participants are not significantly affected by e.g., friends, family and commercials in their decision to participate in fitness programs. This is a surprising finding, considering the general belief that commercials have a strong influence on consumer behavior. It is a finding which definitely needs further investigation. In terms of the internal consistency reliability, the scale was shown to be very promising. The analysis offered preliminary support for the validity and reliability of the scale. The sub-scales showed to have very good values of alpha. Furthermore, male subjects endorsed status/challenge more and health/appearance less than women, which is in line with the findings of the previous studies (e.g., Biddle and Bailey, 1985; Duda and Tappe, 1989; Markland and Hardy, 1993). In addition to evidence for the discriminatory capability of the scale, the enjoyment, relaxation, and challenge scales showed appropriate correlations with the conceptually related scales of the IMI.

The scale developed in the present study can be a useful tool for sport managers and sport development officials in their effort to identify the reasons that drive individuals to take part in fitness. This is of practical importance, as satisfying customers’ needs leads to customer retention, which is amongst the main objectives of every sport organization (Gerson, 1999). Furthermore, the scale developed in the present study can be tested with bigger samples in order to further test aspects related to the construct validity (e.g., demographic differences), and can be used in more theoretical studies in future research. The hierarchical model of leisure constraints (Crawford et al. 1991), for example, might be further advanced by investigating the causal relationships between motivational factors and constraints dimensions, as suggested by Jackson et al. (1993) and Carroll and Alexandris (1997).

[1] Αlexandris, K., Patterns of Recreational Sport Participation among the Adult Population in Greece, “Cyber Journal of Sport Marketing”, 1998, 2, pp. 150-167.

[2] Alexandris, K., Carroll, B., Motives for Recreational Sport Participation in Greece: Implications for Planning and Provision of Sport Services, “European Physical Education Review”, 1997, 3, pp. 129-143.

[3] Beard, G.J., Ragheb, G.M., Measuring Leisure Motivation. “Journal of Leisure Research”,1983, 15, pp. 219-228.

[4] Buonamano, C., Mussino, A., Participation Motivation in Italian Youth Sport, “The Sports Psychologist”, 1995, 9, pp. 265-281.

[5] Carroll B., Alexandris K., Perception of Constraints and Strength of Motivation: Their Relation to Recreational Sport Participation, “Journal of Leisure Research”, 1997, 29, pp. 279-299.

[6] Clough, P., Shepherd, J., Maughan, R., Motives for Participation in Recreational Running, “Journal of Leisure Research”,1990, 21, pp. 279-309.

[7] Crandall, R., Motivations for Leisure, “Journal of Leisure Research”, 1980, 12, pp. 45-54.

[8] Crawford, D., Godbey, G., Reconceptualizing Barriers to Family Leisure, “Leisure Sciences”, 1987, 9, pp. 119-127.

[9] Crawford, D., Jackson, E., Godbey, G., A Hierarchical Model of Leisure Constraints, “Leisure Sciences”, 1991, 13, pp. 309-320.

[10] Davis, C., Fox, J., Brewer, H., Ratunsy, D., Motivations to exercise as a function of personality characteristics, age and gender, “Personality and Individual Differences”, 1995, 19, pp. 165-174.

[11] Deci, L., Ryan, M., Intrinsic Motivation and Self-determination in Human Behavior, New York, Plenum, 1985.

[12] Devellis, R., Theory and Scale Development. Applications, “Applied Social Research Methods Series”, Volume 26, Newbury Park, London: Sage Publications, 1991.

[13] Ellis, D., Witt, A., The Measurement of Perceived Freedom in Leisure, “Journal of Leisure Research”, 1984, 18, pp. 81-95.

[14] Ewert, A., Why People Climb:The Relationship of Participant Motives and Experience Level to Mountaineering, “Journal of Leisure Research”, 1985, 17, pp. 241-250.

[15] Frederick, C., Ryan, R., Differences in Motivation for Sport and Exercise and their Relations with Participation and Mental Health, “Journal of Sport Behavior”, 1993, 16, pp. 124-145.

[16] Gerson, R., Members for Life: Proven Service and Retention Strategies for Health – Fitness and Sports Clubs, Champaign, IL: Human Kinetics Publishers 1999.

[17] Gould, D., Feitz, D., Weiss, M., Motives for Participation in Competitive Youth Swimming, “International Journal of Sport Psychology”, 1985, 16, pp. 126-140.

[18] Graef, R., Csikzentmihalyi, M., Giannino, K., Measuring Intrinsic Motivation in Everyday Life, “Leisure Studies”,1983, 2, pp. 155-168.

[19] Iso-Ahola, E., Motivation for Leisure, (in:) E. Jackson, T. Burton, eds, Understanding Leisure and Recreation, Mapping the Past, Charting the Future, State College, PA: Venture Publishing, 1989, pp. 247-280.

[20] Iso-Ahola, S., Weissinger, E., Leisure and Boredom, “Journal of Social and Clinical Psychology,” 1987, 5, pp. 356-364.

[21] Jackson, E., Crawford, D., Godbey, G., Negotiation of Leisure Constraints, “Leisure Sciences”, 1993, 15, pp. 1-11.

[22] Markland, D., Hardy, L., The Exercise Motivations Inventory: Preliminary Development and Validity of a Measure of Individuals' Reasons for Participation in Regular Physical Exercise, “Personality and Individual Differences”, 1993, 15, pp. 289-296.

[23] Mathes S., McGivern, A., Schneider, C., The Influence of Participation and Gender on Employees' Motives for Involvement in a Corporate Exercise Program, “Journal of Sport Management”, 1992, 6, pp. 1-14.

[24] Ragheb, G.M. & Tate, L.R., A Behavioural Model of Leisure Participation, based on Leisure Attitude, Motivation and Satisfaction, “Leisure Studies”,1993, 12, pp. 61-70.

[25] Silberstein, L.R., Striegel-Moore, R.H., Timko, C., Rodin, J., Behavioral & Psychological Implications of Body Satisfaction: Do men and Women Differ?, “Sex Roles”, 1988, 19, pp. 219-232.

[26] Tabachnick, B., Fidell, L., Using Multivariate Statistics (2nd Edition), New York: Harper Collins Publishers 1989.

[27] Torkildsen, G., Leisure and Recreation Management (3rd Edition). London – Madras: E & FN Spon, 1992.

[28] Wankel, M. & Kreisel, J., Factors Underlying Enjoyment of Youth Sports: Sports and Age Group Comparisons, “Journal of Sport Psychology”, 1985, 11, pp. 355-366.

[29] Weissinger, E., The Development, Reliability, and Validity of a Scale to Measure Intrinsic Motivation in Leisure, “Journal of Leisure Research”, 1995, 27, pp. 379-400.