Effect of perceived sport competence on intentions to exercise among adolescents: mediating or moderating?

Li, Weidong ; Shen, Bo ; Rukavina, Paul B. 等

**********

The prevalence of obesity and physical inactivity has increased to an epidemic scale over the past 20 years in the United States. School physical education (PE) has been identified as a viable avenue to respond to the twin epidemic of obesity and physical inactivity. McKenzie (200l) has argued that school PE curriculum should be health-related, focusing on the development of healthy, physically active lifestyles. With this in mind, an important line of research is to identify motivational strategies to maximize student motivation and engagement in PE and examine how those motivational variables are related to out-of-school exercise motivation and physical activity.

Achievement goals govern and mediate an individual's motivational, affective and behavioral responses in achievement settings (Dweck & Elliot, 1983; Nicholls, 1984a, 1984b, 1989). Students set goals for themselves when entering an achievement context, and there are two goal dimensions operating in an achievement context: a task goal orientation versus an ego goal orientation (Dweck & Elliot, 1983; Nicholls, 1984a, 1984b, 1989). A task goal orientation focuses on mastery learning and self-improvement, whereas an ego goal orientation focuses on social comparisons and demonstrating superiority to others. Compared with ego goal orientations, task goal orientations are more likely to be linked to many adaptive motivational patterns in PE, including being intrinsically motivated (e.g., Goudas, Biddle, & Fox, 1994; Standage, Duda, & Ntoumanis, 2003; Xiang, McBride, Bruene, & Liu, 2007), high levels of effort and persistence (e.g., Agbuga & Xiang, 2008; Thomas & Barron, 2006), increased enjoyment (e.g., Thomas & Barron, 2006), high levels of perceived competence (e.g., Standage et al., 2003; Sproule, Wang, Morgan, McNeil, & McMorris, 2007; Thomas & Barron, 2006), and positive affects (e.g., Ntoumanis & Biddle, 1999).

The achievement goal theory also suggests that ego goal orientations and perceived competence interact to affect motivational, affective, and cognitive outcomes (Nicholls, 1989). Recent research has indicated that individuals who are ego-oriented do not necessarily exhibit maladaptive motivational and behavioral responses (Elliot & Harachiewicz, 1996; Harachiewicz, Barron, & Elliot, 1998). High levels of perceived competence can protect ego-oriented individuals from suffering such negative effects. It is hypothesized that individuals with high ego goal orientations would display adaptive outcomes when their perceived competence is high in the face of difficulties or challenges. An ego goal orientation, especially with low levels of perceived competence, is likely to be related to maladaptive motivational and behavioral responses (e.g., Kavussanu & Roberts, 1996).

Two different interpretations of the role of perceived competence on the basis of Nicholls' theory (1989) have existed in the literature: moderator versus mediator. The terms of moderation and mediation differ from each other in many ways (Baron & Kenny, 1986). A moderator is a variable that influences the direction and/or strength of the relationship between an independent variable and a dependent variable. A mediator is a variable that accounts for the relationship between an independent variable and a dependent variable. A moderating effect can be tested by entering an interaction term between the moderator and the independent variable. A mediating effect can be tested by following the three regression steps recommended by Baron and Kenny (1986). Some researchers have argued that perceived competence moderates the effects of goal orientations on motivational and cognitive patterns. Findings from this line of research in PE settings have demonstrated a mixed pattern of results (e.g., Cury, Sarrazin, & Famose, 1997; Gill & Willams, 2008; Ommundsen & Pedersen, 1999; Standage et al., 2003; Vlachopoulus & Biddle, 1997; Williams & Gill, 1995). For example, Cury et al. (1997) found that perceived ability moderated the effect of ego goal orientations on investment learning (time taken for practice), but failed to moderate the effect of ego goal orientations on effort attributions.

Other researchers have interpreted the role of perceived competence as a mediating effect (Biddle, Soos, & Chatzisarantis, 1999a, 1999b; Lintunen, Valkonen, Leskinen, & Biddle, 1999; Sproule et al., 2007) and the findings from those studies provided evidence supporting a complete mediating effect of perceived sport competence. For example, a study by Biddle et al. (1999b) employed a structural equation modeling technique to test a conceptual model of the mediating relationships among sport ability beliefs, goal orientations, perceived sport competence, and intentions to exercise in Hungarian adolescents. The findings revealed that perceived sport competence completely mediated the relationship between ego goal orientations and intentions, and partially mediated the relationship between task goal orientation and intention to exercise. Biddle et al. (1999b) explained that this partial mediating link may be due to the way that perceived sport competence was defined and measured. They argued that the item statements were task-oriented and could be viewed as self-referenced.

Sports are one of the important components of school PE curriculum and after-school activities for youth. Research has also shown positive associations between perceived competence in PE and leisure-time physical activity intentions (Shen, McCaughtry, & Martin, 2007) and between perceived sport competence and physical activity in adolescents (Sallis, Prochaska, & Taylor, 2000). However, the moderating effect of perceived sport competence on the relationship between achievement goals in PE and exercise motivation and physical activity has been unexplored in the literature. Secondly, the mixed interpretation of the roles of perceived competence guarantee an exploration of whether perceived competence mediates and/or moderates the effects of ego goal orientations on motivational and behavioral responses. Finally, with the call for PE curricula to focus on the development of healthy, physically active lifestyles, research is needed on whether goal orientations in PE can be inductive to the development of healthy, physically active lifestyles among American children and adolescents.

The purpose of the present study was, therefore, to examine whether perceived sport competence mediates or moderates the effects of ego goal orientations on intentions to exercise among urban adolescents. The motivational and cognitive construct of intentions to exercise was selected because this construct is one of the immediate determinants of health related behaviors, such as eating or exercise (Ajzen & Manstead, 2007) and is significantly correlated with physical activity in adolescents (Sallis et al., 2000). The literature on achievement goals has shown that task goal orientations are linked to adaptive behavioral responses regardless of levels of perceived competence and high levels of perceived competence can protect ego-oriented individuals from suffering maladaptive behavioral responses (Elliot & Harachiewicz, 1996; Harachiewicz et al., 1998; Kavussanu & Roberts, 1996). Therefore, we hypothesized that perceived sport competence would completely mediate the effect of ego goal orientations on intentions to exercise, and task goal orientations would directly predict intentions to exercise, as reflected in Figure 1.

[FIGURE 1 OMITTED]

Methods

Participants

Participants were 248 students (106 males and 142 females; 198 African Americans, 7 European Americans, 41 others, and 2 missing) from an urban middle school in the southern United States. The participants' ages ranged from 9 to 15 years (M= 12.1, SD=0.99). Two teachers with 5 and 10 years teaching experience taught PE classes using a direct teaching approach. The PE curriculum covered a unit of physical fitness and a variety of sports. Informed consent was obtained from all participants and their parents according to the University Institutional Review Board.

Instrumentations

Achievement goal orientations. The 13-item Task and Ego Orientation in Sport Questionnaire (TEOSQ; see Duda & Whitehead, 1998) was used to assess participants' task and ego goal orientations. The stem of the questions is "I feel really successful in PE class when ...." An example item for ego goal orientations is "I tan do better than my friends." An example item for task goal orientations is "I learn a new skill by trying hard." The response scale ranges from "strongly disagree" (1) to "strongly agree" (6).

Perceived sport competence. A 5-item questionnaire, based on the Sport Competence items from Fox and Corbin's (1989) Physical Self-Perceptions Profile, was developed to assess participants' perceived sport competence to align with the school PE curriculum. The measurement of perceived sport competence contains items assessing both task and ego goal orientations and all the items were phrased in first person 'T' rather than "some people." Example items include "I feel that I am better than others of my age at sport" and "I feel very confident that I can do really well when it comes to playing sport." The response scale ranges from "strongly disagree" (1) to "strongly agree" (6).

Intentions to exercise. Based on Ajzen (1991), a 3-item questionnaire was used to assess participants' intentions to exercise. Those items included: "I intend to exercise regularly next month," "I will try to exercise regularly next month," and "I plan to exercise regularly next month." As suggested by Ajzen (1991), the response scale was designed to range from "strongly disagree" (1) to "strongly agree" (7).

Procedure

Data were collected over a period of three days, with 80-90 students participating each day. Upon their arrival at school cafeteria, participants were seated at 13 different tables. There were 4-7 participants at each table. A trained graduate assistant instructed participants to complete the demographics, TEOSQ, perceived sport competence, intentions to exercise, and other questionnaires for research purposes. Directions for the completion of questionnaires and all the items of questionnaires were read out to participants using a microphone. Participants were also encouraged to ask any questions. The primary investigator and another trained graduate student supervised the administration of questionnaires and were available to address any questions.

Data Analysis

All negatively worded items in the scale of perceived sport competence were reversed before the data analysis. Internal consistency reliability for all the subscales from the questionnaires was assessed by using the Cronbach's (1951) coefficient alpha. Confirmatory factor analysis with maximum likelihood estimation using the SAS PROC CALIS was conducted to validate the measurements of intentions to exercise and perceived sport competence. The KURTOSIS option was used to compute univariate and multivariate measures of Kurtosis and identify multivariate outliers. Pearson correlations were used to examine the relationships between participants' goal orientations in PE, perceived sport competence, and intentions to exercise.

Both regression model and Structural Equation Models (SEM) are relevant methods in assessing out research questions. The regression model was selected in the present study because of l) its popularity in social psychological research (e.g., Baron & Kenny, 1986; Fritz & MacKinnon, 2007; MacKinnon, 2008; MacKinnon, Krull, & Lockwood, 2000); and 2) being more powerful in handling a non-linear relationship, multicollinearity, outliers, heterogeneity, and polynomial relationships than SEM, which cannot identify and handle those violations of the major assumptions (Gefen, Straub, & Boudreau, 2000; Hair, Anderson, Tatham, & Black, 1998; Neter, Wasserman, & Kutner, 1990).

A moderating effect can be tested by entering an interaction term between the moderator and the independent variable. A mediating effect tan be tested by following the three regression steps recommended by Baron and Kenny (1986). A mediating relation exists if (1) the independent variable significantly predicts the dependent variable; (2) the independent variable significantly predicts the mediator; (3) the mediator significantly predicts the dependent variable after controlling the effect of the independent variable; and (4) to establish a complete mediation, the effect of the independent variable on the dependent variable should be close to zero after controlling the mediator (e.g., Baron & Kenny, 1986).

Recently, researchers have argued that only steps 2 and 3 are essential steps in establishing mediation (e.g., Fritz & MacKinnon, 2007; MacKinnon, 2008; MacKinnon, Krull, & Lockwood, 2000). In the present study, these new recommended two regression steps were used to examine the mediating role of perceived sport competence. Because multiplicative terms can produce high levels of multicollinearity, the scores for perceived sport competence and ego goal orientations were centered. The centered scores for these two variables were calculated by using the original scores from each variable minus the mean of that variable. The product of perceived sport competence and ego goal orientations and the product of perceived sport competence and task goal orientations were computed for each subject by using the centered scores (Jaccard, Turrisi, & Wan, 1990). To examine whether task goal orientations directly significantly predict intentions to exercise, both the moderating and mediating effects of perceived sport competence on the relationship between task goal orientations and intentions to exercise were tested. If test statistics indicates no significant moderating and mediating effects, it would be concluded that task goal orientations are directly related to intentions to exercise.

Moderating effect of perceived sport competence. To test the moderating effect of perceived sport competence, a multiple regression with intentions to exercise as a dependent variable and perceived sport competence, ego goal orientations, and the interaction between those two variables as independent variables was conducted. To test whether perceived sport competence would moderate the effect of task goal orientations on intentions to exercise, a multiple regression with intentions to exercise as a dependent variable and perceived sport competence, task goal orientations, and the interaction between those two variables as independent variables was conducted.

Mediating effect of perceived sport competence. Perceived sport competence would mediate the effect of ego goal orientations on intentions to exercise if (1) ego goal orientations significantly predict perceived sport competence; and (2) perceived sport competence significantly predicts intentions to exercise after controlling ego goal orientations. To establish a complete mediation, the effect of ego goal orientations on intentions to exercise should be zero after controlling perceived sport competence. Two regressions, therefore, were conducted to investigate whether the effect of ego goal orientations on intentions to exercise would be mediated by perceived sport competence. Perceived sport competence was regressed on ego goal orientations to examine whether ego goal orientations would significantly predict perceived sport competence. Intentions to exercise was regressed on both ego goal orientations and perceived sport competence to examine whether perceived sport competence would significantly predict intentions to exercise after controlling ego goal orientations.

Perceived sport competence would mediate the effect of task goal orientations on intentions to exercise if (1) task goal orientations significantly predict perceived sport competence; and (2) perceived sport competence significantly predicts intentions to exercise after controlling task goal orientations. To establish a complete mediation, the effect of task goal orientations on intentions to exercise should be zero after controlling perceived sport competence. Two regressions were conducted to investigate whether the effect of task goal orientations on intentions to exercise would be mediated by perceived sport competence. Perceived sport competence was regressed on task goal orientations to examine whether task goal orientations would significantly predict perceived sport competence. Intentions to exercise was regressed on both task goal orientations and perceived sport competence to examine whether perceived sport competence would significantly predict intentions to exercise after controlling task goal orientations.

Results

Means, standard deviations and Cronbach's alpha for all the variables are presented in Table 1. All subscales demonstrated acceptable levels of internal consistency ranging from 0.67-0.87. Multiple fit indices and Root Mean Square Error of Approximation (RMSEA) were used to determine how well the hypothesized model fit the observed data. According to Hu and Bentler (1999), a cut-off value of .95 was used for the fit indices and a cut-off value of .06 was used for RMSEA.

The confirmatory factor analysis on the scale of perceived sport competence indicated that the hypothesized one factor structure provided a sufficiently good fit to the observed data (CFI = 1; NFI = 0.99; RMSEA = 0). An examination of univariate Skewness (ranging from -1.32 to .14), univariate Kurtosis (ranging from -1.07 to 1.57), and Mardia's multivariate Kurtosis (9.84) indicated some evidence of violations of multivariate normality. The confirmatory factor analysis on the measurement of intentions to exercise indicated that the hypothesized one factor structure provided a sufficiently good fit to the observed data (CFI = 1; NFI = 1; RMSEA = 0). An examination of univariate Skewness (ranging from -1.45 to -1.10), univariate Kurtosis (ranging from .82 to 2.19), and Mardia's multivariate Kurtosis (27.33) indicated evidence of violations of multivariate normality. However, there is growing evidence supporting that the maximum likelihood estimation seems to be fairly robust against moderate violations (Anderson & Gerbing, 1988; Joreskog & Sorbom, 1989; Hatcher, 1994; Hoyle & Panter, 1995). The aggregate scores were then calculated by summing all responses for each scale of the measurements and divided by the relevant number of items per scale.

The correlational analysis indicated that intentions to exercise were positively related to task goal orientations, r (250) = 0.30,p < .0001, and perceived sport competence, r (247) = 0.21, p < .0012. Perceived sport competence was positively associated with task goal orientations, r (244) = 0.14,p < .03, and ego goal orientations, r (245) = 0.34,p < .0001. The correlational matrix is presented in Table 2.

Moderating effects of perceived sport competence

The regression analysis to test whether perceived sport competence would moderate the effect of ego goal orientations on intentions to exercise showed that perceived sport competence positively predicted intentions to exercise, t (241) = 3.54, p < .0005, but not ego goal orientations t (241) = -1.1, p = .27 and the interaction between perceived sport competence and ego goal orientations, t (241) = -.03, p = .98. The total variance accounted for was .05.

The regression analysis to test whether perceived sport competence would moderate the effect of task goal orientations on intentions to exercise showed that perceived sport competence, t (240) = 2.37, p < .02, and task goal orientations, t (240) = 4.47, p < .0001, positively predicted intentions to exercise, but not the interaction between perceived sport competence and ego goal orientations, t (240) = 1.1, p = .27. The total variance accounted for was. 12.

Mediating effect of perceived sport competence

To test the mediating effect of perceived sport competence on ego goal orientations and intentions to exercise, two regressions were conducted. Regressing perceived sport competence onto ego goal orientations showed that ego goal orientations positively predicted perceived sport competence, t (243) = 5.56, p < .0001. The total variance accounted for was .11. Regressing intentions to exercise onto perceived sport competence and ego goal orientations indicated that perceived sport competence positively predicted intentions to exercise, t (242) = 3.55, p < .0005, and ego goal orientations failed to predict intentions to exercise, t (242) = -1.11, p = .27. The regression coefficient for ego goal orientations was close to zero with a value of .07. The total variance accounted for was .05.

To test the mediating effect of perceived sport competence on task goal orientations and intentions to exercise, two regressions were conducted. Regressing perceived sport competence onto task goal orientations showed that task goal orientations positively predicted perceived sport competence, t (248) = 5.02, p <. 0001. The total variance accounted for was .09. Regressing intentions to exercise onto perceived sport competence and task goal orientations indicated that perceived sport competence, t (241) = 2.63, p < .009, and task goal orientation, t (241) = 4.39,p < .0001, positively predicted intentions to exercise. The total variance accounted for was. 11. The regression coefficient for task goal orientations decreased from .47 to .42 after controlling perceived sport competence.

Discussion

The purpose of the present study was to investigate whether perceived sport competence mediates or moderates the effects of ego and task goal orientations on intentions to exercise among urban adolescents. The results of the present study showed that perceived sport competence did not moderate the effects of either task or ego goal orientations on intentions to exercise. Perceived sport competence completely mediated the relationship between ego goal orientations and intentions, and partially mediated the relationship between task goal orientations and intentions. Both the present study and Biddle et al. (1999b) study provided evidence supporting the complete and partial mediation of perceived sport competence on goal orientations. The findings provided some additional empirical evidence supporting that individuals who are ego-involved do not necessarily exhibit maladaptive motivational and behavioral responses (Elliot & Harachiewicz, 1996; Harachiewicz et al., 1998). For those with high levels of perceived competence, they do not surfer such negative effects. The partial mediation of perceived sport competence on task goal orientations may be explained by recent findings from achievement goals using a goal profile approach. For example, Wang, Chatzisarantis, Spray, and Biddle (2002) found that adolescent participants with a profile of high task/high ego/high perceived competence had significantly higher motivation than those with a profile of low task/low ego/low perceived competence. Participants with a profile of moderate task/low ego/moderately low competence had higher motivation than those with a profile of low task/low ego/low perceived competence. Findings from the present study and the Biddle et al. (1999b) may suggest that individuals with profiles of moderate or low task/low ego goal orientations may not display maladaptive motivational and behavioral patterns when they have high levels of perceived competence. The present study took an initial step to examine the mediating or moderating effect of perceived sport competence. Further research is needed to confirm or refute the results of the present study.

The present study suggests that school PE may function as a transmitter of motivation and learning from PE to exercise intentions and behaviors outside of school. A task goal orientation is associated with high levels of intentions to exercise, and perceived sport competence partially mediated the effect of task goal orientations on intentions to exercise. Ego goal orientations are linked with high levels of intentions to exercise through the mechanism of high levels of perceived competence. Research on achievement goals has demonstrated that students set goals for themselves when entering an achievement context and teachers can reinforce or alter those dispositional goals by manipulating the motivational learning climates. Students' goal involvement is a function of their dispositional goals (goal orientations) (Dweck & Elliot, 1983; Nicholls, 1984a, 1984b, 1989) and situational goal cues (the perceived motivational climates) (Ames; 1992; Ames & Ames, 1989; Ames & Archer, 1988). In a task-involved motivational climate, where mastery learning and self-improvement are emphasized, students are more likely to adopt a task goal orientation. On the other hand, in an ego-involved motivational climate, where teachers focus on social comparisons and demonstrating superiority to others, students are more likely to embrace an ego goal orientation. It is suggested that PE teachers create a task-involved and competence-focused motivational climate. Strategies to use include emphasizing task mastery and personal improvement, designing developmentally appropriate task progressions to maximize students' successful experience, instilling the beliefs that ability can be changed, fostering a caring climate, and empowering students by giving leadership opportunities and choices.

The present study was based on the classic achievement goal theory, which provides a foundation for a further investigation on the mediating relations among goal orientations, perceived sport competence, and intentions to exercise using the 2 x 2 approach-avoidance achievement goals approach. Even though the 2 x 2 approach-avoidance approach has been increasingly used to study student motivation and behaviors, many issues still remain unexplored under the classic achievement goal framework. For example, one question is whether and how the task or ego-involved motivational climate can foster avoidance goals. The percentage of variance explained by different independent variables of interest was relatively small in the present study. A further examination of the mediating effects of perceived competence on task and ego goal orientations on motivational and behavioral responses is needed. The present study was cross-section in nature and a causal-effect relationship was relatively arbitrary. Another limitation was the internal-consistency coefficient alpha for the measure of perceived sport competence, which was a bit below what is generally considered as adequate (.70). Future studies should focus on improving the internal consistency reliability coefficient for the measure of perceived sport competence. Based on the findings of the present study and the previous cross-sectional studies, there is a need for experimental studies to investigate the effects of mastery-involved, competence-enhanced climates on students' perceived sport competence, intentions to exercise, and actual exercise behaviors.

References

Agbuga, B., & Xiang, P. (2008). Achievement goals and their relations to self-reported persistence/effort in secondary physical education: A trichotomous achievement goal framework. Journal of Teaching in Physical Education, 27, 179-191.

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211.

Ajzen, I., & Manstead, A. S. R. (2007). Changing health-related behaviors: An approach based on the theory of planned behavior. In K. van den Bos, M. Hewstone, J. de Wit, H. Schut & M. Stroebe (Eds.), The scope of social psychology: Theory and applications (pp. 43-63). New York: Psychology Press.

Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84, 261-271.

Ames, C., & Ames, R. (Eds.). (1989). Research on motivation in education: Vol. 3. Goals and cognition. San Diego: Academic Press.

Ames, C., & Archer, J. (1988). Achievement goals in the classroom: Students' learning strategies and motivation processes. Journal of Educational Psychology, 80, 260-267.

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411-423.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and social psychology, 51, 1173-1182.

Biddle, S., Soos, I., & Chatzisarantis, N. (1999a). Predicting physical activity intentions using goal perspectives and self-determination theory approaches. European Psychologist, 4, 83-89.

Biddle, S., Soos, I., & Chatzisarantis, N. (1999b). Predicting physical activity intentions using a goal perspectives approach: A study of Hungarian Youth. Scandinavian Journal of Medicine and Science in Sport, 9, 353-357.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 207-234.

Cury, F., Sarrazin, P., & Famose, J. P. (1997). Achievement goals, perceived ability and active search for information. European Yearbook of Sport Psychology, 1, 166-183.

Duda, J. L., & Whitehead, J. (1998). Measurement of goal perspectives in the physical domain. In J. L. Duda (ed.), Advances in sport and exercise psychology measurement (pp21-48). Morgantown, WV: Fitness Information Technology.

Dweck, C. S., & Elliott, E. S. (1983). Achievement motivation. In P.H. Mussen (Ed.), Handbook of child psychology (3rd ed.) (Vol. 4, pp. 643-691). New York: Wiley.

Elliot, A. J., & Harachiewicz, J. M. (1996). Approach and avoidance achievement goals and intrinsic motivation: A mediational analysis. Journal of Personality and Social Psychology, 70, 461-475.

Fox, K. R., & Corbin, C. B. (I 989). The Physical Self-Perceptions Profile: Preliminary development and validation. Journal of Sport and Exercise Psychology, 11, 408-430.

Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect. Psychological Science, 18, 233-239.

Gefen, D., Straub, D., & Boudreau, M. (2000). Structural equation modeling and regressions: Guidelines for research practice. Communications of the Association of Information Systems, 4, 1-77.

Gill, D. L., & Williams, L. (2008). Psychological dynamic of sport and exercise. Champaign, IL: Human Kinetics.

Goudas, M., Biddle, S. J. H., & Fox, K. R. (1994). Perceived locus of causality, goal orientations, and perceived competence in school physical education classes. The British Journal of Educational Psychology, 64, 453-463.

Hair, J. F., Jr., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis with readings (5th ed.). Englewood Cliffs, N J: Prentice Hall.

Hatcher, L. (1998). A step-by-step approach to using the SAS system for factor analysis and structural equation modeling. Cary, NC: SAS Institute Inc.

Harackiewicz, J. M., Barron, K. E., & Elliot, A. J. (1998). Rethinking achievement goals: When are they adaptive for college students and why? Educational Psychologist, 33, 1-21.

Hoyle, R. H., & Panter, A. T. (1995). Writing about structural equation models: A second look. In Hoyle R. H. (ed.), Structural equation modeling: Concepts, issues and application (pp 159-176). Thousand Oaks, CA: Sage.

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

Jaccard, J., Turrisi, R., & Wan, C. K. (1990). Interaction effects in multiple regression. Newbury Park, CA: SAGE Publications, Inc.

Joreskog, K.G., & Sorbom, D. (1989). LISREL 7: A guide to the program and application. Chicago: SPSS Inc.

Kavussanu, M., & Roberts, G. C. (1996). Motivation in physical activity contexts: The relationship of perceived motivational climate to intrinsic motivation and self-efficacy. Journal of Sport and Exercise Psychology, 18, 264-281.

Lintunen, T., Valkonen, A., Leskinen, E., & Biddle, S. (1999). Predicting physical activity intentions using a goal perspectives approach: A study of Finnish youth. Scandinavian Journal of Medicine and Science in Sport, 9, 344-352.

MacKinnon, D. P., Krull, J. L., & Lockwood, C. M. (2000). Equivalence of the mediation, confounding, and suppression effect. Prevention Science, 1, 173-181.

MacKinnon, D. P. (2008). Introduction to statistical mediation analysis. Mahwah, NJ: Erlbaum.

McKenzie, T. L. (2001). Back to the future: Health-related physical education. In P. Ward and P. Doutis (Eds.), Physical education for the 21st century (pp. 113-131). University of Nebraska, Lincoln, NE.

Neter, J., Wasserman, W., & Kutner, M. H. (1990). Applied linear statistical models: Regression, analysis of variance, and experimental design. Boston, MA: Irwin.

Nicholls, J. G. (1984a). Conceptions of ability and achievement motivation. In R. Ames & C. Ames (Eds.), Research on motivation in education: Student motivation (Vol. 1, pp. 39-73). New York: Academic Press.

Nicholls, J. G. (1984b). Achievement motivation: conceptions of ability, subjective experience, task choice, and performance. Psychological Review, 91, 328-346,

Nicholls, J. G. (1989). The competitive ethos and democratic education. Cambridge, MA: Harvard Univ. Press.

Ntoumanis, N., & Biddle, S. J. H. (1999). Affect and achievement goals in physical activity: A meta-analysis. Scandinavian Journal of Medicine and Science in Sport, 9, 315-332.

Ommundsen, Y., & Pedersen, B. H. (1999). The role of achievement goal orientations and perceived ability upon somatic and cognitive indices of sport competition trait anxiety. Scandinavian Journal of Medicine and Science in Sport, 9, 333-343.

Sallis, J. F., Prochaska, J. J., & Taylor, W. C. (2000). A review of correlates of physical activity of children and adolescents. Medicine Science and Sport Exercises, 32, 963-975.

Shen, B., McCaughtry, N., & Martin, J. (2007). The influence of self-determination in physical education on leisure-time physical activity behavior. Research Quarterly for Exercise and Sport, 78, 328-338.

Sproule, J., Wang, C. K. J., Morgan, K., McNeill, M., & McMorris, T. (2007). Effects of motivational climate in Singaporean physical education lessons on intrinsic motivation and physical activity intentions. Personality and Individual Differences, 43, 1037-1049.

Standage, M., Duda, J. L., & Ntoumanis, N. (2003). Predicting motivational regulations in physical education: The interplay between dispositional goal orientations, motivational climate and perceived competence. Journal of Sports Sciences, 21, 631-647.

Thomas, J. A., & Barron, K. E. (2006). A test of multiple achievement goal benefits in physical education activities. Journal of Applied Sport Psychology, 18, 114-135.

Vlachopoulos, S., & Biddle, S. J. H. (1997). Modeling the relation of goal orientations to achievement-related affect in physical education: Does perceived ability matter? Journal of Sport & Exercise Psychology, 19, 169-187.

Wang, C. K. J., Chatzisarantis, N. L. D., Spray, C. M., & Biddle, S. J. H. (2002). Achievement goal profiles in school physical education: Differences in self-determination, sport ability beliefs, and physical activity. British Journal of Educational Psychology, 72, 433-445.

Williams, L., & Gill, D. (1995). The role of perceived competence in the motivation of physical activity. Journal of Sport and Exercise Psychology, 17, 363-378.

Xiang, P., McBride, R., Bruene, A., & Liu, Y. (2007). Achievement goal orientation patterns and fifth graders' motivation in physical education running program. Pediatric Exercise Science, 19, 179-191.

Weidong Li

Ohio State University

Bo Shen

Wayne State University

Paul B. Rukavina

Adelphi University

Haichun Sun

University of South Florida

Address Correspondence to: Weidong Li, Ph.D., Assistant Professor, The Ohio State University, College of Education and Human Ecology, School of Physical Activity and Educational Services, Sport and Exercise Education, PAES Building A270, 305 West 17th Avenue, Columbus, OH 43210-1224. Phone: (614) 247-6506. Fax: (614) 688-4885. Email: li.832@osu.edu.

Table 1. Means, Standard Deviations, and Cronbach's Alpha for all the
Variables

                             Means      Standard      Cronbach's
                                       Deviations       Alpha

Intentions                    5.69        1.22           .87

Perceived sport competence    4.47        .94            .67
Task goal orientations        4.98        .80            .84
Ego goal orientations         3.48        1.22           .82

Table 2. Correlational coefficients among all the variables.

                                   1        2         3         4

1. Intentions                     --      .21 *     .30 *      .01
2. Perceived sport competence              --       .14 *     .34 *
3. Task goal orientations                            --        .10
4. Ego goal orientations                                       --