A longstanding research question in the sport psychology literature has been whether a given amount of mental practice prior to performing a motor skill will enhance one's subsequent performance. The research literature, however, has not provided any clear-cut answers to this question and this has prompted the present, more comprehensive review of existing research using the meta-analytic strategy proposed by Glass (1977). From the 60 studies yielding 146 effect sizes the overall average effect size was .48, which suggests, as did Richardson (1967a), that mentally practicing a motor skill influences performance somewhat better than no practice at all. Effect sizes were also compared on a number of variables thought to moderate the effects of mental practice. Results from these comparisons indicated that studies employing cognitive tasks had larger average effect sizes than motor or strength tasks and that published studies had larger average effect sizes than unpublished studies. These findings are discussed in relation to several existing explanations for mental practice and four theoretical propositions are advanced.

The authors present a conceptual model of coaching efficacy and develop a reliable and valid instrument to measure the concept and to examine its hypothesized sources and outcomes. Coaching efficacy is defined as the extent to which coaches believe they have the capacity to affect the learning and performance of their athletes. Using high school coaches, confirmatory factor analysis supported 4 dimensions of the Coaching Efficacy Scale: game strategy, motivation, teaching technique, and character building. Marginal support was also found for 1 general coaching efficacy factor that explained the correlations among the 4 first-order factors. Using a separate sample of high school basketball coaches, coaching efficacy was predicted by a coach's past success, coaching experience, perceived player talent, and social support. In turn, coaching efficacy predicted coaching behavior, player satisfaction, and current success. Results establish preliminary support for the conceptual model of coaching efficacy.Self-efficacy for classroom teaching has been studied extensively and shown to be an important aspect of teaching effectiveness (see, e.g.,

This meta-analysis examined the relationship between self-efficacy and performance in sport. Based on 45 studies (102 correlations), the average correlation between self-efficacy and sport performance was .38. Given the heterogeneity of findings, follow-up univariate and multivariate moderator analyses were conducted. Results indicated that the most important moderator was concordance, thereby highlighting the importance of matching the self-efficacy and performance measures. Additional moderators we examined included the types of self-efficacy measures, the types of performance measures, the nature of the task, and the time of assessments. These variables accounted for approximately 44% of the variance in the self-efficacy-performance relationship. Practical implications of the findings are discussed.

The multidimensional approach to the study of anxiety (Martens, Vealey, & Burton, 1990a) considers subcomponents of anxiety, specifically cognitive anxiety, somatic anxiety, and self-confidence. Much of the research based on this theory has utilized the Competitive State Anxiety Inventory (CSAI-2) (Martens, Burton, Vealey, Bump, & Smith, 1990b). Findings have been inconsistent, with some research suggesting that the three subcomponents have separate relationships with performance and other studies failing to find any relationship between the anxiety subcomponents and performance. This meta-analysis examined the effect of state anxiety as measured by the CSAI-2 (i.e., cognitive anxiety, somatic anxiety, and self-confidence) on athletic performance. Studies were coded for characteristics that could potentially moderate the effects of anxiety on performance (i.e., features of design, subjects, sport). Interdependency between the three subscales was examined using multivariate meta-analytic techniques (Becker & Schram, 1994). Relationships among cognitive anxiety, somatic anxiety, self-confidence, and performance appeared weak. Exploratory modeling showed that self-confidence displayed the strongest and most consistent relationship with performance.

No abstract