KEY WORDS: Goal orientation, mental toughness, performance outcome.

Wushu encompasses all types of Chinese martial art. It began as an art of self defence to protect against wild animals and other humans. It was later used in combat, including war, and then gradually maturing into a fully developed international sport. This study examines goal profiles of wushu athletes using achievement theory framework (Nicholls, 1989) to capture the emphasis laid on task and ego orientations, and determine its relationship with mental toughness and performance outcomes in competition.

Goal perspective theory holds that personal goals influence an individual's thinking, feeling and actions in an achievement situation such as competition (Nicholls, 1989). It is assumed that in a competitive situation, the prevailing achievement goals, both task and ego relate to one's judgment on level of competence and subjectively define successful goal accomplishment (Nicholls, 1989). While task oriented athlete's judgment about success is self-referenced, an ego oriented athlete's judgment about success is on beating other competitors. Duda et al., 1995 indicated that the performance related feedback of an individual's sport ability can have implications for an athlete's motivational orientation in terms of task and ego involvement. Duda et al., 1995 further indicated that performance outcome in a competition helps to ascertain an athlete's improvement (a task conception of ability) and/or are better/worse than others (an ego conception of ability).

Dimensions of task and ego orientations have been reported to be independent in sport settings (Chi and Duda, 1995; Duda and Whitehead, 1998; Robert et al., 1996), a proposal that concurs with Nicholls, 1989 who argued that task and ego orientations are orthogonal constructs. Harwood and Treasure, 2000 report the main effect of goal orientation separately without considering the interaction of task and ego orientations. However, when interactions of task and ego orientations are considered, four goal profiles are found: high task/high ego, low task/low ego, high task/low ego, low task/high ego. In other words, athletes can be high in orientations, low in both, or high in one orientation and low in the other. Hodge and Petlichkoff, 2000 examined goal orientation profiles of male rugby players and indicated that moderate task/high ego profile resulted in a perception of better physical abilities. Hodge and Petlichkoff, 2000 concluded that moderate to high task and ego orientation pattern may compliment each other to adapt in a competitive sport context. Duda, 1997 suggested that most achievement-oriented profiles may consist of both high task and high ego goals. Duda, 1997 further suggested that high task and high ego orientation would provide better opportunities for success. Duda and Treasure, 2001 suggested that individuals with high task and high ego orientations would display adaptive goal profile enabling them to meet the demands of competition.

Therefore, in the present study, in order to avoid imposing restriction on the data, into four arbitrary group structures, and to allow room for moderate responses, we adopted to employ cluster analysis in this study, despite the limitation of sample size.

A factor often associated with successful performance in competition is mental toughness. Mental toughness can be considered as a mental skill factor. Some research findings has identified mental skills as a psychological construct that distinguishes between more and less successful performance across a number of sports; for example, golf (Thomas and Over, 1994), and equestrian, (Meyers et al., 1998). Mental toughness and its importance in competitive sports have been documented in literature (see Goldberg, 1998; Hodge, 1994; Tunney, 1987; Williams, 1988). Loehr (1982; 1986) suggested that fifty percent of success in competitions could be attributed to mental toughness in athletes. On similar lines, Gould et al., 1987 indicated that coaches felt the importance of being mentally tough in achieving success in sports. Norris, 1999 also emphasized the importance of mental toughness in the making of a champion athlete. Gould et al. , 2002 studied the psychological characteristics of Olympic champions, and identified mental toughness as a significant contributor to sports performance enhancement.

A clear definition for mental toughness is still being researched. Loehr, 1986 suggested that mentally tough performers are disciplined thinkers and respond to pressure by remaining relaxed, calm and energized. Loehr, 1986 further suggested that mentally tough athletes have the ability to increase their flow of positive energy in adversities. Although recent research findings identified twelve attributes of mental toughness (see Fourie and Potgieter, 2001; Jones et al., 2002; Middleton et al., 2005), the seven fundamental attributes of mental toughness suggested by Loehr, 1986, show similarities to those identified by recent researchers like Fourie and Potgieter, 2001; Jones et al., 2002 ; Middleton et al., 2005. Therefore in this study we considered the seven fundamental attributes of mental toughness suggested by Loehr, 1986. Specifically, the mental toughness attributes include (1) self-confidence (i.e. belief that on e can perform well and be successful), (2) negative energy control (i.e. to cope with negative emotions such as fear, anger, frustration and temper for achieving success), (3) attention control (i.e. stay focused and to perform well), (4) visualization and imagery control (i.e. creating positive mental images), (5) motivation level (i.e. the energy and willingness to persevere), (6) positive energy control (i.e. energized with fun, joy and satisfaction), and (7) attitude control (i.e. habits of thought and unyielding). Golby and Sheard, 2003 studied mental toughness at different levels of rugby league and reported that the athletes scored significantly higher on two of the seven mental toughness subscales (negative energy control and attention control).

Sports performance in a competition is typically assessed through outcome measure. The outcome measure in Wushu in a competition is subjectively decided based on the style of different movements. A major challenge of the present study is the inclusion of performance outcome measure, considered as winning a medal in the intervarsity competition 2006, for the exclusive purpose of this study. We acknowledge the limitation that medalling in Wushu is a variable affected by different aspects like, an athlete's ability, the opponent's strengths or weakness, subjective bias in evaluation, an athlete's performance quality, level of competition, etc.

An unpublished research by Kuan, 2007 relating goal profile, and mental toughness conducted on 203 state level male athletes, belonging to a particular state of Malaysia revealed that, athletes with goal profile high task/high ego, and high task/moderate ego scored significantly higher on the subscales of mental toughness. To date, no information about the relationship between goal orientation, mental toughness and performance outcome exists in Wushu. To test the above relation the purpose of this study, were 1) to determine if mental toughness differed as a function of goal profile; 2) to determine if athletes with successful performance outcome displayed any particular pattern of goal profile; 3) to assess if athletes with successful performance outcome differed significantly in mental toughness.

Descriptive statistics
Means standard deviations and Cronbach's alpha for entire sample are presented in Table 1. Positive correlations were observed between goal orientation (task and ego) with subscales of mental toughness (Table 2). The correlation between task and ego(r =. 36, p < .05), task and motivation level (r = .45, p < .05), task and positive energy control (r =.54, p < .05), task and attitude control (r =.36, p < .05), ego and negative energy control (r = .36, p < .05) were moderate, but significant.

Age, gender and competitive experience
Previous research has indicated differences in goal orientations between gender and individuals of different ages (e.g., Newton and Duda, 1993; Petlichkoff, 1993; White and Zellner, 1996). However, Cumming et al., 2002 reported no differences across ages, gender and competitive standards in their study. In this study, separate MANOVAs were conducted to determine whether any differences occurred across age, gender and years of competitive experiences. No significant age differences among the participants (Pillai's Trace = 0.32, F_9,26 = 1.36, p = 0.258, eta² = 0.32), between the gender (Pillai's Trace = 0.40, F_9,26 = 1.94, p = 0.091, eta² = 0.40) and years of competitive experiences (Pillai's Trace = 0.37, F_9,26 = 1.66, p = 0.149, eta² = 0.37) were found. Therefore, the data were collapsed across age, gender and year of competitive experiences.

Goal profiles
To identify the different clusters based on task and ego orientation scores, goal profiles were generated using a combination of hierarchical method and non hierarchical k-means cluster analysis (see Hodge and Petlichkoff, 2000; Wang and Biddle, 2001). All variables were standardized using z-scores as a standard procedure of cluster analysis. Two cases in this sample were detected as outliers (|z-scores| < 3.00) and were deleted from further analyses. Using dendrogram and agglomeration schedule from the hierarchical method of cluster analysis, a three-cluster solution was found suitable. Additional analysis using k-mean clustering method showed that there were lack of participants in each profile group from four-cluster and five-cluster solutions and the lack of significant difference among the two-cluster solution indicated that the three-cluster solution was the most robust. The stability testing on the three-cluster solution was done using two-thirds random sample. The re-cluster analysis indicated a stable cluster pattern, with approximately 88.5% of participants maintaining their original cluster membership.

The final clusters identified were high task and moderate ego (HT/ME), moderate task and low ego (MT/LE) and, moderate task and moderate ego (MT/ME) (Table 3).

Goal profile and mental toughness
Table 4 illustrated the mean and standard deviation of the dependent measures in each group profiles. The results show that athletes in cluster 1 high-task/moderate-ego scored higher mean on five variables viz. negative energy control, visualization and imagery control, motivational level, positive energy control and attitude control. To determine the effect of goal profile on the fundamental areas of mental toughness, MANOVA was done. The results revealed significant multivariate effect for goal profile on the dependent measures (Pillai's Trace = 0.62, F_14,60 = 1.94, p = 0.039, eta² = 0.31). Further univariate analyses revealed significant results for the negative energy control (F_2,35 = 5.81, p = 0.007) and positive control energy (F_2,35 = 3.60, p = 0.038). Tukey post-hoc tests indicated that Wushu athletes with high task and moderate ego (cluster 1) scored significantly higher on negative energy control than athletes in moderate task and low ego (cluster 2) (p = 0.005). Furthermore, athletes in high task and moderate ego (cluster 1) also scored significantly higher on positive energy control than athletes in moderate task and moderate ego (cluster 3) (p = 0.035).

In order to examine whether successful performance outcome in competition was a function of goal profiles, Chi-square (χ²) test was adopted. No significant differences were found among athletes with different goal profiles and medallist status, χ² = 0.98, p = 0.612. This indicated that successful outcome at the competition were not merely a function of goal profiles. Further to determine whether mental toughness variables had any influence on successful performance outcomes, we performed univariate t-test between medallist and non medallist. The results indicated a significant difference between medallist and non medallist in self-confidence (p = 0.001) and negative energy control (p = 0.042). The medallist scored significantly higher on self confidence (mean = 21.82, SD = 2.72) and negative energy control (mean = 19.59, SD = 2.32) than the non medallist (self confidence-mean = 18.76, SD = 2.49; negative energy control mean = 18.14, SD = 1.91).

• Mental toughness can be influenced by certain goal profile combination.
• Athletes with successful outcomes in performance (medallist) displayed greater mental toughness.