ArticlePDF Available

Abstract and Figures

Previous research indicates the viability of a distinction between cognitive and somatic components of the anxiety response, and multidimensional anxiety scales have proven useful in relating cognitive and somatic anxiety to behavioral outcomes. This article describes the development and validation of a sport-specific measure of cognitive and somatic trait anxiety. The Sport Anxiety Scale measures individual differences in Somatic Anxiety and in two classes of cognitive anxiety, Worry and Concentration Disruption. Both exploratory and confirmatory factor analyses supported these dimensions in several different athlete samples. Psychometric properties of the Sport Anxiety Scale are described, as are its relations with other psychological measures and with precompetition affective state measures. In the last of the four studies reported, scores on the Concentration Disruption scale were negatively related to the performance of college football players over the course of a season. The studies suggest that the Sport Anxiety Scale may be useful in defining sport-related anxiety more sharply and assessing how the cognitive and somatic anxiety components relate to performance and other outcome measures in sport.
Content may be subject to copyright.
Smith, Smoll, and Grossbard are with the Department of Psychology, University of Washington, Seattle,
Washington 98195-1525, and Cumming is now with the School for Health, University of Bath, Bath,
England, BA2 7AY.
Journal of Sport & Exercise Psychology, 2006, 28, 479-501
© 2006 Human Kinetics, Inc.
Measurement of Multidimensional Sport
Performance Anxiety in Children and
Adults: The Sport Anxiety Scale-2
Ronald E. Smith, Frank L. Smoll, Sean P. Cumming,
and Joel R. Grossbard
University of Washington
This article describes the development and validation of the Sport Anxiety Scale-2
(SAS-2), a multidimensional measure of cognitive and somatic trait anxiety in
sport performance settings. Scale development was stimulated by ndings that
the 3-factor structure of the original Sport Anxiety Scale (SAS; Smith, Smoll, &
Schutz, 1990) could not be reproduced in child samples and that several items on
the scale produced conicting factor loadings in adult samples. Alternative items
having readability levels of grade 4 or below were therefore written to create a
new version suitable for both children and adults. Exploratory and conrmatory
factor analyses replicated the original SAS factor structure at all age levels, yielding
separate 5-item subscales Somatic Anxiety, Worry, and Concentration Disruption
in samples as young as 9 to 10 years of age. The SAS-2 has stronger factorial
validity than the original scale did, and construct validity research indicates that
scores relate to other psychological measures as expected. The scale reliably pre-
dicts precompetition state anxiety scores and proved sensitive to anxiety-reduction
interventions directed at youth sport coaches and parents.
Key Words: sport anxiety measurement, reliability, factorial and construct
The study of anxiety, its antecedents, its relations with other psychological
variables, and its consequences has a long history of theoretical and empirical
attention within sport psychology. Cognition and arousal are widely considered to
be different components of the anxiety response, and a distinction has long been
made between cognitive and somatic anxiety (Burton, 1998; Davidson & Schwartz,
1976; Deffenbacher, 1977; Smith, Smoll, & Wiechman, 1998). Moreover, although
they interact with one another, cognitive and somatic anxiety can at times be
elicited by different antecedents (Burton, 1998; Morris & Engle, 1981; Morris &
Liebert, 1973), and they can be differentially related to performance, depending
480 Sport Anxiety Scale-2
on the nature of the task (e.g., Deffenbacher, 1980; Gould, Petlichkoff, Simons,
& Vevera, 1987; Smith et al., 1990). Recent research indicates that different brain
regions are involved in different aspects of anxiety, specically, anticipatory
anxiety (worrying) and anxious arousal during a task, supporting still further the
cognitive-somatic distinction (Heller, Schmidtke, Nitschke, Koven, & Miller, 2002;
Hoffman et al., 2005).
Theoretical advances stimulated the development of new measuring instru-
ments to assess the construct of trait anxiety. During the 1980s and thereafter,
sport-specic trait anxiety has frequently been assessed using the Sport Competi-
tion Anxiety Test (SCAT; Martens, 1977), a unidimensional measure that does
not distinguish between or measure differences in somatic and cognitive anxiety.
Although the SCAT has proven to be a valuable research instrument, it measures
primarily somatic anxiety (Smith et al., 1990). It is therefore limited in its useful-
ness for investigating cognitive aspects of trait anxiety. In 1990, cognitive-affective
models of anxiety and empirical ndings concerning differential antecedents and
consequences of cognitive and somatic anxiety prompted the development of a
new sport-specic multidimensional trait anxiety measure, the Sport Anxiety Scale
(SAS; Smith et al., 1990). Developed and cross-validated using both exploratory
and conrmatory factor analysis, the 21-item SAS measures individual differences
in somatic anxiety and in two aspects of cognitive anxiety, namely, worry and con-
centration disruption (Dunn, Causgrove Dunn, Wilson, & Syrotuik, 2000; Smith et
al., 1990). Differential relations of the scales with performance measures have also
been reported. For example, Smith et al. (1990) found that concentration disruption
was the strongest negative predictor of performance in college football players.
Since its development, the SAS has proven useful to researchers in a variety of sport
contexts and appears to be a reliable and valid measure of cognitive and somatic
sport performance anxiety (nCresswell & Hodge, 2004; Giacobbi & Weinberg,
2000; Johnson, Ekengren, & Andersen, 2005; Smith, Ptacek, & Patterson, 2000).
Researchers have also been interested in studying performance anxiety in
children. To extend measurement of competitive trait anxiety downward on the age
continuum, Martens (1977) developed a children’s form of the Sport Competition
Anxiety Test (the SCAT-C). Like the adult version of the SCAT, this unidimensional
measure proved to be a reliable and valid instrument and it has been used in many
studies to assess the antecedents and consequences of anxiety in children (see
Martens, Vealey, & Burton, 1990). Though useful as a global measure of anxiety,
the SCAT-C, like its adult counterpart, does not allow for the assessment of separate
cognitive and somatic components of anxiety.
Indications that the SAS may not be appropriate for younger age groups
appeared when Smith, Smoll, and Barnett (1995) used the SAS as an outcome
measure (together with the SCAT-C) in a study involving a coach-training interven-
tion designed to reduce situational sources of stress and thereby lower performance
trait anxiety. The children in the study ranged in age from 9 to 12 years. Following
data collection, a factor analysis of the SAS was carried out to ensure that it was
appropriate to use its three subscales as dependent variable measures. The analysis
yielded an uninterpretable 5-factor solution with numerous cross-loadings, indi-
cating a failure to replicate the 3-factor structure so consistently found in older
samples. It was therefore necessary to use the total score as the outcome measure.
Although the intervention resulted in highly signicant reductions in trait anxiety
Sport Anxiety Scale-2 481
on both the SAS total score and the SCAT-C (which were highly correlated), it was
not possible to assess effects of the intervention on the somatic anxiety, worry, and
concentration-disruption components of sport performance anxiety. In unpublished
data derived from 10- to 12-year-old children from a youth basketball program,
the SAS’s factor structure broke down again in similar fashion, this time with 6
uninterpretable factors (Everett & Barnett, 1995).
Failure to replicate the 3-factor SAS model in younger samples suggested two
major possibilities, one methodological, the other developmental. At a method-
ological level, it seemed possible that the items, originally developed using high
school and college athlete samples, were too difcult for some younger respondents
to understand, and that poor comprehension of item content was responsible for
the breakdown of the SAS factor structure. In a previous study using the SAS in
a younger sample, Weiss, Ebbeck, and Horn (1997), anticipating this potential
problem, deleted 3 of the 21 SAS items and rewrote 6 others “to enhance their
comprehension for children” (p. 56). Unfortunately, Weiss et al. did not factor-
analyze their adapted scale to determine whether their modications resulted in
the assumed 3-factor structure, so it is not clear whether they were successful in
remedying the problem encountered by Smith et al. (1995). To assess potential
readability problems in younger samples, we therefore subjected each SAS item
to a Flesch-Kincaid readability assessment (Harrison, 1980) and found that many
of the items had reading levels above the 9th grade. We therefore concluded that
the high reading level of the SAS items renders the scale inappropriate for younger
Failure to reproduce the original factor structure in younger children may
also reect developmental aspects of emotional self-perception. A well-dened
factor structure derived from a self-report measure indicates that respondents
are perceiving distinctions among items and responding differentially to them in
ways that produce clusters of items that share common variance and therefore are
assumed to have common psychological meaning (Nunnally & Bernstein, 1994).
Therefore, another potential reason for the breakdown of the 3-factor SAS structure
is that children’s emotional self-perception capabilities do not allow them to dif-
ferentiate between the three aspects of subjectively experienced anxiety indexed
by its items.
There are few studies in the literature that address this question of cognitive-
affective discrimination. Most developmental research has focused on the ability of
children to draw distinctions between discrete emotions, rather than on their ability
to make cognitive-affective distinctions. For example, in a cross-sectional study of
children in the 3rd, 5th, and 7th grades, Turner and Barrett (2003) performed conr-
matory factor analyses of scores on the Revised Children’s Manifest Anxiety Scale
(Reynolds & Richmond, 1985) and the Children’s Depression Inventory (Kovacs,
1980/1981). At all grade levels, a 2-factor anxiety/depression model exhibited a
strong t, indicating that children as young as age 7 differentiated between the sub-
jective experiences of anxiety and depression. However, Clark and Watson’s (1991)
tripartite model, which postulates separate factors of physiological hyperarousal,
negative affectivity, and low positive affect, also ts the data well at all three age
levels. Because the negative affectivity factor has a signicant number of worry
items (but also noncognitive arousal items), this nding suggests a possible per-
ceptual distinction between physiological and cognitive aspects of these emotions
482 Sport Anxiety Scale-2
at even the 3rd-grade level. A study of age-related worry by Muris, Merckelbach,
Meesters, and van den Brand (2002) also showed that worry emerged as a cognitive
response to stress as early as age 3 and became more elaborated with increasing
chronological age, but this study did not relate worry to somatic anxiety.
Few studies have explored the cognitive-affective components of anxiety from
a developmental perspective. In one instance, however, White and Farrell (2001)
administered the Revised Children’s Manifest Anxiety Scale to children between
the ages of 10 and 14 and used conrmatory factor analysis (CFA) to test several
models, all of which posited separate arousal and worry factors. Their analyses
provided evidence of cognitive-somatic differentiation. However, the average
participant in this study was nearly 12 years old and no analyses were done by
age group, so that we cannot be certain of cognitive-somatic differentiation in the
younger age groups. Moreover, this question has not been explored within the con-
text of performance anxiety. The availability of a trait scale with age-appropriate
items could be a useful research tool in determining whether and at what age level
cognitive-somatic differentiation in emotional experience emerges in children’s
sport-related anxiety reactions.
Although the major reason we undertook to develop a new scale was the
assessment of multidimensional anxiety in children, recent developments involving
the SAS prompted us to expand its range of potential application to older samples.
Results of several studies indicate that the factorial validity of the SAS is not as
sound as originally suggested. Analyses by Dunn et al. (2000) and by Prapavessis,
Maddison, and Fletcher (2005) replicated the basic 3-factor structure, but called
into question the factorial integrity of the Concentration Disruption scale. Speci-
cally, two of the items on that scale either cross-loaded or loaded more strongly on
the Worry scale. Utilizing item response theory analyses, Prapavessis et al. (2005)
also found that one of the items on the Somatic scale had marginal measurement
properties. Moreover, their CFAs, while supporting the 3-factor structure of the SAS,
yielded a goodness-of-t index (GFI) of .88 and a non-normed t index (NNFI)
of .81, well below established standards for acceptable t (Hu & Bentler, 1999).
Although a revised scoring system for the SAS improved model t slightly (CFI
= .92, RMSEA = .063; Smith, Cumming, & Smoll, 2006a), we were hopeful that
the new scale would exhibit stronger psychometric properties and a better model
t. Our major focus was on developing a child-appropriate scale, but we saw no
reason why reducing the item reading level would not be of potential benet in
assessing older athletes as well.
In this article, we describe the development and psychometric properties of
the Sport Anxiety Scale-2 (SAS-2). Our goal was to provide researchers with a
reliable and valid multidimensional measure of sport performance anxiety that
would mirror the factor structure of the original SAS, but measure its dimensions
more precisely. Such a measure would enable researchers to measure individual
differences in somatic anxiety, worry, and concentration disruption; to study the
antecedents and consequences of cognitive and somatic performance anxiety in
children and adults; and to measure multidimensional anxiety in longitudinal stud-
ies that begin in childhood. Because the scale was developed within the context
of a coach and parent intervention project in which performance anxiety was one
of several outcome variables, we desired a relatively brief scale that could be used
as part a battery of outcome measures that would not overwhelm young children.
Sport Anxiety Scale-2 483
A by-product of scale construction was the ability to address an important devel-
opmental question, namely, whether and when children beyond the age of 9 years
discriminate between somatic and cognitive aspects of the anxiety response.
Both child and college-age athletes were involved in the development of the
new scale. A total of 1,038 child athletes (571 males and 467 females) ranging
in age from 9 to 14 years (M = 11.5 years, SD = 1.51) participated in the SAS-2
scale-development and validation phases. The majority of the participants were
Caucasian (78%), along with smaller numbers of Asian Americans (9%), African
Americans (6%), Hispanics (4%), and individuals reporting more than one ethnic-
ity (3%). The sample included 277 children 9 and 10 years of age, 418 between 11
and 12 years, and 342 at ages 13 and 14 years, with similar proportions of males
and females at each age level. The samples were drawn from several community
basketball programs in Seattle and from volleyball, soccer, and hockey summer
camps in Seattle, New York, and Boston. One sample of 188 athletes was used
in the process of item selection, and an independent sample of 850 children was
used for factorial validation of the scale using CFA. Other subsamples (specied
below), were used to assess test-retest reliability, relations with other scales, and
to test predictive validity in relation to state anxiety.
To assess the psychometric properties of the SAS-2 for older athletes and to
correlate the subscale and total scores of the new scale with the SAS in the age
population in which the original scale was developed, we selected from a sample
of 1,294 college students enrolled in an introductory psychology class 593 college
freshman students (237 males and 356 females) who were currently involved in
organized athletic activities ranging from intramural and club sports to intercol-
legiate sports, and/or who had participated in high school varsity sports during
the past 3 to 9 months. This sample had a mean age of 18.36 (SD = 3.17). Ethnic
group composition was 59% Caucasian, 33% Asian-American, and 8% African
American or “other.”
Children’s data were collected in group sessions within the activity context by
trained research assistants following the obtaining of signed consent by parents/
guardians. College-age participants were administered the SAS and the SAS-2 in
counterbalanced order during group sessions under anonymous conditions.
Our goal was to develop a brief instrument having 5 to 7 items on each of the
three somatic and cognitive subscales, providing researchers with a short instru-
ment that could be administered as part of a larger test battery. A rational-theoretical
(construct-based) strategy was used to generate new items written to represent each
SAS subscale’s underlying construct, but at a reading appropriate for younger chil-
dren. The underlying theoretical model was a cognitive-affective model advanced
by Smith (1996; also see Smith et al., 1998; Smith & Smoll, 2004), which posits a
3-component model with a higher-order global anxiety latent variable. For purposes
484 Sport Anxiety Scale-2
of generating items, the somatic construct involved various indices of autonomic
arousal centered in the stomach and muscles. The worry construct involved con-
cerns about performing poorly and the resulting negative consequences. Finally, the
concentration disruption construct involved difculties in focusing on task-relevant
cues. The constructs were narrowly dened in order to avoid confounding of content
(e.g., we avoided items like, “I’m concerned that I won’t be able to concentrate”)
that sometimes produced cross-loadings on the SAS Concentration Disruption and
Worry factors (Dunn et al., 2000; Prapavessis et al., 2005).
Each new item was subjected to a Flesch-Kincaid reading level analysis using
the Microsoft Word 2003 program and was retained if it was at or below grade 4.0
on that measure and if its content seemed similar to a corresponding SAS item.
After a preliminary screening during which 13 young athletes between the ages
of 8 and 11 were asked to read potential items and identify any they did not fully
understand, an initial pool of 30 items was generated (10 each for the Somatic,
Worry, and Concentration Disruption subscales). Participants responded on a 4-
point extent-of-experience scale containing the following anchors: 1 (not at all),
2 (a little bit), 3 (pretty much), and 4 (very much). The instructions had a reading
grade level of 3.6.
Exploratory Factor Analyses
Exploratory factor analyses (EFA) played an integral role in item selection. These
items were administered to the sample of 188 boys and girls between the ages of 9
and 14 (M = 11.31, SD = 1.42) who were participating in a community basketball
Following logarithmic transformation of item scores to decrease positive
skewness of item distributions, principal axis factor analyses were carried out
with rotation to oblimin (oblique) solutions. Oblique rotation allows for correlated
factors, as we might expect with components of anxiety, which are known to be
correlated. Factor analysis revealed three clear factors with eigenvalues exceeding
1.00 and a distinct elbow following the third factor on a scree plot. These three
factors corresponded to the Somatic, Worry, and Concentration Disruption sub-
scales of the SAS. All item loadings exceeded .50 on their factor and less than .30
on other factors. In additional to an overall analysis involving the entire sample,
supplementary analyses were conducted in the 9-to-11 and 12-to-14 age groups.
Selection of items based on their factor loadings at each age level, their judged
content validity, and their contributions to scale reliability resulted in a 15-item
SAS-2 containing three subscales, each consisting of 5 items. To approximate
simple factor structure as closely as possible, we retained items only if they had
oblique factor loadings of at least .60 on their primary factor and loadings lower
than .20 on the other factors. The items selected for the 15-item scale ranged in
Flesch-Kincaid reading scores from grade 0.5 to grade 3.9, with a mean reading
level of grade 2.3. The entire scale, including items, instructions, and response
category labels, had a reading level of grade 2.4.
A principal axis factor analysis with oblique rotation based on these 15 items
yielded the factor structure shown in Table 1 for the combined child sample. The
Sport Anxiety Scale-2 485
Table 1 Exploratory Factor Analysis of the SAS-2 with Oblique
Rotation for Children Ages 9 to 14
SAS-2 Item
2. My body feels tense
.12 .18
6. I feel tense in my stomach
10. My muscles feel shaky
12. My stomach feels upset
.04 .02
14. My muscles feel tight because
I am nervous
.03 .14
3. I worry that I won’t play well .01
5. I worry that I will let others down .09
8. I worry that I will not play my best -.05
9. I worry that I will play badly .00
11. I worry that I will mess up during the game .05
1. It is hard to concentrate on the game
4. It is hard for me to focus on what I am
supposed to do
.09 .10
7. I lose focus on the game .09
13. I cannot think clearly during the game .18 .04
15. I have a hard time focusing on what my coach
tells me to do
Note. n = 188. Item factor loadings .60 are boldfaced.
three unrotated factors accounted for 64% of the item response variance. The
same three factors, all with factor loadings exceeding .60 and accounting for more
than 60% of the response variance, also occurred at the 9-to-11 and 12-to-14 age
Because, by convention, EFAs and CFAs should not be performed on the
same sample, we elected to conduct a CFA as the primary analysis to test model
t in the college sample. However, for the reader who might be curious about EFA
factor loadings in this age group, a follow-up principal axis analysis with oblimin
rotation yielded a factor structure similar to that shown in Table 1, with item factor
loadings ranging from .64 to .92 on the primary factors, and no loading on another
factor exceeding .20. As expected on theoretical and empirical grounds, the SAS-
2 subscale scores, derived by summing raw scores on the individual items, were
substantially correlated with one another in the child sample (Somatic with Worry
= .64; Somatic with Concentration Disruption = .62; Worry with Concentration
Disruption = .63).
In the college sample, the corresponding interscale correlations were somewhat
lower (Somatic with Worry = .55; Somatic with Concentration Disruption = .35;
Worry with Concentration Disruption = .47). These results are similar to those
obtained with the original SAS (Smith et al., 1990) and with other anxiety measures
containing cognitive and somatic scales (nMorris et al., 1981; Sarason, 1984), and
486 Sport Anxiety Scale-2
they are consistent with cognitive-affective models of anxiety, which posit causal
relations among the subcomponents of anxiety (Smith et al., 1998).
Confirmatory Factor Analyses
In the theoretical model underlying the SAS (Smith et al., 1998; Smith & Smoll,
2004), performance anxiety is regarded as a global construct that has three related
somatic and cognitive subcomponents. Accordingly, both the SAS and the SAS-2
are designed to provide separate scores on each of the subscales, plus a total score
that reects the global construct. Various researchers have chosen to use the total
score, the subscale scores, or both in their research (Smith et al., 1998). Accord-
ingly, a model involving one somatic and two cognitive factors and a higher-order
anxiety factor was the theoretically preferred model. We also evaluated the t of
the data to two other models: a 3-factor model reecting the somatic anxiety, worry,
and concentration-disruption components without the second-order global (total
score) factor, and a global single-factor model. These analyses were carried out
in a sample of 850 children (M = 11.23, SD = 1.87) participating in a variety of
community and summer-camp sports and not utilized in the EFA phase, and with
the college freshman sample of 593 athletes. To test scalar t to the hypothesized
model, CFAs were conducted on the total child sample and the college sample.
As part of another study focusing on the developmental issue (Grossbard, Smith,
Smoll, & Cumming, 2005), CFAs were also conducted on the 9 and 10, 11 and
12, and 13 and 14 year-old groups to explore the developmental issue regarding
cognitive-somatic discrimination and to assess factorial invariance across these
age groups.
Maximum likelihood estimation using the Amos 5.0 program (Arbuckle, 2003)
was used in all analyses. Items were specied to load on only one factor each. A
variety of commonly reported t indices, including model chi-square, the compara-
tive t index (CFI), the goodness of t (GFI) index, the Bentler-Bonnet non-normed
t index (NNFI), and the root mean square error of approximation (RMSEA) were
used to assess model t (see Hu & Bentler, 1999). The latter indices have been
developed to address perceived inadequacies of the chi-square test, particularly its
tendency to yield signicant results with large sample sizes, therefore increasing
the likelihood of Type II error (Bollen, 1989). The NNFI and the RMSEA are less
affected by sample size and are therefore preferred by many experts (e.g., Fan,
Thompson, & Wang, 1999). Because the item distributions were positively skewed,
as is typical of anxiety scales, we performed a logarithmic transformation of item
scores to better meet the CFA assumption of multivariate normality prior to the
analyses (Tabachnick & Fidell, 2001). No modication analyses were performed on
the data, although there was clearly an opportunity to improve model t by doing
so. This produced more conservative tests of the models but facilitates replicability
and comparisons by other investigators (Byrne, 2001).
Conrmatory factor analysis results for each age group are presented in Table
2. Using criteria recommended by Hu and Bentler (1999), acceptable model t was
found in each age group. At each age level, the data conformed well to both models
tested, with comparative t indices exceeding .95. For the combined child sample,
the CFI was .97, NNFI was .96, and RMSEA was .05. It thus appears that the 3-
factor model of anxiety, with or without a higher-order global anxiety component,
Sport Anxiety Scale-2 487
ts the data quite well. Acceptable t indices for both models were also obtained
in the college sample, indicating similarly high factorial validity for this age group.
Because some researchers may prefer to also use the total score as a general index
of sport performance anxiety, we present the standardized coefcients for the 3-
factor/higher-order model in Figure 1.
Factorial invariance refers to the extent to which a factor structure model
exhibits consistency across measurement periods or groups (Meredith, 1993). Of
major interest was whether the 3-factor structure of the SAS-2 exhibited structural
stability across age groups. Using Amos 5.0, we tested whether the standardized
item coefcients for the three factors, the variance related to the three subfactor
residuals, and the regression coefcients between the latent variables differed sig-
nicantly as a function of age. To test invariance, we compared the t characteristics
of nonconstrained (in which parameters were free to vary) and constrained models
when tested simultaneously across the three children’s age groups (Byrne, 2001).
The models did not differ signicantly from one another, χ
(6) = 10.62, p > .05,
indicated age-related consistency in the SAS-2 factor structure.
Table 2 Confirmatory Factor Analysis Goodness-of-Fit Statistics
for Alternative Models for Children Ages 9 to 14 and for College-Age
90% CIs
9 to 10 years old
Three factors 87 126.89* .97 .97 .044 .028–.059
Higher order 89 133.81* .97 .97 .042 .025–.058
11 to 12 years old
Three factors 87 188.66** .94 .95 .060 .048–.071
Higher order 89 192.70** .94 .95 .060 .048–.071
13 to 14 years old
Three factors 87 184.31** .94 .95 .064 .051–.077
Higher order 89 185.69** .94 .95 .065 .052–.078
9 to 14 years old
Three factors 87 270.48** .96 .96 .050 .043–.057
Higher order 89 285.96** .96 .96 .051 .045–.058
College sample
Three factors 87 303.12** .95 .95 .65 .057–.073
Higher order 89 315.55** .94 .95 .066 .058–.073
Note. NNFI = Non-normed t index (Tucker-Lewis index); CFI = Bentler comparative t index; RMSEA
= root mean square error of approximation; 90% CI = 90% condence interval for RMSEA.
n = 155;
n = 431;
n = 264;
based on Grossbard et al., 2005;
n = 850;
n = 593.
*p < .01, **p < .001.
488 Sport Anxiety Scale-2
Figure 1 Conrmatory factor analysis of the SAS-2 items in relation to a 3-factor
model with a second-order global anxiety factor. The values in the gure are standardized
Sport Anxiety Scale-2 489
Finally, we tested an alternative (null) single-factor model that posits no dis-
tinctions between the three cognitive-somatic factors. This model achieved a poor
t for all age groups, yielding CFIs and RMSEAs of .86 and .10, respectively, in
the 9-to-10 age group, .76 and .13 in the 11-to-12 group and .73 and .15 in the 13-
to-14 group. These indices are below acceptable t levels (Hu & Bentler, 1999).
Clearly, the null single-factor model is not a statistically tenable alternative to either
3-factor model shown in Table 2.
Normative Data
Means and standard deviations for the children aged 9 to 10, 11 to 12, and 13 to
14 derived from the Grossbard et al. (2005) developmental study are presented in
Table 3 as normative data. Descriptive statistics for the college sample are also
presented. In all age groups, mean scores on the Worry scale tended to be higher
than those on the other two scales, and Somatic scores exceeded Concentration
Disruption. Scores tended to be higher for college student athletes than for the
children on all but the Concentration Disruption scale. As is typically found when
anxiety scales are administered to nonclinical samples, score distributions exhibited
a positive skew.
Scale Reliability
Internal consistency and test-retest reliability analyses were carried out for the nal
15-item SAS-2. Cronbach’s alpha served as the measure of internal consistency,
assessed in the total sample of 1,038 children. For total score based on all 15 items,
Table 3 Means and Standard Deviations of SAS-2 Scores for Child
and College-Age Groups
SAS-2 Scale
Age Group
Somatic Anxiety 8.29
Worry 9.05
Total Score 24.88
Note. SDs in parentheses. Subscale scores can range from 5 to 20; total anxiety score can range from
15 to 60. Data on 9- to 14-year-old groups based on Grossbard et al., 2005.
n = 277,
n = 418,
n = 343,
n = 593.
490 Sport Anxiety Scale-2
alpha = .91 (95% CI = .90–.92). Subscale reliability coefcients were .84 (CI =
.82–.85) for Somatic, .89 (CI = .87–-.90) for Worry, and .84 (CI = .82–.85) for
Concentration Disruption. Total score alpha coefcients exceeded .89 for all age
groups. For the subscales, coefcients ranged from .81 to .92 for all age groups
with the exception of Concentration Disruption at age 11 to 12, which yielded a
coefcient of .74. In the college sample, the alpha coefcients for the Somatic,
Worry, and Concentration Disruption scales were .89 (95% CI = .87–.90), .91 (CI
= .90–.92), and .84 (CI = .82–.86), respectively, and the total score alpha was .91
(CI = .90–.92). Thus, the SAS-2 exhibits acceptable internal consistency at both
the total score and subscale levels, and its reliability is quite similar to that found
for the SAS in older samples (Smith et al., 1990).
Test-retest reliability was assessed in a sample of 21 competitive gure skat-
ers ranging in age from 10 to 18 years (M = 12.6 years, SD = 1.75). The athletes
were retested 1 week after the initial administration. Test-retest coefcients were
.76 for Somatic, .90 for Worry, .85 for Concentration Disruption, and .87 for total
score, indicating acceptable measurement stability.
Construct Validity
Acceptable reliability and factorial validity do not ensure that a scale is measuring
the construct it is designed to measure. The underlying construct must be embedded
in a nomological network that species relations with other theoretically related
and unrelated constructs (Cronbach & Meehl, 1955). This entails assessing both
the convergent and discriminant aspects of construct validity (Campbell & Fiske,
1959). Convergent validity was assessed by correlating the SAS-2 with other mea-
sures with which the scale was expected to correlate, and discriminant validity was
assessed validity through correlations with theoretically unrelated variables.
Because a major objective of our work was to develop a multidimensional
scale that would be valid for child samples, we focused particularly on the validity
of the scale in this population. In various subgroups of children aged 9 to 14 years
old, we obtained measures of achievement-related goals, coach-initiated motiva-
tional climate, self-esteem, and social desirability, all of which have been shown
to be related to anxiety in previous research. Relations with goal orientations were
assessed in a sample of 189 male and female basketball players ranging in age
from 10 to 14 (M = 11.52 years, SD = 1.65). Relations with motivational climate,
self-esteem, and social desirability were measured in a sample of 572 of athletes
drawn from community programs and sport camps (M = 11.27 years, SD = 2.14).
We also assessed the predictive validity of the SAS-2 by relating it to state anxiety
reactions in future competitive situations. Finally, we assessed the scales sensitivity
to an intervention that has been shown to decrease children’s trait anxiety.
Correlations With the SAS. If the SAS-2 is measuring the same constructs
as the original SAS, we should expect it to correlate highly with the SAS. We
therefore correlated the two measures in the college sample (n = 593) described
above. This analysis was restricted to the college sample because of the lack of
factorial validity of the original SAS for children. As shown in Table 4, the SAS-2
subscales correlated far more highly with their corresponding SAS scales than they
did with other SAS scales. Total scores on the two scales correlated at .90. The
lowest correlation involved the corresponding Concentration Disruption scales.
Sport Anxiety Scale-2 491
This may be attributable to the previously cited factorial shortcomings of the SAS
Concentration Disruption scale and the likelihood that the SAS-2’s scale is a more
valid measure of this construct. The SAS-2 scales may therefore be regarded as
appropriate substitutes for the original SAS in adult samples.
Achievement Goal Orientations. The nature of achievement goals and their rela-
tion to anxiety have received considerable empirical attention (see Chi [2004] and
Duda & Hall [2001] for reviews). In both educational and sport settings, research
inspired by achievement goal theory has shown that an ego goal orientation is posi-
tively associated with performance anxiety, whereas a mastery or task orientation
is negatively related to performance anxiety (Bandalos, Finney, & Geske, 2003;
McGregor & Elliott, 2002; Vealey & Campbell, 1988). Table 4 shows relations
between the SAS-2 and the task and ego achievement goal orientation subscales
of the Perception of Success Questionnaire (POSQ; Roberts & Treasure, 1995) in
a sample of 189 child athletes. As predicted, SAS-2 subscale and total scores were
positively and signicantly correlated with POSQ ego orientation scores and nega-
tively correlated with task orientation scores at levels approximating those found in
earlier studies with other anxiety measures, including the SAS. On a child-specic
measure of the avoidance (fear of failure) variant of achievement-goal orientation,
expected to be positively related to performance anxiety, Cumming, Smith, Smoll,
and Grossbard (2006) reported correlations of .34, .30, .39, and .40 with the Somatic,
Table 4 Correlations of SAS-2 Subscales and Total Score With
Other Variables
SAS-2 Scale
Sport Anxiety Scale
Somatic 593 .90* .57* .35* .79*
Worry 593 .86* .51* .77*
Concentration Disruption 593 .69* .58*
Total score 593 .90*
POSQ Goal Orientation
Task 189
.25*0 .10 0 .29* .24* 0
Ego 189 .23* .21* .27* .27*
Motivational Climate
Mastery 572
.28* .29* 0 .32* .34* 0
Ego 572 .30* .28* .35* .35*
Self-Esteem 563
.40* .47* 0 .50* .53* 0
Social Desirability 563
.13* .20* 0 .16* .19* 0
Data from Smith et al., 2006b.
*p < .01.
492 Sport Anxiety Scale-2
Worry, Concentration Disruption, and total score, respectively, further supporting
the convergent validity of the SAS-2 scales in relation to achievement goals.
Motivational Climate. Previous research has shown that the nature of the achieve-
ment environment created by signicant adults, such as teachers and coaches, is
related to performance anxiety (Papaioannou & Kouli, 1999; Walling, Duda, &
Chi, 1993; White, 1998). Ego-oriented motivational settings, where emphasis is
placed on besting and comparing oneself with others, are associated with higher
performance anxiety. In contrast, mastery (task)-oriented climates are associated
with lower anxiety. Table 4 shows correlations between athletes’ SAS-2 scores
and scores on the Motivational Climate Scale for Children (Smith, Cumming, &
Smoll, 2006b), designed to measure the motivational climate initiated by youth
sport coaches. As expected on theoretical and empirical grounds, Smith et al.
reported that SAS-2 scores were negatively related to mastery (task) climate scores
and positively related to ego climate scores. When analyzed at a team-mean rather
than individual-athlete level, several of the correlations increased in magnitude.
Most notably, the correlations between team-level ego motivational climate and
mean anxiety level of the team was .47 for Concentration Disruption and .41 for
SAS-2 total score.
Self-Esteem. Trait anxiety has consistently exhibited negative relations with
measures of self-esteem (Brown, 1998; Wylie, 1979). To test this relation using
the SAS-2, we administered the anxiety scale and the Washington Self-Description
Questionnaire (WSDQ; Smoll, Smith, Barnett, & Everett, 1993), a measure of global
self-esteem, to 563 child athletes. As shown in Table 4, the SAS-2 subscales and
total score exhibited the expected negative relations with WSDQ scores.
State Anxiety. Performance trait anxiety is regarded as a predisposition to experi-
ence high anxiety states under conditions of threat (Smith et al., 1998; Spielberger,
1966). Thus, an athlete who is high in sport-specic trait anxiety would be expected
to experience high levels of somatic arousal, worry, and/or concentration disruption
when exposed to stressful competitive sport situations.
To assess the predictive validity of the SAS-2 in a preliminary fashion, 10- to
12-year-old athletes (n = 28) from ve youth basketball teams were administered
the SAS-2 at the beginning of the season and then were administered a state anxiety
scale before an important late-season game 10 weeks later. The state anxiety measure
was adapted from the SAS-2 to allow us to assess relations between the SAS-2 trait
scales and corresponding state subscales (e.g., “I am worrying that . . .” rather than
“I worry that . . .”). The items were answered on the same scale as the SAS-2, but
in terms of “how you feel right now.” Moderate-to-high predictive relations were
observed between the trait and state scales. Somatic trait and Somatic state scales
correlated .38 (p < .05). The corresponding trait-state correlations for the cognitive
scales of Worry and Concentration Disruption were .74 (p < .001) and .46 (p < .01),
respectively. Total scores on the trait and state measures correlated .64 (p < .001).
Although replications with larger samples and in diverse sports are needed, these
results offer initial support for the predictive validity of the SAS-2.
Responsiveness to Anxiety-Reduction Interventions for Coaches and
Parents. Smith, Smoll, and Curtis (1979) developed a coach-training program
that is designed to help coaches create a more positive and less stressful competitive
Sport Anxiety Scale-2 493
sport situation for young athletes. Coach Effectiveness Training (CET) provides
coaches with guidelines designed to create a socially supportive environment
through frequent use of positive reinforcement, encouragement, and technical
instruction, while discouraging the use of punitive behaviors. Coaches are also
encouraged to create a task or mastery-oriented motivational climate, which also
has anxiety-reduction properties.
Smith et al. (1995) showed signicant reductions in performance anxiety
as measured by the SAS total score and by the Sport Competition Anxiety Test
for Children (Martens, 1977) among children who played for coaches who were
exposed to the CET intervention. Children who played for an untrained control
group showed slight increases in anxiety. Because the SAS factor structure could
not be replicated in this sample of 10- to 12-year-old children, it was not possible
to assess reductions in the multiple dimensions of anxiety in this study.
To assess the effects of the intervention on SAS-2 total and subscale scores,
an experimental group of 20 basketball coaches was administered an updated
version of CET called the Mastery Approach to Coaching, which focuses more
explicitly than CET did on promoting a mastery-oriented motivational climate. A
control group of 16 coaches received no training. Children who played for the two
groups of coaches were administered the SAS-2 prior to and 10 weeks later near
the end of the season.
Results revealed statistically signicant (p < .005) Time × Conditions interac-
tions on all SAS-2 subscales and on total score (Smith, Smoll, & Cumming, 2006).
Children exposed to trained coaches exhibited reductions in anxiety scores over
the course of the season, whereas children who played for the untrained coaches
showed increases on all subscales and total score as competitive pressures increased
near the end of the season. In the intervention condition, signicant reductions
occurred for SAS-2 total score (p < .01), Somatic Anxiety (p < .01), and Worry (p
< .025), but the decrease on Concentration Disruption was not signicant. Thus,
the SAS-2 appears to be sensitive to an intervention that has been shown in earlier
research to reduce performance trait anxiety.
In a second study in which mastery-promoting motivational climate interven-
tions were directed at both the coaches and parents of young athletes, a similar pat-
tern of signicant Time × Condition interactions were found for SAS-2 total score
and all subscales (Smoll, Smith, & Cumming, 2006). Again, signicant reductions
in total score, Somatic Anxiety, and Worry occurred in the intervention condition,
but the decrease in Concentration Disruption was not signicant. It thus appears
that Concentration Disruption is less affected by motivational climate interventions
than are the other scales.
Social Desirability. Because it is socially undesirable to endorse anxiety items
(Edwards, 1970), discriminant validity requires that variance on an anxiety measure
not be attributable solely to socially desirable responding. To assess the relation
of the SAS-2 to social desirability response set, we administered an 18-item ver-
sion of the Children’s Social Desirability Scale (Crandall, Crandall, & Katkovsky,
1965) to our large validation sample. Based on previous research, we expected the
SAS-2 to correlate moderately and negatively with social desirability. Instead, we
obtained low negative correlations not exceeding −.20 between the SAS-2 subscales
and social desirability. These correlations are somewhat lower than those obtained
between the adult SAS and the Marlowe-Crowne Social Desirability Scale, an adult
494 Sport Anxiety Scale-2
measure of the same social desirability construct (Smith et al., 1990). Results thus
indicate that the SAS-2 is minimally inuenced by social desirability response set
and support its discriminant validity.
Perceived Competence. Athletes at all levels of self-perceived competence
can exhibit trait anxiety, and previous research has shown low negative relations
between fear of failure, as measured by the Performance Failure Appraisal Inventory
(PFAI), and perceived competence (Conroy, Willow, & Metzler, 2002). Given the
conceptual convergence of the fear of failure and performance anxiety constructs
(and the correlation of .50 between the PFAI and the SAS reported by Conroy et al.),
we should expect a similar pattern of results for the SAS-2. We therefore admin-
istered the SAS-2 and a 9-point measure of self-rated competence in basketball
(ranging from very poor to among the best) to a sample of 570 youth basketball
players. Low negative correlations were observed between anxiety and perceived
competence. Somatic Anxiety correlated −.07, whereas the cognitive Worry and
Concentration Disruption exhibited correlations of −.16 and −.06, respectively.
The SAS-2 total score correlated −.11 with perceived competence in basketball.
This result provides further evidence of discriminant validity in that the SAS-2 is
measuring something other than perceived competence.
The scale used in our research, together with the scoring key for the SAS-2
subscales, is presented in the appendix to this article.
Development of the SAS-2 was prompted by several issues relating to the facto-
rial validity of the original SAS. First, several studies suggested that at least three
items on the original measure had major measurement shortcomings (Dunn et al.,
2000; Prapavessis et al., 2005). Second, although the original SAS exhibited sat-
isfactory t indices in CFAs, we felt a need to improve its t to the hypothesized
3-factor model. A nal impetus was a failure to nd the usual 3-factor structure
of the SAS when the scale was administered to younger athletes in the 10- to 12-
year-old range. This failure raised two questions, one methodological, the other
theoretical. The methodological question related to the applicability of the SAS to
samples younger than those for which it was developed. Reading level analyses
revealed that many of the SAS items had reading level scores above the 9
One possibility, therefore, was that the 3-factor structure failed to replicate because
of item-comprehension difculties in younger samples.
Failure to replicate the 3-factor SAS factor structure also raised the theo-
retically interesting question of whether the three-component cognitive-affective
model of anxiety applies to younger children’s experiences of anxiety. Perhaps
children do not fully differentiate between cognitive and somatic components of
anxiety until some point in adolescence. Surprisingly, we found that with a few
exceptions (e.g., Turner & Barrett, 2003), little developmental research had been
done on cognitive-somatic discrimination in children’s self-perceptions of their
emotional reactions.
Development of a new version of the SAS with age-appropriate reading levels
has helped address both the methodological and the theoretical issues. Within the
9- to 14-year-old range, and in the college sample, the new measure yielded a
factor structure that replicated the three-component structure of the original SAS.
Sport Anxiety Scale-2 495
At even the youngest levels of our age sample, CFA revealed a good t between
children’s item responses and the 3-factor model, with or without a higher-order
general performance anxiety factor. Moreover, the factor structure remained invari-
ant across age groups. It thus appears that, at least down to 9 years of age, children
do indeed differentiate between the experiential aspects of anxiety that correspond
to the Somatic, Worry, and Concentration Disruption factors. Moreover, the SAS-2
subscales and the scale as a whole have high internal consistency and acceptable
test-retest reliability over a period of up to 3 months.
Conrmatory factor analyses strongly supported both a 3-factor model and
a 3-factor model with a higher-order (total score) factor. Thus, researchers can
justiably use the three subscale scores, the total score, or all four scores in their
empirical work. Especially encouraging are the high factor loadings and a complete
absence of cross-loadings at all age levels, as well as the substantial increase in
CFA t indices compared with results derived from the original SAS. We should
note that, although modication indices we applied could have improved model t
(particularly the RMSEA index), we elected to report unmodied CFAs, resulting
in more conservative tests of model t.
Although the SAS-2 subscales are substantially correlated with one another,
as cognitive-affective theories would predict, it is worth noting that the 3-factor
solution with similarly high loadings was also found when an EFA orthogonal
(varimax) rather than an oblique rotation was performed on the SAS-2 items. This
is important because factor scores generated from an orthogonal rotation are essen-
tially uncorrelated. Using orthogonal factor scores as either predictor or outcome
variables can help clarify the independent roles of cognitive and somatic anxiety
for theoretical purposes (see Smith [1989] for a discussion).
Although additional work relating the SAS-2 to other measures is clearly
needed, preliminary results are promising. The scales exhibited low correlations
with a measure of social desirability, and they were essentially unrelated to self-
perceived competence, providing evidence of discriminant validity. Results bearing
on convergent validity were also encouraging. In the college sample, correlations
between the SAS and the SAS-2 were high enough to conclude that the two measures
are tapping the same constructs. The fact that the respective Concentration Disrup-
tion scales correlated less substantially than the other scales may be attributable
to suboptimal items in the original SAS scale that have cross-loaded with Worry
in some studies (Dunn et al., 2000; Prapavessis et al., 2005). On psychometric
grounds, the SAS-2 thus appears to be an improvement over the SAS.
Relations between SAS-2 scores and achievement goal constructs were con-
sistent with theoretical expectations and previous results. At the level of individual
athletes, ego and avoidant achievement orientations were positively associated
with anxiety and task orientation was negatively related. The type of motivational
climate created by coaches was also associated with differences in anxiety. At both
the athlete and team level, the more ego-oriented the motivational climate was
judged to be, the higher were the levels of somatic and cognitive anxiety reported
by the athletes. We should note, however, that these results are correlational in
nature, and all measures are based on athlete reports, so that causal inferences
cannot be made with certainty. Theoretically, we would expect that an ego-oriented
motivational climate would increase the potential threat value of the athletic situ-
ation and increase anxiety (Duda & Hall, 2001; Roberts & Treasure, 1995), but in
496 Sport Anxiety Scale-2
the absence of athlete-independent measures of the motivational climate (such as
observational measures), we cannot rule out the possibility that anxious athletes
tend to view athletic situations as more ego-oriented, or that some third variable is
responsible for the relation between motivational climate and anxiety. This topic is
clearly deserving of future empirical attention and will require independent sources
of data on motivational climate.
Two other sets of results provided evidence for the validity of the SAS-2. First,
the SAS-2 successfully predicted state anxiety scores collected 10 weeks later.
This is a key validity nding, given that trait anxiety is viewed as a predisposition
to experience state anxiety in challenging or threatening situations (Spielberger,
1966). The new state anxiety measure derived from the SAS-2 also allowed us
to test the predictive power of each of the subscales. Although all of the SAS-2
subscales predicted corresponding state anxiety components at a statistically sig-
nicant level, the strongest predictive power was seen for the Worry subscale and
the SAS-2 total score.
The SAS-2 also proved to be sensitive to a coach-training intervention previ-
ously shown to decrease trait anxiety in young athletes (Smith et al., 1995). In a
more recent experimental study (Smith, Smoll, et al., 2006), children who played
for trained coaches exhibited a decrease in performance anxiety over the course
of the sport season, whereas children who played for untrained coaches increased
in anxiety over the same period, paralleling the results shown in the earlier study.
All of the subscales and the total score exhibited this signicant Time × Conditions
interaction. Similar results were obtained in a second study, this one involving
complementary interventions directed at both coaches and parents (Smoll et al.,
2006). It thus appears that the SAS-2 may be useful as an outcome measure in
research designed to evaluate anxiety-reduction interventions in sport.
In summary, reliability and validity studies indicate that the SAS-2 has good
psychometric properties. Further, there is evidence for factorial, convergent, dis-
criminant, and predictive validity. From a reading-level perspective, the measure
appears appropriate for use with children down to age 8 or 9 and with older popula-
tions as well. The SAS-2 can be used to extend multidimensional anxiety research
downward to younger age groups while measuring the same anxiety components
in older populations. The instrument seems suitable not only for basic research on
the cognitive and somatic aspects of anxiety, but also for assessing the efcacy of
interventions designed to reduce anxiety. Moreover, the instrument may be help-
ful in tailoring interventions to individual athletes who differ in their patterns of
somatic and cognitive anxiety. For example, an athlete with a high somatic anxiety
component might be particularly responsive to arousal-control interventions such
as relaxation training, whereas one high in concentration disruption might prot
maximally from a more cognitively oriented attention-control training approach.
Several limitations and unanswered questions should be noted. As in all
instances of instrument development, replication of results in future studies is
needed. As an example, the factorial issues involving several items in the original
SAS were not apparent in the samples used in the scales original development. Only
when the factor structure was studied in new samples (Dunn et al., 2000; Prapaves-
sis et al., 2005) did the item-loading disparities appear. Although our samples were
large ones containing both males and females, several age levels, and involving
Sport Anxiety Scale-2 497
several sports, additional research is needed within other athletic populations. Also
absent in our construct validation studies is information on relations with athletic
performance measures. There is a need for such studies in view of well-documented
relations of anxiety with performance outcomes (Burton, 1998; Smith et al., 1998).
More research is also needed with older-age samples, as our validity studies focused
on child athletes. Given the positive results on the CFA t indices in the college
sample, we are cautiously optimistic that the SAS-2 will be useful in the study of
sport performance anxiety in adult populations as well. We should note, however,
that the college sample represented a range of competitive sport levels and, in some
cases, athletes who had not competed at a high level for 3 to 9 months. Although
such a sample is suitable for correlational analyses on a trait measure like ours,
the validity of SAS-2 scores in relation to other variables and in other adult sport
populations requires future empirical attention. As Nunnally and Bernstein (1994)
note, validation of an instrument is a continuing process, not an end point, and much
remains to be done to extend the validity ndings reported here.
The problems encountered in using the SAS with athletes younger than the
high school and college athletes used in its development, and the success of the
SAS-2 in remedying these problems, illustrate the importance of assessing reading
level in existing measures and devising instruments that are age-appropriate for
younger populations. Many of the measures commonly used in sport psychology
research, like the SAS, were developed using college-age and adult samples. Our
recommendation is that researchers exercise caution in applying such instruments
(even those with high face validity) to younger athlete populations without assess-
ing reading level and the ability of children to understand item content. In using
multidimensional scales, it is also important to apply factor analysis to ensure that
the dimensions measured by the scale are reproducible in the younger age group.
Failure to do so may yield misleading results if the scale in question is inappropri-
ately applied to a child sample.
This research was supported in part by Grant 2297 from the William T. Grant Foundation.
We express our appreciation to the following for their assistance in data collection:
Erica Coppel, Polo DeCano, Kira Elste, Christopher Harris, Leslie Lombardo, Kim
Matz, Cheree Monroe-Wilson, Olivia Morrow, Tori Nutsch, Dana Ryan, Jason Victor,
and Nathalie Walker.
Arbuckle, J.L. (2003). AMOS 5.0 [Computer software]. Chicago: Smallwaters.
Bandalos, D.L., Finney, S.J., & Geske, J.A. (2003). A model of statistics performance based
on achievement goal theory. Journal of Educational Psychology, 95, 604-616.
Bollen, K.A. (1989). Structural equations with latent variables. New York: Wiley.
Brown, J.D. (1998). The self. New York: McGraw-Hill.
Burton, D. (1998). Measuring competitive state anxiety. In J.L. Duda (Ed.) (1998). Advances
in sport and exercise psychology measurement (pp. 129-148). Morgantown, WV: Fit-
ness Information Technology.
498 Sport Anxiety Scale-2
Byrne, B.M. (2001). Structural equation modeling with AMOS: Basic concepts, applications,
and programming. Mahwah, NJ: Erlbaum.
Campbell, D.T., & Fiske, D.W. (1959). Convergent and discriminant validation by the
multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
Chi, L. (2004). Achievement goal theory. In T. Morris & J. Summers (Eds.), Sport psy-
chology: Theories, applications, and issues (pp. 152-174). Sydney: John Wiley
Clark, L.A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psycho-
metric evidence and taxonomic implications. Journal of Abnormal Psychology, 100,
Conroy, D.E., Willow, J.P., & Metzler, J.N. (2002). Multidimensional fear of failure mea-
surement: The Performance Failure Appraisal Inventory. Journal of Applied Sport
Psychology, 14, 76-90.
Crandall, V.C., Crandall, V.J., & Katkovsky, W. (1965). A children’s social desirability
questionnaire. Journal of Consulting Psychology, 29, 27-36.
Cronbach, L.J., & Meehl, P.E. (1955). Construct validity in psychological tests. Psychologi-
cal Bulletin, 52, 281-302.
Cumming, S.P., Smith, R.E., Smoll, F.L., & Grossbard, J.R. (2006). Measurement of achieve-
ment goals in youth sports: The Children’s Achievement Orientation Scale. Manuscript
submitted for publication.
Davidson, R.J., & Schwartz, G.E. (1976). The psychobiology of relaxation and related states:
A multi-process theory. In D. Mostofsky (Ed.), Behavioral control and modication of
physiological activity (pp. 399-442). Englewood Cliffs, NJ: Prentice-Hall.
Deffenbacher, J.L. (1977). Relationship of worry and emotionality to performance on the
Miller Analogies Test. Journal of Educational Psychology, 69, 191-195.
Deffenbacher, J.L. (1980). Worry and emotionality in test anxiety. In I.G. Sarason (Ed.), Test
anxiety: Theory, research, and applications (pp. 111-128). Hillsdale, NJ: Erlbaum.
Duda, J.L., & Hall, H.K. (2001). Achievement goal theory in sport: Recent extensions and
future directions. In R. Singer, H. Hausenblas, & C. Janelle (Eds.), Handbook of sport
psychology (2nd ed., pp. 417-443). New York: Wiley.
Dunn, J.G.H., Causgrove Dunn, J., Wilson, P., & Syrotuik, D.G. (2000). Reexamining the
factorial composition and factor structure of the Sport Anxiety Scale. Journal of Sport
& Exercise Psychology, 22, 183-193.
Edwards, A.E. (1970). The measurement of personality traits by scales and inventories.
New York: Holt, Rinehart, & Winston.
Everett, J.J., & Barnett, N.P. (1995). [Performance anxiety in recreational youth sports].
Unpublished raw data.
Fan, X.B., Thompson, B., & Wang, L. (1999). Effects of sample size, estimation method,
and model specication on structural equation modeling t indices. Structural Equation
Modeling: A Multidisciplinary Journal, 6, 56-83.
Giacobbi, P.R. Jr., & Weinberg, R.S. (2000). An examination of coping in sport: Individual
trait anxiety and situational consistency. The Sport Psychologist, 14, 42-62.
Gould, D., Petlichkoff, L., Simons, J., & Vevera, M. (1987). The relationship between Com-
petitive State Anxiety Inventory-2 subscale scores and pistol shooting performance.
Journal of Sport Psychology, 9, 33-42.
Grossbard, J.R., Smith, R.E., Smoll, F.L., & Cumming, S.P. (2005). Cognitive-affective dis-
crimination in children: A developmental perspective on the self-perception of anxiety.
Unpublished manuscript, University of Washington.
Harrison, C. (1980). Readability in the classroom. Cambridge, UK: Cambridge Educa-
Heller, W., Schmidtke, J.I., Nitschke, J.B., Koven, N.S., & Miller, G.A. (2002). States,
traits, and symptoms: Investigating the neural correlates of emotion, personality, and
psychopathology. In D. Cervone & W. Mischel (Eds.), Advances in personality science
(pp. 106-126). New York: Guilford Press.
Sport Anxiety Scale-2 499
Hoffman, S.G., Moscovitch, D.A., Litz, B.T., Kim, H., Davis, L.L., & Pizzagalli, D.A.
(2005). The worried mind: Autonomic and prefrontal activation during worrying.
Emotion, 5, 464-475.
Hu, L., & Bentler, P.M. (1999). Cutoff criteria for t indexes in covariance structure analy-
sis: Conventional criteria versus new alternatives. Structural Equation Modeling: A
Multidisciplinary Journal, 6, 1-55.
Johnson, U., Ekengren, J., & Andersen, M.B. (2005). Injury prevention in Sweden: Helping
soccer players at risk. Journal of Sport & Exercise Psychology, 27, 32-38.
Kovacs, M. (1980/1981). Rating scales to assess depression in school-aged children. Acta
Paedopsychiatrica, 46, 305-315.
Martens, R. (1977). Sport Competition Anxiety Test. Champaign, IL: Human Kinetics.
Martens, R. Vealey, R.S., & Burton, D. (1990). Competitive anxiety in sport. Champaign,
IL: Human Kinetics.
McGregor, H.A., & Elliot, A.J. (2002). Achievement goals as predictors of achievement-
related processes prior to task engagement. Journal of Educational Psychology, 94,
Meredith, W. (1993). Measurement invariance, factor analysis and factorial invariance.
Psychometrika, 58, 525-543.
Morris, L.W., & Engle, W.B. (1981). Assessing various coping strategies and their effects
on test performance and anxiety. Journal of Clinical Psychology, 37, 165-171.
Morris, L.W., & Liebert, R.M. (1973). Effects of negative feedback, threat of shock, and
level of trait anxiety on the arousal of two components of anxiety. Journal of Consult-
ing Psychology, 20, 321-326.
Muris, P., Merckelbach, H., Meesters, C., & van den Brand, K. (2002). Cognitive develop-
ment and worry in normal children. Cognitive Therapy and Research, 26, 775-787.
nNewton, M., Duda, J.L., & Yin, Z. (2000). Examination of the psychometric properties
of the Perceived Motivational Climate in Sport Questionnaire-2 in a sample of female
athletes. Journal of Sport Sciences, 18, 1-16.
Nunnally, J.C., & Bernstein, I.H. (1994). Psychometric theory (3rd ed.). New York:
Papaioannou, A., & Kouli, O. (1999). The effect of task structure, perceived motivational
climate, and goal orientations on students’ task involvement and anxiety. Journal of
Applied Sport Psychology, 11, 51-71.
Prapavessis, H., Maddison, R., & Fletcher, R. (2005). Further examination of the factor
integrity of the Sport Anxiety Scale. Journal of Sport & Exercise Psychology, 27,
Reynolds, C.R., & Richmond, B.O. (1985). Revised Children’s Manifest Anxiety Scale. Los
Angeles: Western Psychological Services.
Roberts, G.C., & Treasure, D.C. (1995). Achievement goals, motivational climate, and
achievement strategies and behaviors in sports. International Journal of Sport Psy-
chology, 26, 64-80.
Sarason, I.G. (1984). Stress, anxiety, and cognitive interference: Reactions to tests. Journal
of Personality and Social Psychology, 46, 929-938.
Smith, R.E. (1989). Conceptual and statistical issues in research involving multidimensional
anxiety scales. Journal of Sport & Exercise Psychology, 11, 452-457.
Smith, R.E. (1996). Performance anxiety, cognitive interference, and concentration enhance-
ment strategies in sport. In I.G. Sarason, B.R. Sarason, & G.R. Pierce (Eds.), Cognitive
interference: Theories, methods, and ndings. Hillsdale, NJ: Erlbaum.
Smith, R.E., Cumming, S.P., & Smoll, F.L. (2006a). Factorial integrity of the Sport Anxiety
Scale: A methodological note and revised scoring recommendations. Journal of Sport
& Exercise Psychology, 28, 109-112.
Smith, R.E., Cumming, S.P., & Smoll, F.L. (2006b). Measurement of motivational climate
in youth sports: The motivational climate scale for children. Manuscript submitted
for publication.
500 Sport Anxiety Scale-2
Smith, R.E., Ptacek, J.T., & Patterson, E. (2000). Moderator effects of cognitive and somatic
trait anxiety on the relation between life stress and physical injuries. Anxiety, Stress,
and Coping, 13, 269-288.
Smith, R.E., & Smoll, F.L. (2004). Anxiety and coping in sport: Theoretical models and
approaches to anxiety reduction. In T. Morris & J.J. Summers (Eds.), Sport psychology:
Theories, applications, and issues (2nd ed., pp. 294-321). Sydney, Australia: Wiley.
Smith, R.E., Smoll, F.L., & Barnett, N.P. (1995). Reduction of children’s sport performance
anxiety through social support and stress-reduction training for coaches. Journal of
Applied Developmental Psychology, 16, 125-142.
Smith, R.E., Smoll, F.L., & Cumming, S.P. (2006). Effects of coach training on changes
in athletes’ cognitive and somatic sport performance anxiety. Manuscript submitted
for publication.
Smith, R.E., Smoll, F.L., & Curtis, B. (1979). Coach Effectiveness Training: A cognitive-
behavioral approach to enhancing relationship skills in youth sport coaches. Journal
of Sport Psychology, 1, 59-75.
Smith, R.E., Smoll, F.L., & Schutz, R.W. (1990). Measurement and correlates of sport-spe-
cic cognitive and somatic trait anxiety: The Sport Anxiety Scale. Anxiety Research,
2, 263-280.
Smith, R.E., Smoll, F.L., & Wiechman, S.A. (1998). Measurement of trait anxiety in sport.
In J.L. Duda (Ed.), Advances in sport and exercise psychology measurement (pp. 105-
128). Morgantown, WV: Fitness Information Technology.
Smoll, F.L., Smith, R.E., Barnett, N.P., & Everett, J.J. (1993). Enhancement of children’s
self-esteem through social support training for youth sport coaches. Journal of Applied
Psychology, 78, 602-610.
Smoll, F.L., Smith, R.E., & Cumming, S.P. (2006). Effects of coach and parent training on
performance anxiety in young athletes: A systemic approach. Manuscript submitted
for publication.
Spielberger, C.D. (1966). Theory and research on anxiety. In C.D. Spielberger (Ed.), Anxiety
and behavior (pp. 1-17). New York: Academic Press.
Tabachnick, B., & Fidell, L. (2001). Using multivariate statistics. New York: Harper Col-
Turner, C.M., & Barrett, P.M. (2003). Does age play a role in the structure of anxiety and
depression in children and youths? An investigation of the tripartite model in three age
cohorts. Journal of Consulting and Clinical Psychology, 71, 826-833.
Vealey, R.S., & Campbell, J.L. (1988). Achievement goals of adolescent gure skaters:
Impact on self-condence, anxiety, and performance. Journal of Adolescent Research,
3, 227-243.
Walling, M.D., Duda, J.L., & Chi, L. (1993). The Perceived Motivational Climate in Sport
Questionnaire: Construct and predictive validity. Journal of Sport & Exercise Psychol-
ogy, 15, 172-183.
Weiss, M.R., Ebbeck, V., & Horn, T.S. (1997). Children’s self-perceptions and sources of
physical competence information: A cluster analysis. Journal of Sport & Exercise
Psychology, 19, 52-70.
White, K.S., & Farrell, A.D. (2001). Structure of anxiety symptoms in urban children:
Competing factor models of the Revised Children’s Manifest Anxiety Scale. Journal
of Consulting and Clinical Psychology, 69, 333-337.
White, S.A. (1998). Adolescent goal proles, perceptions of the parent-initiated motivational
climate, and competitive trait anxiety. The Sport Psychologist, 12, 16-28.
Wylie, R.C. (1979). The self-concept (Vol. 2). Lincoln, NE: University of Nebraska Press.
Sport Anxiety Scale-2 501
Many athletes get tense or nervous before or during games, meets or matches. This happens even to pro
athletes. Please read each question. Then, circle the number that says how you USUALLY feel before or
while you compete in sports. There are no right or wrong answers. Please be as truthful as you can.
Before or while I compete in sports:
Not At
A Little
It is hard to concentrate on the game.
1 2 3 4
My body feels tense.
1 2 3 4
I worry that I will not play well.
1 2 3 4
It is hard for me to focus on what I am supposed to
1 2 3 4
I worry that I will let others down.
1 2 3 4
Before or while I compete in sports:
Not At
A Little
I feel tense in my stomach.
1 2 3 4
I lose focus on the game.
1 2 3 4
I worry that I will not play my best.
1 2 3 4
I worry that I will play badly.
1 2 3 4
My muscles feel shaky.
1 2 3 4
Before or while I compete in sports:
Not At
A Little
I worry that I will mess up during the game.
1 2 3 4
My stomach feels upset.
1 2 3 4
13. I cannot think clearly during the game. 1 2 3 4
14. My muscles feel tight because I am nervous. 1 2 3 4
15. I have a hard time focusing on what my coach tells
me to do.
1 2 3 4
Scoring Key. Somatic: Items 2, 6, 10, 12, 14; Worry: Items 3, 5, 8, 9, 11; Concentration
Disruption: Items 1, 4, 7, 13, 15.
Appendix: Sport Anxiety Scale-2
... Athletes are able to reduce their anxiety with themselves, motivational encouragement from parents and coaches, and control anxiety well. Smith et al., (1990) analyzed the factors of the anxiety scale The Sport Anxiety Scale (SAS), with the result that somatic anxiety, and two classes of cognitive anxiety namely, worry and concentration disorders can measure individual anxiety. Gould et al., (2002) Mottaghi et al., (2013) and Bali (2015) shows that there is a significant negative relationship between anxiety and athlete performance. ...
... Competing performance instruments were adopted from (Cohn, 1991) with a validity value of 0.863 and a reliability of 0.953, self-confidence was adopted from (Amir, 2015) with a validity of 0.614 and a reliability of 0.87, motivation was adopted from (Pelletier et al., 1995) with a validity of 0.70 and reliability 0.85, and anxiety was adopted from (Smith et al., 1990) with validity 0.838 and reliability 0.74. The data analysis technique used is a prerequisite test which consists of a normality test, linearity test, and multicollinearity test, while hypothesis testing uses partial and simultaneous tests. ...
Full-text available
This study aims to analyze the effect of self-confidence, motivation, and anxiety on the performance of basketball athletes. This type of research is quantitative with an ex post facto approach. The population in this study were high school basketball athletes in Baubau City, totaling 119 people (68 male, 51 female). The instrument used is a questionnaire. The data analysis technique used the F test. The results showed that (1) there was a positive effect of self-confidence on competitive performance in high school basketball athletes in Baubau City. (2) there is a positive motivational effect on competitive performance in high school basketball athletes in Baubau City. (3) there is a negative effect on match performance on high school basketball athletes in Baubau City. (4) There is an effect of self-confidence, motivation and anxiety on the competitive performance of high school basketball athletes in Baubau City.
... High-performance sports are extremely competitive, and performance failure is the most prominent situational stress for athletes (Smith et al., 1990). Many elite athletes experience fear of failure (FF) when participating in such sports (Correia and Rosado, 2018), which is a negative emotion experienced when an individual engages in achievement-oriented activities and predicts they will not meet certain set goals (Conroy et al., 2001). ...
Full-text available
Controlling coaching style is a key predictor of athletes’ fear of failure, but the mediating mechanisms underlying this relationship require further exploration. Based on the self-determination theory (SDT) and the hierarchical model of intrinsic and extrinsic motivation (HMIEM), this study investigated the effect of a controlling coaching style on athletes’ fear of failure, as well as the chain mediating effects of basic psychological needs and sport commitment. A questionnaire survey of 252 active athletes was administrated using scales for fear of failure, controlling coaching style, sport commitment, and basic psychological needs. The results indicated that a controlling coaching style was negatively correlated with basic psychological needs and indirectly affected athletes’ fear of failure via basic psychological needs and constrained commitment. The chain mediating effect of basic psychological needs on enthusiastic commitment was not significant, but it was for constrained commitment. In conclusion, the impact of a controlling coaching style on athletes’ fear of failure via basic psychological needs is manifested through the strengthening of constrained commitment rather than the weakening of enthusiastic commitment. These findings contribute to a deeper understanding of why and how a controlling coaching style influences athletes’ fear of failure. Coaches should seek more scientific and effective ways to instruct their athletes.
... The Sports Anxiety Scale (SAS) was developed by Smith et al. (1990) as a multidimensional measure of competitive trait anxiety experienced by athletes before or during competition. This 21-item questionnaire assesses somatic anxiety (nine items), worry (seven items), and concentration disruption (five items). ...
Full-text available
This study investigated the effects of psychological skills training (PST) in shooters psychophysiologically using heart rate variability (HRV) in addition to psychological questionnaires and participant interviews. Five junior pistol shooters participated in an 8-week PST program consisting of a group session per week followed by individual counseling. Before and after PST, we collected electrocardiography data during rest, mental imagery of sport-related crisis situations, and successful performance, to analyze differences in HRV indices. Participants also responded to the Psychological Skills Inventory for Archery and Shooting (PSIAS), Intrinsic Motivation Inventory (IMI), Sports Anxiety Scale (SAS), and Trait Sport Confidence Inventory (TSCI). Results showed that the perceived competence (pre: 2.52 ± 0.95, post: 3.36 ± 0.73, p = 0.049) and trait sport confidence (pre: 4.94 ± 1.17, post: 6.60 ± 0.65, p = 0.049) significantly improved after PST. The analysis of HRV indicated that the ratio of low-frequency power to high-frequency power (LF/HF ratio) decreased significantly during imagery of crisis (pre: 3.4 ± 2.3, post: 1.014 ± 0.71, p = 0.038) and success (pre: 1.933 ± 0.917, post: 0.988 ± 0.572, p = 0.046), reflecting a strengthened autonomic nervous system’s responsiveness to stress. Our findings illustrate that PST can help athletes better cope with psychologically disturbed situations during competition, by providing psychophysiological evidence through HRV changes.
... The sport emotion questionnaire (SEQ) emerged as one inventory that has gained prominence in evaluating a broad spectrum of emotions experienced during competition across different athletes from a multi-dimensional perspective (Jones et al., 2005), despite the existence of several other unidimensional scales like Sport Anxiety Scale (SAS; Smith et al., 1990), Competitive State Anxiety Inventory-2 (CSAI-2; Martens et al., 1990) and Competitive Aggressiveness and Anger Scale (CAAS; Maxwell and Moores, 2007). The SEQ was developed and validated considering several theories. ...
Full-text available
Background: Despite the widespread use of the sports emotion questionnaire (SEQ) in several studies, it is surprising that only a few have explicitly tested the validity and utility of the instrument in non-western populations. Besides, the issue of dimensionality and the latent structure of the instrument remain inconclusive given that several authors have revealed different factor structures across diverse populations. The central concern is whether the items on the various dimensions, proposed for the original SEQ, offer adequate information to their respective expected subscale or otherwise. This study assessed the underlying latent structure of the SEQ using confirmatory and bifactor multidimensional item response (MIRT) models. Methods: Through a well-designed validation study 300 athletes from three West African countries, participating in the 2018 West African University Games were surveyed to respond to the SEQ. The data were analyzed using first, a 5-factor confirmatory factor analysis (CFA) via the MIRT model and second, a bifactor MIRT analysis. Results: The results revealed that items on the SEQ were fairly good in measuring the construct under the respective domains of the instrument. However, the outcome of the bifactor model showed that the majority of the items on the SEQ explained common variance in relation to the general factor other than the specific domains (5-dimensions). Conclusion: Findings of the bifactor model question whether the sub-dimensions of the SEQ are needed since most of the items on the SEQ explained larger variances in the general factor than any of the five domains. It is concluded that instruments like SEQ should be scored for a general factor and not as sub-dimensions. Further investigations are encouraged by scholars within the area to probe the dimensionality of the SEQ.
... Les résultats d'études dans le domaine du sport montrent une relation en U inversée entre la performance et l'anxiété somatique, un lien linéaire négatif entre la performance et l'anxiété cognitive et un lien positif entre la performance et la confiance en soi, confirmant les hypothèses à la base, des théories sur l'anxiété multidimensionnelle (Burton, 1988;Hardy, 1996). Cependant, d'autres études ont tenté de répliquer les résultats précédents et les liens attendus n'ont pas toujours été confirmés, surtout le lien entre l'anxiété somatique et la performance (Bird et Horn, 1990;Caruso et al., 1990;Gould et al., 1987;Krane, 1990;Smith et al., 1990). Même si les résultats ne sont pas toujours constants, cette partie de la littérature nous rappelle l'importance de considérer que le sens des relations avec la performance pourrait changer selon ses composantes somatique, cognitive et de la confiance en soi, et que les relations ne sont peut-être pas toujours linéaires. ...
... all of whom completed a four-week programme. A staggering number of 13 measures were employed in this, including such recognised scales as the original SAS(Smith, Smoll, & Schutz, 1990), the MPS and the DFS-2. The results of the MSPE programme seemed to offer promise with regards to increasing flow, mindfulness and certain aspects of confidence. ...
Full-text available
This thesis aimed, firstly, to test the psychological effects of two psychological interventions designed to enhance the psychological strength of competitive athletes; and, secondly, to possibly determine which intervention might be considered superior to the other. The third aim of this thesis was to explore the results of the findings in terms of both their shared and differential theories. The two psychological approaches in question were Rational-Emotive Behavioural Therapy (REBT) – supplemented with adjunctive Mental Skills Training (MST) techniques and the Mindfulness-Acceptance-Commitment (MAC) approach. To test the psychological strengthening effects of both interventions and to compare the results, a non-equivalent pre-test post-test control group design was followed which utilised a control group along with purposive sampling. Various dependent variables were investigated using seven instruments. Variables tested included: - mental wellbeing, irrational beliefs, mental toughness, achievement motivation and competitive anxiety, as well as anger and frustration levels and subjective performance. The instruments used to capture these variables were: - The Mental Health Continuum-Short Form (MHC-SF); Shortened General Attitudes and Beliefs Scale (SGABS); the Sport Mental Toughness Questionnaire (SMTQ); the Sport Competition Anxiety Test (SCAT); as well as one five-point Likert question written by the researcher and one question from the Psychological Skills Inventory (PSI). Seventy-one male adolescent rugby players, from three different schools were sourced to take part in this study. Thus 32 (Mage = 17.33; SD = 6= .729) participants from one high school completed the seven-session REBT-Based MST intervention. 21 (Mage = 17.23; SD = .669) from another school completed the seven-module MAC intervention and 18 (Mage = 15.52; SD = .326) from an additional school were used as a control group school. All three group participants filled out the instrument questionnaires prior to undergoing the various interventions and just following it. Each intervention’s sessions were about an hour to an hour and fifteen minutes long. The control group just filled out the instrument questionnaires in about the same time frame that the other schools had their pre- and post-intervention assessments. The results demonstrated that generally both interventions improved the psychological strength of their players more so than the control group, however, the REBT intervention decreased irrationality and improved subjective performance measures more significantly when compared to the other treatment group.
... Cognitive is a range in a person's brain that exhibits exercise symptoms and anxiety disorders that affect competition activities, such as inability to concentrate, thinking about irrelevant things, and distracting concentration by negative thoughts (Smith et al. 1990) In sports and games, psychological and physical factors play an important role in determining the level of performance. Many players who perform well during training or practice may suffer from performance anxiety on the day of the game. ...
... With the Competition Anxiety Inventory (WAI-T; Brand et al., 2009; see also Sport Anxiety Scale, Smith et al., 1990) we assessed athletes' sport-specific competition anxiety. The WAI-T consists of three subscales with four items each: Worry (e.g., "I am concerned about choking under pressure. ...
Full-text available
In this article, we conducted the first meaningful study of irrational beliefs (IBs) in a German sample of athletes. Moreover, we investigated associations between IBs and potential general as well as sport-specific markers of mental health in German athletes. As general markers, we considered psychological distress and wellbeing in addition to IBs, and as sport-specific markers, we considered anxiety, perfectionism, and athletic identity. To achieve this, our first step was to translate and validate a specific measure of irrational beliefs, namely the Irrational Performance Beliefs Inventory (iPBI). The iPBI is a performance-relevant measure that captures specific IB, taking into account the situational circumstances of the target population, namely operators in different performance context (academia, sports, business, medicine, etc.). Its theoretical basis is largely Ellis’ work on rational and irrational beliefs. We developed a short and a long version of the iPBI, which both capture four core IBs (i.e., demandingness, awfulization, low frustration tolerance, and depreciation). Factorial validity was confirmed by a confirmatory factor analysis (comparative fit index = 0.92) with data from 234 athletes. Both versions of the newly developed iPBI showed good internal consistency (Cronbach’s alpha > 0.77) and retest reliability (intra-class correlation coefficients >0.71). Results of the correlational analyses indicated low positive relationships between IBs and athletes’ psychological distress, and low negative relations between IBs and wellbeing. In terms of sport-specific markers, there were low to moderate correlations with IBs. This study also examined the differences in IBs between females and males, individual and team sport athletes, and across three different performance levels. Implications of these findings are discussed along with approaches for future research and applied work.
Full-text available
This study aims to assess the validity and reliability of the Polish version of the Sport Anxiety Scale-2, as well as to determine the relationship between anxiety and goal orientation among high-performance and recreational athletes. A total of 519 athletes aged M = 22.83 (SD = 4.92) participated in the study, including 266 males and 253 females. 242 athletes trained professionally and 277 recreationally. The Sport Anxiety Scale-2 (SAS-2) was used to assess anxiety levels, while the Task and Ego Orientation in Sport Questionnaire (TEOSQ) and the Perception of Success Questionnaire (POSQ) enabled to assess athletes’ goal orientation. Confirmatory factor analysis showed a good fit of the model to the data for the Polish version of the Sport Anxiety Scale-2 (CFI = 0.945, RMSEA = 0.072). The models obtained during analysis of high-performance and recreational athletes, women and men, also presented a satisfactory fit to the data (CFI 0.932–0.946). The configural, metric, scalar and strict measurement invariances were demonstrated for high-performance and recreational athletes as well as among women and men. High internal consistency coefficients (alpha 0.81–0.91) and a high test–retest reliability indexes were reported (ICC 0.74–0.87). Women presented higher level of competitive anxiety than men. A positive relationship between competitive anxiety and athletes’ ego orientation was also presented. This relationship concerned particularly women practicing sport recreationally.
Full-text available
The purpose of the present study was to provide further evidence for the factor structure and composition of the Sport Anxiety Scale (SAS; Smith, Smoll, & Schutz, 1990) using a sample of competitive male rugby players (N = 570). Three models were tested using both confirmatory factor analytic and polytomous item-response theory procedures: Smith et al's original model; Dunn et al.'s (2000) alternative model in which Items 14 and 20 were originally designed to measure Concentration Disruption load on the Worry factor (Model A); and Model B (the removal of Item 1). Results showed that Models A and B provided similar fits to the data. Overall these findings argue for the utilization of Model B to improve model fit and maintain conceptual clarity. Our findings suggest that the factor structure and composition of the SAS needs further examination and possible refinement before researchers can feel more confident about the effectiveness of the instrument's psychometric properties.
Full-text available
The purposes of the present investigation were to examine the coping responses of different subgroups of athletes (e.g., high and low trait anxious athletes), and to assess the consistency of athlete's coping behaviors across situations. Two-hundred and seventy-three athletes completed the Sport Anxiety Scale (SAS) by Smith, Smoll, & Schutz (1990) and coping assessments in trait and state versions of the sport adapted COPE (MCOPE) by Crocker and Graham (1995). The state coping measures assessed coping responses of situations for which the athletes actually experienced. The results of three separate, doubly multivariate, repeated measures, MANOVA's showed that high trait anxious athletes responded to stressful situations using different coping behaviors (e.g., denial, wishful thinking, and self-blame) than the low trait anxious athletes. In addition, coping appears to be more stable than situationally variable as Pearson correlational coefficients computed between the three measures ranged from 0.53 to 0.80. The results are discussed with regard to theoretical, research, and applied issues.
For the past five years we have been engaged in a program of research whose ultimate goal has been the development and evaluation of therapeutic methods for reducing anxiety. A basic assumption underlying our work has been that the successful evolution of such strategies will be facilitated by advances in our knowledge about the nature of anxiety itself. Consequently the majority of the research has attempted to identify basic conditions (environmental and subject) that serve to maintain or reduce the anxiety response.
Little League Baseball coaches were exposed to a preseason training program designed to assist them in relating more effectively to children. Empirically derived behavioral guidelines were presented and modeled, and behavioral feedback and self-monitoring were used to enhance self-awareness and to encourage compliance with the guidelines. Trained coaches differed from controls in both overt and player-perceived behaviors in a manner consistent with the behavioral guidelines. They were also evaluated more positively by their players, and a higher level of intrateam attraction was found on their teams despite the fact that they did not differ from controls in won-lost records. Children who played for the trained coaches exhibited a significant increase in general self-esteem compared with scores obtained a year earlier; control group children did not. The greatest differences in attitudes toward trained and control coaches were found among children low in self-esteem, and such children appeared most sensitive to variations in coaches' use of encouragement, punishment, and technical instruction.