Content uploaded by Deborah Feltz
Author content
All content in this area was uploaded by Deborah Feltz
Content may be subject to copyright.
1
Self-efficacy Beliefs of Athletes, Teams, and Coaches
Deborah L. Feltz Cathy D. Lirgg
Michigan State University University of Arkansas
(2001). In R. N. Singer, H. A. Hausenblas, & C. Janelle (Eds.), Handbook of Sport
Psychology, 2
nd
ed. (pp. 340-361). New York: John Wiley & Sons.
Direct correspondence to: Deborah L. Feltz, Department of Kinesiology, 134 I. M. Sports-
Circle, Michigan State University, East Lansing, MI 48824. dfeltz@msu.edu
2
Chapter: Self-efficacy Beliefs of Athletes, Teams, and Coaches
The self-efficacy construct is one of the most influential psychological constructs
thought to affect achievement strivings in sport (Feltz, 1988). Gould and his colleagues
found that self-efficacy and team efficacy were chief among that factors that US Olympic
athletes reported to influence their performance at the Nagano Olympic games (Gould,
Greenleaf, Lauer, & Chung, 1999). Bandura (1977, 1986, 1997) defined self-efficacy as the
belief one has in being able to execute a specific task successfully (e.g., a pitcher striking out
a batter) in order to obtain a certain outcome (e.g., self-satisfaction or coach recognition).
Since the first publication of the self-efficacy concept (Bandura, 1977), there have been over
60 research articles published on self-efficacy related specifically to sport performance
(Moritz, Feltz, Mack, & Fahrbach, in press). This chapter provides an overview of the self-
efficacy concept and its measurement, a review of relevant research on athletes, athletic
teams, and coaches, and future directions for research in this field.
Self-efficacy Theory
Bandura’s (1977) theory of self-efficacy theory was developed within the framework of
social cognitive theory. Although, originally, the theory was proposed to account for the
different results achieved by diverse methods used in clinical psychology for the treatment of
anxiety, it has since been expanded and applied to other domains of psychosocial functioning
including health and exercise behavior (McAuley, 1992; McAuley & Mihalko 1998; O'Leary,
1985), and sport and motor performance (Feltz, 1988).
Self-efficacy beliefs are not judgments about one's skills, objectively speaking, but rather
about one's judgments of what one can accomplish with those skills (Bandura, 1986). In other
words, self-efficacy judgments are about what one thinks one can do, not what one has. These
judgments are a product of a complex process of self-appraisal and self-persuasion that relies on
cognitive processing of diverse sources of efficacy information (Bandura, 1990). Bandura
(1977, 1986) categorized these sources as past performance accomplishments, vicarious
experiences, verbal persuasion, and physiological states. Others have added separate categories
for emotional states and imaginal experiences (Maddux, 1995; Schunk, 1995).
Performance accomplishments have proved to be the most influential source of efficacy
information because they are based on one's own mastery experiences (Bandura, 1997). One’s
mastery experiences affect self-efficacy beliefs through the cognitive processing of such
information. If one has repeatedly viewed these experiences as successes, self-efficacy beliefs
will increase; if these experiences were viewed as failures, self-efficacy beliefs will decrease.
Furthermore, the self-monitoring or focus on successes should provide more encouragement and
enhance self-efficacy more than the self-monitoring of one’s failures. One must be careful,
however, not to become complacent by one’s success. Bandura (1997) suggests that letdowns
after easy successes and intensifications after failure are common sequences in competitive
struggles. The continued setting of challenging goals and the positive reactions to substandard
performances help to elevate the intensity and level of motivation.
3
The influence of past performance experiences on self-efficacy beliefs also depends on
the perceived difficulty of the performance, the effort expended, the amount of guidance
received, the temporal pattern of success and failure, and the individual’s conception of a
particular “ability” as a skill that can be acquired versus an inherent aptitude (Bandura, 1986;
Lirgg, George, Chase, & Ferguson, 1996). Bandura has argued that performance
accomplishments on difficult tasks, tasks attempted without external assistance, and tasks
accomplished with only occasional failures carry greater efficacy value than tasks that are easily
accomplished, tasks accomplished with external help, or tasks in which repeated failures are
experienced with little sign of progress. Miller (1993) found a negative relationship between
high self-efficacy perceptions of competitive swimmers and their motivation when they were
given unchallenging goals.
Efficacy information can also be derived through a social comparison process with
others. This process involves observing the performance of one or more other individuals,
noting the consequence of their performance, and then using this information to form judgments
about one’s own performance (Maddux, 1995). Vicarious sources of efficacy information are
thought to be generally weaker than performance accomplishments; however, their influence on
self-efficacy can be enhanced by a number of factors. For example, the less experience people
have had with performance situations, the more they will rely on others in judging their own
capabilities. The effectiveness of modeling procedures on one’s self-efficacy judgments has also
been shown to be enhanced by perceived similarities to a model in terms of performance or
personal characteristics (George, Feltz, & Chase, 1992; Weiss, McCullagh, Smith, & Berlant,
1998).
One particular mode of modeling influence that has been suggested to enhance one’s
sense of efficacy and performance in sport is self-modeling (Dowrick, 1991; Franks & Maile,
1991). Self-modeling consists of the individual repeatedly observing the correct or best parts of
his or her own past performance, and using that as a model for future performance (Dowrick &
Dove, 1980). Bandura (1997) suggests that self-modeling affects performance through its impact
on efficacy belief. The little research in sport on this topic is equivocal (Singleton & Feltz, 1999;
Winfrey & Weeks, 1993). Winfrey and Weeks (1993) found no effects on self-efficacy or
balance-beam performance using self-modeling videotapes with female gymnasts. However,
they did not measure self-efficacy according to Bandura’s recommended procedures and had a
very small sample. Singleton and Feltz (1999), using a 5-item, skill specific self-efficacy scale,
found that collegiate hockey players exposed to several weeks of self-modeling videotapes
showed greater shooting accuracy and higher self-efficacy for game performance compared to
controls.
Persuasive techniques are widely used by coaches, managers, parents, and peers in
attempting to influence an athlete’s self-perceptions of efficacy. These techniques include verbal
persuasion, evaluative feedback, expectations by others, self-talk, positive imagery, and other
cognitive strategies. Self-efficacy beliefs based on persuasive sources are also likely to be
weaker than those based on one's accomplishments, according to the theory. However, Bandura
(1997) indicates that the debilitating effects of persuasory information are more powerful than
the enabling effects. Individuals tend to avoid challenging activities in which they have been
4
persuaded that they lack the capabilities or they give up quickly. It is harder to instill strong
beliefs of self-efficacy by persuasory means only.
The extent of the persuasive influence on self-efficacy has also been hypothesized to
depend on the prestige, credibility, expertise, and trustworthiness of the persuader. Coaches are
usually believed to be credible sources of their athletes’ capabilities. In addition to providing
inspirational messages, effective coaches also structure activities for their athletes that bring
success and avoid placing them prematurely in situations that are likely to bring repeated failures
(Bandura, 1997). Credible coaches also encourage their athletes to measure their successes in
terms of self-improvement rather than outcome.
Efficacy information can also be obtained from a person’s physiological state or
condition. Physiological information includes autonomic arousal that is associated with fear and
self-doubt or with being psyched-up and ready for performance, as well as one's level of fitness,
fatigue, and pain (in strength and endurance activities). Physiological information has been
shown to be a more important source of efficacy information with respect to sport and physical
activity tasks than in the case of nonphysical tasks (Chase, Feltz, Tully, & Lirgg, 1994; Feltz &
Riessinger, 1990).
Similar to physiological information, one’s emotional state can be an additional source of
information in forming efficacy perceptions. Positive affect, such as happiness, exhilaration, and
tranquility, are more likely to enhance efficacy judgments than are negative affective states, such
as sadness, anxiety, and depression (Maddux & Meier, 1995; Treasure, Monson, & Lox, 1996).
Schunk (1995) suggested that emotional symptoms that signal anxiety might be interpreted by an
individual to mean that he or she lacks the requisite skills to perform a certain task, which in
turn, influences efficacy judgments.
Lastly, Maddux (1995) introduced imaginal experiences as a separate source of efficacy
information. People can generate efficacy beliefs by imagining themselves or others behaving
successfully or unsuccessfully in anticipated performance situations. Bandura (1997) refers to
this as cognitive self-modeling (or cognitive enactment) and describes it as a form of modeling
influence. Imagining oneself winning against an opponent has been shown to raise efficacy
judgments and endurance performance (Feltz & Riessinger, 1990). Other cognitive simulations,
such as mental rehearsal strategies have also been shown to enhance competition efficacy beliefs
and competitive performance (Garza & Feltz, 1998).
These categories of efficacy information, based on Bandura’s theory of self-efficacy
(1977, 1986, 1997), are not mutually exclusive in terms of the information they provide, though
some are more influential than others. How various sources of information are weighted and
processed to make judgments on different tasks, in different situations, and for individuals’ skills
is still unknown. The consequences of these judgments, however, have been shown to determine
people's levels of motivation, as reflected in the challenges they undertake, the effort they
expend in the activity, and their perseverance in the face of difficulties (Bandura, 1997). In
addition, individuals’ self-efficacy judgments also have been shown to influence certain thought
patterns (e.g., goal intentions, worries, causal attributions) and emotional reactions (e.g., pride,
shame, happiness, sadness) that also influence motivation (Bandura, 1997).
5
Furthermore, the relationship between self-efficacy judgments and performance
accomplishments is believed to be temporally recursive: "Mastery expectations influence
performance and are, in turn, altered by the cumulative effect of one's efforts." (Bandura, 1977,
p. 194). Bandura (1990) has emphasized the recursive nature of the relationship between self-
efficacy and thought patterns as well. Figure 1 presents the relationships between the major
sources of efficacy information, efficacy judgments, and consequences as predicted by Bandura's
theory and the additional determinants proposed by Maddux (1995).
Bandura (1977, 1986, 1997) has provided some qualifiers to the predictiveness of self-
efficacy judgments. Self-efficacy beliefs are a major determinant of behavior only when people
have sufficient incentives to act on their self-perception of efficacy and when they possess the
requisite skills. Self-efficacy beliefs will exceed actual performance when there is little incentive
to perform the activity or when physical or social constraints are imposed on performance. Some
people may have the necessary skill and high self-efficacy beliefs, but no incentive to perform.
According to Bandura, discrepancies between efficacy beliefs and performance will also occur
when tasks or circumstances are ambiguous or when one has little information on which to base
efficacy judgments, such as when one is first learning a skill.
Self-efficacy expectations should not be confused with outcome expectations. Outcome
expectancies are defined as the belief that certain behaviors will lead to certain outcomes. Self-
efficacy, on the other hand, is the belief in one's ability to successfully perform the behavior in
question successfully (Bandura, 1977). In essence, outcome expectations are concerned with
beliefs about one's environment and efficacy expectations are concerned with beliefs about one's
competence. Some sport psychology researchers confuse performance markers, such as
“winning an event,” with an outcome expectation. Bandura (1997) describes the three major
forms that outcome expectations can take: physical effects, social effects, and self-evaluative
effects. “Behavior and the effects it produces are different classes of events.” (p. 22). That is,
examples of physical outcome effects are positive/negative sensory experiences; examples of
social outcome effects are approval/disapproval and monetary compensation/deprivation of
privileges; and examples of self-evaluative outcome effects are self-sanctions/self-satisfaction.
An athlete’s position in a competition or winning does not fit this class of effects. An athlete’s
position in a competition --first, second, third, etc.-- is a performance marker. Feltz and Chase
(1998) have labeled this “competitive” or “comparative” efficacy. An outcome expectation of
winning a competitive event might be a high level of self-satisfaction, approval from one’s
coach, and money or a trophy. Although both self-efficacy beliefs and outcome expectations can
influence behavior in sport situations according to Bandura (1997), outcome expectations are
highly dependent on self-efficacy judgments and thus do not predict much beyond what is
predicted by self-efficacy.
The Measurement of Self-Efficacy
Bandura (1977, 1986, 1997) has always advocated using self-efficacy measures that are
specific to particular domains of functioning, rather than ones that assess global expectations of
performance. This means using a microanalytic approach, which requires a detailed assessment
of the level, strength, and generality of self-efficacy beliefs. Level of self-efficacy is defined as
6
one’s belief about the magnitude or level of performance possible. Strength is defined as the
certainty that one can attain a given level of performance. Generality refers to the number of
domains in which an individual believes he/she is efficacious. Measures of generality of self-
efficacy are rarely included in research studies on sport.
A microanalytic approach allows one to analyze the degree of congruence between self-
efficacy and performance at the level of individual tasks (Bandura, 1997). Analyzing the degree
of congruence involves a computation of the percentage of items for which an efficacy judgment
and performance agree. As Wurtele (1986) noted, this type of analysis has not been conducted in
studies in sport psychology. Rather, researchers in sport psychology have typically correlated
aggregate self-efficacy level or strength scores with aggregate performance scores (Feltz &
Chase, 1998).
In sport studies, self-efficacy measures typically are constructed by listing a series of
tasks that vary in difficulty, complexity, or stressfulness. These are called hierarchical self-
efficacy measures. Participants are asked to designate (yes or no) the tasks they believe they can
perform (efficacy level). For each task designated as “yes,” they rate their degree of certainty
(efficacy strength) that they can execute it on a near-continuous scale from total uncertainty to
total certainty, with scale ranges from zero to 10, in one-unit increments, or zero to 100, in 10-
unit increments. Most hierarchical scales are constructed by listing tasks in increasing order of
difficulty, such as landing your most difficult figure skating jump from 1 out of 10 to 10 out of
10 times (Garza & Feltz, 1998).
In constructing nonhierarchical scales, a conceptual analysis of the subskills needed to
perform in a given domain is conducted, along with a contextual analysis of the level of
situational demands. Such a scale for wrestling might include escape, get reversal, get back
points, pin opponent, not get take down, get take down by throw, get take down single leg, ride
opponent, get take down double leg, and not be pinned (Treasure et al., 1996). Researchers
using nonhierarchical scales should determine and report the internal consistency if they are
using an aggregated score to represent self-efficacy (Feltz & Chase, 1998).
Some research studies have used one-item questions in which participants rate how
certain they are of their performance or beating an opponent’s performance. However, one-item
scales have been subject to problems with reliability and validity, especially in competitive
situations when several factors influence the outcome (Feltz & Chase, 1998). In these situations,
the correlations between self-efficacy and performance outcome have been shown to be much
smaller than when using multiple items (Moritz et al., in press).
Although most of the self-efficacy scales have been constructed for a specific study,
Ryckman and his colleagues (Ryckman, Robbins, Thornton, & Cantrell, 1982) constructed the
Physical Self-Efficacy Scale (PSE), with two subscales, to provide a more generalized measure
of self-efficacy in the sport and physical activity realm. The Perceived Physical Ability subscale
(PPA) measures perceptions about one’s general physical ability, and a Physical Self-
Presentation Confidence subscale (PSC) reflects perceived efficacy in the display of physical
skills. Rather than being on a probability scale, items are on a 6-point Likert scale, with response
alternatives ranging from strongly agree to strongly disagree. Although predictive validity has
7
been found for the PSE in competitive sports contexts (Gayton, Matthews, Borchstead, 1986),
others have found task-specific scales to be better predictors of specific tasks (LaGuardia &
Labbé, 1993; McAuley & Gill, 1983; Slanger & Rudestam, 1997). The concept of the PSE as a
self-efficacy measure has also been questioned because the items were not developed within a
goal-striving context and seem to represent more of a self-concept measure (Feltz & Chase,
1998; Maddux & Meier, 1995).
Two other measures that have been used to tap self-appraisals of people’s capability in
sport are Vealey’s (1986) sport confidence measure and the self-confidence subscale of the
Competitive Sport Anxiety Inventory-2 (CSAI-2: Martens, Burton, Vealey, Bump, & Smith,
1990). Sport confidence is a more broadly defined concept that assesses one’s trait and state
perceptions to be successful in one’s sport. Self-confidence, as measured on the CSAI-2, also
has a broader focus regarding one’s capability to perform successfully in competition. Details
regarding the use of these measures are described in Vealey’s chapter [this volume] and Hardy
and Woodman’s chapter [this volume].
Regardless of how the self-efficacy measure is constructed, it is most useful in explaining
motivated behavior and sport performance when the measures have been constructed within the
tenets of the theory. Thus, research participants should have the proper incentives to perform,
measures that are specific to the performance domain should be used, self-efficacy and
performance measures should be concordant, and self-efficacy and performance measures should
be assessed closely in time. Of particular importance, Bandura (1997) stated that a proper
assessment of the structure of the relationship between efficacy beliefs and action requires that
both measures should match. In a meta-analysis of the relationship between self-efficacy and
sport performance, Moritz et al. (in press) found that when the self-efficacy and performance
measures were not concordant, the correlations between the two were not as strong (r = .26) as
when both measures tapped similar capabilities (r = .43). A lack of concordance would be
evident if one assessed wrestling moves microanalytically (i.e., escapes, get reversals, pin
opponents, etc.), but then used the wrestler’s overall score as the performance measure.
The time lapse from self-efficacy assessment to performance is also important according to
Bandura (1986). If self-efficacy and performance measures are not assessed closely in time,
one’s efficacy beliefs could be altered by an intervening experience (Bandura 1986). Wiggins
(1998) found, however, that efficacy expectations for athletes remained very stable within 24 hr.
to competition. For a more thorough discussion on how to construct self-efficacy scales and the
measurement issues surrounding self-efficacy, the reader is directed to Feltz and Chase (1998).
Self-efficacy Research on Athletes
Much of the evidence for the effectiveness of self-efficacy as an influential
mechanism in sport performance comes from studies using nonathlete populations and
contrived settings (Feltz, 1992). That research has demonstrated consistent evidence that
people’s perceptions of their performance capability significantly affect their motivation and
performance (Feltz, 1994). The research reviewed in this section contains only studies that
examined the self-efficacy of athletes using self-efficacy scales. Studies using solely sport
8
confidence (Vealey, 1986) or CSAI-2 measures are not included here because they are
covered in other chapters.
Of the 45 studies that were reviewed in the Moritz et al. (in press) meta-analysis on
self-efficacy, only 10 studies examined the self-efficacy and sport performance relationship
with athletes. We found an additional 15 studies that either used children and youth (Moritz
et al. did not include samples under 16 years of age), did not employ a performance measure,
or were published after the meta-analysis was conducted. A summary of all 24 studies is
contained in Table 1.
Studies ranged from a focus on youth and high school athletes to extreme sport
athletes and athletes with disabilities. Feltz and Chase (1998) caution researchers about the
format and appropriateness of measures used in assessing self-efficacy in children. The
typical format with measures of strength and level of self-efficacy, may be too difficult for
children under 9 years of age. In addition, children under 9 years are not as accurate in their
self-assessments of capabilities as are older children and adults. Some of the studies we
found for this chapter used athletes younger than 9 years of age (Lee, 1982, 1986; Watkins,
Garcia, & Turek, 1994; Weiss, Wiess, & Klint, 1989; Winfrey & Weeks, 1993.
In terms of measures used, all but two studies (Gayton et al., 1986; Ryckman &
Hamel, 1993) used task specific self-efficacy scales. However, approximately only one-half
of the studies measured self-efficacy in accord with Bandura’s (1977) recommendations and
administering within 24 hours of performance measures. For studies that examined
performance, 11 used competitive outcomes, such as finish times, win/loss percentages, and
scores on judged competitions. In other studies, specific contests were constructed, such as
penalty shooting contests, in which number of shots made constituted the performance
measure.
Most of the studies (n = 18) that investigated self-efficacy beliefs of athletes
examined the self-efficacy-performance relationship. Some of these studies also made
comparisons with other predictors of performance. Most of these studies showed a
significant and at least moderate relationship between self-efficacy and performance.
Investigations that showed low correlations between the two measures either used a
nontraditional measure of self-efficacy, had a long time-lag between measures, or had a low
concordance between their self-efficacy and performance measure. For instance, in the Lee
(1988) study of collegiate female hockey players, there was an unspecified time period
between self-efficacy and performance, self-efficacy was not assessed prior to matches, and
the self-efficacy scale (based on individual hockey skills) was not concordant with the
performance measure (based on team winning percentage).
In addition to examining the relationship of self-efficacy to performance, 14 studies
also compared self-efficacy beliefs with other predictors of performance. Various other
predictors included general self-efficacy measures, response-outcome, valance, anxiety,
worry, affect, perceived control, personal goals and goal importance, competitive orientation,
sport confidence, and past performance/experience/training history. Most of these studies
found self-efficacy beliefs to have predictive superiority over other variables or have similar
9
predictive strength. For instance, George (1994) found self-efficacy and anxiety (cognitive
and somatic) to equally predict hitting performance in the first of a 9-game series. Kane et
al. (1996) found that prior performance predicted the percentage of wrestling matches won,
but that self-efficacy contributed most strongly to overtime performance. LaGuardia &
Labbé (1993) demonstrated that runners’ predicted times, training mileage, and self-efficacy
beliefs all predicted pace times for 3 races.
In only three studies were other variables found to be stronger predictors of
performance than self-efficacy beliefs. Lee (1982) found that although gymnasts’ self-
efficacy was more related to performance than previous performance, only the coach’s
estimate of the gymnast’s performance and, to a lesser extent, the number of previous
competitions were significant predictors of performance score. Wurtele (1986), however,
pointed out a number of methodological problems with the study that limit generalizations:
“(1) Subjects were quite young (ages 7-12 years) and may not have understood the task; (2)
not all of the subjects had been in previous competition; (3) only subjects’ strength was
assessed, and not level of self-efficacy; and (4) self-efficacy judgment were made 1 week
prior to competition.” (p. 292). Although the measurement of self-efficacy level (or
magnitude) is not essential (Feltz & Chase, 1998), the use of a public estimation of one’s
performance score as a measure of self-efficacy strength is of questionable validity.
In a subsequent study, Lee (1988) found team goal setting to have a stronger direct
relationship with the teams’ winning percentages than players’ self-efficacy beliefs. In
addition to the problems noted earlier in this chapter with the way in which self-efficacy was
measured, Feltz and Lirgg (1998) have demonstrated that team performance is more strongly
related to team beliefs than to individual beliefs.
Lastly, Watkins et al. (1994) did not find self-efficacy beliefs of youth baseball
players to predict batting cage performance as well as previous performance. Baseball hitters
in this study performed under invariant conditions for four trials. As Bandura points out
(1997), the predictiveness of prior performance is inflated under this condition and is not
realistic to batting under competitive conditions. When batting was performed within
baseball games, George (1994) found self-efficacy beliefs, but not prior performance to
predict subsequent performance.
Some studies have examined the antecedents of self-efficacy judgments.
Performance variables such as prior performance, training history, playing experience, have
been investigated as predictors of self-efficacy expectations in accord with Bandura’s (1986,
1997) predictions, as well as cognitive variables such as anxiety, affective states, competitive
orientation, goal importance, and trait sport confidence. All of the studies that investigated
performance variables as predictors of self-efficacy found strong relationships between the
two measures (George, 1994; Haney & Long, 1995; Kane, Marks, Zaccaro, & Blair, 1996;
Okumabua, 1986; Watkins et al., 1994). In addition, the studies that have measured this
relationship over trials, through path analysis (e.g., George, 1994; Haney & Long, 1995;
Kane et al., 1996), have found support for the recursive pattern that Bandura (1977)
emphasized between performance and self-efficacy. Even so, performance variables
typically were found to be stronger predictors of self-efficacy than self-efficacy was of
10
performance, which supports previous path analyses with nonathletes (e.g., Feltz, 1982; Feltz
& Mugno, 1983), and corroborates the findings in the meta-analysis by Moritz et al. (in
press). Given the complex nature of sport performance, however, self-efficacy should not be
expected to be as strong of a variable in the efficacy – performance relationship (Bandura,
1986, 1990). If performance measures are used where factors beyond one’s control are
partially responsible for the performance score, such as contact percentage, winning
percentage, and finish place, self-efficacy will not be as strong of a predictor of performance
as performance is of self-efficacy (Feltz, 1992).
The cognitive variables most strongly associated with self-efficacy expectations of
athletes are anxiety, positive and negative affective states, one’s goal orientation to win, and
trait sport confidence. George (1994) and Treasure et al. (1996) found significant negative
relationships between self-efficacy and state anxiety (cognitive and somatic). Treasure and
his colleagues also found self-efficacy to be negatively correlated with negative affect (e.g.,
jittery, nervous, upset) and positively correlated with positive affect (e.g., alert, determined,
inspired). Thus, not only do more efficacious athletes have lower levels of cognitive and
somatic anxiety prior to competition, they maintain a more positive affective state, as
Treasure et al. suggested.
The competitive orientations of athletes (i.e., desire to win or perform better than
others or perform well relative to one’s own standard) have been thought to be related to their
efficacy expectations (Martin & Gill, 1991; 1995a, 1995b). In particular, outcome goals,
based on a win orientation, are reasoned to undermine self-efficacy expectations because
they are considered less controllable and flexible than performance goals. Performance
goals, based on a goal orientation, are suggested to enhance efficacy expectations (Martin &
Gill, 1991). In a series of studies, Martin and Gill examined the competitive orientations and
self-efficacy beliefs for ‘placing’ (outcome) and for ‘finish time’ (performance) of distance
runners. They found that a win orientation was positively associated with efficacy beliefs for
placing; whereas, a goal orientation was positively associated with beliefs for finish time.
However, the outcome efficacy – win orientation relationship was much stronger than the
performance efficacy – goal orientation. In their second study (Martin & Gill, 1995a), they
also found that runners with a strong win orientation chose important place goals that also
predicted outcome efficacy beliefs. The results suggest that rather than outcome goals being
negatively associated with self-efficacy, they may be based on realistic appraisals of one’s
capability compared with other competitors. The authors also admit that their performance
time efficacy measure was not conceptually consistent with their goal importance measure.
Lastly, we found only three studies that applied interventions with athletes to enhance
self-efficacy expectations. Interventions are typically based on one or more sources of efficacy
information within Bandura’s (1977) theory. Singleton and Feltz (1999) investigated the use of
self-modeling techniques to enhance the self-efficacy beliefs and back-hand shots of collegiate
hockey players. As mentioned earlier in this chapter, they found that players exposed to several
weeks of self-modeling videotapes showed greater shooting accuracy and higher self-efficacy for
game performance compared to controls. A second study, with a much smaller sample, also
investigated the use of self-modeling techniques with gymnasts, but failed to find self-efficacy or
11
performance effects (Winfrey & Weeks, 1993). They also failed to measure self-efficacy
appropriately.
The third study involved the use of two selected mental practice techniques in an effort to
enhance the self-efficacy beliefs, competition confidence, and performance ratings of
competitive figure skaters (Garza & Feltz, 1998). Junior figure skaters, who were members of
the United States Figure Skating Association, were randomly assigned to one of two mental
practice interventions (drawing one’s freestyle routine on paper or walking through one’s routine
on the floor) or a stretching control group. The home-based interventions took place over 4
weeks and included procedural reliability and manipulations checks. Upon completion of the
intervention training, the skaters competed in their club’s annual competition. Coaches rated
their skaters’ current skating ability prior to the intervention and after the competition.
Self-efficacy was measured by constructing individualized figure-skating self-efficacy
scales to emphasize the skaters’ own current skating levels of ability in the areas of jumps, spins,
and steps/connecting moves. Skaters were asked “What is the most difficult jump or
combination jump, spin or spin combination, and step/connecting move in your skating routine?”
Skaters were then asked to rate their confidence in performing each skill from 1 out of 10 to 10
out of 10 times on an 11-point probability scale. Competition self-confidence was measured
using the self-confidence subscale of Martens et al.’s (1990) CSAI-2.
Both mental practice groups significantly improved their performance ratings and their
competition confidence compared to the stretching control group. All groups improved in their
self-efficacy judgments, including the stretching group, but the walk through group showed
higher improvements in spin self-efficacy compared to the other two groups. The authors noted
that self-efficacy assessment was not concordant with the treatment intervention. That is, the
intervention was designed to improve one’s entire freestyle routine rather than just jumps, spins,
and connecting moves.
It is surprising that so few intervention studies have been conducted with self-efficacy as
a dependent variable. Perhaps the reason is due to the emphasis on performance as the primary
variable in competitive sport. Nonetheless, research is needed to examine other promising
interventions to enhance and maintain self-efficacy beliefs over time. As Schunk (1995) noted,
studies are typically conducted over brief periods and may not examine maintenance of self-
efficacy beliefs at all.
Overall, the research on the self-efficacy beliefs of athletes has shown self-efficacy to be
a reliable predictor of sport performance and useful in combination with other cognitive and
training variables in accounting for performance variance. High self-efficacy expectations have
also been shown to be accompanied by low precompetitive anxiety, positive affect, strong goal
importance and high personal goals, and high trait sport confidence in athletes. In studies where
self-efficacy was not found to be a significant predictor of performance, and where interventions
were not fully successful in enhancing efficacy beliefs, measurement problems were readily
apparent.
12
Collective Efficacy Research on Teams: An Extension of Self-Efficacy Theory
Although many research studies have examined the relationship between a
performer’s self-efficacy and subsequent performance, only recently has the relationship
between a group’s collective confidence and its performance been studied. Sport coaches
and spectators alike are often baffled by teams who are composed of talented individuals but
who perform poorly. In contrast, some overachieving teams frequently are characterized by a
togetherness that overshadows any individual performer. Other overachieving teams win in
spite of within-group problems. The confidence group members have in their collective
abilities (collective efficacy) may begin to explain these inconsistencies.
Definitions. Conceptual distinctions need to be made between collective efficacy as
defined by Bandura (1997) and other related constructs. Bandura (1997) defines collective
efficacy as a group’s shared beliefs in its capacities to organize and execute actions to
produce a desired goal. Therefore, collective efficacy, as well as self-efficacy, is seen as
task-specific. Bandura asserts that merely summing a group’s individual assessments of
personal efficacy is insufficient to represent the coordinative dynamics of its members. In
other words, groups may be composed of high or low efficacious persons; however, how
members perceive their group’s ability as a whole is more salient than how they perceive
their individual capabilities. According to Zaccaro, Blair, Peterson, and Zazanis (1995),
because groups inherently require coordination, interaction, and integration, a summing of
individuals’ judgments about their individual abilities ignores these components. Collective
efficacy refers not only to how well each and every group member can use his or her
individual resources, but also how well those resources can be coordinated and combined.
Although Bandura (1995) considers perceptions of a team’s capability to perform a
task to encompass the coordination and interaction influences operating within a team, some
authors consider these resources to measure to separate factors of collective efficacy
percpetions (Mischel & Northcraft, 1997; Paskevich, 1995; Zaccaro et al., 1995). Mischel
and Northcraft, for instance, define collective task efficacy as “members’ beliefs that their
group has the task-related knowledge, skill, and abilities (KSAs) to successfully perform a
specific task,” and collective interdependence efficacy as “members’ beliefs that their group
has the knowledge, skills, and abilities (KSAs) to interact effectively in performing a specific
task.” (p. 184). These separate dimensions are also hypothesized to be influenced by
different moderators. Perceived task complexity is proposed to moderate collective task
efficacy; whereas, perceived task interdependence is proposed to moderate collective
interdependence efficacy.
A related concept to collective efficacy, group “potency,” has been defined as the shared
belief of a group that it can be effective (Guzzo, Yost, Campbell, & Shea, 1993). However,
group potency suggests generalized beliefs whereas collective efficacy is task-specific (Mulvey
& Klein, 1998). While collective efficacy is typically a measure of individuals, those individuals
are, by necessity, influenced by other group members. Collective efficacy, then, may have both
individual and group level components (Kenny & LaVoie, 1985; Zaccaro, Zazanis, Diana, &
Greathouse, 1994).
13
Because Bandura (1997) places the construct of collective efficacy at the group level, the
averaging of individual data for use as group means can be arguable. For example, Gibson,
Randel, and Early (1996) use the term “group efficacy” to denote a group’s consensus about that
group’s abilities. Group efficacy, in this sense, would be comprised of one rating, agreed upon
by all members of the group. The drawback to this method is that social persuasion by a few
leaders within the group may lead to a forced consensus that is not representative of most of the
group’s members (Bandura, 1997). However, Rousseau (1985) suggests that perceptions at the
level of the individual can be aggregated to a higher level construct and the mean used to
represent this collective interpretation when the two variables are functionally equivalent. This
condition is met when perceptual consensus has been demonstrated (James, 1982; Kozlowski &
Hattrup, 1992). Perceptual consensus exists when group members perceive the team or their
abilities within the team to function in the same way. Within-group differences in collective
efficacy may be the result of self-efficacy beliefs, personalities of the individuals in the group, or
different perceptions or exposure to group stimuli within the group (Watson & Chemers, 1998).
If within-group variabilities are not taken into account, aggregating data at the individual level to
represent a higher level of analysis may result in aggregation bias (James, 1982). Therefore,
studies in collective efficacy should first consider the research question in order to determine the
proper level of analysis. Consensus should be demonstrated if a group-level analysis is deemed
appropriate (see Feltz & Chase, 1998, for a complete discussion of measurement issues for
collective efficacy).
However, Bandura (1997) suggests that in groups where interdependence among group
members is low (for example, a golf team), an aggregate of individual efficacies may have
sufficient predictive power for group outcomes. When interdependence is high (e.g., a
basketball team), an aggregate of individuals’ judgments about group efficacy would be the
better predictor. Some evidence, using sport tasks, exists to support this contention (Moritz,
1998). Bandura also contends that individuals who play different positions in the group may
view that group’s efficacy differently, based on those positions. It would be rare for a group to
have unanimity of beliefs across members. However, as the group continues together sharing
experiences and outcomes, collective efficacy beliefs should reflect group consensus over time.
Zaccaro et al. (1994) suggest that the degree to which collective efficacy is made at the
group level is dependent upon whether team members have a sufficient base of common
experiences. Results from Zaccaro et al. (1994) support this premise. They assigned Army
soldiers to teams of 10-12 persons and asked them to complete a series of physical exercises that
required substantial coordination of movement. Results showed that collective efficacy beliefs
become more homogeneous within the teams over their lifespan. In sport, there may be new team
members from season to season, but brand new teams form much less frequently. Thus, most
sports team have some shared congruence at the onset of a season. Watson and Chemers (1998)
studied 28 Division III basketball teams and found collective efficacy to be stable from
beginning to end of season, but they also found smaller within-group variance at the end of the
season than at the beginning.
Sources of collective efficacy. Because collective efficacy is rooted in self-efficacy
(Bandura, 1997), at least some of the sources of collective efficacy should be similar to self-
efficacy. Of course, these sources should be focused at the group level. Thus, enactive mastery
14
experiences would be based on team masteries, vicarious experience might involve watching a
similar team in a similar situation, verbal persuasion would be directed to the group, and
physiological and affective states might involve perceptions of the group’s nervousness. While
these may indeed affect individuals’ perceptions of their team’s efficacy, other influences may be
important.
Watson and Chemers (1998) suggest that three group level influences are most important:
(a) group composition, (b) previous group experiences, and (c) leader’s effectiveness. First, the
group’s composition may contribute to high or low perceptions of collective efficacy. The
authors reason that composition influences actually could go either way. Large groups may
experience coordination difficulties and those difficulties may be reflected in low perceived
collective efficacy. However, large groups may also contain more resources, which may
strengthen collective efficacy beliefs. If coordination is the problem, collective efficacy may
increase across a season as the team learns to work together (Watson & Chemers, 1998; Zaccaro
et al., 1994).
Past experience has been shown to be the strongest source of efficacy for individuals.
Likewise, a group’s previous experiences should have a powerful effect on a team’s collective
efficacy. Using structural equation modeling, Riggs and Knight (1994) tested the effects of a
group’s success or failure in a work environment on personal and collective efficacy as well as
personal and collective outcome expectancy. They found that success/failure played a direct and
dominant role in all four variables. They believe that these results suggest that “success breeds
success and that failure must surely be difficult to overcome” (p. 762).
Watson and Chemers (1998) added leader effectiveness to the list of sources of collective
efficacy. They suggest that a group’s collective efficacy will be influenced by exceptional
leadership (Shamir, House, & Arthur, 1992). Leaders have the opportunity to contribute to their
team’s smooth functioning, and to eliminate or minimize coordination problems for
performance. They can also enhance efficacy by modeling confidence. A well-respected leader
may verbally be able to persuade his or her charges that they indeed have the resources necessary
to achieve a goal. By contrast, a negative coach could demoralize a team by constantly belittling
the group.
George and Feltz (1995) speculate that spectators or the media may similarly provide
relevant feedback to teams that may influence their collective efficacy. A booing home crowd or
negative hometown newspaper may be as demoralizing as the coach who constantly berates his
or her team, whereas a supportive home crowd, even in times where the going is tough, may lift
that team’s confidence in itself. It is obvious that research in discerning sources of collective
efficacy is much needed so that coaches can use the information to strengthen their team’s
confidence levels.
Collective efficacy research in sport. To date, only a few studies have been conducted
for the specific purpose of studying the relationship between collective efficacy and performance
in sport. In the most extensive study, Feltz and Lirgg (1998) followed six intercollegiate male
ice hockey teams across the season. Individual and collective efficacy were assessed before each
game; team performance statistics from each game were also obtained. Results were in
15
agreement with Bandura’s (1997) suggestion that collective efficacy, rather than aggregated self-
efficacy, should hold more predictive power in relation to team performance for highly
interdependent teams, as collective efficacy emerged as the stronger predictor of team
performance. In addition, when wins and losses were analyzed across a season, collective
efficacy was affected by performance outcome but not self-efficacy. Team efficacy increased
after a win and decreased after a loss.
Spink (1990) was primarily interested in the relationship between team cohesion and
collective efficacy. He recruited volleyball players playing in a volleyball tournament for either
elite teams or recreational teams. They were asked to complete the Group Environment
Questionnaire (Widemeyer, Brawley, & Carron, 1985), a cohesion measure, as well as
responding to questions devised to measure collective efficacy. Individuals were asked what
placing they expected for their teams and also how confident they were in those placings. Elite
and recreational teams were similarly confident in their ratings. Results showed that, for elite
teams only, high collective efficacy teams scored higher on Individual Attractions to the Group-
Task (e.g., an individual’s feelings toward involvement with the group’s task, productivity,
goals, and objectives) and the shared social interests of the team than did low efficacy teams. No
differences between high and low collective efficacy groups were found among the recreational
players. Spink also found that high collective efficacy teams placed higher than did low
collective efficacy teams. Spink argued that the difference in the finding between elite and
recreational teams could have been a result of greater the emphasis on winning by the elite teams
(the reward was monetary for the elite tournament only). He suggested that group goals may
moderate the relationship between collective efficacy and team cohesion.
Paskevich (1995) also examined the collective efficacy and cohesion relationship tp
performance in volleyball teams. His collective efficacy scales were moe elaborate than that of
Spink’s 1990), including eight scales, and efficacy values were measured over the course of a
season. Results showed that perceived collective efficacy and cohesion increased over the
course of the season and that collective efficacy mediated the relationship between task-oriented
cohesion and team performance at early season but not later season. There was also evidence for
the independent effects of collective efficacy and cohesion on performance. The mediation
effect supports Bandura’s (1986, 1997) contention that collective efficacy acts as a mediator
between cohesion and performance. However, as Paskevich noted, the independent effects of
these variables on performance at different points in the season suggests that a more complex
relationship was operating.
Watson and Chemers (1998) measured 28 male and female intercollegiate basketball
team members concerning their collective and self-efficacy beliefs and their optimism. Team
captains, or other team leaders, were also asked to rate their leadership confidence. Measures
were taken before the season began and before post-season play. Previous team performance
(last year’s won-loss record), season team and individual performances, and leader evaluations
made by team members were also examined. Before the season, players who had higher
optimism scores also had higher collective efficacy beliefs. By the end of the season, this
relationship was not apparent. Also, at the beginning of the season, collective and self-efficacy
were positively related, but only for high self-efficacy teams. Low self-efficacy teams showed a
negative relationship between collective and self-efficacy. However, at the end of the season,
16
this relationship was positive. Beginning-of-season collective efficacy predicted end-of-season
collective efficacy. In addition, Watson and Chemers also found that beginning efficacy
expectations predicted end-of-season performance. Finally, leader evaluation was positively
related to collective efficacy, but more so for teams that were unsuccessful in the previous
season; in previously unsuccessful teams, players who believed they had effective leaders were
more confident in their teams. This last finding may be especially relevant to coaches who find
themselves inheriting losing teams. If leadership abilities are apparent to their charges, they may
be also increasing the collective efficacy of their teams.
Two additional studies on collective efficacy used contrived teams or tasks to examine,
experimentally, the collective efficacy and performance relationship. Using a novel physical
task, Hodges and Carron (1989) assigned individuals to teams and gave bogus feedback on a
hand dynamometer task concerning the team’s ability. One team was led to believe that they
were inferior in team strength to a confederate group; the other team was led to believe that they
were superior. Team members were then shown the competitive task in which they would
participate: a medicine ball task where groups would be asked to hold the ball up with one arm as
long as possible with that arm fully extended at shoulder level. A manipulation check
confirmed that this bogus manipulation was enough to affect collective efficacy, as the inferior
team recorded lower collective efficacy scores than did the superior team before the task was
even attempted. After one trial of the task, both teams were told that they had been beaten by
their respective confederate teams. However, after this failure, the high collective efficacy team
actually improved their performance on a second trial while the low collective efficacy team
showed a decrement in performance. Similar to self-efficacy, high efficacious teams may be
more likely to put forth more effort in the face of failure to achieve a goal than would low
efficacious teams.
Lichacz and Partington (1996) also manipulated collective efficacy. They created three-
and four-member groups composed either of members of basketball or rowing teams (true teams)
or ad hoc groups (non-team members). Subjects were asked to participate in a rope-pulling task,
where individual pulls and group pulls could be recorded. Collective efficacy was manipulated
by telling teams that their collective pulls were either 10% below standards set by high level
athletes (low efficacy) or 10% above standards set by non-athletes (high efficacy). Results
showed that high efficacy groups rated their collective efficacy higher than did low efficacy
groups. In terms of performance, an interaction between group history (true versus ad hoc
teams) and performance feedback was found. For all groups, except the rowers, high efficacy
teams outperformed low efficacy teams. However, the two groups of rowers (high and low
collective efficacy) did not differ in performance. The authors suggest that a task that is both
salient and challenging to experienced performers may, in fact, motivate them to do their best
work. That is, in terms of task characteristics, pulling may be more similar to rowing
performance than to basketball performance. However, it is possible that preexisting efficacy
beliefs may not have been tapped, especially in the case of the rowers, and those beliefs may
have influenced the results of this study.
In an effort to examine the relationships among self-efficacy, collective efficacy, and
team performance in both more and less interdependent tasks, Moritz (1997) randomly assigned
participants in bowling classes to two-person teams. For half of the teams, the team score was
17
represented by the sum of their two scores (less interdependent). The other half of the teams
performed “Scotch Bowling,” where bowlers alternated balls and the team scores were reflected
by one score for the team (more interdependence). However, each bowler started in alternating
frames, whether or not they were the last person to bowl in the previous frame. The performance
measure used in the analyses was the average number of pins dropped on each first ball divided
by 10 frames. Individual efficacy was an aggregate measure of both bowlers’ efficacy scores.
Consensus analyses were conducted to ensure that this aggregation was justified. For collective
efficacy (or ‘group efficacy’ as used by Gibson et al., 1996), both bowlers together agreed upon
a team efficacy rating. Results showed that the predictiveness of collective efficacy to
performance was moderated by task type (i.e., bowling condition). For the less interdependent
condition, collective efficacy was not a predictor of team performance; however, it was for the
more interdependent condition. Task type did not moderate the relationship between aggregated
self-efficacy and team performance. For more interdependent tasks, then, collective efficacy is a
stronger predictor of team performance than it is for less interdependent tasks, at least with two-
persons teams.
Self-Efficacy Research on Coaches
In addition to the paucity of research on collective efficacy in sport, few studies have
investigated the role that coaches play in building the efficacy beliefs of their athletes and teams
nor the efficacy beliefs of coaches themselves to carry out their roles. Three studies have
examined the strategies that coaches use most to develop self-efficacy in athletes (Gould, Hodge,
Peterson, & Giannini, 1989; Weinberg, Grove, & Jackson, 1992; Weinberg & Jackson, 1990).
At the elite coaching level, intercollegiate wrestling coaches and United States national coaches
reported encouraging positive as opposed to negative self-talk, modeling confidence themselves,
using instruction and drills to ensure performance improvements, and using rewarding statements
liberally to be most the effective ways to enhance self-efficacy in their athletes (Gould et al.,
1989). High school and age-group coaches reported using similar techniques to enhance self-
efficacy and also reported using verbal persuasion (Weinberg et al., 1992; Weinberg & Jackson,
1990). These strategies are all based on the major sources of efficacy information as identified
in Bandura’s (1977) theory: performance accomplishments, vicarious experiences (modeling),
and verbal and self-persuasion. As the authors of this research have noted, however,
observations of coaches were not conducted to determine the actual use of the self-efficacy
techniques or whether these techniques were effective in enhancing the confidence of their
athletes and improving performance.
The coach’s efficacy expectations of the athlete or team may also play a role in
determining the efficacy beliefs of their athletes. When US Olympic athletes were asked to list
the best coaching actions to enhance athletes’ performance, providing support and confidence
were ranked second (Gould et al., 1999). Chase, Lirgg, and Feltz (1997) specifically examined
the relationship between coaches’ efficacy for their teams and team performance. Coaches of
four intercollegiate women’s basketball teams were queried before their games as to their
confidence in their teams’ abilities to perform specific basketball skills (i.e., shoot field goals and
free throws, rebound, commit turnovers, etc.). Coaches were also asked to rate the importance
they placed on these skills, the perceived control they felt over the outcome, and opponent
ability. Coaches who had higher efficacy beliefs for their teams perceived themselves to have
18
higher control over their teams’ outcomes. Also, the higher the perceived ability of the
opponent, the lower the coach’s efficacy in her team. In terms of coach’s efficacy in the team
and team performance, only free throws and turnover performance could be predicted
.
A second purpose of the study was to determine what coaches used as a basis in forming
their efficacy judgments of their teams. Inductive content analysis was used to identify both high
and low efficacy sources. Factors that resulted in high efficacy expectations included good past
game and practice performances, favorable comparison with opponents, return of an injured
player, and hearing negative comments from players on the opposing team. Coaches also
identified good performance preparation by either themselves, their staff, or their players as
contributing to high efficacy expectations in their teams. One interesting finding was that many
coaches cited past poor performance as a reason that they were confident in their teams because
they believed in their teams’ ability to bounce back. Low efficacy factors were similar to those
of high efficacy factors: past poor game and practice performance, injured or tired players, and
comparisons to better opponents. Other factors that contributed to a coach’s low efficacy
expectation for their team included coaches’ perceptions that the players themselves had low
efficacy and a team’s inconsistent prior performances. The researchers reasoned that if indeed
players are aware of the efficacy expectations coaches have for their teams, a situation similar to
the Pygmalion Effect might occur. According to this effect, a coach first forms expectations of
his or her team. He or she then acts in ways that are consistent with those expectations. Athletes
then perceive and interpret those actions and respond in a way that reinforces the original
expectations. If this happens, coaches with low efficacy expectations for their teams may
inadvertently be contributing to low player efficacy while those believing their teams are capable
may convey that attitude to their players.
Another line of research is the examination of the efficacy beliefs of coaches in their own
coaching. As Bandura (1997) suggests, the development of resilient self-efficacy in athletes is
heavily influenced by the managerial efficacy of coaches. Coaching efficacy has been defined as
the extent to which coaches believe they have the capacity to affect the learning and performance
of their athletes (Feltz, Chase, Moritz, & Sullivan, 1999). Feltz et al. (1999) conceptualized a
model of coaching efficacy based on Bandura’s (1977, 1986, 1997) writings and Denham and
Michael’s (1981) model of teacher efficacy. Their concept of coaching efficacy comprised four
dimensions: motivation, technique, game strategy, and character building efficacy. Motivation
efficacy was defined as the confidence coaches have in their ability to affect the psychological
skills and motivational states of their athletes. Technique efficacy was defined as the belief
coaches have in their instructional/diagnostic skills. Game strategy efficacy was defined as the
confidence coaches have in their ability to coach during competition and lead their team to a
successful performance. Lastly, character-building efficacy involved the confidence coaches
have in their ability to influence a positive attitude towards sport in their athletes.
In line with Bandura’s concept of self-efficacy, Feltz et al. (1999) proposed that the four
dimensions of coaching efficacy are influenced by one’s past performance and experience (e.g.,
coaching experience, coaching preparation, previous won-lost record), the perceived ability of
one’s athletes, and perceived social support (e.g., school, community, and parental support).
They also proposed, in turn, that coaching efficacy has an influence on one’s coaching behavior,
player satisfaction of the coach, the performance of one’s athletes (as measured by winning
19
percentage in their study), and player efficacy levels. Figure 2 illustrates the model of coaching
efficacy as conceptualized by Feltz et al.
In addition to the model, Feltz et al. (1999) developed the Coaching Efficacy Scale (CES)
to measure the multidimensional aspects of coaching efficacy. They found the psychometric
properties of the CES to be sound. Their confirmatory factor analysis supported the four factor
solution structure and, in addition, marginal support for one overall coaching efficacy factor
using various global fit indices.
Feltz et al. (1999) also tested the proposed sources and outcomes of CES using high
school basketball coaches. They found support for their model of coaching efficacy, in that past
winning percentage, years in coaching, perceived team ability, community support and parental
support were significantly predictive of coaching efficacy. The most important sources of
coaching efficacy were years of coaching experience and community support. They also found
that higher efficacy coaches had significantly higher winning percentages, greater player
satisfaction, used more praise and encouragement behaviors, and used fewer instructional and
organizational behaviors than lower efficacy coaches. However, the sources of coaching
efficacy accounted for only 13% of coaching efficacy beliefs, and their correlational design did
not allow for tests of causal effects between any of the variables within the model
This study was followed by two additional studies that provided support for the concept
of coaching efficacy (Chase, Hayashi, & Feltz, 1999; Malete & Feltz, in press). In the first, 12 of
30 high school basketball coaches observed in the ‘outcome’ portion of the Feltz et al. (1999)
study were randomly selected from the high and low efficacy groups and interviewed to identify
sources of coaching efficacy information from a coach’s perspective. Major sources of efficacy
themes supported and delineated the sources presented in the Feltz et al. (1999) model.
Basketball coaches reinforced the importance of coaching development in terms of education,
preparation, philosophy, experience, and knowledge of the game. Coaches also identified
information supplied by their players in terms of players’ confidence in them, players’ enjoyment
of the sport, and player development. A coach’s past success or performance accomplishments
may be more related to player development, which is more under the coach’s control, than to
won-lost records. The third major efficacy source theme was self-assessment, in terms of
analyzing one’s own coaching performance and one’s leadership skills. This supports Bandura’s
(1997) contention that past performance, by itself, does not provide sufficient information to
judge one’s ability. Self-appraisal of one’s effectiveness includes assessments of one’s effort,
task difficulty, and situational circumstance. This requires the integration of multiple sources of
efficacy information.
Although Feltz et al. (1999) assessed coaching experience in terms of years in
coaching, they did not assess the extent of one’s coaching preparation. Malete and Feltz (in
press) examined the effect of participation in a 12-hour coaching education program on
coaches’ perceived coaching efficacy. Results showed a small but significant improvement
in coaching efficacy based on the educational program compared to the efficacy levels of
control coaches. This study adds further support for coaching preparation as a source of
coach efficacy information. The most effective coaching education programs should be ones
that use approaches that help increase one’s confidence in coaching (e.g., mastery
20
experiences, challenging and reachable goals, observational learning, and simulated learning
components).
Barber (1998) also examined the sources of coaching efficacy information and
coaching efficacy levels of male and female high school coaches within a perceived
competence framework. Using a Sources of Coaching Competence Information Scale and a
Perceived Coaching Competence Questionnaire, developed specifically for the study, Barber
found that male and female coaches showed few differences in preferences for sources of
coaching competence information. Female coaches placed greater importance on the
improvement observed in their athletes and improvement they observed in their own
coaching skills as sources of coaching competence than did male coaches, but all coaches
viewed these as the top two sources. In terms of perceived coaching competence, of the
seven competency areas surveyed, the only gender difference was in teaching sport skills,
where female coaches perceived themselves to be more competent than male coaches.
Barber (1998) was also interested in coaches’ perceptions of factors that might
influence a future decision to discontinue coaching. Of the three categories of reasons
offered – ‘time demands’, ‘perceptions of coaching competence’, and lack of administrative
support’, - gender differences were found on two. Males cited ‘lack of administrative
support’ as a more important reason for retiring from coaching; whereas, females reported
‘low perceived coaching competence’ as more important. This finding suggests the
importance of developing and maintaining coaching efficacy in terms of coaching
motivation, especially for women.
Future Directions for Research
Since Feltz’s (1992) commentary on self-efficacy and motivation in sport, more research
has moved from laboratory settings to field settings with athletes in competition. However, as
previously called for, more research is needed in how athletes process multidimensional efficacy
information; the study of efficacy beliefs over time and in different situations; efficacy beliefs
regarding the cognitive and emotional aspects of performance; the resiliency of efficacy beliefs;
how various interventions can enhance efficacy beliefs; and a comprehensive examination of
efficacy beliefs in teams that would include individual beliefs, team beliefs, and beliefs of
coaches and leaders (Feltz, 1992, 1994).
Research has not been conducted on how athletes process multidimensional efficacy
information and the heuristics they use in weighting and integrating these sources of information
in forming their efficacy perceptions. Athletes across situations and in different sports may vary
in the importance they place on different sources of efficacy information. For instance, as
mentioned earlier in this chapter, physiological information was a more important source of
efficacy information for female collegiate athletes than was social comparison or persuasive
information (Chase et al., 1994). However, how these athletes derived the weightings of their
sources and how they integrated them into an efficacy judgment was not determined. That is,
was the information available used in an additive way? Did some information override other
sources? Answers to such questions as whether or how a coach’s persuasive techniques can
outweigh an athlete’s or team’s previous performance defeats in forming efficacy expectations
21
for subsequent performance would be of great importance to coaches. The use of qualitative
analyses may be necessary to determine some of these answers.
The majority of the research on self- and collective efficacy in sport has been approached
in a static way. Athletes, however, usually perform over time and across seasons. Many athletes
are also members of teams, which are dynamic in nature (Carron & Hausenblas, 1998). The
sources of efficacy information may change over time for individual athletes and teams, and the
influence of self- and collective efficacy perceptions, in combination with other cognitions, may
change. For instance, a recent gold medalist at the Nagano Olympic games reported that
knowing he was the strongest and fittest person in the event had always been his source of
efficacy information in the past, but that was not the case at these Olympic games. He, therefore,
worked on his mental skills to provide him with the level of efficacy he needed (Gould et al.,
1999).
Athletic performance is influenced by cognitive and emotional skills as well as physical
skills. Some athletes have stronger perceptions of efficacy than others in the mental aspects of
performance. As Bandura (1997) has noted, athletic efficacy involves control of disruptive
thinking and affective states as well as physical performance. Furthermore, Gould and his
colleagues (Gould et al., 1999) found that successful Olympic performance required extensive
planning and flexibility to deal with numerous unexpected events and distractions. Research is
needed to examine the influence on performance of efficacy beliefs regarding one’s
attention/concentration skills, one’s ability to set and work toward goals, one’s ability to manage
stress and disruptive thought processes, and one’s ability to make the right decisions,
unhesitatingly.
Bandura (1997) has also suggested that athletes must have a resilient sense of self-
efficacy to sustain perseverant effort in the face of failure and competitive pressure. According
to Bandura, experience with failures and setbacks helps in developing this robust sense of
personal efficacy. Future research might examine how different patterns of success and failure
influence the development of a robust sense of efficacy. In addition, Bandura notes that some
individuals and teams recover from setbacks more quickly that others. Knowing how and why
some individual athletes and teams are able to regain their sense of efficacy more quickly than
others would be a valuable information for designing interventions that would help efficacy
recovery.
As stated previously in this chapter, few interventions studies have been conducted with
athletes and teams to enhance their efficacy perceptions in their physical or mental performance.
Two procedures based on Bandura’s (1977) sources of efficacy information are worthy of
examination. One uses computer technology and the other is based on social comparison
information. The use of computer graphics and virtual reality technology is becoming more
popular as a teaching tool among coaches. However, whether these techniques can enhance and
maintain efficacy beliefs overtime has not yet been investigated. Before athletic programs invest
large sums of money in expensive equipment, they should determine if the technology has any
long lasting influence on efficacy beliefs.
22
The use of social comparison information also has not been investigated with athletes.
Whether upward comparisons have a negative effect on self-efficacy beliefs or a motivating and
challenging effect to surpass the comparative standard has not been tested. Evidence from
nonathletes suggests that upward comparisons have negative effects (George et al., 1992).
However, athletes, who generally have a more robust sense of efficacy may use the upward
comparative information as a challenge. There is some evidence that high self-esteem
individuals are more likely to make upward comparisons (Collins, 1996). Coaches and sport
psychologists would benefit from knowing what specific models or comparative others athletes
rely on to build their confidence, when left to their own choices, and how they use that
information.
In terms of collective efficacy specifically, a comprehensive examination of efficacy
beliefs in teams that would include individual beliefs, team beliefs, and beliefs of coaches and
leaders is needed along with other team related variables in order to better understand the
dynamics of teams. Sources of collective efficacy that are unique at the level of team needs
further investigation as does how the collective efficacy beliefs of team members change as team
membership changes and team leadership changes (Mischel & Northcraft, 1997). In addition,
the concepts of collective task and collective interdependence efficacy, as separate dimensions
could be examined in relation to their proposed moderators: task complexity and task
interdependence perceptions (Mischel & Northcraft).
Further research could include other conceptual and theoretical perspectives of group
motivation. For example, relationships between collective efficacy and team attributions, desire
for team success, team goals, communication in teams, and team cohesion have received little to
no attention. These variables should be examined over the course of a competitive season.
Lastly, the influence of coaches on the collective efficacy judgments of athletic teams
warrants further investigation. What characteristics of coaches and what coaching behaviors
provide the strongest efficacy beliefs in athletes and teams? In addition to the confidence that
coaches have in their players and teams, their own perceived managerial efficacy may influence
the team’s performance. Research outside of sport suggests that there is evidence for this
influence (Wood & Bandura, 1989).
23
References
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84(2), 191-215.
Bandura, A. (1986). Social foundation of thought and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (1990). Perceived self-efficacy in the exercise of personal agency. Journal of
Applied Sport Psychology, 2, 128-163.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Barber, H. (1998). Examining gender differences in sources and levels of perceived
competence in interscholastic coaches. The Sport Psychologist, 12, 237-252.
Barling, J., & Abel, M. (1983). Self-efficacy beliefs and tennis performance. Cognitive
Therapy and Research, 7, 265-272.
Carron, A. V., & Hausenblas, H. A. (1998). Group dynamics in sport, 2
nd
ed.
Morgantown, WV: Fitness Information Technology.
Chase, M. A., Feltz, D. L., Tully, D. C., & Lirgg, C. D. (1994). Sources of collective and
individual efficacy in sport. Journal of Sport and Exercise Psychology, 16, S18.
Chase, M. A., Hayashi, S., & Feltz, D. L. (1999). Sources of coaching efficacy: The
coaches’ perspectives. Unpublished manuscript, Michigan State University.
Chase, M. A., Lirgg, C. D., & Feltz, D. F. (1997). Do coaches’ efficacy expectations for
their teams predict team performance? The Sport Psychologist, 11, 8-23.
Collins, R. L. (1996). For better or worse: The impact of upward social comparison on
self-evaluations. Psychological Bulletin, 119, 51-69.
Denham, C.H., & Michael, J.J. (1981). Teacher sense of efficacy: A definition of the
construct and a model for further research. Educational Research Quarterly, 5, 39-63.
Dowrick, P. W. (1991). Practical guide to using video in the behavioral sciences. New
York: Wiley.
Dowrick, P. W., & Dove,C. (1980). The use of modeling to improve the swimming
performance of spina bifida children. Journal of Applied Behavior Analysis, 13, 51-56.
Feltz, D. L. (1982). Path analysis of the causal elements in Bandura’s theory of self-
efficacy and an anxiety-based model of avoidance behavior. Journal of Personality and Social
Psychology, 42, 764-781.
Feltz, D. L. (1988). Self-confidence and sports performance. In K. B. Pandolf (Ed.)
Exercise and Sport Sciences Reviews, (pp. 423-457). New York: MacMillan.
Feltz, D. L. (1992). Understanding motivation in sport: A self-efficacy perspective. In G.
C. Roberts (Ed.), Motivation in sport and exercise (pp. 107-128). Champaign, IL: Human
Kinetics.
Feltz, D. L. (1994). Self-confidence and performance. In D. Druckman, & R. A. Bjork
(Eds.), Learning remembering, believing. (pp. 173-206). Washington, D. C.: National Academy
Press.
Feltz, D. L., & Chase, M. A. (1998). The measurement of self-efficacy and confidence in
sport. In J. L. Duda (Ed.), Advancements in sport and exercise psychology measurement (pp. 63-
78). Morgantown, WV: Fitness Information Technology.
Feltz, D. L., Chase, M. A., Moritz, S. E., & Sullivan, P. J. (1999). Development of the
multidimensional coaching efficacy scale. Journal of Educational Psychology, 91,
24
Feltz, D. L., & Lirgg, C. D. (1998). Perceived team and player efficacy in hockey.
Journal of Applied Psychology, 83, 557-564.
Feltz, D. L., & Mugno, D. A. (1983). A replication of the path analysis of the causal
elements in Bandura’s theory of self-efficacy and the influence of autonomic perception. Journal
of Sport Psychology, 5, 263-277.
Feltz, D. L., & Riessinger, C. A. (1990). Effects on in vivo emotive imagery and
performance feedback on self-efficacy and muscular endurance. Journal of Sport and Exercise
Psychology, 12, 132-143.
Franks, I. M. & Maile, L. J. (1991). The use of video in sport skill acquisition. In P. W.
Dowrick (Ed.), Practical guide to using video in the behavioral sciences (pp. 231-243). New
York: John Wiley and Sons.
Garza, D. L. & Feltz, D. L. (1998). Effects of selected mental practice on performance,
self-efficacy, and competition confidence of figure skaters. The Sport Psychologist, 12, 1-15.
Gayton, W. F., Matthews, G. R., & Borchstead, G. N. (1986). An investigation of the
validity of the physical self-efficacy scale in predicting marathon performance. Perceptual and
Motor Skills, 63, 752-754.
Geisler, G.W.W., & Leith, L.M. (1997). The effects of self-esteem, self-efficacy, and
audience presence on soccer penalty shot performance. Journal of Sport Behavior, 20, 322-337.
George, T. R. (1984). Self-confidence and baseball performance: A causal examination
of self-efficacy theory. Journal of Sport and Exercise Psychology, 16, 381-399.
George, T. R., & Feltz, D. L. (1995). Motivation in sport from a collective efficacy
perspective. International Journal of Sport Psychology, 26, 98-116.
George, T. R., Feltz, D. L., & Chase, M. A. (1992). The effects of model similarity on
self-efficacy and muscular endurance: A second look. Journal of Sport and Exercise Psychology,
14, 237-248.
Gibson, C. B., Randel, A., & Earley, P. C. (1996). Understanding group efficacy: An
empirical test of multiple assessment models. Paper presented at the Society for Industrial and
Organizational Psychology Conference, San Diego, CA.
Gill, D. L., & Deeter, T. E. (1988). Development of the sport orientation questionnaire.
Research Quarterly for Exercise and Sport, 59, 191-202.
Gould, D., Greenleaf, C., Lauer, L., & Chung, Y. (1999). Lessons from Nagano. Olympic
Coach, 9 (3), 2-5.
Gould, D., Hodge, K., Peterson, K., & Giannini, J. (1989). An exploratory examination
of strategies used by elite coaches to enhance self-efficacy in athletes. Journal of Sport and
Exercise Psychology, 11, 128-140.
Guzzo, R. A., Yost, P. R., Campbell, R. J, & Shea, G. P. (1993). Potency in groups:
Articulating a construct. British Journal of Social Psychology, 32, 87-106.
Haney, C. J., & Long, B. C. (1995). Coping effectiveness: A path analysis of self-
efficacy, control, coping, and performance in sport competitions. Journal of Applied Social
Psychology, 25, 1726-1746.
Hodges, L., & Carron, A. (1992). Collective efficacy and group performance.
International Journal of Sport Psychology, 23, 48-59.
James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journal of
Applied Psychology, 67, 219-229.
Kane, T. D., Marks, M. A., Zaccaro, S. J., & Blair, V. (1996). Self-efficacy, personal
goals, and wrestlers' self-regulation. Journal of Sport and Exercise Psychology, 18, 36-48.
25
Kenny, D. A., & LaVoie, L. (1985). Separating individual and group effects. Journal of
Personality and Social Psychology, 48, 339-348.
Kozlowski, S., & Hattrup, K. (1992). A disagreement about within-group agreement:
Disentangling issues of consistency versus consensus. Journal of Applied Psychology, 77, 161-
167.
LaGuardia, R., & Labbé, E.E. (1993). Self-efficacy and anxiety and their relationship to
training and race performance. Perceptual and Motor Skills, 77, 27-34.
Lee, C. (1982). Self-efficacy as a predictor of performance in competitive gymnastics.
Journal of Sport Psychology, 4, 405-409.
Lee, C. (1986). Efficacy expectations, training performance, and competitive
performance in women's artistic gymnastics. Behaviour Change, 3, 100-104.
Lee, C. (1988). The relationship between goal setting, self-efficacy, and female field
hockey team performance. International Journal of Sport Psychology, 20, 147-161.
Lichacz, F. M., & Partington, J. T. (1996). Collective efficacy and true group
performance. International Journal of Sport Psychology, 27, 146-158.
Lirgg, C. D., George, T. R., Chase, M. A., & Ferguson, R. H. (1996). Impact of
conception of ability and sex-type of task on male and female self-efficacy. Journal of Sport and
Exercise Psychology, 18, 426-343.
Maddux, J.E. (1995). Self-efficacy theory: An introduction. In J.E. Maddux (Ed.), Self-
efficacy, adaptation, and adjustment: Theory, research, and application (pp. 3-33). New York:
Plenum.
Maddux, J.E., & Meier, L.J. (1995). Self-efficacy and depressionIn J.E. Maddux (Ed.),
Self-efficacy, adaptation, and adjustment: Theory, research, and application (pp. 143-172). New
York: Plenum.
Malete, L., & Feltz, D. L. (in press). The effect of a coaching education program on
coaching efficacy. The Sport Psychologist.
Martens, R., Burton, D., Rivkin, F. & Simon, J. (1980). Validity and reliability of the
Competitive State Anxiety Inventory-1. Psychology of motor behavior and sport-1979 (pp. 91-
99). Champaign, IL: Human Kinetics.
Martens, R., Burton, D., Vealey, R. S, Bump, L., & Smith, D. E. (1990). Competitive
State Anxiety Inventory-2. In R. Martens, R. S. Vealey, & D. Burton, Competitive anxiety in sport
(pp. 117-213). Champaign, IL: Human Kinetics.
Martin, J .J., & Gill, D. L. (1991). The relationships among competitive orientation,
sport-confidence, self-efficacy, anxiety, and performance. Journal of Sport and Exercise
Psychology, 13, 149-159.
Martin, J. J., & Gill, D. L. (1995a). The relationships of competitive orientations and self-
efficacy to goal importance, thoughts, and performance in high school distance runners. Journal
of Applied Sport Psychology, 7, 50-62.
Martin, J. J., & Gill, D. L. (1995b). Competitive orientation, self-efficacy and goal
importance in Filipino marathoners. International Journal of Sport Psychology, 26, 348-358.
McAuley, E. (1992). Self-referent thought in sport and physical activity. In T. S. Horn
(Ed.), Advances in Sport Psychology, (pp. 101-118). Champaign, IL: Human Kinetics,
McAuley, E., & Gill, D. (1983). Reliability and validity of the physical self-efficacy in a
competitive sport setting. Journal of Sport Psychology, 5, 185-191.
26
McAuley, E., & Mihalko, S. L. (1998). Measuring exercise-related self-efficacy. In J. L.
Duda (Ed.), Advancements in sport and exercise psychology measurement (pp. 371-390).
Morgantown, WV: Fitness Information Technology.
Miller, M. (1993). Efficacy strength and performance in competitive swimmers of
different skill levels. International Journal of Sport Psychology, 24, 284-296.
Moritz, S. E. (1997). The effect of task type in the relationship between efficacy beliefs
and performance. Unpublished doctoral dissertation, Michigan State University.
Moritz, S. E., Feltz, D. L., Mack, D., & Fahrbach, K. (in press). The relation of self-
efficacy measures to sport performance: A meta-analytic review. Research Quarterly for
Exercise and Sport.
Mischel, L. J., & Northcraft, G. B. (1997). “I Think We Can, I Think We Can…”; The
role of efficacy beliefs in group and team effectiveness. In B. Markovsky, M. J. Lovaglia, & E. J.
Lawler (Eds.), Advances in group processes, vol. 14, (pp. 177-197). Greenwich, CT: JAI Press.
Mitchell, T. R., Hopper, H., Daniels, D., George-Falvy, J., James, L. R. (1994),
Predicting self-efficacy and performance during skill acquisition. Journal of Applied
Psychology, 4, 506-517.
Mulvey, P. W., & Klein, H. J. (1998). The impact of perceived loafing and collective
efficacy on group goal processes and group performance. Organizational Behavior and Human
Decision Processes, 74, 62-87.
Okwumabua, T.M. (1986). Psychological and physical contributions to marathon
performance: An exploratory investigation. Journal of Sport Behavior, 8, 163-171.
O'Leary, A. (1985). Self-efficacy and health. Behavior Therapy and Research, 23, 437-
452. (1985).
Pajares, F. (1997). Current directions in self-efficacy research. In M. Maehr & P. R.
Pintrich (Eds.), Advances in motivation and achievement. Volume 10, (pp. 1-49). Greenwich, CT:
JAI Press.
Paskevich, D. M. (1995). Conceptual and measurement factors of collective efficacy in its
relationship to cohesion and performance outcome. Unpublished doctoral dissertation.
University of Waterloo.
Riggs, M. L., & Knight, P. A. (1994). The impact of perceived group success-failure on
motivational beliefs and attitudes: A causal model. Journal of Applied Psychology, 79, 755-766.
Rousseau, D. M. (1985). Issues in organizational research: Multi-level and cross level
perspectives. Research in Organizational Behavior, 7, 1-37.
Ryckman, R.M., & Hamel, J. (1993). Perceived physical ability differences in the sport
participation motives of young athletes. International Journal of Sport Psychology, 24, 270-283.
Ryckman, R., Robbins, M., Thornton, B., & Cantrell, P. (1982). Development and
validation of a physical self-efficacy scale. Journal of Personality and Social Psychology, 42,
891-900.
Schunk, D.H. (1995). Self-efficacy and education and instruction. In J.E. Maddux (Ed.),
Self-efficacy, adaptation, and adjustment: Theory, research, and application, (pp. 281-303).
New York: Plenum
Shamir, B., House, R. J., & Arthur, M. B. (1992). The motivational effects of charismatic
leadership: A self-concept based theory. Organizational Science, 4, 577-594.
Singleton, D. A., & Feltz, D. L. (1999). The effect of self-modeling on shooting
performance and self-efficacy among intercollegiate hockey players. Unpublished manuscript,
Michigan State University.
27
Slanger, E., & Rudestam, K. E. (1997). Motivation and disinhibition in high risk sports:
Sensation seeking and self-efficacy. Journal of Research in Personality, 31, 355-374.
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the State-Trait
Anxiety Inventory (Self-Evaluation Questionnaire). Palo Alto, CA: Consulting Psychologists
Press.
Spink, K.S. (1990). Group cohesion and collective efficacy of volleyball teams. Journal
of Sport and Exercise Psychology, 12, 301-311.
Treasure, D.C., Monson, J., & Lox, C.L. (1996). Relationship between self-efficacy,
wrestling performance, and affect prior to competition. The Sport Psychologist, 10, 73-83.
Vealey, R. S. (1986). Conceptualization of sport-confidence and competitive orientation:
Preliminary investigation and instrument development. Journal of Sport Psychology, 8, 221-246.
Watkins, B., Garcia, A.W., & Turek, E. (1994). The relation between self-efficacy and
sport performance: Evidence from a sample of youth baseball players. Journal of Applied Sport
Psychology, 6, 21-31
Watson , C. B., & Chemers, M. M. (1998). The rise of shared perceptions: A multilevel
analysis of collective efficacy. Paper presented at the Organizational Behavior Division for the
Academy of Management Meeting, San Diego, CA.
Widmeyer, W. N., Brawley, L. R., & Carron, A. V. (1985). The measurement of
cohesion in sport teams: The Group Environment Questionnaire. London, Ontario: Sports
Dynamics.
Wiggins, M. S. (1998). Anxiety intensity and direction: Preperformance temporal
patterns and expectations in athletes. Journal of Applied Sport Psychology, 10, 201-211.
Weinberg, R., Grove, R., & Jackson, A. (1992). Strategies for building self-efficacy in
tennis players: A comparative analysis of Australian and American coaches. The Sport
Psychologist, 6, 3-13.
Weinberg, R. S., & Jackson, A. (1990). Building self-efficacy in tennis players: A
coach’s perspective. Journal of Applied Sport Psychology, 2, 164-174.
Weiss, M. R., McCullagh, P., Smith, A. L., & Berlant, A. R. (1998). Observational
learning and the fearful child: Influence of peer models on swimming skill performance and
psychological responses. Research Quarterly for Exercise and Sport, 69, 380- 394.
Weiss, M.R., Wiese, D.M., & Klint, K.A. (1989). Head over heels with success: The
relationship between self-efficacy and performance in competitive youth gymnastics. Journal of
Sport and Exercise Psychology, 11, 444-451
Winfrey, M. L. & Weeks, D. L. (1993). Effects of self-modeling on self-efficacy and
balance beam performance. Perceptual and Motor Skills, 77, 907-913.
Wood, R., & Bandura, A. (1989). Impact of concept of ability on self-regulatory
mechanisms and complex decision making. Journal of Personality and Social Psychology, 56,
407-415.
Wurtele, S.K. (1986). Self-efficacy and athletic performance: A review. Journal of Social
and Clinical Psychology, 4, 290-301.
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.
Zaccaro, S. J., Zazanis, M., Diana, M, & Greathouse, C. (1994). The antecedents of
collective efficacy over a team’s lifespan. Paper presented at the Society for Industrial and
Organizational Psychology Conference, Nashville, TN.
28
Study Sample Purpose Self-efficacy Measure Performance
Measure Results
Barling & Abel
(1983) 32 league & 8
nonleague tennis
athletes
M age = 26.6 yr.
USA.
Examine relationship
between SE, response-
outcome, valence, and
tennis performance
10-item SE strength on
5 pt. Scale for tennis
skills
(α = N/A)
TTP: 3 hr after
performance rating
37-item external
rating scale – 12
categories.
1. Efficacy strength related to
12 performance categories. 2.
Lower correlations with
response-outcome and
valence.
Garza & Feltz
(1998) 27 female members of
US Figure Skating
Assoc.
Pre-preliminary -
novice
M age = 12.37 yr.
USA
Intervention to compare
effectiveness of mental
practice (MP) techniques on
SE, competition self-
confidence (CSAI-2), and
performance
Three 10-item SE
strength on 11-pt.
probability scales
(jumps, spins, moves).
(hierarchical)
TTP: 1 week after
competition
3 external 6-pt.
rating scales: 16-
item jump scale, 10-
item spin scale, 5-
item moves scale
1. Both MP techniques
improved performance and
competition confidence
(CSAI-2).
2. All groups improved in SE.
Gayton et al.
(1986) 33 marathon runners
(22 men, 11 women)
M age = 38.6 yr.
USA
Test the validity of the PSE PSE and Perceived
Physical Ability (PPA)
subscale.
(α = N/A)
TTP: less than 1 hr.
Finish time PSE and PPA were related to
finish time.
George (1994) 25 collegiate and 28
high school baseball
athletes.
M age = 20.7 yr.
college; 17.3 yr. high
school
USA
Examine SE – performance
relationship and cognitive
and somatic state anxiety
(CSAI-2) over 9-game
period
4-item SE strength on
11-pt. probability scale
for hitting.
(hierarchical)
TTP: 15-20 min.
Contact percentages
for 9 games 1. SE predicted hitting
performance in 5 games.
2. Performance predicted SE
in 6 games.
3. Anxiety and SE predicted
performance in Game 1.
4. Lower levels of anxiety
were related to stronger SE in
7 games. (used path analysis)
Geisler &
Leith (1997) 40 male current and
former collegiate soccer
athletes
M age = 23.8 yr.
Canada
Examine self-efficacy, self-
esteem, and audience effects
on soccer penalty shooting
performance
1-item SE on 10-pt.
Scale on comparative
ability in penalty shots
TTP: several weeks
10 penalty shots
against a goal
keeper
1. Dichotomized SE had no
effect on performance.
2. Dichotomized self-esteem
had no effect on performance.
3. No audience effect on
performance.
29
Haney & Long
(1995) 178 female athletes
M age = 18.7yr.;
20.4yr. (basketball;
field hockey/soccer)
Canada
Examine a model of coping
effectiveness: Relationships
among SE, control, somatic
anxiety (CSAI-2),
engagement and
disengagement coping, and
performance
Two 4-item SE
strength on 101 pt.
probability scales for
shots listed for field
hockey/soccer and
basketball.
(hierarchical)
TTP: 5 min.
Shooting contest (2
rounds):
1. Number of free
throws or penalty
shots.
2. Performance
satisfaction
1. Years playing experience
predicted SE and perceived
control.
2. SE predicted Round 1
performance , but not Round
2.
3. Round 1 performance
predicted control and SE.
(used path analysis)
Kane et al.
(1996) 216 high school
wrestlers
M age = N/A
USA
Examine the relationships
among SE, personal goals,
and wrestling performance
10-item SE strength on
7-pt. scale for
wrestling moves.
(α = .80)
TTP: N/A
1. Prior performance
2. Win percentage
3. Overtime sudden
death performance
4. Performance
satisfaction
1. Prior performance predicted
SE.
2. SE did not predict win %.
3. SE predicted overtime
performance and satisfaction.
(used path analysis)
LaGuardia &
Labbé (1993) 47 club runners
(33 men, 14 women)
16 college track athletes
(10 men, 6 women)
M age = N/A
(all over 19 yr.)
USA
Compare predictive power
of task-specific SE, general
SE, predicted time, and
training mileage on running
performance and examine
the anxiety (STAI)-SE
relationship
1. 14-item SE on 7-pt.
probability scale for
running.
2. PSE & PPA
(α = N/A)
TTP: 1 hr.
Pace times for 3
races
(1 mile to 10K)
1. Running SE, but not PSE,
predicted pace times in all 3
races..
2. PSE, but not running SE,
was related to STAI.
Lee (1982) 14 female gymnasts
M age = 9.7 yr.
Australia
Compare predictive power
of SE and previous
competitive scores on
competition performance
Public estimation of
score on each of 5
apparatus (1-10 pts.)
TTP: 7 days prior
M score of best 3
apparatus
performances on 10-
pt. scales.
1. Gymnasts’ expectancies
related to performance more
than previous performance.
2. Coaches’ expectancies
related to performance more
than gymnasts’ expectancies.
Lee (1986) 16 female gymnasts
M age = 10.9 yr.
Australia
Compare predictive power
of SE, previous competition
score, and training
performance on competition
performance
Public estimation of
score (1-10 pts.) on
uneven bars
TTP: 2 weeks prior
Judged score on 10-
pt. scale on uneven
bars
1. Training performance
related to competition score.
2. SE and previous score not
related to competition score.
Lee (1988) 96 college female field
hockey athletes on 9
teams
M age = 21 yr.
USA
Examine relationships
among SE, goal-setting, and
team performance.
Number of items: N/A
10-pt. probability scale
for SE strength and
level for hockey skills
(α = N/A)
TTP: Distant
Team won/lost
percentage 1. SE strength, but not level,
related to team winning
percentage.
2. Team goal-setting had
stronger direct relationship
with winning percentage than
SE strength level.
30
Martin & Gill
(1991) 73 male high school
middle and long
distance runners
M age = 16 yr.
USA
Examine relationships
among SE, competitive
orientation (SOQ and COI),
sport confidence (TSCI and
SSCI), cognitive anxiety
(CSAI-2) and performance
1. 6-item placement
SE (strength) on 101-
pt. probability scale
2. 6-item performance
time SE (strength) on
101 pt. scale
(hierarchical)
TTP: 25-35 min.
1. Finish time for
1/2, 1, or 2 mile,
standardized across
events
2. Finish place
1. TSCI predicted placement
SE.
2. Only placement SE
predicted finish time and
finish place.
3. Competitive orientation
(SOQ)was weak predictor of
performance time SE.
Martin & Gill
(1995a) 86 high school distance
runners
(38 women, 48 men)
M age = 16 yr.
USA
Examine relationships
among SE, competitive
orientation (SOQ), goal
importance, goal thoughts,
and performance
1. 6-item placement
SE on 101-pt.
probability scale
2. 6-item performance
time SE on 101-pt.
scale
(hierarchical)
TTP: 25-35 min.
1. Finish time for
1/2, 1, or 2 mile,
standardized across
events
2. Finish place
1. Win orientation and place
goal importance predicted
placement SE.
2. Time goal importance
predicted performance time
SE.
3. Placement SE predicted
finish place. (used path
analysis)
Martin & Gill
(1995b) 41 male marathon
runners
M age = 32.2 yr.
Philippines
Examine relationships
among SE, sport confidence
(TSCI),competitive
orientation (SOQ), and goal
importance
1. 6-item placement
SE on 101-pt.
probability scale
2. 6-item performance
time SE on 101-pt.
scale
(hierarchical)
TTP: 1-3 days
None 1. TSCI correlated with
placement SE.
2. Placement SE correlated
with place and time goal
importance.
3. Time SE correlated with
time goal importance.
Martin &
Mushett (1996) 78 athletes with
disabilities competing
at cerebral palsy games
in England
(34 women, 44 men)
M age = 23.4 yr.
Australia, Canada,
Great Britain
Examine relationships
among social support, SE,
and athletic satisfaction
1-item SE on 101
probability scale for
ability to train to
achieve one’s potential
None SE correlated with listening
support, emotional support,
and technical challenge
support
McAuley &
Gill
(1983)
52 female collegiate
gymnasts
M age = N/A
USA
Compare predictive power
of task-specific and general
SE on gymnastic
performance
1. 4 SE strength scales
(vault, beam, floor,
bars), each with 7
items (hierarchical)
2. PSE (α = .72)
Subscales:
PPA (α = .76);
Individual scores for
each event on 10-pt.
scale
Task-specific SE scales were
better predictors of
performance than PSE scales.
31
PSPC (α = .42)
TTP: less than 1 hr.
Miller
(1993) 84 club-level
competitive swimmers
(42 mean, 42 women)
M age = 14.38
Canada
Compare SE, skill level, and
motivation on swimming
performance in
experimental design,
manipulating SE into high
and low levels; and examine
the SE – motivation
relationship
SE strength on 100-pt.
probability scale.
Number of items =
N/A.
(α = .N/A)
TTP: 3 min.
200m individual
medley. Simulated
competition
1. High SE faster than low SE
swimmers.
2. No effect on performance
for skill level or motivation.
3. Negative relationship
between high SE and
motivation.
Okumabua
(1986) 90 marathon runners
(82 men, 8 women)
M age = 35.5
USA
Examine relationships
among SE, associative
cognitive strategy use,
expected pain, training
history, past performance,
and race performance.
9-item SE strength and
level on a 100-pt.
probability scale for
the marathon task.
(hierarchical)
TTP: approx. 3 days
Finish time 1. SE strength was the
strongest predictor of finish
time, followed by past
performance, expected pain,
and training history.
2. SE strength and level were
related to training history and
past performance.
Ryckman &
Hamel (1993) 123 Grade 9 high
school athletes
(61 women, 62 men)
M age = 14.34 yr.
USA
Examine PPA and sport
participation motives PPA None High PPA athletes rated skill
development, team affiliation,
and having fun as more
important reasons for sport
participation than low PPA
athletes.
Singleton &
Feltz (1999) 23 male ice hockey
athletes
M age = N/A
Range = 18-23
USA
Intervention to examine
effect of self-modeling on
SE and goal shooting
performance
5-item SE strength on
10-pt. probability scale
for performing
shooting skills in
competition
(α = .80)
TTP: immediate
5 backhand shots at
each of four targets
in each corner of
goal. Total shots =
20 at each of 3 time
periods
Self-modeling group showed
greater shooting accuracy and
stronger SE than control.
Slanger &
Rudestam
(1997)
40 male participants in
extreme sports of
skiing, rock climbing,
white water kayaking,
stunt flying
(20 extreme risk takers,
Compare extreme, high risk
sports, and moderate risk
sports participants on
general SE, task-specific
SE, sensation seeking, death
anxiety, and
1. Physical risk SE
strength scale (α =
.91) with 3 error
focused subscales:
trivial, harmful, fatal.
Each scale contained 6
None Physical risk SE was the only
variable that distinguished
between extreme and high risk
participants.
32
20 high risk takers)
20 trained athletes in
moderate risk sports
M age = N/A
repression/senitization. items on 101-pt.
probability scale.
Trivial (α = .90);
Harmful (α = .89);
Fatal (α = .92);
2. PSE 3. SES
TTP: not relevant
Treasure et al.
(1996) 70 male high school
wrestlers
M age = 16.03
USA
Examine the relationships
among SE, performance,
anxiety (CSAI-2) and affect
prior to competition.
10-item SE strength on
101-pt. probability
scale on wrestling
maneuvers
(α = N/A)
TTP: 15 min.
1. Win-loss
2. Number of points
scored
1. SE sig. related to
precompetition positive affect
and anxiety.
2. SE sig. related to both
performance measures.
3. SE was only sig. predictor
of winners and losers
compared with positive affect,
anxiety, wrestling experience,
and age.
Watkins et al.
(1994) 205 male youth
baseball players at
sports camp
M age = 12.5 yr.
USA
Examine the relationship
between SE and baseball
hitting performance.
6-item SE strength on
10-cm visual analog
scale
(hierarchical)
TTP: immediate
Hitting performance
in batting cage over
4 trials
1. SE did not predict
performance.
2. Previous performance
predicted SE and subsequent
performance.
Weiss et al.
(1989) 22 male youth
gymnasts at state
tournament
M age = 11.5 yr.
USA
Examine the relationships
among SE, competitive
anxiety (CSAI-C), worry
cognitions, experience, and
performance
Estimation of score on
each of 6 apparatus
TTP: 2 hr.
Judges’ scores on
high bar, horse,
floor, bars, rings,
vault, and all-around
SE only sig. predictor of
performance
Winfrey &
Weeks (1993) 11 female youth
gymnasts, intermediate
level
M age = N/A
Range = 8-13 yr.
USA
Intervention to examine
effect of self-modeling on
SE and balance beam
performance
9-item SE on 9-pt.
scale for balance
beam, modified from
SSCI
(α = .82-.97)
TTP: immediate
Judged balance
beam skill tests
across 4 time
periods
No effect for SE or
performance
Note. COI = Competitive Orientation Inventory (Vealey, 1986)CSAI-2 = Competitive State Anxiety Inventory-2 (Martens et a., 1990). CSAI-C = Competitive
State Anxiety Inventory – Children (Martens, Burton, Rivkin, & Simon, 1980). N/A = Not available. PPA = Perceived Physical Ability Subscale (Ryckman et
al., 1982). PSE = Physical Self-efficacy Scale (Ryckman et al., 1982). PSC = Physical Self=Presentation Confidence Subscale (Ryckman et al., 1982). SE =
Self-efficacy. SOQ = Sport Orientation Questionnaire (Gill & Deeter, 1988). SSCI = State Sport Confidence Inventory (Vealey, 1986). = STAI = Spielberger
State-Trait Anxiety Inventory (Spielberger, Gorsuch, & Lushene, 1970). TSCI = Trait Sport Confidence Inventory (Vealey, 1986). TTP = Time to performance.
33
Figure Captions
Figure 1. Relationship between sources of efficacy information, efficacy judgments, and
consequences. Note. From
Advances in sport and exercise psychology measurement
(p.
64), by J. L. Duda (Ed.), 1998, Morgantown, WV: Fitness Information Technology.
Copyright by Fitness Information Technology. Adapted with permission.
Figure2. Conceptual model of coaching efficacy. Note. From “A conceptual model of
coaching efficacy: Preliminary investigation and instrument development,” by D. L.
Feltz, M. A. Chase, S. E. Moritz, and P. J. Sullivan, 1999,
Journal of Educational
Psychology, 91
, p.2. Copyright 1999 by American Psychological Association. Reprinted
with permission.
34
Performance
accomplishments
Vicarious
experience
Verbal
persuasion
Physiological
states
Confidence
expectations
Behavior
Choice
Effort
Persistence
Thought
Patterns
Goals
Worry
Attributions
35
Sources of Coaching
Efficacy Information
Extent of Coaching
experience/preparation
Prior success
(won-lost record)
Perceived skill
of athletes
School/community support
Coaching
Efficacy
Dimensions
Game strategy
Motivation
Technique
Character
building
Outcomes
Coaching behavior
Player/team
satisfaction
Player/team
performance
Player/team
efficacy
