The Influence of Achievement Motivation and Chess

SPORT Psychology Journal of Sport & Exercise Psychology, 2007, 29, 561-583 © 2007 Human Kinetics, Inc.

The Influence of Achievement Motivation and Chess-Specific Motivation on Deliberate Practice Anique B.H. de Bruin, Remy M.J.P. Rikers, and Henk G. Schmidt Erasmus University Rotterdam Although the importance of high motivation to engage in deliberate practice has been acknowledged, no research has directly tested this hypothesis. Therefore, the present study examined this relation in adolescent elite chess players by means of a questionnaire. In addition, to provide an explanation for dropout among promising chess players, differences in motivation between persistent and dropout chess players were analyzed. Competitiveness and the will to excel proved to be predictors of investments in deliberate practice. Moreover, achievement motivation and chess-specific motivation differed to a certain extent between persisters and dropouts. Our results suggest that motivation to engage in deliberate practice not only contains elements of the will to improve performance, but also of the will to attain exceptional levels of performance. Key Words: deliberate practice, expertise, chess, achievement motivation

The positive relationship between accumulated hours of deliberate practice and excellent performance has been shown in a large number of studies in sports (e.g., Baker, Deakin, & Côté, 2005; Duffy, Baluch, & Ericsson, 2004; Helsen, Starkes, & Hodges, 1998; Hodge & Deakin, 1998; Hodges, Kerr, Starkes, Weir, & Nananidou, 2004), chess (Charness, Krampe, & Mayr, 1996; Charness, Tuffiash, Krampe, Reingold, & Vasykova, 2005), and music (Sloboda, Davidson, Howe, & Moore, 1996). Typically, in these studies, the number of hours individuals had spent on deliberate practice was assessed and related to their current expertise level. According to Ericsson, Krampe, and Tesch-Römer (1993), deliberate practice is defined as practice activities that are primarily directed at performance improvement and, therefore, require an adequate level of difficulty. Moreover, deliberate practice activities involve informative feedback and include ample opportunity for reflection and correction of errors. Practice activities that are performed merely for their inherent enjoyment (e.g., practicing for fun), are not considered deliberate and, as such, do not foster progress on the road to excellence.

The authors are with the Department of Psychology, Erasmus University Rotterdam, the Netherlands. 561

562 De Bruin, Rikers, and Schmidt

Since Ericsson and colleagues’ original article (Ericsson et al., 1993), studies in other fields have led to some domain-specific adaptations of the characteristics of deliberate practice. For example, in team sports, deliberate practice is better characterized as team or group practice rather than practice alone (Helsen et al., 1998). Moreover, in contrast to what was found in the domain of music, deliberate practice in team sports is generally considered enjoyable (Hodges et al., 2004). However, one of the preconditions of deliberate practice that is considered crucial in all domains of expertise is the motivation to improve performance. This type of motivation is usually referred to as achievement motivation (Ames, 1992). Although the nature of achievement motivation and its variations across individuals has been studied in diverse contexts, it has never been related directly to investments in deliberate practice. One can imagine that in order to maintain a demanding schedule of daily deliberate practice, often without external rewards, high achievement motivation is indispensable. Ericsson and colleagues (1993) acknowledge this by describing that, in the literature, the most cited condition for optimal learning is the motivation of students to attend to a task and improve performance. According to Ericsson et al. (1993), individuals are motivated to practice, because practice improves performance. For example, expert performers are often seen to decrease practice intensity off season (Ericsson et al., 1993). Research examining the performance motives of experts in relation to time engaged in deliberate practice can provide insight into the required motivational style that is needed for long-term dedication to deliberate practice. Although this has never been directly tested, Ericsson and colleagues’ view of the role of achievement motivation on deliberate practice corresponds mainly with what is termed a mastery orientation of motivation (Helmreich & Spence, 1978). In contrast to a competitive orientation, individuals striving toward mastery of a domain are motivated because they are challenged by difficult tasks and have high internal standards of excellence. In contrast, individuals with a competitive orientation are motivated by a desire to win in competition against others (Helmreich & Spence, 1978). In other theories of motivation, similar dimensions of achievement motivation have been described. For instance, Dweck and Leggett (1988) and Elliott and Dweck (1988) distinguish between learning goals and performance goals. The former can be viewed as goals that are directed at increasing one’s competence (comparable to mastery), whereas the latter refer to goals that aim at receiving positive judgments of one’s competence. Dweck and colleagues associated performance goals with lower learning outcomes than learning goals. Furthermore, Ames (1992) states in the achievement goal theory that a task-involved climate (i.e., a climate that fosters the need to master a task, and make progress) is positively correlated with intrinsic motivation. By contrast, an ego-involved climate (i.e., a climate that stimulates the need to demonstrate high ability, or be perceived as competent) is not, or even negatively correlated with intrinsic motivation. In experimental studies, similar effects of ego-involved situations on intrinsic motivation have been revealed. For example, when an external reward is promised (Pritchard, Campbell, & Campbell, 1987\bb\) or when the competitive situation is pressuring (Reeve & Deci, 1996), task-oriented motivation diminishes and the amount of time participants spend on a relevant task during a free-choice period decreases (Deci, 1971). Several studies have provided evidence that a competitive situation has a detrimental effect on intrinsic motivation to perform a particular

Achievement Motivation and Chess-Specific Motivation 563

task (e.g., Deci, Betley, Kahle, Abrams, & Porac, 1981; Reeve, Olson, & Cole, 1985; Vallerand & Reid, 1984). Similar results have been found in applied settings, such as in sports. Even though sports contain a certain extrinsic component (e.g., prizes, awards, social pressure), amateur athletes report to be directed more by intrinsic motives, such as the need to improve oneself (Vallerand, Deci, & Ryan, 1987). The extrinsic rewards are believed to negatively affect intrinsic motivation because they are used to control people and, therefore, undermine individuals’ sense of self-determination (Deci & Ryan, 1985). That is, individuals lose their feeling of autonomy, which, in turn, diminishes intrinsic motivation and increases chances of dropping out (Sarrazin, Vallerand, Guillet, Pelletier, & Cury, 2002; Van Yperen & Duda, 1999). Although these studies underline the importance of a mastery-oriented training environment, there are reasons to believe that, in high-performing individuals, certain ego-involved factors may also contribute to keeping up levels of achievement motivation. Research on the personality structure of high-performing individuals in sports (e.g., Olympic champions) shows that they are characterized by specific traits that are found to a lesser extent in nonexperts. They are, for example, highly confident of their abilities, exemplify mental toughness, and are very competitive (Brewer, Van Raalte, & Linder, 1993; Gould, Dieffenbach, & Moffett, 2002). One can imagine that in environments where direct competition and performance feedback are omnipresent, an interest in competition is a prerequisite to remain motivated. A field were this is particularly the case is chess. Elite chess players are constantly confronted with competition and performance by receiving trimonthly chess ratings. Recent research on the personality of young chess players has shown that they score lower on the Big Five factor agreeableness and higher on intellect/ openness. The latter was especially the case for the subgroup of high-performing young chess players (Bilalić, McCleod, & Gobet, 2007). However, because Bilalić and colleagues (2007) did not measure achievement motivation, it is unclear whether the lower agreeableness corresponded with higher competitiveness. In sum, it is possible that not only a mastery orientation but also competitiveness might play a role in achievement motivation of high-performing individuals in a field such as chess. To further examine this, a study is needed that compares mastery and competitiveness within high-performing individuals, and relates it to investments in deliberate practice. An issue related to the question of which factors determine motivation of highperforming individuals is what explains dropout among this population. Research into this matter is relevant because it might lead to prevention of dropout among promising individuals and provides insight into the factors that determine success on the road to expertise. Factors as sport enjoyment, personal investment and social constraints have been related to sport commitment and dropout in previous studies (e.g., the sport commitment model; Scanlan, Carpenter, Schmidt, Simons, & Keeler, 1993). Moreover, the lack of a competitive achievement orientation could provide a possible explanation for dropout of those who were once promising in their field. One of the explanations for dropout could be that these individuals differ from those who persist in how much they strive toward and are motivated by competition. In a few studies, it has been shown that ego-involved climates caused those who eventually dropped out to develop lower perceived competence and feelings of low autonomy (Guillet, Sarrazin, Carpenter, Trouilloud, & Cury, 2001\bb\; Pelletier,


Fortier, Vallerand, & Brière, 2001). In contrast, those who persisted demonstrated a high level of self-determination and showed no decrease in motivation (Vallerand & Bissonnette, 1992). The extent to which a dislike for competition might lead to dropout has never been directly studied. Given the abovementioned competitive personality profiles of high-performing individuals (e.g., Brewer et al., 1993; Gould et al., 2002), it could be possible that those who do not succeed on the road to excellence are characterized by too little competitiveness. To test this hypothesis, a study is needed that controls for differences in training environment and further external factors, and that examines differences in the mastery and competitiveness aspects of achievement motivation between persisters and dropouts. In this study, we pursued three goals. First, given the lack of research relating motivation to deliberate practice, we examined the effect of achievement motivation and motivation to engage in deliberate practice on accumulated hours of deliberate practice and performance in a group of young elite chess players. The domain of chess was selected because the positive relationship between deliberate practice and performance has been well documented in this field and because objective performance indicators (i.e., chess ratings) are readily available (e.g., Charness et al., 1996, 2005). Moreover, in contrast to physical sports, little is known about the motivation of elite chess players. General achievement motivation, not specific to chess, was assessed by use of the Work and Family Orientation Questionnaire (WOFO; Helmreich & Spence, 1978; Spence & Helmreich, 1983). In this measure, achievement motivation is viewed as a multidimensional construct, not specific to certain tasks or situations, and contains three scales: work, mastery, and competitiveness. The work scale reflects the willingness to work hard and perform well, whereas the mastery scale measures to what extent individuals are driven by a challenge to perform well on difficult tasks. The competitiveness scale indicates the degree to which individuals look for competition, and strive toward winning. More details about the WOFO are provided in the Methods section. The decision to engage in a certain task for an extended period of time is determined not only by general achievement motivation, but also by situational motivation that the specific task evokes (e.g., Guay, Vallerand, & Blanchard, 2000; Vallerand, 1997). Therefore, to map participants’ chess-specific motivation to engage in deliberate practice, we designed an instrument that assessed individuals’ need for performance improvement and need for competition in chess. Although some items referred to rather stable motives (“I want to become a professional chess player”), most inquired about motives that could fluctuate over time (e.g., “When I start practicing chess, I first stop and think about what I need to improve”). Using the path analysis technique of structural equation modeling (Kline, 1998), we examined the structural relationships between the WOFO, the deliberate practice motivation questionnaire, accumulated hours of deliberate practice, and performance level as indicated by current chess rating. The advantage of structural equation modeling above traditional regression analysis is that it allows for structural modeling with more than one dependent variable. Therefore, we were able to simultaneously study the effect of achievement motivation on deliberate practice, and the effect of deliberate practice on chess performance. This allowed us to analyze a statistical model that would provide a more complete picture of the relation between motivation, deliberate practice, and chess performance. We hypothesized that both the measure of achievement motivation (WOFO) and the chess-specific measure


(Deliberate Practice Motivation Questionnaire [DPMQ], described in the next section) would positively contribute to investments in deliberate practice, and thus, chess performance. Given the emphasis that previous research has laid on personal performance improvement on the road to expertise (e.g., Ericsson et al., 1993), we were interested in the mastery measure of the WOFO. Secondly, in view of the potential importance of competition in elite chess players, and taking into account previous research on the personality of high-performing individuals (e.g., Brewer et al., 1993; Gould et al., 2002), we hypothesized that the competition subscales of the WOFO and the DPMQ would also contribute to deliberate practice and chess performance. Finally, in view of the relatively low inherent enjoyment of deliberate practice compared with other everyday activities, we hypothesized that a model that depicted motivation as the result of deliberate practice and chess performance would exemplify a lower fit to the data than a model that described motivation as the cause of deliberate practice and chess performance. The third goal of this study was to shed further light on the possible role of certain personality characteristics as a cause for dropout in promising young individuals. We hypothesized that dropouts would demonstrate lower achievement motivation, and lower deliberate practice motivation. Therefore, we analyzed differences in achievement motivation and chess-specific deliberate practice motivation between a group of young elite chess players, who were still in the selection of the Dutch national chess training, and a similar group of chess players who had dropped out of the Dutch national chess training.

Methods Sample A total of 81 (30 girls, 51 boys) adolescent chess players participated (Mean age = 16.19 years, SD = 2.75, range 12–23 years). These chess players were either in the 2003-2004 selection of the national chess training of the Dutch Chess Federation (59 participants) or were selected in earlier years but had at some point decided to quit (24 participants) (mean chess rating for the complete group at time of test = 1,944, SD = 259, range 1,488–2,505). The average Dutch chess rating of a competitive adult chess player is around 1,700. Although there are no official titles associated with specific ratings, someone with a chess rating higher than 2,000 is considered a strong player in the Netherlands. This holds for less than 10% of those who receive a rating. The Dutch Chess Federation selects about 10 young adolescents per year from the top-performing chess players in the Netherlands, based on chess ratings and information from regional coaches. Of the 24 chess players who had quit between 1999 and 2004, two were not willing to participate. Of the 59 participants tested in June 2004 (92.2% of the total national training group at that time), 11 persons voluntarily decided not to return to the national training after the summer, and thus were categorized as dropouts. The group that still received national training (hereafter referred to as persisters) consisted of 48 participants (18 females, 30 males, mean age = 15.13 years, SD = 2.14), whereas the group that had quit the national training during or before the summer of 2004 (hereafter referred to as dropouts) consisted of 33 participants (12 females, 21 males, mean age = 17.77 years, SD = 2.85). The mean age at which the dropouts had quit was


16.12 years (SD = 2.02). All participants received a small financial compensation after completion of the study.

Materials Participants completed a paper-and-pencil questionnaire that contained two sections. The first part of the questionnaire inquired about participants’ engagement in deliberate practice. For every year since starting to play chess seriously, they were asked to rate how many hours of deliberate practice they had engaged in. Deliberate practice is traditionally defined as the amount of time individuals engage in practice alone that is performed with the goal to improve skill (Ericsson et al., 1993). In chess, this has been translated as the amount of time engaged in individual, serious study of the game of chess (Charness et al., 1996, 2005). In this study, we defined deliberate practice as the accumulated number of hours of serious chess play and serious chess study alone. The former form of practice, in contrast to Ericsson’s definition of deliberate practice (Ericsson, 1996), also contained serious chess games played in training or in tournaments. For this young sample—having only limited experience playing serious chess games—this activity would also be considered deliberate practice because it is valuable to improve performance. This argument was substantiated by the high correlation between serious chess play and chess rating in our sample (r = .54, N = 79, p < .001), which was somewhat higher than the correlation between serious chess study and rating (r = .48, N = 78, p < .001). The latter form of practice includes individually analyzing chess games and chess books or magazines. Internet chess games were not taken into account in the serious chess play measure because, for this specific sample, these are considered playful interaction. All participants estimated their practice hours per year from the start of their chess career until the current year. Part two of the questionnaire consisted of 26 items that assessed participants’ motivation to engage in chess activities that are specifically directed at improving performance, being a core characteristic of deliberate practice (Ericsson et al., 1993). The items either described participants’ need to improve performance in chess (i.e., “I want to become a chess grandmaster”), or participants’ need to win in chess (i.e., “To me, winning is the most important thing in chess”). This questionnaire was designed to map participants’ motivation to perform deliberate practice, and shall be referred to as the DPMQ (Deliberate Practice Motivation Questionnaire; see Table 2 for the complete list of items). All items were rated on a forced-choice scale from 1 (completely disagree) to 5 (completely agree). Some of the questions were reversely scored, so that for every item a high rating corresponded with scoring high on the underlying factor. The dropouts were asked to answer the questions from the perspective when they still participated in the national training. This was further stimulated by rephrasing the items where necessary (e.g., “I preferred playing chess against friends above making the chess exercises for the national training”). The reliability of this questionnaire was fairly high (Cronbach’s alpha = .84; a value of .60 or higher is considered acceptable). The third part of the questionnaire consisted of the Work and Family Orientation Questionnaire (WOFO; Helmreich & Spence, 1978; Spence & Helmreich, 1983). Spence and Helmreich (1983) originally designed the questionnaire to study


individual differences in achievement motives and to compare motivational systems of men and women. The reliability and validity of the WOFO as a measure of general achievement motivation have been tested and confirmed in several studies on, for example, sensation seeking (Schroth & Lund, 1997\bb\), sports (Gill, 1988), and aging (Die, Seelbach, & Sherman, 1987). The WOFO Questionnaire assesses four dimensions of achievement motivation: work (the desire to work hard and perform well on a task), mastery (having a preference for challenging tasks to meet internal standards of excellence), competitiveness (the enjoyment of personal competition and the desire to win and be better than others), and personal unconcern (the lack of concern with negative reactions of others). Because Spence and Helmreich (1983) report that the personal unconcern scale is of little value in achievement research, we administered only the first three scales of the questionnaire. These three scales consisted together of 19 items (work: 6, mastery: 8, and competitiveness: 5). All items were rated on a 5-point scale (1 = completely disagree, 5 = completely agree). Whereas Part 2 of our questionnaire aimed to measure chess-specific motivation, the WOFO is assumed to assess intrinsic achievement motivation as a general personality trait. Therefore, the dropouts did not complete the WOFO by looking back at when they participated in the national training, but filled it in from the perspective of their current situation. Moreover, because this questionnaire measured achievement motivation in a wide variety of settings, these items did not need to be adapted to the domain of chess. Participants’ most recent chess ratings were collected with help of the Dutch Chess Federation. For the dropouts, the most recent chess rating is probably not the most representative of their chess career because it is measured after the peak of their chess performance. Training and chess playing environment have since changed considerably between persisters and dropouts, and comparisons between these groups are therefore less informative. It is obvious that the dropouts have decreased practice hours and have less motivation to play chess. What is more interesting is whether this was also the case when they were still in the national training. Therefore, we analyzed the dropouts’ data up until the moment they left the national training. For the same reason, instead of collecting the most recent chess rating, the chess rating at time of quitting the national training was collected for the dropouts. Correlation between the dropouts’ current rating and their rating at time of quitting was r = .96 (N = 28, p < .001). The Dutch ratings are calculated in a manner similar to those of the World Chess Federation (FIDE). However, the Dutch ratings are also based on Dutch tournaments, which are generally not taken into account in the FIDE ratings. Therefore, the Dutch ratings tend to be somewhat lower than the FIDE ratings. Because we were not interested in the absolute level of the ratings but instead in the variation in ratings between individuals, we did not transform the Dutch ratings to FIDE ratings.

Procedure The persisters completed the questionnaire individually during a national chess-training weekend in June 2004. Because the dropouts no longer attended the national training, a research assistant visited them at home and asked them to fill out the questionnaire. Participants were informed that the goal of the study was


to gain insight into factors that determine the development of chess expertise. We also explained that the Dutch Chess Federation had been involved in the design of the questionnaire and that they had provided their approval.

Analysis Chess Ratings and Deliberate Practice. The cumulative number of hours spent

on serious chess study alone and on serious play against others was calculated by multiplying the weekly estimates by 52 and summing the total hours for each year up until the time of the study. To estimate reliability of the retrospective estimates, a subset of the participants (36 individuals: 20 persisters, 16 dropouts, 44.4% of the original sample) completed an Internet diary about their engagement in chess activities. We calculated the mean weekly hours of serious chess study alone and serious chess play based on three consecutive weeks of diary reports. These means were correlated, using a Pearson product–moment correlation, with the retrospective weekly estimate of serious chess play and serious chess study alone for the current year of practice in the questionnaire. For serious chess play, the correlation was r = .74 (N = 36, p < .01), whereas, for serious chess study alone, it was r = .60 (N = 36, p < .01). These correlations are comparable to those found in previous studies (e.g., Hodges & Starkes, 1996; Hodges et al., 2004). That the serious chess play estimates revealed a higher correlation could be ascribed to the more planned nature of serious chess play compared to serious chess study. Serious chess play always requires a chess partner, and, therefore, has to be done at specifically planned time periods. For serious chess play, mean weekly retrospective estimate was 5.36 hr (SD = 5.05), whereas the diary average for this variable was 6.31 hr (SD = 6.10). For serious chess study alone, mean weekly retrospective estimate was 4.38 hr (SD = 6.23), and the diary average was 3.70 hr (SD = 3.79). Internal Validity of the Deliberate Practice Motivation Questionnaire. Four

types of statistical analyses were performed on the data. First, the correlations between the subscales of the WOFO, and between the DPMQ and the WOFO were calculated. Secondly, to analyze the underlying factorial structure of the DPMQ, exploratory factor analysis (principal components estimation) was applied. Because this measure was designed to assess chess players’ need for performance improvement and need for competition, the factorial analysis was aimed at testing the validity of these two factors. Differences Between Persisters and Dropouts in Chess Rating and Motivation.

Differences between persisters and dropouts in accumulated hours of deliberate practice, and chess ratings were computed using independent t tests. We controlled for age in these analyses, because the dropouts were generally older than the persisters. Because rapid development of the brain and cognitive abilities occurs during adolescence (Gathercole, Pickering, Ambridge, & Wearing, 2004; Luna, Carver, Urban, Lazar, & Sweeney, 2004), age might have affected their ability to perform concentrated deliberate practice and improve in chess, although this factor is not related to their characteristics of being dropouts. Controlling for age prevented that developmental aspects only related to age could influence the results. Independent t tests between persisters and dropouts were also run on deliberate practice motivation (DPMQ) and achievement motivation (WOFO). These analyses were performed


on the subscales of these questionnaires as well. Effect sizes were calculated by means of Cohen’s d (Cohen, 1977). Fit of Structural Models of Motivation, Deliberate Practice, and Chess Rating.

Using a structural equation modeling approach, the relationships between deliberate practice motivation (DPMQ), achievement motivation (WOFO), deliberate practice, and current chess rating were causally examined (Byrne, 2001). Amos 4.0 was used as statistical program (Arbuckle & Wothke, 1999). Structural equation modeling combines multiple regression and path analysis to enable testing of causal relations in a hypothetical model based on the covariance and variance structure in a data set. The advantage of structural equation modeling over regular multiple regression is that it allows for the use of more than one dependent variable, making the analysis of causal models with more than one structural path between variables possible. In our analysis, the criterion (or dependent) variables were deliberate practice and chess rating (current rating for persisters, rating at time of quitting for dropouts), where deliberate practice also served as a predictor of chess rating. The primary independent variables were the two measures of motivation mentioned above. The Amos software produces several goodness-of-fit criteria indicating how well the tested model accounts for the observed variance structure. In our analysis, both absolute and incremental fit indices were used. As to the absolute fit indices, we analyzed the chi-square value and its p value, to assess whether there was a difference between the hypothetical model and the covariance structure of the data set. As the root mean square error of approximation (RMSEA) appears to be minimally influenced by sample size, this index was also included. The RMSEA represents the root square of the chi-square divided by the number of degrees of freedom. This fit index also assesses the extent to which the hypothesized model and the underlying data structure correspond. Its value is required to be smaller than .05 to be considered acceptable. Because the overall chi-square value is sensitive to sample size, these values should be interpreted with caution. Therefore, we also report information on incremental fit indices. The incremental indices we used were the comparative fit index (CFI), which compares the fit of the particular model under test with a model in which none of the variables are related. A CFI of .90 or higher indicates that the tested model fits the data well. As we were interested in comparison of fit across models, we also included Akaike’s information criterion (AIC) into our analyses (Akaike, 1987\bb\). Akaike’s information criterion creates a composite measure of goodness of fit and complexity by forming a weighted sum of the two. The model with the lowest AIC represents the best fit. The fit of a tested model was deemed adequate if all fit indices had values below the cutoff scores, and when the AIC was relatively low compared with other models. Several structural models were tested and compared as to their level of fit to the data. Given the large body of evidence that describes the monotonic effect of deliberate practice on current performance (for an overview, see Starkes & Ericsson, 2003), we depicted chess rating as the result of deliberate practice in all models. Model 1a analyzed the relationship between the three dimensions of achievement motivation, measured with the WOFO, accumulated hours of deliberate practice, and current chess rating. Because we were interested in the specific effect of competitiveness on deliberate practice and because Model 1a revealed that this relationship had the highest path coefficient (see Figure 1), a second model was tested that separately studied the effect of competitiveness on deliberate practice and chess rating (Model 1b).

Figure 1 — Structural models tested to examine the relation between achievement motivation, deliberate practice motivation, accumulated hours of deliberate practice, and current chess rating. 570


Several studies have shown that, in competitive situations, winning has a positive effect on perceived competence and intrinsic motivation compared with losing (i.e., McAuley & Tammen, 1989; Reeve & Deci, 1996; Vallerand & Reid, 1984). Translated to chess, winning or losing is not only limited to individual chess games, but is also more strongly expressed in the trimonthly chess ratings that chess players receive. This would mean that fluctuation of chess ratings can have an influence on chess-specific deliberate practice motivation as well. To test this hypothesis, Model 2 included deliberate practice motivation either as the cause of deliberate practice, and, hence, chess rating (Model 2a), or as the consequence of deliberate practice and chess rating (Model 2b). Finally, in Model 3 we examined to what extent achievement motivation and deliberate practice motivation concurrently contribute to accumulated hours of deliberate practice, and hence, chess rating. Because our goal was to study the relationship between motivation, deliberate practice, and chess performance in general, we did not discriminate the structural equation modeling analyses between persisters and dropouts but aggregated their data. Because the practice and chess performance data concerned only the time when persisters and dropouts attended the national training, differences in these variables after having dropped out cannot have influenced our data.

Results Chess Ratings and Deliberate Practice Mean chess ratings, mean deliberate practice hours, and mean scores on the DPMQ and WOFO are presented in Table 1. Details of the items of the DPMQ are reported in Table 2 and are described below. The correlation between accumulated hours of serious chess study alone and most recent chess rating was r = .45 (p < .001), whereas the correlation between accumulated hours of serious chess play against others and most recent chess rating was r = .42 (p < .001). Independent t tests did not yield any significant differences between persisters and dropouts in the mean number of hours of serious chess play and serious chess study. When controlling for the effect of age, the mean of all chess ratings of the persisters was higher than the mean of the dropouts’ chess ratings, F(1, 80) = 5.07, MSE = 30185.19, p < .05, ηp2 = .06. When controlling for participants’ age at the time of the specific measurement, the difference between persisters’ and dropouts’ ratings at time of entry in the national training, and at time of test, was not significant, both Fs < 1. Because recent analyses have revealed that chess ratings have risen over the last decades, it is possible that dropouts who started the national training years ago might have lower chess ratings than persisters who entered the chess training recently. Therefore, we performed an independent t test that compared persisters’ and dropouts’ chess rating at time of training entry. Some of the chess players were very young, so they did not receive chess ratings at that point. Our analysis revealed no difference between persisters and dropouts, t(45) = −1.14, p = .26, d = 0.36. Analysis of the development of chess ratings over time, and the concurrent influence of gender is reported elsewhere (De Bruin, Smits, Rikers, & Schmidt, 2007).


Table 1 Mean Number of Hours of Deliberate Practice, Chess Ratings, Item Averages on the (Subscales of the) WOFO and the Deliberate Practice Motivation Questionnaire (DPMQ). Standard Deviations Are in Parentheses. Item

Persisters

Dropouts

2,604.73 (2067.78)

2,050.90 (1270.66)

2,089.02 (2774.16)

1,291.36 (1363.23)

Chess rating at entry of training

1,698.77 (215.56)

1,769.12 (176.37)

Most recent chess rating

1,936.64 (279.40)

1,957.50 (223.81)

Mean of all chess ratings*

1,829.51 (241.48)

1,729.70 (185.67)

Mean number of hours of serious chess play Mean number of hours of serious chess study alone

DPMQ Will to excel** (20 items)

2.47 (0.60)

1.86 (0.45)

Competition (5 items)

2.53 (0.74)

2.76 (0.84)

2.48 (0.50)

2.04 (0.39)

Total DPMQ** (25 items)

WOFO Work (6 items)

3.85 (0.56)

3.85 (0.38)

Mastery* (8 items)

3.31 (0.52)

3.06 (0.54)

Competitiveness (5 items)

3.51 (0.95)

3.17 (0.80)

Total WOFO* (19 items)

3.53 (0.45)

3.34 (0.39)

Note. Difference between persisters and dropouts is significant at *p < .05 and **p < .01.

Internal Validity of the Deliberate Practice Motivation Questionnaire To gain further insight into the structure of the components underlying these items, an exploratory principal components factor analysis was conducted. Because we were unsure whether to expect a correlation between the measure of performance improvement and the measure of competition, we ran both an orthogonal factor analysis (principal components analysis) and a similar analysis that contained an oblique rotation method (direct oblimin, delta = 0). Moreover, because the instrument consisted of 26 items and therefore the variance explained as represented per eigenvalue is low, the cutoff for the eigenvalues was set to 2. In both analyses, two factors were identified. Visual inspection led us to conclude that the factor loadings of the correlated factor analysis were not higher than the loadings of the orthogonal solution, so we decided to proceed with the simpler orthogonal solution. Also, from a theoretical point of view, the mastery-related performance improvement approach is not necessarily correlated with a competitive approach. The first factor had an eigenvalue of 6.51 and explained 25.04% of the variance.


The second factor had an eigenvalue of 2.45 and explained 9.42% of the variance. Twenty items predominantly loaded onto the first factor, whereas five items loaded primarily onto the second factor. One item loaded low on both factors (“To improve in chess, analyzing played games is better than playing new games,” factor loading 1 = .143, factor loading 2 = .271). One item loaded moderately on both factors (“Improving performance is the most important thing to me when playing chess,” factor loading 1 = .424, factor loading 2 = .346). These items were left out of further analyses. The remainder of the items and their factor loadings are shown in Table 2. Analysis of the content of the items that loaded differently on the two factors led us to conclude that the first factor primarily measured “the will to excel in chess.” This factor includes the will to improve performance, but stretches further and also contains elements of the desire to reach exceptional levels of performance. This is mainly indicated by the items that loaded highest on this factor (“I want to become a chess grandmaster” and “I would give up my other hobbies in order to improve in chess if I have to”). The second factor was mainly focused at winning in competition (“My main goal when entering a chess tournament is to gain points for my chess rating” and “To me, winning is the most important thing in chess”). As to the reliability of the two factors, for the factor Will to Excel, Cronbach’s alpha = .88, for the factor Competition, Cronbach’s alpha = .63. Although the overall reliability of the deliberate practice motivation questionnaire was high, further research is needed to determine its usefulness in other fields and other populations.

Differences Between Persisters and Dropouts in Motivation To analyze differences between persisters and dropouts on the two separate factors and on the overall score of the DPMQ, item averages were calculated per factor and for the overall DPMQ. The mean item averages of the questionnaires and their subscales are presented in Table 1. First, we analyzed the overall difference between persisters and dropouts on the DPMQ. Persisters scored significantly higher than dropouts on this overall measure, t(79) = 4.30, p < .001, d = 1.00. Separating analyses over the two subscales of the DPMQ, the persisters scored significantly higher than the dropouts on the factor Will to Excel, t(79) = 5.27, p < .001, d = 1.17, but not on the Competition scale, t(79) = −1.33, p = .20, d = 0.30. A similar pattern was found for the WOFO. Dropouts scored significantly lower than persisters, t(79) = 1.98, p < .05, d = 0.45. Specifically, persisters scored significantly higher than dropouts on the mastery scale, t(79) = 2.06, p < .05, d = 0.46. No difference was observed for the competitiveness scale, t(79) = 1.67, p = .10, d = 0.38. Also, no effect was found for the work scale, t(79) = 0.01. The correlation between the WOFO and the DPMQ was relatively high (r = .42, p < .001). The correlations between the DPMQ, the WOFO, and their subscales are presented in Table 3.

Fit of Structural Models of Motivation, Deliberate Practice, and Chess Rating In Figure 1, the models tested and their corresponding path coefficients are shown. Because this is the first time that the DPMQ has been used on a highly representative, but fairly small sample and because of the skewed distribution of items between the two factors of the DPMQ, we decided to utilize the total DPMQ score instead

574

.712 .529 .779 .748 .643 .470 .443 .788 .567 .660 .471

I work on coming up with variations in chess openings

I would give up my other hobbies in order to improve in chess if I have to

I want to become a professional chess player

I’ve wanted to become a professional chess player for a long time

I can not imagine ever quitting playing chess

I think I am able to become a very good chess player

I want to dedicate as much time as possible to playing chess

If I would know that I could earn a living playing chess, I would dedicate more time to it

I have to play chess every day

When I start practicing chess, I first stop and think about what I need to improve

Scale—Will to excel

Will to excel

I want to become a chess grandmaster

Item

−.070

−.112

−.154

−.080

.209

−.276

.064

−.075

.030

−.014

−.044

Competition

Table 2 Items of the Deliberate Practice Motivation Questionnaire (DPMQ) and Their Respective Factor Loadings (Represented in Columns 2 and 3)

575

.433 .741 .400 .353 .462

Even though the chess exercises I make are difficult, I enjoy making them

I want to be involved in chess in as many ways as possible

I replay complex parts of a chess game in my head

I prefer playing chess against friends above making the chess exercises for the national training

When I loose a chess game, I practice harder afterwards

.729 .531 .544

.012 .118 −.162

To me, winning is the most important thing in chess I learn more from making the chess exercises for the national training than from playing chess against friends If I lose against a strong player who taught me a lot when analyzing the game together afterwards, I don’t mind having lost (mirrored score)

.633

.094

.581

.320

.055

.253

.047

−.085

−.177

−.053

My main goal when entering a chess tournament is to gain points for my chess rating

When I enter a chess tournament, gaining experience is more important to me than winning (mirrored score)

−.180

.524

After a played game, I replay it to analyze what I could have done better

Scale—competition

.412

After a played game, I take notes about what I should improve next time


Table 3 Correlations Between the WOFO, the DPMQ, and Their Subscales (Work, Mastery, and Competitiveness of the WOFO, Will to Excel and Competition of the DPMQ). Work (W) Mastery (M) Competitiveness (C1) Total WOFO Will to excel (E) Competition (C2)

W —

M .47**

C1 .04

WOFO .62**

E .20

C2 −.12

.22*

.81**

.41**

−.16

.67**

.24*

.32**

.41**

.05 −.10

DPMQ .16 .36** .33** .42** .95** .21

Total DPMQ *p < .05, **p < .01.

of the two factor scores in the structural equation analyses. Preliminary analysis revealed highly significant correlations between some of the subscales of the WOFO (see Table 3). Therefore, we entered the correlation between work and mastery, and the correlation between competitiveness and mastery into the model. Given that the correlation between work and competitiveness was zero, this path was not taken into the model. This is represented in Figure 1 by the two-ended arrow between work and mastery, and competitiveness and mastery, which signifies the estimation of a covariance. The path coefficient between accumulated hours of deliberate practice and chess rating was fairly high (.57). This was also found for the relationship between the DPMQ and accumulated hours of deliberate practice (.50) (Model 2a). The path coefficients between the scales of the WOFO (Model 1a) and the accumulated hours of deliberate practice were somewhat lower (between .10 and .30). The critical ratios of the path coefficients of the work and mastery scale were 1.142 and 0.792, respectively. Given that these were both smaller than 1.96, this indicates that they were nonsignificant. Therefore, we decided to test a less constrained model that did not include the paths between work and deliberate practice, and mastery and deliberate practice (Model 1b). Because the adapted Model 1b is less constrained, the difference in χ2 should be significant in order to signal a significant improvement of the model fit compared with the more constrained Model 1a. However, the difference in χ2 between the model with (Model 1a) and the model without (Model 1b) these paths was 3.54 at 2 degrees of freedom. Because this change in χ2 is not significant at a cutoff score of .05 and because these constructs are theoretically relevant for explaining variation in deliberate practice, we decided not to exclude the work and mastery paths from the final model (Model 1a). Although the path coefficients do not provide much information about the statistical plausibility of the theoretical models tested, the goodness-of-fit criteria do (Table 4). The goodness-of-fit indices reveal that the model, which takes into

577

χ2 p (χ2)

7.67

6

0.26

0.39

0.04 16.59

1 1

0.00

0.85

DPMQ, deliberate practice, and chess rating

4.13

4

WOFO, deliberate practice, and chess rating

df

0.98†

1.00†

1.00†

1.00†

CFI

Note. Fit index indicates †adequate fit or ††near adequate fit of the model compared with cutoff score.

Combined model: WOFO, DPMQ, deliberate practice, and chess rating Model 3 (WOFO, DPMQ, deliberate practice, and chess rating) 6 6.18 0.40 1.00†

Model 2a (DPMQ, deliberate practice, and chess rating) Model 2b (Deliberate practice, chess rating, and DPMQ)

Model 1a (WOFO, deliberate practice, and chess rating) Model 1b (Competitiveness, deliberate practice, and chess rating)

Model name

0.02†

0.44

0.00†

0.06††

0.02†

RMSEA

48.20

32.59

16.04

35.67

36.13

AIC

Table 4 Fit Indices for the Tested Models on Achievement Motivation (WOFO), Deliberate Practice Motivation (DPMQ), Accumulated Hours of Deliberate Practice, and Chess Rating


account all subscales of the WOFO (Model 1a), provides a better explanation for the data. Although the CFI is acceptable for all models, only for Model 1a is the RMSEA smaller than .05. The model that only relates competitiveness to deliberate practice (Model 1b) also provides a reasonable fit, given the RMSEA of .06. The AICs of the two models are alike, and do not lead to a preference of one model over the other. As to the causal relationship between motivation, deliberate practice, and chess rating, the results support our hypothesis: The model that depicts deliberate practice motivation (DPMQ) as the cause of deliberate practice and chess ratings (i.e., Model 2a) provides a better fit than the model that regards deliberate practice motivation as the result of chess ratings (i.e., Model 2b). This is reflected by both the AIC and the RMSEA index, which are both considerably lower for model 2a than for model 2b (Table 4). For both models, the CFI is considered adequate. Given the correlation between the DPMQ and the WOFO (r = .42, N = 81, p < .01), we were interested in the cumulative effect of motivation as measured by the WOFO and the DPMQ on deliberate practice and chess rating. A model including both these instruments allowed us to assess to what extent achievement motivation and deliberate practice motivation both influence deliberate practice. This model contained the relationship between the WOFO (per subscale) and the DPMQ. Moreover, given the relatively high correlation between the mastery and the competitiveness subscale of the WOFO on one side and the DPMQ on the other side, these paths were also included. The fit of this model (Model 3 in Figure 1) was adequate: The RMSEA was 0.02 and the CFI was higher than 0.90. A comparison of the path coefficients of this model to the previous models reveals that the subscales of the WOFO now contribute less to variance in deliberate practice, and that variance is largely explained by the DPMQ. As we had no model to compare this final model with, the AIC provides little information about fit.

Discussion This study was conducted to shed light on one of the conditions that is theorized to enable deliberate practice, namely, motivation. Our findings indicate that, in a group of elite young chess players, general achievement motivation and chess-specific deliberate practice motivation both contributed significantly to variations in accumulated hours of deliberate practice. Moreover, the lack of fit of the reversed model that depicted motivation as the result of deliberate practice and performance supports Ericsson and colleagues’ (1993) hypothesis that motivation can be viewed as a precondition for engaging in deliberate practice. Our findings suggest that the directly related measure of chess-specific motivation more strongly contributes to deliberate practice in chess than the more distant measure of achievement motivation. That is, the path between deliberate practice motivation and deliberate practice has a higher coefficient and a better fit than the path between achievement motivation and deliberate practice. However, the model that only included achievement motivation (specifically competition) also provided a reasonable fit. This suggests that achievement motivation in itself also exerts a relevant influence on investments in deliberate practice.


Concerning the conflicting findings in previous research about the influence of competition on motivation, and, therefore, on time on task (e.g., Deci et al., 1981; Reeve et al., 1985; Vallerand & Reid, 1984), our results suggest that in high-performing adolescent chess players, besides mastery, competitiveness also contributes to a small extent to optimizing investments in deliberate practice. Not only are chess players confronted with winning or losing after every played game, but they are also provided with performance feedback every 3 months by receiving an updated chess rating. Under these circumstances, those who are not motivated to do their utmost and exemplify low achievement motivation might invest less time in deliberate practice. However, the relatively small difference in competitiveness between persisters and dropouts suggests that this factor does not reliably predict dropout, although it might influence variation in deliberate practice across persisters and dropouts. Our data do provide support for the hypothesis that variance in mastery, and not competitiveness, is related to dropout. For example, the chess dropouts demonstrated lower achievement motivation than the persisters, which could mainly be attributed to a lower mastery orientation. Moreover, the dropouts showed a lower will to excel on the deliberate practice motivation questionnaire. The conceptual relatedness of mastery and the will to excel (i.e., both exemplify the need to achieve mastery and to perform well on a task) suggests that the effect found on the latter subscale cannot be primarily ascribed to the higher number of items this scale consists of, but represents a difference between persisters and dropouts in the underlying mastery motive that these scales measure. It should be noted that, when the persisters and dropouts’ data were aggregated in the structural equation modeling, the relation between mastery and deliberate practice was moderate to low. However, that does not exclude the possibility that levels of mastery can differentiate persisters and dropouts. Our study reveals that besides an ego-involved training environment, as was found in previous research (e.g., Deci et al., 1981; Reeve et al., 1985; Vallerand & Reid, 1984), certain personality features such as low mastery achievement motivation might contribute to chances of dropping out. Given the lack of fit of the model that depicted motivation as the result of chess ratings, the lower achievement motivation observed in the dropouts is probably not the result of lower performance, but is, as shown in the models tested in this study, the cause. Although this is the first time the deliberate practice motivation questionnaire was used, the results provide information about the nature of elite chess players’ motivation to engage in deliberate practice. That is, the underlying factors of the questionnaire suggest that, in elite chess players, motivation is explained by the will to improve performance, but moreover by elements of the will to reach exceptional levels of performance. Having spent hundreds of hours practicing deliberately, these elite chess players are possibly driven more by long-term achievement goals, than by short-term concrete behavior that they wish to improve. This suggests that Ericsson’s characterization of deliberate practice as pursuing a performance improvement goal (Ericsson et al., 1993; Ericsson, 1996) also contains aspects of the will to attain an exceptional level of performance. Research in fields other than chess is needed to shed further light on this issue.


In sum, this is the first study that directly relates variance in motivation to investments in deliberate practice. In order to engage in high levels of deliberate practice, achievement motivation and chess-specific motivation appear to play a pivotal role. From both a training and a selection perspective, these findings can provide useful implications. For instance, in view of the strong relation between motivation and investment in deliberate practice, motivation should be regularly monitored in individuals to optimize levels of deliberate practice. These interventions should be primarily aimed at enhancing chess-specific motivation, as that is considered situational, and can therefore be influenced by environmental factors. Such an intervention could include focusing training on aspects the trainee especially likes, or resolving particular personal or situational issues that might depress motivation. Note that our results emphasize that engagement in deliberate practice is largely determined by chess-specific motivation. This underlines the idea that the motivation to attain expertise in chess is not innate and unchangeable, but can be stimulated by the environment. That is, it is unlikely that individuals are born with a high motivation to play chess. Instead, this motivation will develop as a result of environmental factors, or an interaction of environmental and personality factors. To determine how and through which influences chess-specific motivation develops, research is needed that analyzes this factor over time in young chess players who were recently introduced to the field and have too little experience to have reached a high level of motivation. However, whether the more stable personality factor achievement motivation can be improved through intervention remains subject of further study. Furthermore, if the current results are replicated in future research, it could be relevant to include measures of achievement motivation and domainspecific motivation (in this case, chess-specific motivation) in the selection process for high-level training, as these predict to what extent an individual is willing to commit himself or herself to extended hours of deliberate practice. Before doing so, some issues require further attention. First, similar analyses in other domains besides chess can provide insight into the question to what extent our results are specific to the domain of chess, where individual competition and performance feedback are omnipresent. For example, the relationship between achievement motivation and deliberate practice could be less straightforward in team sports. Second, the retrospective nature of our study did not allow for an analysis of the development of motivation and deliberate practice over time. Achievement motivation is generally considered a stable trait, not easily changed over time. However, chess-specific motivation possibly varies over time as the result of several influences, which could ultimately affect performance (see also Deci, 1971). It could be that the chess players’ motivation fluctuates as a result of particular positive or negative experiences. To provide insight into this issue, future research should take into account the longitudinal development of motivation and concurrent investments in deliberate practice. Moreover, our sample consisted solely of young elite chess players. It would also be interesting to compare motivation of chess players across skill levels and determine to what extent differences in motivation across skills levels influences engagement in deliberate practice. Acknowledgments The authors would like to thank Jeroen Bosch and Esther de Kleuver for their help in designing the questionnaire and recruiting participants, and Mirjam Laging for her help in


collecting the data. The authors would also like to thank Manon de Jong for designing the Internet diary.

References Ames, C. (1992). Achievement goals, motivational climate, and motivational processes. In G. Roberts (Ed.), Motivation in sport and exercise (pp. 161-176). Champaign, IL: Human Kinetics. Arbuckle, J.L., & Wothke, W. (1999). Amos 4.0 user’s guide. Chicago, IL: SmallWaters Corporation. Baker, J., Deakin, J., & Côté, J. (2005). On the utility of deliberate practice: Predicting performance in ultra-endurance triathletes from training indices. International Journal of Sport Psychology, 36, 225-240. Bilali´c, M., McLeod, P., & Gobet, F. (2007). Personality profiles of young chess players. Personality and Individual Differences, 42(6), 901-910. Brewer, B.W., Van Raalte, J.L., & Linder, D.E. (1993). Athletic identity: Hercules muscle or Achilles heel? International Journal of Sport Psychology, 24(2), 237-254. Byrne, B.M. (2001). Structural equation modeling with AMOS: Basic concepts, applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates. Charness, N., Krampe, R., & Mayr, U. (1996). The role of practice and coaching in entrepreneurial skill domains: An international comparison of life-span chess skill acquisition. In K.A. Ericsson (Ed.), The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games (pp. 51-80). Mahwah, NJ: Lawrence Erlbaum Associates. Charness, N., Tuffiash, M., Krampe, R., Reingold, E., & Vasykova, E. (2005). The role of deliberate practice in chess expertise. Applied Cognitive Psychology, 19(2), 151-165. Cohen, J. (1977). Statistical power analysis for the behavioural sciences (rev. ed.). New York: Academic Press. De Bruin, A.B.H., Smits, N., Rikers, R.M.J.P., & Schmidt, H.G. (2007). Deliberate practice predicts performance over time: A linear mixed models analysis. Manuscript submitted for publication. Deci, E.L. (1971). Effects of externally mediated rewards on intrinsic motivation. Journal of Personality and Social Psychology, 18, 105-115. Deci, E.L., Betley, G., Kahle, J., Abrams, L., & Porac, J. (1981). When trying to win: Competition and intrinsic motivation. Journal of Personality and Social Psychology, 7, 79-83. Deci, E.L., & Ryan, R.M. (1985). Intrinsic motivation and self-determination in human behavior. New York: Plenum. Die, A.H., Seelbach, W.C., & Sherman, G.D. (1987). Achievement motivation, achieving styles, and morale in the elderly. Psychology and Aging, 2(4), 407-408. Duffy, L.J., Baluch, B., & Ericsson, K.A. (2004). Dart performance as a function of facets of practice amongst professional and amateur men and women players. International Journal of Sport Psychology, 35, 232-245. Dweck, C.S., & Leggett, E.L. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95(2), 256-273. Elliott, E.S., & Dweck, C.S. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54, 5-12. Ericsson, K.A. (1996). The acquisition of expert performance: An introduction to some of the issues. In K.A. Ericsson (Ed.), The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games (pp. 1-50). Mahwah, NJ: Lawrence Erlbaum Associates. Ericsson, K.A., Krampe, R.Th., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363-406.


Gathercole, S.E., Pickering, S.J., Ambridge, B., & Wearing, H. (2004). The structure of working memory from 4 to 15 years of age. Developmental Psychology, 40, 177-190. Gill, D.L. (1988). The relationship of competitiveness and achievement orientation to participation in sport and nonsport activities. Journal of Sport & Exercise Psychology, 10, 139-150. Gould, D., Dieffenbach, K., & Moffett, A. (2002). Psychological characteristics and their development in Olympic champions. Journal of Applied Sport Psychology, 14(3), 172-204. Guay, F., Vallerand, R.J., & Blanchard, C. (2000). On the assessment of situational intrinsic and extrinsic motivation: The Situational Motivation Scale. Motivation and Emotion, 24, 175-213. Guillet, E., Sarrazin, P., Carpenter, P.J., Trouilloud, D., & Cury, F. (2002). Predicting persistence or withdrawal in female handballers with social exchange theory. International Journal of Sport Psychology, 37(2), 92-104. Helmreich, R.L., & Spence, J.T. (1978). The Work and Family Orientation Questionnaire: An objective instrument to assess components of achievement motivation and attitudes toward family and career. Journal Supplement Abstract Service, Catalog of Selected Documents in Psychology, 8, 35. Helsen, W.F., Starkes, J.L., & Hodges, N.J. (1998). Team sports and the theory of deliberate practice. Journal of Sport & Exercise Psychology, 20, 12-34. Hodge, T., & Deakin, J.M. (1998). Deliberate practice and expertise in the martial arts: The role of context in motor recall. Journal of Sport & Exercise Psychology, 20, 260279. Hodges, N.J., Kerr, T., Starkes, J.L., Weir, P.L., & Nananidou, A. (2004). Predicting performance times from deliberate practice hours for triathletes and swimmers: What, when, and where is practice important? Journal of Experimental Psychology: Applied, 10(4), 219-237. Hodges, N.J., & Starkes, J.L. (1996). Wrestling with the nature of expertise: A sport-specific test of Ericsson, Krampe, and Tesch-Römer’s (1993) theory of “deliberate practice.” International Journal of Sport Psychology, 27, 400-424. Kline, R.B. (1998). Principles and practice of structural equation modeling. New York: The Guildford Press. Luna, B., Carver, K.E., Urban, T.A., Lazar, N.A., & Sweeney, J.A. (2004). Maturation of cognitive processes from late childhood to adulthood. Child Development, 75, 13571372. McAuley, E., & Tammen, V.V. (1989). The effects of subjective and objective competitive outcomes on intrinsic motivation. Journal of Sport & Exercise Psychology, 11, 84-93. Pelletier, L.G., Fortier, M.S., Vallerand, R.J., & Brière, N.M. (2001). Associations among perceived autonomy support, forms of self-regulation, and persistence: A prospective study. Motivation and Emotion, 25(4), 279-306. Pritchard, R.D., Campbell, K.M., & Campbell, D.J. (1977). Effects of extrinsic financial rewards on intrinsic motivation. Journal of Applied Psychology, 62, 9-15. Reeve, J., & Deci, E.L. (1996). Elements of the competitive situation that affect intrinsic motivation. Personality and Social Psychology Bulletin, 22(1), 24-33. Reeve, J., Olson, B.C., & Cole, S.G. (1985). Motivation and performance: Two consequences of winning and losing a competition. Motivation and Emotion, 9, 291-298. Sarrazin, P., Vallerand, R., Guillet, G., Pelletier, L., & Cury, F. (2002). Motivation and dropout in female handballers: A 21-month prospective study. European Journal of Social Psychology, 32, 395-418. Scanlan, T.K., Carpenter, P.J., Schmidt, G.W., Simons, J.P., & Keeler, B. (1993). An introduction to the sport commitment model. Journal of Sport & Exercise Psychology, 15, 1-15.


Schroth, M.L., & Lund, E. (1994). Relationships between need achievement, sensation seeking and cognitive performance. Personality and Individual Differences, 16(6), 861-867. Sloboda, J.A., Davidson, J.W., Howe, M.J.A., & Moore, D.G. (1996). The role of deliberate practice in the development of performing musicians. British Journal of Psychology, 87, 287-309. Spence, J.T., & Helmreich, R.L. (1983). Achievement-related motives and behaviors. In J.T. Spence (Ed.), Achievement and achievement motives: Psychological and sociological approaches (pp. 10-74). San Francisco, CA: W.H. Freeman and Company. Starkes, J., & Ericsson, K.A. (2003). Expert performance in sports: Advances in research on sport expertise. Champaign, IL: Human Kinetics. Vallerand, R.J. (1997). Toward a hierarchical model of intrinsic and extrinsic motivation. In M.P. Zanna (Ed.), Advances in experimental social psychology (Vol. 27, pp. 271-360). New York: Academic Press. Vallerand, R.J., & Bissonnette, R. (1992). Intrinsic, extrinsic, and amotivational styles as predictors of behaviour: a prospective study. Journal of Personality, 60(3), 599-620. Vallerand, R.J., Deci, E.L., & Ryan, R.M. (1987). Intrinsic motivation in sport. In K. Pandolf (Ed.), Exercise and sport science review (pp. 389-425). New York: Macmillan. Vallerand, R.J., & Reid, G. (1984). On the causal effects of perceived competence on intrinsic motivation: A test of cognitive evaluation theory. Journal of Sport Psychology, 6, 94-102. Van Yperen, N.W., & Duda, J.L. (1999). Goal orientations, beliefs about success, and performance improvement among young elite Dutch soccer players. Scandinavian Journal of Medicine & Science in Sports, 9(6), 358-364. Manuscript submitted: July 5, 2006 Revision accepted: May 12, 2007