The Importance of Analyzing Position-Specific Self-Efficacy.

Weigand, Daniel A. ; Stockham, Kimberley J.

The purposes of this study were to assess perceptions of position-specific and cross-skill self-efficacy in a team sport and to assess the effect of competition level on skill-specific self-efficacy. Data were collected from 110 British amateur players (M age = 26.93, SD = 5.8) competing in either a 1st, 2nd, or 3rd Division team of the Godfrey Davis Women's Field Hockey League. One hour prior to a league game players completed measures including 8 items assessing the magnitude and strength (Bandura, 1986) of skill-specific self-efficacy. Based on the results of a pilot study, the 8 skills were considered to be task-specific to forward (3 items), defensive (3 items), and midfield (2 items) positions. Results of 3 X 3 (Position by Division) between subjects MANOVA/ANOVAs and Scheffe tests indicated that forwards scored significantly (p [less than] .05) higher on the forward-specific self-efficacy scores than either midfielders or defense. Furthermore, 1st Division athletes scored significantly (p [less than] .05) higher on the forward-specific self-efficacy scores than either 2nd or 3rd Division athletes. There were no significant differences found between athletes on the measure of cross-skill self-efficacy. Findings are discussed with regard to the need to examine skill-specific, rather than generalized, self-efficacy in the relationship between self-efficacy and performance in sport.

According to Roberts (1992), "The theory of self-efficacy has been the most extensively used theory for investigating motivational issues in sport and exercise" (p. 11). Self-efficacy theory is a social-cognitive approach to behavior that takes into account behavioral, physiological, and cognitive factors, as well as environmental influences. According to McAuley (1992), the theory provides "[ldots]a common mechanism through which people demonstrate control over their own motivation and behavior" (p. 109) because the theory focuses on the role of self-referent thought in relation to psychosocial functioning. Bandura (1977, 1997) suggested that self-efficacy affects choice of activities, effort expenditure, persistence, and achievement. Therefore, because self-efficacy can be broadly defined as the beliefs that individuals have in their capability to engage successfully in a course of action sufficient to satisfy the situational demands (McAuley, 1992), self-efficacy can be seen to be an important determinant of physical activity and sport behavior. Athletes who possess higher amounts of self-efficacy are more likely to choose tasks they can accomplish, work harder, persist longer when they experience difficulties, and achieve at a higher level than those with lower levels of self-efficacy (Schunk, 1995).

Research has shown that self-efficacy is a determinant of educational, social, clinical, and health-related behaviors (see O'Leary, 1985; and Schunk, 1989). Furthermore, according to recent reviews (Feltz, 1992; McAuley, 1992), self-efficacy theory also shows considerable promise for explaining motivation and behavior in sport and exercise contexts. It seems that repeated successes raise an individual's efficacy appraisals and also increases motivation, while repeated failures can lower these appraisals, which also may decrease motivation (Bandura, 1977). Individual success in team sports, such as field hockey, is often skill (or position) specific. For example, defensive players may define success in terms of their ability to keep the other team from scoring or being able to consistently hit the ball a long distance. Conversely, forwards in field hockey need to be accurate at shots on goal and have the ability to dribble the ball successfully down the field. Thus, is appears that different skills are requir ed for different positions that are largely dependent upon the particular abilities individuals possess. Team-sport athletes are therefore likely to develop certain skills (those more required for certain desired positions) at the loss of other, less relevant skills. Consequently, their self-efficacy for certain skills required by specific positions may be higher than for other positions.

Although the findings from sport-related efficacy research are compelling, the measurement of self-efficacy has been inconsistent (Feltz & Chase, in press). According to Bandura (1986), the measurement of self-efficacy cognitions should be carried out in a microanalytical fashion by assessing an efficacy task specifically along three dimensions: level, strength, and generality. The level of self-efficacy concerns the expected performance attainment of individuals (e.g., whether athletes think they can perform a task or not), while strength is the certainty with which they expect to successfully attain the task (e.g., on a 0-100 point scale). Generality, however, refers to the number of domains in which individuals feel that they are efficacious. Thus, someone who has a high perception of self-efficacy in running may have an equally high degree of self-efficacy in biking.

In contrast, Bandura's conception of generality also suggests that a high level of self-efficacy in one domain does not necessarily result in a high level of self-efficacy in another domain (McAuley 1992). That is, there is a potential for specificity of self-efficacy perceptions. For example, an individual with high expectations in tennis may not have similar expectations in badminton. Although both sports use a racquet to hit an object over a net, the biomechanics and motor skills require sufficiently diverse abilities that self-efficacy perceptions may significantly differ. In a related manner, team-sport athletes may have developed skill-specific expertise required for certain positions, at the expense of skill development necessary to be successful at other positions. For instance, point guards in basketball may develop expertise in dribbling and passing the ball, whereas centers may not be able to dribble or pass as well as guards, but they should be able to rebound better than guards.

In summary, based on Bandura's (1986) suggestions, it appears that self-efficacy in sport or exercise should be measured by evaluating: (a) the number of tasks that individuals can expect to perform leading up to a target behavior, (b) the confidence with which they expect to successfully attain each aspect of the target behavior, and (c) the number of domains or skills in which they consider themselves efficacious. Thus, the first purpose of this study was to assess perceptions of cross-skill and skill-specific (i.e., position-specific) self-efficacy in British Women's Field Hockey. In field hockey, forwards need to be able to demonstrate highly accurate shots at the goal and an ability to dribble the ball successfully. Conversely, defensive players are required to successfully defend short corner shots and be able to push the ball a long distance. The role of midfielders is a combination of the roles of forwards and the defense; they need to be able to attack as well as defend. Considering the different ro les that players have while on the pitch, it can be assumed that field hockey players may have different perceptions of their abilities to perform task-specific skills according to the positions they play. Therefore, in support of the specificity aspect of Bandura's generality principle, it was hypothesized that athletes' self-efficacy would be significantly higher for the skills relevant to their position than for the skills relevant to other positions. Moreover, it also was expected that there would be no significant differences between athletes on a measure of cross-skill self-efficacy.

An additional purpose of this study was to assess the effect of competition level on skill-specific self-efficacy. According to Bandura (1977), performance accomplishments provide the most dependable and influential source of self-efficacy. Thus, it is possible that athletes competing at higher levels of ability might have higher performance expectations than those competing at lower levels of ability. Furthermore, athletes at higher levels of competition might have developed their position-specific skills more so than those who compete at lower levels. Therefore, the final hypothesis was that athletes competing at higher levels of competition would have higher position-specific self-efficacy than those competing at lower levels of competition.

Method

Participants and Procedures

Data were collected from 110 female athletes (Mage = 26.93, SD = 5.8) who trained and played amateur British field hockey at least twice a week throughout the season. Participants played for either a 1st Division (n = 36), 2nd Division (n = 38), or 3rd Division (n = 36) team. Athletes also were classified as either forwards (n = 37), midfielders (n = 36), or defense (n = 37). After obtaining informed consent from each participant, questionnaires were administered approximately one hour before a game of the Godfrey Davis Women's League. The questionnaires required approximately 15 minutes to complete. Respondents were encouraged to ask the researcher about any questions that arose during administration of the questionnaire. The participants also were reminded that their participation was voluntary and that all responses would be anonymous.

Measures

Self-Efficacy. Based on the recommendations of Locke and Latham (1990), an instrument was designed to assess cross-skill and skill-specific self-efficacy in field hockey. A pilot study was conducted to evaluate skills required for effective performance at forward, midfield, and defense positions. Interviews with 10 female 1st Division hockey players were conducted by the second author (a collegiate field hockey player) approximately one hour before a league game. Based on responses from the participants in the pilot study, self- efficacy was assessed by having players record the level to which they thought they were capable of performing eight items considered to be major components of the game and task- specific to forward (3 items), defensive (3 items), and midfield (2 items) positions. Skills that were considered to be most relevant to forwards included open-field scoring, dribbling, and penalty shots. The items related to defensive positions were defending short corners, tackling, and hitting the ball fo r distance. Pushing the ball and lifting an aerial ball successfully were the skills identified as important for midfielders. Participants were first required to indicate (in Column A) whether or not they felt capable (Yes or No) of executing each skill at four ascending levels of difficulty (e.g., I can hit the ball 16, 25, 50, or 75 yards), then were required (in Column B) to rate their certainty of performing each level from 0% (Extremely Uncertain) to 100% (Absolute Certain). They also were instructed that a "No" response in Column A equaled 0% in Column B. Cross-skill self-efficacy was scored by calculating a mean score including all eight items. Higher scores represented greater self-efficacy. This cross-skill scale generated an acceptable internal consistency: Cronbach's (1951) alpha = .78. Skill-specific self-efficacy scores were the values generated by the position-specific items separately. Only two of these three scales generated acceptable internal consistency: forward-specific alpha = .70 and def ense-specific alpha = .67 (Weiss et al. [1985] identified alpha coefficients of .60 as acceptable for the internal consistency of a sport scale, therefore the defense-specific scale was considered acceptable). The midfield-specific scale was not considered to be internally consistent (alpha = .24) and was therefore excluded from additional analyses.

Results

Descriptive statistics for the cross-skill and position-specific self-efficacy scores are presented in Table 1. Results of a 3 X 3 (Position by Division) between subjects ANOVA on the cross-skill self-efficacy scores revealed no significant main effects for position, F (2, 101) = 0.44, p [greater than] .05, nor division, F (2, 101) = 0.47, p [greater than] .05, nor was there an interaction effect, F (4, 101) = 0.46, p [greater than] .05. In contrast, results of a 3 X 3 (Position by Division) between subjects MANOVA on the forward- and defense-specific self-efficacy scores revealed significant main effects for position, F (4, 200) = 7.69, p [less than] .001, and division, F (4, 200) = 2.83, p [less than] .05, but not an interaction, F (8, 200) = 0.78, p [greater than] .05. Results of subsequent 3 X 3 (Position by Division) univariate ANOVAs revealed that the significant main effects were generated on the forward-specific self-efficacy scores, [F.sup.Position] (2, 101) = 8.09, p [less than] .001; [F.sup.Division] (2, 101) = 4.21, p [less than] .05. Specifically, post hoc Scheffe tests indicated that forwards scored significantly (p [less than] .05) higher (M = 62.55) on the forward-specific self-efficacy scores than either midfielders (M = 48.30) or defense (M = 45.27); 1st Division athletes scored significantly (p [less than] .05) higher (M= 60.01) on the forward-specific self-efficacy scores than either 2nd (M = 48.06) or 3rd (M = 48.46) Division athletes. In summary, these results indicate that, in comparison to a cross-skill measure of self-efficacy, significant differences between positions and Divisions emerged only when position-specific self-efficacy was assessed.

Discussion

The purposes of this study were to assess perceptions of position-specific and cross-skill self-efficacy in a team sport and to assess the effect of competition level on skill-specific self-efficacy. Results of analyses of variance and post hoc tests provide partial support for the hypothesis that position-specific efficacy expectations were significantly higher for the skills relevant to specific positions than for the skills relevant to other positions. Specifically, self-efficacy perceptions for forwards were significantly higher than either midfielders or those on defense for the skills of open-field scoring, dribbling, and taking penalty shots, skills that most field hockey experts would agree are more important for forwards than for others. Furthermore, in support of the second hypothesis, there were no significant differences found between athletes on a measure of cross-skill self-efficacy.

Results of analyses of variance and post hoc tests also provide partial support for the final hypothesis. Specifically, 1st Division players had significantly higher self-efficacy perceptions than either 2nd or 3rd Division athletes for those skills deemed to be most appropriate for forwards.

The results of this investigation seem to support Bandura's (1986) belief that the measurement of self-efficacy, at least in a team sport, should be carried out in a microanalytical fashion by assessing task-specificity. That is, rather than measuring cross-skill efficacy expectations, including various sport-specific skills, one should take into account the potential ability differences between individuals at different positions and assess skill-specific self-efficacy. Furthermore, athletes competing at different levels of competition may have diverse performance expectations that relate more to specific, than general, task-related abilities. These findings have implications for applied sport psychologists.

Mental skills training techniques should address each of Bandura's (1977, 1989, cited in Feltz, 1992) four primary sources of self-efficacy information (performance accomplishments, vicarious experiences, verbal persuasion, and physiological states) with a focus on developing skill-specific self-efficacy. Some of these areas have been developed by Schunk (1995) in a comprehensive review of self-efficacy, motivation, and performance, although not with specific reference to skill-specific self-efficacy. First, Schunk emphasized the importance of assessing how procedures affect self-efficacy and motivation. That is, in addition to assessing how interventions affect performance outcomes, educators and coaches should be providing instruction that "includes periods of self-directed mastery or independent practice where learners practice skills on their own" (p. 132). This should not only facilitate skill acquisition, but also enhance self-efficacy. The current findings suggest that in team sports these self-direct ed mastery experiences should be skill-specific. Athletes need to develop certain abilities to successfully perform certain roles on a team. Each role is therefore likely to generate specific efficacy expectations. Performance accomplishments are often skill-specific in a team sport (although they also could be team oriented, e.g., team effort, team communication, team work, cf Zaccaro, Blair, Peterson, & Zazanis, 1995). If sport psychologists are going to facilitate enhanced self-efficacy for athletes on a team, then they should recommend that coaches provide opportunities for individuals to develop skills relevant to specific roles and abilities.

Schunk (1995) also indicated that the research on self-efficacy in education and sport suggests that peers should be employed as models. Vicarious experiences are important sources of information and the research seems to suggest that, although adults (e.g., teachers and coaches) make good models for teaching skills, peers may make better models. Self-efficacy may be better enhanced by models who are more similar to the observers than those more expert on the skills being developed. Thus, sport psychologists working with teams may want to recommend that coaches identify individuals who are sufficiently competent in the skills being developed to use as models for those who are not as well developed. The present results further suggest that in team sports these models should be demonstrating skills that are specific to certain roles. By viewing similar others demonstrating competent role-specific skill execution, the observers should better concentrate on what is being said and shown and consequently improve t heir performance and enhance their self-efficacy.

Another of Schunk's (1995) recommendations was that teachers and coaches should provide specific and credible feedback to learners. According to Schunk, research in sport and education has shown that specific "feedback that denotes how performance has improved is likely to raise self-efficacy and motivation" (p. 132). It is not enough for coaches to tell their athletes that they have "done well" unless the athletes know what specifically they have done well. Moreover, coaches will not facilitate self-efficacy by telling athletes that they are improving if the athletes believe they are struggling. Sport psychologist should therefore suggest to coaches that verbal persuasion should be realistic and targeted toward specific skills that are developing or need development. If athletes are aware of their strengths and weaknesses, they will have more realistic expectations about their abilities and understand where they should be putting forth more or less effort in practice.

The last of Schunk's (1995) suggestions is that coaches and teachers should use goals to foster commitment. According to Bandura (1989, cited in Feltz, 1992), one source of efficacy information is emotional arousal and research (Locke & Latham, 1990) has shown that goals motivate people to exert effort necessary to meet task demands and select appropriate strategies for task completion, which should reduce anxiety and enhance self-efficacy. Athletes who set and attain goals may experience an increase in self-efficacy which may facilitate setting more difficult goals. Achieving these more difficult goals will again enhance self-efficacy. Thus, as athletes observe goal progress, they are becoming more skillful, and their self-efficacy should continue to increase as well. The current findings also suggest that in team sports goals should be skill-specific. That is, athletes should be setting goals that focus on the skills necessary for effective performance according to the position in which they want to develo p skills. One caveat needs discussing. Although the research seems to suggest that goals may be effective in sport, the literature also contains studies with conflicting or inconsistent results on the effects of goal setting in sport (see Locke, 1991, 1994; Weinberg & Weigand, 1993, 1996). Thus, additional research is needed to better understand the influence that goals have on self-efficacy and performance in sport.

In conclusion, although the results of this study should be taken as preliminary, the findings seem to suggest that self-efficacy should be considered relative to role-specific skills, at least in team sports, rather than as a measure of cross-skill self-efficacy across the various skills necessary for success in a particular sport. Sport psychologists should be aware of this when designing mental skills training packages and recommend that coaches enhance performance and self-efficacy by focusing on the skills relevant to positions and roles.

Daniel A. Weigand, Ph.D., is a Principal Lecturer of Sport & Exercise Psychology in the School of Physical Education, Sport, & Leisure at De Montfort University, Bedford. Kimberley J. Stockham is a graduate of De Montfort University, Bedford.

This investigation is based on an undergraduate honours degree study at De Montfort University, Bedford, completed by Kimberley J. Stockham under the supervision of Daniel A. Weigand, Ph.D.

References

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215.

Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice-Hall.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16,297-334.

Feltz, D. L. (1992). Understanding motivation in sport: A self-efficacy perspective. In G. C. Roberts (Ed.), Motivation in Sport and Exercise (pp. 93-105). Champaign Il: Human Kinetics.

Feltz, D. L., & Chase, M. A. (in press). The measurement of self-efficacy and confidence in sport. In J. Duda (Ed.), Advancements in sport and exercise psychology measurement. Morgantown, WV: Fitness Information Technology.

Locke, E. A. (1991). Problems with goal-setting research in sports-and their solution. Journal of Sport & Exercise Psychology, 13, 311-316.

Locke, E. A. (1994). Comments on Weinberg and Weigand. Journal of Sport & Exercise Psychology, 16,212-215.

Locke, E. A., & Latham, G. P. (1990). A Theory of goal-setting and task performance. Englewood Cliffs, NJ: Prentice Hall.

McAuley, E. (1992). Understanding exercise behavior: A self-efficacy perspective. In G. C. Roberts (Ed.), Motivation in sport and exercise (pp. 107-127). Champaign Il: Human Kinetics.

O'leary, A. (1985). Self-efficacy and health. Behavior and Research Therapy, 23,437-451.

Roberts, G. C. (1992). Motivation in sport and exercise: Conceptual constraints and convergence. In G. C. Roberts (Ed.), Motivation in sport and exercise (pp. 3-29). Champaign Il: Human Kinetics.

Schunk, D. H. (1989). Self-efficacy and achievement behaviors. Educational Psychology Review, 1, 173-208.

Schunk, D. H. (1995). Self-efficacy, motivation, and performance. Journal of Applied Sport Psychology, 7, 112-137.

Weinberg, R., & Weigand, D. (1993). Goal setting in sport and exercise: A reaction to Locke. Journal of Sport & Exercise Psychology, 15, 88-96.

Weinberg, R. S., & Weigand, D. A. (1996). Let the discussions continue: A reaction to Locke's comments on Weinberg and Weigand. Journal of Sport & Exercise Psychology, 18,89-93.

Weiss, M.R., Bredemeier, B.J., & Shewchuk, R. (1985). An intrinsic/extrinsic sport motivation scale for the youth sport setting: A confirmatory factor analysis. Journal of Sport Psychology, 7,75-91.

Zaccaro, S. J., Blair, V., Peterson, C., & Zazanis, M. (1995). Collective efficacy. In J. E. Maddux (Ed.), Self-efficacy adaptation and adjustment: Theory, research and application (pp. 308-330). New York: Plenum.

 Means and Standard Deviations Across Positions and Divisions on
 Cross-Skill and Position-Specific Self-Efficacy
 Forwards [a] Midfielders [b]
Self-efficacy M SD M SD
Cross-skill 64.56 14.24 56.87 15.17
Forward-specific 62.55 [12] 19.09 48.30 [1] 21.53
Defense-specific 74.58 16.46 74.81 17.24
 1st Division [e] 2nd Division [f]
Cross-skill 64.06 12.99 58.11 12.87
Forward-specific 60.01 [12] 19.39 48.06 [1] 19.04
Defense-specific 76.70 12.28 75.88 15.72
 Defense [c]
Self-efficacy M SD F [d]
Cross-skill 59.79 14.00 0.44
Forward-specific 45.27 [2] 22.02 8.09 [***]
Defense-specific 81.14 13.04 1.92
 3rd Division [g]
Cross-skill 59.20 17.56 0.47
Forward-specific 48.46 [2] 25.69 4.21 [*]
Defense-specific 77.88 19.21 0.13
Note. Midfield-specific self-efficacy scores are excluded due
to the scale's lack of internal consistency. Variables are
percentages of certainty of executing the skills. Similar
subscripts indicate groups that are significantly (p [less
than] .05) different.
(a.)n = 37.
(b.)n = 36.
(c.)n/ = 37.
(d.)(*.)p [less than] .05,
(**.)p [less than] .01,
(***.)p [less than] .001.
(e.)n = 36.
(f.)n = 38.
(g.)n = 36.