Abstract: This study aimed at testing Harwood, Hardy and Swain’s (2000) proposed 3-dimensional model of achievement goals in sport. The three dimensions are pure task, self-referenced ego involvement and norm-referenced ego involvement. Their proposal came about in the context of criticisms of Nicholl’s (1984, 1987) popular 2-dimensional task-ego goal orientation theory in sport and its related instruments, namely, Task and Ego Orientation Sport Questionnaire and Perception of Success Questionnaire. Harwood, Hardy and Swain questioned (a) the rationale of the differentiation of effort and ability in competitions; (b) the appropriateness of using the two questionnaires as the items have failed to effectively discriminate among individuals who are truly high, low or moderate in task or ego involvement.
Two studies were conducted. Study 1 was the development of a 9-item Achievement Goal Questionnaire (AGQ). Reliability and validity of the questionnaire were assessed. The objective of Study 2 was to assess the nature of achievement goal orientation and involvement states of four elite table tennis players (M=2; F=2) before, during and after competition. Four important matches were selected and videotaped for interviews, conducted within three days of an important match. Any changes in the achievement goal orientation and involvement states were noted when the players recalled experiencing an adverse or good situation during the match in question. Qualitative method was used to analyze the interview content.
Confirmatory factor analysis resulted in a clear 3-factor solution. Test-retest (2 weeks apart) statistics showed good reliability of each of the three factors. Consistent results of the coaches’ evaluation of the four players and the players’ responses to the AGQ lent further support to the convergent validity of the questionnaire. Findings of the four players in Study 2 suggested: (1) there was an inconsistency between their achievement goal orientation and involvement states before an important match, reflecting the influence of individual characteristics and the qualities of particular opponents; (2) before and during competition goal involvement states showed a multiple-goal tendency, players could emphasize on a single goal as well as two or three goals at the same time; (3) after competition goal involvement states tended to be affected by competition process and result; and (4)sub-goal involvement states (characterized by a lack of strong achievement goal motivation) was also observed when players experienced major setbacks in competition.
Results of this study supported Harwood, Hardy and Swain’s 3-dimensional model of achievement goals. A reliable and valid questionnaire was developed to test the model. The questionnaire should, however, be further refined by using it for other sports and a wider variety of athletes. In addition, the study showed the nature of goal orientations and goal involvement states of players before, during and after competition. Qualitative and quantitative methods were used for data analysis. The study also implied that an understanding of the nature of achievement goals in athletes may better help them cope with competition stress. For example, it is hypothesized that an athlete with self-referenced ego involvement (rather than norm-referenced) might be more likely to interpret a stressful event (e.g., competition) as a challenge than a threat or harm, then he/she might be able to cope with stress more effectively. Perhaps, athletes can be trained to appraise their achievement goals differently when under different competition stresses

Key words: achievement goal, goal orientation, goal involvement states, table tennis player