Journal of Sports Sciences, 2001, 19, 397± 409

Empirical links between achievement goal theory and self-determination theory in sport NIKOS NTOUMANIS* School of Leisure and Sport Studies, Leeds Metropolitan University, Beckett Park Campus, Leeds LS6 3QS, UK

Accepted 12 February 2001

The aim of this study was to examine the empirical links between achievement goal theory and self-determination theory in sport. Addressing theoretical and methodological limitations of previous research, the study tested the independent and interactive eþ ects of goal orientations and perceived competence on seven motivational variables with diþ erent degrees of self-determination. Regression analyses of data collected from 247 British university students showed that task orientation predicted motivational variables with high self-determination. In contrast, ego orientation predicted motivational variables with low self-determination. Perceived competence predicted both high self-determined and low self-determined motivational variables. A signi® cant interaction emerged between task and ego orientations in predicting external regulation. The results suggest the adaptive role of task orientation in facilitating self-determined motivation in sport. However, the ® ndings are not conclusive, as the variance explained in most analyses was relatively small. Suggestions are oþ ered for a more comprehensive empirical testing of the links between the two theories. Keywords: goal orientations, interaction, perceived competence, self-determination theory.

Introduction Two well-known theories of motivation are achievement goal theory (Nicholls, 1989) and self-determination theory (Deci and Ryan, 1985; Ryan and Deci, 2000). These theories have extensive applications in sport and exercise and have facilitated our understanding of motivated behaviour and related cognitive, aþ ective and behavioural outcomes (for reviews, see Vallerand and Losier, 1999; Duda, in press). Although other theorists (e.g. Dweck and Leggett, 1988) have developed similar achievement goal theories, the work of Nicholls is reviewed here, as it has been the most in¯ uential in sport and exercise. The present study tries to provide empirical evidence for the links between key constructs of achievement goal theory and self-determination theory. Speci® cally, it aims to show how diþ erent achievement criteria and perceived competence predict independently and in interaction diþ erent degrees of self-determination. Such evidence can be used to facilitate future theoretical eþ orts to integrate the two theories. From an applied perspective, an under* e-mail: [email protected]

standing of how diþ erent achievement criteria and perceived competence predict diþ erent degrees of selfdetermination can help practitioners to oþ er a more comprehensive framework of psychological support. This paper presents a brief overview of the two theories and discusses their theoretical links. Furthermore, I review the theoretical and methodological limitations of empirical studies that have looked at the relationships between goal orientations and self-determination. I show how these limitations can be addressed to advance our understanding of the two theories.

Achievement goal theory Achievement goal theory postulates that, in the achievement setting of sport, two goal orientations are evident: task and ego. These goal orientations are linked to two diþ erent conceptions of ability. The ® rst is the undiþ erentiated (or less diþ erentiated) conception of ability, evident in most individuals below the age of about 12 years, whereby eþ ort, luck and task diý culty cannot be distinguished from ability (or they are partly distinguished) as causes of outcomes (Nicholls, 1989).

Journal of Sports Sciences ISSN 0264-0414 print/ISSN 1466-447X online Ó http://www.tandf.co.uk/journals

2001 Taylor & Francis Ltd

398 For example, high eþ ort is assumed to result in more learning, which is interpreted as indicative of high ability. The second conception of ability, the differentiated conception, is observed at the age of about 12 years, when, through various cognitive stages, eþ ort, luck and task diý culty are diþ erentiated as causes of outcomes. Nicholls (1989) also labelled the diþ erentiated conception of ability as `ability as current capacity’ , because children who hold this conception realize that eþ ort can increase performance only up to the limits of their present capacity on the task at hand. After the age of about 12 years, individuals can adopt the diþ erentiated or undiþ erentiated conception of ability depending on the salience of situational cues (Nicholls, 1989). If the situational emphasis is on competition and the public evaluation of skills, a differentiated conception of ability will be observed. In contrast, when the situational focus is on learning and individual improvement, individuals will use a less diþ erentiated conception. A diþ erentiated conception of ability leads to normative evaluations and standards (e.g. winning), because individuals strive to achieve success by demonstrating superior ability; an ego goal orientation is then salient. Alternatively, a less diþ erentiated conception of ability leads to self-referenced evaluations (e.g. personal improvement and learning), because the emphasis is on success through high eþ ort and learning; a task goal orientation is then evident (Nicholls, 1989). As Duda (in press) has emphasized, goal orientations should not be confused with one’ s goal-setting eþ orts; goal orientations are dispositional tendencies re¯ ecting diþ erent ways of cognitively processing achievement in a given activity. Duda (in press) also distinguished between goal orientations and states of goal involvement. Speci® cally, task involvement and ego involvement represent transitory and situationspeci® c goal states that are the outcomes of the interaction between dispositional goal orientations and motivational climates (i.e. perceptions of situational evaluative cues; see Ntoumanis and Biddle, 1999). A review of the literature by Duda and Whitehead (1998) showed that variations in goal orientations can lead to observed diþ erences in individuals’ cognition, aþ ect and behaviour. These variations have been more or less observed across samples diþ ering in competitive standard, sex, age and culture. For example, high task orientation has been related to pro-social views about the role of sport (Duda, 1989), high enjoyment and interest (Hom et al., 1993), fairplay (Lee et al., 1999) and successful coping with stress (Ntoumanis et al., 1999). Based on Nicholls’ (1989) theory, it has been assumed (but not tested) that these adaptive motivational patterns are evident in all individuals with high task orientation irrespective of their perceived com-

Ntoumanis petence. This is because those with high task orientation do not judge their self-worth based on the adequacy of their ability and the demonstration of superiority. In contrast, according to Nicholls (1989), perceived competence plays an important moderating role in determining the motivational outcomes of ego orientation. This is because individuals with high ego orientation tend to be more preoccupied with the adequacy of their ability. When they perceive themselves as highly competent, they exhibit motivational patterns similar to those with high task orientation (Nicholls, 1989; Hardy, 1997). In contrast, when ego orientation is coupled with low perceived ability, negative outcomes are observed, for example a low investment in learning (Cury et al., 1997) and personally uncontrollable attributions (Vlachopoulos and Biddle, 1997). When an individual with a high ego orientation and low perceived competence also has low task orientation, negative motivational outcomes are even more profound (Fox et al., 1994). Despite the prediction by Nicholls (1989) that ego orientation and perceived competence interact to predict motivational outcomes, such interactive eþ ects have rarely been tested in the literature (for a recent criticism, see Hardy, 1997).

Self-determination theory Variations in achievement goals have been linked to diþ erences in self-determination. Before the available evidence is reviewed, it is informative to present brie¯ y self-determination theory. According to Deci and Ryan (1991), central elements of their theory are the active organism and the social context. The ® rst term refers to how one integrates `new experiences and regulatory processes with one’ s intrinsic self’ (p. 239). Selfdetermination theory argues that, throughout their lives, individuals strive to achieve integration and cohesion of new ideas and interests both within themselves and with others. This is referred to as the `organismic integration process’. The social context is also important because it can support or undermine these integrative processes. Social factors (e.g. cooperation, provision of choice) will facilitate the organismic integration process and will promote self-determination if they satisfy three innate psychological needs for autonomy, competence and relatedness. Deci and Ryan (1991) de® ned the need for autonomy as individuals’ eþ orts to determine their own behaviour. The need for competence was described as individuals striving to experience eþ ectance. Lastly, the need for relatedness was de® ned as individuals’ attempts to have a satisfying and coherent involvement with others. Deci and Ryan (1985) argued that people do not always internalize and integrate external values and

399

Achievement goals and self-determination experiences. In such circumstances, motivation is less self-determined and more controlling or amotivating. Deci and Ryan (1985) proposed a self-determination continuum to describe motivational variables with diþ erent degrees of self-determination. From higher to lower self-determination, these are: intrinsic motivation, extrinsic motivation (integrated regulation, identi® ed regulation, introjected regulation and external regulation) and amotivation. Intrinsically motivated behaviour has the highest selfdetermination, occurs without the incentive of external rewards (e.g. trophies) and is undertaken out of interest in the activity itself rather than the outcomes of the activity. Three dimensions of intrinsic motivation have been identi® ed by Pelletier et al. (1995), which represent the same degree of self-determination. The ® rst is the motivation to know; this describes circumstances in which individuals ® nd satisfaction while learning new things or skills. The second dimension is the motivation to accomplish; this describes the pleasure experienced while individuals strive to improve on already learned skills. Vallerand and Fortier (1998) emphasized that this dimension describes the process of accomplishment rather than the end result. The third dimension of intrinsic motivation refers to the motivation to experience stimulation and pleasant sensations (e.g. a `runner’s high’). Extrinsic motivation refers to activities that are carried out as a means to an end and not for their own sake (Deci and Ryan, 1991). Extrinsic motivation is also multidimensional and comprises four dimensions. The ® rst, integrated regulation, represents the most self-determined form of the internalization process. It refers to behaviours that are performed out of choice to harmonize and bring coherence to diþ erent parts of the self. For example, one may choose to participate in sport, have a good diet and a balanced social life to achieve the overarching goal of a healthy lifestyle. Unfortunately, none of the existing sport motivation questionnaires assesses this dimension. The second dimension, identi® ed regulation, describes behaviours that are highly valued and performed out of choice but, unlike integrated regulation, the choice is limited to the particular activity. For example, some individuals may participate in sport because they believe that this is a good way to develop their ® tness. Both integrated and identi® ed regulation represent self-determined forms of behaviour; however, they are extrinsic, because individuals perform them to achieve personal goals and not for their inherent appeal (Deci and Ryan, 1991). The third dimension of extrinsic motivation is introjected regulation, which refers to behaviours that individuals perform to achieve social recognition or avoid internal pressures and feelings of guilt. The fourth

dimension, external regulation, describes behaviours regulated through external means, such as rewards or constraints. Both introjected regulation and external regulation are considered to be controlling or low selfdetermined types of motivation. The last type described by Deci and Ryan (1991) is amotivation, or a lack of intrinsic or extrinsic motivation. Amotivation is evident when individuals perceive no contingencies between their actions and the end result and question whether they should still be involved in a particular activity. It is viewed as a non-self-determined type of motivation. Reviews by Vallerand and Fortier (1998) and Vallerand and Losier (1999) in sport have indicated that self-determined regulation is more likely to lead to positive cognitive (e.g. high concentration), aþ ective (e.g. enjoyment) and behavioural (e.g. persistence) outcomes, compared to controlling and amotivated regulation.

Links between the two theories There is both theoretical and empirical evidence to show that the two goal orientations are linked to different types of motivation. As discussed earlier, Deci and Ryan (1991) argued that factors which satisfy the needs for autonomy, competence and relatedness will promote self-determined types of motivation. I argue here that high task orientation can ful® l one or more of these needs and, therefore, can enhance self-determined motivation. In contrast, high ego orientation, especially when coupled with low perceptions of competence, is not conducive to the satisfaction of these needs. Speci® cally, there is evidence in the literature to suggest that the controlling features of ego orientation undermine autonomy and foster an external locus of causality, whereas task orientation facilitates the autonomy of behaviour (Ryan, 1982; Brunel, 1999). When individuals have high ego orientation, they are more interested in the anticipated outcomes of a particular activity rather than in the activity itself (Nicholls, 1989). The attainment of social approval and rewards, the demonstration of superior ability and other expected outcomes will control their behaviour. In such circumstances, individuals are less likely to ® nd personal satisfaction in the inherent aspects of an activity and in the process (eþ ort) of task mastery. Therefore, ego orientation is more likely to relate to controlling than to self-determined types of motivation. In support of this argument, Brunel (1999) reported that ego orientation was positively related to introjected regulation and external regulation in 160 French undergraduate students enrolled on a badminton course. Furthermore, Ryan (1982) showed that ego involvement undermined intrinsic interest in hidden-® gures puzzles.

400 In contrast, task orientation facilitates the autonomy of behaviour, because when individuals are taskinvolved, their motivation to perform a task is derived from its intrinsic properties and not from its expected outcomes (e.g. attainment of social approval, demonstration of normative competence). In such circumstances, individuals experience satisfaction while they strive to learn or master sport skills. In an academic context, Butler (1987) showed that mastery (task) goals promoted intrinsic motivation by fostering challengeseeking and task persistence. In sport, positive relationships between task orientation and the diþ erent forms of intrinsic motivation have been reported by Brunel (1999). Task orientation can satisfy not only the need for autonomy, but also the need for competence. Duda (1992, in press) argued that individuals with high task orientation are less likely to feel incompetent in sport than those with high ego orientation. This is because the former use self-referenced criteria to judge their competence. These criteria are more controllable and, therefore, more achievable. In contrast, perceptions of competence are fragile when one has high ego orientation, because competence is construed on the basis of normative criteria (e.g. winning), which are more diý cult and not under one’s control. Of course, it could be suggested that individuals with high ego orientation and high perceived competence can be self-determined, as sport achievement satis® es their need for competence. However, Deci and Ryan (1985) argued that high perceived competence will lead to selfdetermined types of motivation only under conditions of autonomy. As mentioned above, individuals with high ego orientation are not likely to feel or behave autonomously because their conception of ability as capacity will lead to a preoccupation with the adequacy of their competence, which will control their behaviour. Compared to ego orientation, task orientation is unlikely to undermine the need for relatedness. Although at present there is no direct evidence to support this claim, it is logical to assume that the constant interindividual comparison promoted by ego orientation (Duda, 1992) is unlikely to strengthen social links among athletes. In contrast, task orientation downplays normative comparisons and rivalry among athletes and, therefore, it should not undermine their social aý liations.

Similarities and diþ erences between the two theories Goal achievement theory and self-determination theory share certain characteristics. Brie¯ y, both are social cognitive theories of motivation which emphasize that

Ntoumanis the way individuals construe the meaning of an activity will in¯ uence the quality of their engagement in it. However, as Butler (1989) and Ryan and Deci (1989) argued, each theory focuses on a diþ erent body of meaning and perceptions that should be viewed as complementary rather than contradictory. Speci® cally, achievement goal theory focuses primarily on the eþ ects of task and ego involvement on performance and preferences for task diý culty, whereas selfdetermination theory examines the eþ ects of goal involvement on intrinsic motivation. Both theories emphasize the role of social factors (e.g. cooperation, competition, social evaluation, normative feedback) as antecedents of achievement-related behaviour. Achievement goal theory examines how perceptions of task- and ego-promoting climates, created by signi® cant others (e.g. parents, teachers, coaches), interact with dispositional goals to in¯ uence cognition, aþ ect and behaviour in achievement contexts. In contrast, self-determination theory examines how social factors ± that is, human and non-human factors in social environments (Vallerand, 1997) ± impact on human motivation through the mediating variables of competence, autonomy and relatedness. Both theories underline the role of perceived competence in guiding achievement behaviour. However, goal achievement theory distinguishes between a diþ erentiated and a less diþ erentiated conception of ability. In contrast, self-determination theory views competence as a unitary human need, the satisfaction of which will promote self-determined motivation. As a result, self-determination theory fails to explain how social contexts impact on motivation by promoting one rather than another conception of competence (Butler, 1989). On the other hand, the exclusive focus of achievement goal theory on perceptions of competence may lead to an incomplete understanding of motivation in achievement contexts, such as sport. In such contexts, individuals strive also to have a choice in their actions (i.e. a need for autonomy) and feel aý liated to others in a meaningful way (i.e. a need for relatedness). Goal orientations and motivational types can be located within the same stage of generality in Vallerand’ s (1997) hierarchical model of motivation. Vallerand presented a motivational sequence in which social factors in¯ uence diþ erent types of motivation through the three needs for competence, autonomy and relatedness. In turn, the motivational types predict various cognitive, aþ ective and behavioural outcomes. This motivational sequence is posited to exist at three stages of generality: global (i.e. trait), contextual (i.e. context-speci® c) and situational (i.e. state). Most previous studies, as well as the present one, have examined goal orientations and types of motivation at the contextual stage of generality. It should be noted, however, that goal orientations are

401

Achievement goals and self-determination not an explicit part of Vallerand’s (1997) model. This is a limitation of the model, as self-determination theory posits the in¯ uence of both social and individual factors on human motivation. Ferrer-Caja and Weiss (2000) presented a model in which goal orientation mediated the in¯ uence of social factors on the variables of autonomy and perceived competence. Their results show that task orientation mediated the in¯ uence of learning social climate on perceived competence and autonomy. In turn, the latter two variables predicted intrinsic motivation. In contrast, ego orientation mediated the in¯ uence of performance social climate on perceived competence. These ® ndings provide preliminary evidence for the co-existence of goal orientations, perceived competence and self-determination at the same stage of generality, with goal orientations having an antecedent role. As the research evidence in this section shows, the two theories have the potential to explain motivated behaviour in achievement contexts by complementing each other. Indeed, both Butler (1989) and Ryan and Deci (1989) have encouraged attempts to integrate the two theories. However, before such initiatives are undertaken, it is important to ascertain the empirical links between some of the key variables of the two theories. Unfortunately, as the discussion in the next section shows, previous research that has tried to measure such links has been aþ ected by conceptual and methodological limitations.

Empirical testing to date Several studies have looked at the relationship between goal orientations and intrinsic motivation in sport and physical education (PE). Duda et al. (1995) showed that intrinsic interest and eþ ort in sport were positively related to task orientation and negatively related or unrelated to ego orientation in two samples of undergraduate students. Vlachopoulos and Biddle (1996) and Dorobantu and Biddle (1997), in the context of British and Romanian PE respectively, reported that the most intrinsically motivated students were those with high task orientation irrespective of their ego orientation. Furthermore, Newton and Duda (1999) examined potential interactions between goal orientations, motivational climates and perceptions of competence in a sample of junior female volleyball players. None of the interactions were signi® cant. Main eþ ects showed that enjoyment or interest in volleyball was positively predicted by a task-promoting motivational climate and perceived competence and negatively predicted by ego orientation. Eþ ort or importance was positively predicted by task orientation and perceived competence and negatively predicted by ego orientation. Lastly,

Ferrer-Caja and Weiss (2000) used a sample of highschool PE students to test a model in which task and ego goal orientations in¯ uenced intrinsic motivation through the mediation of perceived competence and autonomy. The results for both the male and female sample showed strong direct positive paths linking task orientation with intrinsic motivation and small direct negative paths linking ego orientation with intrinsic motivation. Task orientation also had very small indirect paths through perceived competence and self-determination. However, these studies have several limitations. First, with the exception of Ferrer-Caja and Weiss (2000), they used the Intrinsic Motivation Inventory (McAuley et al., 1989), which measures antecedents (competence) and outcomes (enjoyment or interest, eþ ort, low tension or pressure) of intrinsic motivation rather than intrinsic motivation by itself (Markland and Hardy, 1997; Vallerand and Fortier, 1998). Ferrer-Caja and Weiss (2000) used a measure that, despite the authors’ claims, assessed both intrinsic and extrinsic motivation. Secondly, these studies viewed intrinsic motivation as a unidimensional as opposed to a multidimensional construct. Thirdly, relationships between goal orientations and identi® ed regulation, introjected regulation, external regulation and amotivation were overlooked. Fourthly, Ferrer-Caja and Weiss (2000) assigned a mediator role to perceived competence, although it has a moderator role in Nicholls’ (1989) theory (for a distinction between mediators and moderators, see Baron and Kenny, 1986). Goudas et al. (1994) overcame some of these limitations by using a conceptually appropriate multidimensional measure of motivation (i.e. a PE adaptation of the Academic Self-Regulation Questionnaire; Ryan and Connell, 1989). However, Goudas et al. (1994) combined the diþ erent motivational types into a selfdetermination index. Using this index can be problematic, because it obscures the unique antecedents and outcomes of the diþ erent types of motivation (Vallerand, 1997). The most comprehensive empirical test of the links between achievement goal theory and self-determination theory in sport was carried out by Brunel (1999). However, this study (like all previously mentioned studies, with the exception of Newton and Duda, 1999) can be criticized for not assessing the moderating role of perceived competence.

The present study This study was designed to address the limitations of, and expand on, previous research in several ways. First, it examined whether goal orientations and perceived competence could predict each of the diþ erent

402 motivational types. Speci® cally, I hypothesized that task orientation would positively predict intrinsic types of motivation and identi® ed regulation, but would not predict introjected regulation, external regulation or amotivation. Similar hypotheses were made for perceived competence; the only exception was that perceived competence was expected to negatively predict amotivation. In contrast, I hypothesized that ego orientation would positively predict introjected regulation and external regulation, but would not predict intrinsic forms of motivation, identi® ed regulation or amotivation. I expected to observe a similar pattern of predictions for both males and females. Although diþ erences have been shown between the sexes in the mean levels of goal orientations, perceived competence and motivational types (e.g. Lirgg, 1991; Duda et al., 1995; Fortier et al., 1995; Li et al., 1996), from a theoretical viewpoint, the pattern of associations between these variables should not diþ er across the two sexes. For example, task orientation should positively predict the three intrinsic types of motivation in both males and females. The second way in which the present study expanded on previous research was by looking at the interactive eþ ects of goal orientations and perceived competence on the diþ erent types of motivation. Whereas previous studies (e.g. Duda et al., 1995; Vlachopoulos and Biddle, 1996; Dorobantu and Biddle, 1997) on goal orientations overlooked diþ erent levels of perceived competence, this study took this variable into account. According to achievement goal theory (Nicholls, 1989), perceived competence can moderate the impact of ego orientation on achievement-related variables. I expected that when ego orientation was high, more controlling motivation (i.e. introjected and external regulation) and less amotivation would be reported when perceived competence was high rather than low. I did not expect that, under conditions of high ego orientation, perceived competence would impact on self-determined motivation (i.e. intrinsic motivation types and identi® ed regulation). Individuals with high ego orientation and high perceived competence are likely to be driven in sport by rewards and external recognition and not by their desire to learn new skills and work hard on their weaknesses. In contrast, those with high ego orientation who doubt their ability are likely to be amotivated and to lack intrinsic or extrinsic motivation. No signi® cant interactions between task orientation and perceived competence were expected, as these were not theorized by Nicholls (1989). Hardy (1997, 1998) has criticized the literature for examining the two goal orientations in isolation from each other, despite Nicholls’ (1989) theorizing that the two goal orientations are orthogonal and can be found to diþ erent extents within the same person. Hardy (1997, 1998) pointed out that the orthogonal nature of

Ntoumanis the two goal orientations implies that they can interact with each other to predict various motivational outcomes. Because of this argument, it was important to examine in this study whether task and ego goal orientations interact to predict motivational types with diþ erent degrees of self-determination. Both Hardy (1998) and Duda (in press) have suggested the use of moderated hierarchical regression analysis to test for interaction eþ ects, instead of the four-factor one-way analysis of variance, which has traditionally been used in the literature. I expected that, under conditions of high task orientation, greater controlling motivation would be observed when ego orientation was high rather than low. No diþ erences in amotivation were expected. Both high task/low ego and high task/high ego individuals are expected to be motivated; however, the former will depend less on controlling motivation because of their weaker ego orientation. Furthermore, under conditions of high task orientation, no signi® cant diþ erences were expected in self-determined motivation as an outcome of diþ erent levels of ego orientation. That is, the interaction between ego orientation and task orientation in predicting self-determined types of motivation was not expected to be signi® cant. One could object to this hypothesis and argue that, because of its association with controlling motivation, ego orientation could potentially reduce the self-determination of a predominantly task-oriented individual. However, this argument would assume a pattern of strong negative relationships between controlling and self-determined types of motivation. This is unlikely to be the case, as distant types of motivation in the self-determination continuum are not highly correlated (see Li and Harmer, 1996, on the simplex structure of the selfdetermination continuum). Extending the two-way interactions and based on similar analysis by Newton and Duda (1999), three-way interactions were also tested between goal orientations and perceived competence. I expected that high ego orientation would predict the greatest controlling motivation and amotivation, and the least selfdetermined motivation, when perceived competence and task orientation were low. When individuals depend heavily on comparative criteria, make little or no use of self-referenced criteria and doubt their ability, they are unlikely to be self-determined in sport.

Methods Participants The participants were 268 British university athletes (154 males, 108 females; 6 did not specify their sex)

403

Achievement goals and self-determination from a wide variety of sports, studying at a large university in the North of England. The age range of the sample was 18± 36 years (20.4 ± 2.7 years; mean ± s), but approximately 90% of the participants were aged 18± 23 years. Most were of a relatively high competitive standard. Speci® cally, 45% of the sample had achieved their highest competitive standard competing for a district or county team, and a further 22% had achieved their highest competitive standard competing for a regional or national team. Measures The two goal orientations were Goal orientations. measured with the Task and Ego Orientation in Sport Questionnaire (Duda and Nicholls, 1992). This questionnaire has seven items that measure task orientation and six items that measure ego orientation. When completing the questionnaire, participants are asked to think of when they feel most successful in their sport and then indicate their agreement with items describing task and ego orientation. An example of a task orientation item is `Something that I learn makes me want to go and practise more’; an example of an ego orientation item is `Others mess things up and I don’t’ . Responses are reported on 5-point scales ranging from 1 (`strongly disagree’) to 5 (`strongly agree’). An extensive review by Duda and Whitehead (1998) has shown that the Task and Ego Orientation in Sport Questionnaire is a well-established measure of goal orientations with good reliability and validity. In this study, the alpha coeý cients were 0.76 for task orientation and 0.84 for ego orientation. Seven motivational types were Motivational types. assessed using the multidimensional Sport Motivation Scale (Pelletier et al., 1995). When completing this scale, athletes are asked why they practise their sport; they then respond to items re¯ ecting the diþ erent types of motivation. Three diþ erent types of intrinsic motivation are measured: intrinsic motivation to know (e.g. `for the pleasure of discovering new training techniques’), intrinsic motivation to accomplish (e.g. `for the pleasure I feel while improving some of my weak points’) and intrinsic motivation to experience stimulation (e.g. `for the pleasure I feel in living exciting experiences’). The scale also measures three types of extrinsic motivation: identi® ed regulation (e.g. `because it is a good way to learn lots of things which could be useful to me in other areas of my life’ ), introjected regulation (e.g. `because I must do sports to feel good about myself’ ) and external regulation (e.g. `because people around me think it is important to be in shape’ ). Lastly, the scale measures amotivation (e.g. `I used to have good reasons for doing sports, but now I am asking

myself if I should continue doing it’ ). All types of motivation have four items and are measured on 5-point scales ranging from 1 (`does not correspond at all’ ) to 7 (`corresponds exactly’ ). The reliability and validity of the Sport Motivation Scale have been shown in an extensive review by Vallerand and Fortier (1998). In the present study, the alpha coeý cients were 0.84 for intrinsic motivation to know, 0.75 for intrinsic motivation to accomplish, 0.72 for intrinsic motivation to experience stimulation, 0.72 for identi® ed regulation, 0.75 for introjected regulation, 0.74 for external regulation and 0.77 for amotivation. The perceived competence of Perceived competence. the athletes was assessed using ® ve items taken from the Intrinsic Motivation Inventory (McAuley et al., 1989). The ® ve items were slightly modi® ed because, in the original study, they referred to a discrete basketball task. An example item is `I think I am pretty good in my sport’ . The ® ve items are measured on 5-point scales ranging from 1 (`strongly disagree’) to 7 (`strongly agree’). McAuley et al. (1989) reported relatively high standardized item loadings and a Cronbach’s alpha of 0.84 for the perceived competence subscale. In the present study, the alpha was 0.81. Procedure All athletes were approached through their coaches or the University Athletic Union. The three questionnaires were administered on a single occasion in the order they are presented in the `Measures’ section. Data collection lasted about 3 weeks and was carried out by the author at mutually convenient times. All athletes were asked to sign an informed consent form. During administration of the questionnaires, the athletes were informed they could withdraw at any time. Furthermore, they were reassured that their responses would be kept in strict con® dence and would not be available to any third party.

Results Preliminary analyses At the beginning, all univariate and multivariate outliers were removed using Boxplots and the Mahalanobis distance criterion respectively. Thus, the ® nal sample included 247 athletes. Table 1 presents the means and standard deviations for all the variables in the study. The results show that the athletes had a relatively good motivational pro® le. Speci® cally, they reported high task orientation, perceived competence and selfdetermined motivation, moderate ego orientation and

404

Ntoumanis Table 1. Descriptive statistics for all variables (mean ± s) Variables Task orientation Ego orientation Perceived competence IM-knowledge IM-accomplishment IM-stimulation Identi® ed regulation Introjected regulation External regulation Amotivation

4.07 ± 0.44 3.13 ± 0.85 5.12 ± 0.88 4.59 ± 1.07 5.07 ± 0.92 5.00 ± 0.95 4.33 ± 1.08 3.98 ± 1.29 3.56 ± 1.16 2.00 ± 0.93

Note: Task orientation and ego orientation were measured on 5-point scales; all other variables were measured on 7-point scales. Abbreviation: IM = intrinsic motivation.

controlling motivation, and low amotivation. It was not a purpose of this study to examine diþ erences in goal orientations, perceived competence and diþ erent types of motivation between the sexes. Nevertheless, to assess the diþ erences between males and females, a one-way multivariate analysis of variance was carried out. The results showed that males were signi® cantly higher than females on ego orientation, perceived competence, external regulation, intrinsic motivation to accomplish, intrinsic motivation to experience stimulation and identi® ed regulation. However, eþ ect sizes (g2) showed that these diþ erences were of very little practical signi® cance, as they explained at most 8% of the variance (mean g2 = 0.03). Furthermore, Fisher’s transformation formula showed no signi® cant diþ erences in the size of correlations among the variables in the two sexes. This ® nding indicates that the relationships were largely independent of the sex of the individual. Moderated hierarchical regression analyses A series of moderated hierarchical regression analyses was conducted to determine whether the two goal orientations and perceived competence could predict independently and in interaction the seven motivational types. In each of the regressions, task orientation, ego orientation and perceived competence were ® rst entered individually to ascertain their unique contribution to the amount of variance explained. Next, three two-way interactions were entered individually: ego orientation with perceived competence, task orientation with perceived competence, and task orientation with ego orientation. Then, the three-way interaction term of these variables was entered in the last step. Following

Fig. 1. Signi® cant interaction between task and ego goal orientations in predicting external regulation. r , high ego orientation; j , low ego orientation.

the suggestions by Aiken and West (1991) and Jaccard et al. (1990), goal orientations and perceived competence were standardized before the analyses to avoid multicollinearity between these variables (i.e. ® rst-order terms) and their interaction terms. A signi® cant interaction eþ ect was inferred by a signi® cant increment in R2 when an interaction term was entered in the equation. As recommended by Aiken and West (1991), a step-down hierarchical procedure was used. This approach begins with the full regression equation; non-signi® cant terms are then omitted sequentially in stages beginning with the highest-order term in the equation. When evaluating the results of moderated hierarchical regression analysis, Jaccard et al. (1990) recommended the use of unstandardized regression coeý cients, which should be interpreted in the spirit of standardized scores. A summary of the ® nal regressions on each dependent variable is reported in Table 2. One two-way interaction was signi® cant: external regulation was predicted by the interaction between task and ego goal orientations. To determine the form of the interaction, two regression lines were plotted to depict the regression of task orientation on external regulation as a function of low (1 standard deviation below the mean) and high (1 standard deviation above the mean) ego orientation (Aiken and West, 1991). As Fig. 1 shows, when task orientation was low, diþ erent levels of ego orientation did not impact on external regulation. However, when task orientation was high, high ego orientation predicted signi® cantly higher external regulation than when ego orientation was low. The results also showed several signi® cant main eþ ects. All self-determined motivation types (i.e. the three types of intrinsic motivation and identi® ed regulation) were positively predicted by task orientation and perceived competence. Intrinsic motivation to

405

Achievement goals and self-determination Table 2. Summary of ® nal moderated hierarchical regression analyses R2 change

F change

b

t

IM-knowledge Task orientation Perceived competence

0.10 0.05

28.7*** 15.7***

0.34 0.25

5.35*** 3.96***

IM-accomplishment Task orientation Perceived competence

0.20 0.10

59.5*** 36.7***

0.40 0.30

7.71*** 6.06***

IM-stimulation Task orientation Ego orientation Perceived competence

0.13 0.03 0.06

36.1*** 7.7** 17.9***

0.34 0.15 0.24

6.01*** 2.77** 4.23***

Identi® ed regulation Task orientation Perceived competence

0.04 0.06

10.3*** 17.4**

0.22 0.28

3.22** 4.18***

Introjected regulation Ego orientation

0.03

7.3**

0.22

2.71**

0.13 0.07 0.05

36.6*** 21.4*** 16.2***

0.41 0.31 0.24

6.05*** 4.63*** 4.03***

0.08

20.6***

-0.26

-4.54***

Predictors

External regulation Ego orientation Perceived competence Task orientation ´ ego orientation Amotivation Perceived competence

Abbreviation: IM = intrinsic motivation. ** P < 0.01, *** P < 0.001.

experience stimulation was also positively predicted by ego orientation. The two controlling motivation types (i.e. introjected regulation and external regulation) were positively predicted by ego orientation. External regulation was also positively predicted by perceived competence. Lastly, amotivation was negatively predicted by perceived competence. However, it should be noted that the variance explained in these regressions was relatively small (mean R2 = 0.16).

Discussion This study examined how key constructs of achievement goal theory (Nicholls, 1989) and self-determination theory (Deci and Ryan, 1985) are linked in sport. Although the conceptual links between these two theories have previously been described (Duda, 1992), their empirical testing to date has been insuý cient. Achievement goal theory argues that two goal orientations, namely task and ego, are evident in sport. These orientations re¯ ect diþ erent criteria for cognitively processing achievement in a given activity (Duda, in press). Perceived competence has an important but distinct role in both theories. Achievement goal theory

suggests that normatively referenced perceived competence can moderate the in¯ uence of ego orientation on motivational outcomes. In self-determination theory, competence is conceived as a need, the satisfaction of which can lead to self-determined motivation. A review of the literature shows that achievement goals, notions of competence and intrinsic and extrinsic motivation are common themes in both theories, although there are certain diþ erences in the conceptualization and the emphasis placed upon these variables by each theory. Therefore, to examine empirical links between achievement goal theory and self-determination theory, this study focused on the independent and interactive eþ ects of diþ erent goal orientations and levels of perceived competence on types of motivation with varying degrees of selfdetermination. Independent eþ ects of goal orientations and perceived competence on diþ erent types of motivation Regression analyses showed that task orientation positively predicted all three types of intrinsic motivation as well as identi® ed regulation. The positive links between task orientation and self-determined types

406 of motivation are in accordance with Nicholls’ (1989) theorizing and with results of previous research (e.g. Duda et al., 1995; Vlachopoulos and Biddle, 1996; Dorobantu and Biddle, 1997). This ® nding is also in accordance with the hypotheses of this study. There is a tendency for athletes with high task orientation to feel self-determined, because they enjoy the challenge of learning a new skill and improving on their weaknesses (Duda, 1992). The reasons that predominantly task-oriented athletes have to participate in sport are not likely to be related to avoidance of guilt (i.e. introjected regulation) or the attainment of extrinsic rewards (i.e. external regulation). This hypothesis was also supported in the present study, as like the results reported by Brunel (1999), task orientation did not predict introjected and external regulation. Furthermore, task orientation did not predict amotivation. This makes sense, because task orientation represents an adaptive achievement orientation that guides individuals to set controllable and achievable goals. These goals will encourage individuals to stay motivated and committed in sport. High perceived competence was positively related to self-determined motivation. This is in accordance with Deci and Ryan’ s (1985) theorizing that the more competent individuals perceive themselves at some activity, the more intrinsically motivated they will feel (provided that they also feel autonomous to perform the activity). However, perceived competence positively predicted external regulation. Pelletier et al. (1995) also reported a positive relationship between external regulation and perceived competence. Although this ® nding was not expected, a closer look at the items in the external regulation sub-scale of the Sport Motivation Scale shows that they measure the extent to which individuals participate in sport for the `prestige of being an athlete’ and to `show others how good I am at my sport’ . Therefore, individuals with perceptions of high competence will be more likely to associate competence with these extrinsic outcomes. It appears that competence can control athletes’ behaviour when it becomes the means by which they can please signi® cant others or demonstrate their superiority. As expected, perceived competence negatively predicted amotivation. Pelletier et al. (1995) also reported a negative relationship between these two variables. This ® nding makes conceptual sense, because both achievement goal theory and self-determination theory predict that high perceived competence will sustain and increase one’s motivation. In contrast, amotivation has been linked by Vallerand (1997) to learned helplessness, characterized, among other things, by lack of ability beliefs. In accordance with the hypotheses of this study, ego orientation positively predicted introjected regulation and external regulation. This is probably because indi-

Ntoumanis viduals with high ego orientation participate in sport mainly out of extrinsic reasons, such as to demonstrate competence and superiority, and to attain rewards and acknowledgment from signi® cant others (Nicholls, 1989; Duda et al., 1995). Extrinsic goals can potentially control their behaviour and undermine their selfdetermination in sport. Indeed, this was the case in the present study, as ego orientation was not a signi® cant predictor of most types of self-determined motivation. The only exception was that ego orientation positively predicted intrinsic motivation to experience stimulation. This was not expected and it may be attributed to the possibility that athletes with high ego orientation will report positive sensations and excitement when they are able to meet their criteria for success (i.e. winning and outperforming others). Lastly, amotivation was not predicted by ego orientation. This makes sense theoretically, because ego orientation motivates individuals to strive and achieve certain comparative goals. Brunel (1999) reported relatively similar results with ego orientation predicting external regulation only from the self-determination continuum. Interactive eþ ects of goal orientations and perceived competence on diþ erent types of motivation A further aim of this study was to look at potential interactive eþ ects of task orientation, ego orientation and perceived competence on types of motivation with diþ erent degrees of self-determination. Based on Nicholls’ (1989) argument that perceived competence moderates the achievement patterns of ego orientation, interactions between ego orientation and perceived competence were expected. Furthermore, drawing from Hardy’ s (1997, 1998) criticism of the achievement goal literature for its failure to acknowledge the orthogonality of goal orientations, this study tested interactive eþ ects between task orientation and ego orientation. Furthermore, extending the two-way interactions, potential three-way interactions between ego orientation, task orientation and perceived competence were tested. The interactions between ego orientation and perceived competence and the three way-interaction between task orientation, ego orientation and perceived competence were not signi® cant. These results are contrary to the predictions of achievement goal theory and the hypotheses of this study and can be partially attributed to the high perceived competence of the participants (85% scored 4 or above on a 7-point scale). This restriction in the range of scores can reduce the statistical power of an interaction eþ ect (Finney et al., 1984; Aguinis and Stone-Romero, 1997). Future studies should overcome this limitation by using samples with more diverse competitive standards. It

407

Achievement goals and self-determination is also possible that the non-signi® cant interactions involving perceived competence could have been due to the wording of the items in the Intrinsic Motivation Inventory (e.g. `I think I am pretty good in my sport’ ; `I am pretty skilled at my sport’ ). Such items do not make explicit whether perceived competence is assessed based on comparative or self-referenced criteria. From an achievement goal theory viewpoint, perceived competence can play a moderating role only when it is construed on the basis of comparative criteria. From a self-determination perspective, one could argue that perceived competence was not a moderator in this study because the athletes may not have felt autonomous. Deci and Ryan (1985) argued that autonomy will mediate the eþ ects of perceived competence on self-determined motivation. Although there was no evidence in this study to support the claim that the athletes did not feel autonomous, it would be interesting to examine in the future whether there is a case for mediated moderation (Baron and Kenny, 1986). That is, it is worth determining whether the moderating eþ ect of perceived competence on motivated behaviour, predicted by achievement goal theory, is mediated by perceptions of autonomy. Of the two hypothesized interactions between task orientation and ego orientation, only one was signi® cant. Contrary to what was expected, the interaction between the two goal orientations in the prediction of introjected regulation was not signi® cant. For the most extreme form of extrinsic motivation, external regulation, the interaction was signi® cant. As predicted, when task orientation was high, lower external regulation was observed when ego orientation was low than when it was high. In contrast, when task orientation was low, diþ erent degrees of ego orientations did not impact on external regulation. This ® nding shows that a combination of high task and low ego orientation may be needed for lower levels of extreme extrinsic motivation. From a theoretical viewpoint, this makes sense. When individuals have high task orientation and low ego orientation, they are unlikely to participate in sport to pursue extrinsic goals, such as the demonstration of superior ability and the attainment of social approval and rewards (Duda, 1992). However, when high task-oriented individuals also have high ego orientation, such extrinsic goals become important for sport participation. In accordance with the hypothesis of this study, ego orientation was not in¯ uential in determining the selfdetermination of high-task oriented individuals, as the interactions between the two goal orientations were not signi® cant. High task orientation predicted high self-determination irrespective of ego orientation. This provides further support to the potentially important

role of high task orientation in safeguarding selfdetermination in sport (Duda, 1992, in press).

Conclusion In conclusion, the results of this study suggest the adaptive role of high task orientation in promoting self-determination in sport. However, the results should be viewed as preliminary, because there are no similar ® ndings in the literature to make appropriate comparisons. That is, no previous studies have tested interactions between goal orientations and perceived competence on all seven motivational types. Furthermore, the cross-sectional nature of the research design precludes any causal inferences regarding the links between the two theories. It should also be noted that the variance explained in the regression analyses was small and, therefore, the empirical framework of this study should be replicated and extended before any conclusions are drawn. This study could be improved on in several ways. Future research should examine empirical links between the two motivational theories using participants with more diverse competitive standards or from other physical activity contexts, such as physical education classes and ® tness clubs. In this way, variations in perceived competence will be greater than in the present study and, therefore, the interplay of goal orientations and perceived competence should be more evident. Also, perceived competence measures should be developed that will unambiguously assess perceived normative competence. Furthermore, longitudinal studies are needed to test the potentially dynamic interrelationship of the two theories and the possibility of reciprocal mechanisms. In extending this study, future research could explore the links of the two theories at the situational stage of generality (Vallerand, 1997) by looking at whether states of goal involvement can predict self-determined and controlling situational motivation. Furthermore, a more comprehensive integrative model of the two theories is needed that will take into account perceptions of autonomy and relatedness, besides those of competence. Autonomy and relatedness are not discussed by achievement goal theory; however, they are very important in the self-determination theoretical framework and can play an important role in the competitive context of sport (Vallerand and Losier, 1999). Attempts to integrate the two theories can oþ er a more comprehensive understanding of motivation in physical activity by showing the mechanisms through which diþ erent achievement criteria can facilitate or hinder self-regulation and selfmotivation.

408 References Aguinis, H. and Stone-Romero, E.F. (1997). Methodological artifacts in moderated multiple regression and their eþ ects on statistical power. Journal of Applied Psychology, 82, 192± 206. Aiken, L.S. and West, S.G. (1991). Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage. Baron, R.M. and 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. Brunel, P. (1999). Relationship between achievement goal orientations and perceived motivational climate on intrinsic motivation. Scandinavian Journal of Medicine and Science in Sports, 9, 365± 374. Butler, R. (1987). Task-involving and ego-involving properties of evaluation: Eþ ects of diþ erent feedback conditions on motivational perceptions, interest, and performance. Journal of Educational Psychology, 79, 474± 482. Butler, R. (1989). On the psychological meaning of information about competence: A reply to Ryan and Deci’s comment on Butler (1987). Journal of Educational Psychology, 81, 269± 272. Cury, F., Biddle, S.J.H., Sarrazin, P. and Famose, J.P. (1997). Achievement goals and perceived ability predict investment in learning a sport task. British Journal of Educational Psychology, 67, 293± 309. Deci, E.L. and Ryan, R.M. (1985). Intrinsic Motivation and Self-Determination in Human B ehavior. New York: Plenum Press. Deci, E.L. and Ryan, R.M. (1991). A motivational approach to self: Integration in personality. In Nebraska Symposium on Motivation. Vol. 38: Perspectives on Motivation (edited by R. Deinstbier), pp. 237± 288. Lincoln, NB: University of Nebraska Press. Dorobantu, M. and Biddle, S.J.H. (1997). The in¯ uence of situational and individual goals on the intrinsic motivation of Romanian adolescents towards physical education. European Yearbook of Sport Psychology, 1, 148± 165. Duda, J.L. (1989). Relationship between task and ego orientation and the perceived purpose of sport among high-school athletes. Journal of Sport and Exercise Psychology, 11, 318± 335. Duda, J.L. (1992). Motivation in sport settings: A goal perspective approach. In Motivation in Sport and Exercise (edited by G.C. Roberts), pp. 57± 91. Champaign, IL: Human Kinetics. Duda, J.L. (in press). Goal perspective research in sport: Pushing the boundaries and clarifying some misunderstandings. In Advances in Motivation in Sport and Exercise (edited by G.C. Roberts). Champaign, IL: Human Kinetics. Duda, J.L. and Nicholls, J.G. (1992). Dimensions of achievement motivation in schoolwork and sport. Journal of Educational Psychology, 84, 290± 299. Duda, J.L. and Whitehead, J. (1998). Measurement of goal perspectives in the physical domain. In Advances in Sport and Exercise Psychology Measurement (edited

Ntoumanis by J.L. Duda), pp. 20± 47. Morgantown, WV: Fitness Information Technology. Duda, J.L., Chi, L., Newton, M.L., Walling, M.D., and Catley, D. (1995). Task and ego orientation and intrinsic motivation in sport. International Journal of Sport Psychology, 26, 40± 63. Dweck, C.S. and Leggett, E.L. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95, 256± 273. Ferrer-Caja, E. and Weiss, M.R. (2000). Predictors of intrinsic motivation among adolescent students in physical education. Research Quarterly for Exercise and Sport, 71, 267± 279. Finney, J.W., Mitchell, R.E., Cronkite, R.C. and Moos, R.H. (1984). Methodological issues in estimating main and interactive eþ ects: Examples from coping/social support and stress ® eld. Journal of Health and Social B ehavior, 25, 85± 98. Fortier, M.S., Vallerand, R.J., Briere, N.M. and Provencher, P.J. (1995). Competitive and recreational sport structures and gender: A test of their relationship with sport motivation. International Journal of Sport Psychology, 26, 24± 39. Fox, K., Goudas, M., Biddle, S., Duda, J. and Armstrong, N. (1994). Children’ s task and ego goal pro® les in sport. British Journal of Educational Psychology, 64, 253± 261. Goudas, M., Biddle, S. and Fox, K. (1994). Achievement goal orientations and intrinsic motivation in physical ® tness testing with children. Pediatric Exercise Science, 6, 159± 167. Hardy, L. (1997). The Coleman Roberts Griý th Address: Three myths about applied consultancy work. Journal of Applied Sport Psychology, 9, 277± 294. Hardy, L. (1998). Responses to the reactants on three myths in applied consultancy work. Journal of Applied Sport Psychology, 10, 212± 219. Hom, H.L., Duda, J.L. and Miller, A. (1993). Correlates of goal orientations among young athletes. Pediatric Exercise Science, 5, 168± 176. Jaccard, J., Turrisi, R. and Wan, C.K. (1990). Interaction Eþ ects in Multiple Regression. Newbury Park, CA: Sage. Lee, M., Whitehead, J., Ntoumanis, N. and Hatzigeorgiadis, A. (1999). The Eþ ect of Values, Achievement Goals, and Perceived Ability on Moral Attitudes in Youth Sport. Report for the Economic and Social Research Council. Swindon: ESRC. Li, F. and Harmer, P. (1996). Testing the simplex assumption underlying the Sport Motivation Scale: A structural equation modeling analysis. Research Quarterly for Exercise and Sport, 67, 396± 405. Li, F., Harmer, P. and Acock, A. (1996). The Task and Ego Orientation in Sport Questionnaire: Construct equivalence and mean diþ erences across gender. Research Quarterly for Exercise and Sport, 68, 228± 238. Lirgg, C.D. (1991). Gender diþ erences in self-con® dence in physical activity: A meta-analysis of recent studies. Jour nal of Sport and Exercise Psychology, 8, 294± 310. Markland, D. and Hardy, L. (1997). On the factorial and construct validity of the Intrinsic Motivation Inventory: Conceptual and operational concerns. Research Quarterly for Exercise and Sport, 68, 20± 32.

Achievement goals and self-determination McAuley, E., Duncan, T.E. and Tammen, V.V. (1989). Psychometric properties of the intrinsic motivation inventory in a competitive sport setting: A con® rmatory factor analysis. Research Quarterly for Exercise and Sport, 60, 48± 58. Newton, M. and Duda, J.L. (1999). The interaction of motivational climate, dispositional goal orientations, and perceived ability in predicting indices of motivation. Inter national Jour nal of Sport Psychology, 30, 63± 82. Nicholls, J.G. (1989). The Competitive Ethos and Democratic Education. Cambridge, MA: Harvard University Press. Ntoumanis, N. and Biddle, S.J.H. (1999). A review of motivational climate in physical activity. Journal of Sports Sciences, 17, 643± 665. Ntoumanis, N., Biddle, S.J.H. and Haddock, G. (1999). The mediating role of coping strategies on the relationship between achievement goals and aþ ect in sport. Anxiety, Stress and Coping, 12, 299± 327. Pelletier, L.G., Fortier, M.S., Vallerand, R.J., Tuson, K.M. Brière, N.M. and Blais, M.R. (1995). Toward a new measure of intrinsic motivation, extrinsic motivation, and amotivation in sports: The Sports Motivation Scale (SMS). Journal of Sport and Exercise Psychology, 17, 35± 53. Ryan, R.M. (1982). Control and information in the intrapersonal sphere: An extension of cognitive evaluation theory. Journal of Personality and Social Psychology, 43, 450± 461. Ryan, R.M. and Connell, J.P. (1989). Perceived locus of causality and internalization: Examining reasons for acting

409 in two domains. Journal of Personality and Social Psychology, 57, 749± 761. Ryan, R.M. and Deci, E.L. (1989). Bridging the research traditions of task/ego involvement and intrinsic/ extrinsic motivation: Comment on Butler (1987). Journal of Educational Psychology, 81, 265± 268. Ryan, R.M. and Deci, E.L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68± 78. Vallerand, R.J. (1997). Toward a hierarchical model of intrinsic and extrinsic motivation. In Advances in Experimental Social Psychology (edited by M. Zanna), pp. 271± 360. New York: Academic Press. Vallerand, R.J. and Fortier, M.S. (1998). Measures of intrinsic and extrinsic motivation in sport and physical activity: A review and critique. In Advances in Sport and Exercise Psychology Measurement (edited by J.L. Duda), pp. 81± 101. Morgantown, WV: Fitness Information Technology. Vallerand, R.J. and Losier, G.F. (1999). An integrative analysis of intrinsic and extrinsic motivation in sport. Jour nal of Applied Sport Psychology, 11, 142± 169. Vlachopoulos, S. and Biddle, S.J.H. (1996). Achievement goal orientations and intrinsic motivation in a track and ® eld event in school physical education. European Physical Education Review, 2, 158± 164. Vlachopoulos, S. and Biddle, S.J.H. (1997). Modeling the relation of goal orientations to achievement-related aþ ect in physical education: Does perceived ability matter? Journal of Sport and Exercise Psychology, 19, 169± 187.