645060

research-article2016
SAP0010.1177/0081246316645060South African Journal of PsychologyMabweazara et al.

South African Journal of Psychology 1­–11 © The Author(s) 2016

Reprints and permissions: sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0081246316645060 sap.sagepub.com

Article
Predicting swimming performance
using state anxiety

Smart Z Mabweazara, Lloyd Leach

and Barry S Andrews

Abstract
Competitive state anxiety is a common response to stressful competitive sports situations that
could affect athletic performance. The effects of state anxiety on swimming performance need
further inquiry. The aim of the study was to determine the component of state anxiety that best
predicts swimming performance. A quantitative, cross-sectional study design that made use of the
Competitive State Anxiety Inventory-2 to measure precompetitive state anxiety was used. A total
of 61 male high school swimmers whose age ranged between 14 and 19 years (M = 16.16, standard
deviation = 1.66 years) completed the Competitive State Anxiety Inventory-2 1 hr before competing
in a 50-m individual swimming event. Performance was evaluated using finishing position. Due to
the relatively short duration of the 50-m event, the available literature would suggest that Somatic
Anxiety would have a greater effect on Performance—there is not enough time to allow cognitive
anxiety to have a detrimental impact on performance. Thus, it was hypothesized that somatic
rather than cognitive anxiety will best predict swimming performance. It emerged that both
cognitive (b = .787; p < .001) and somatic anxieties (b = .840; p < .001) can independently predict
swimming performance. However, when both cognitive and somatic anxieties were regressed
onto swimming performance, somatic anxiety partially dominated cognitive anxiety (b = .626;
p < .001) and became the significant predictor of swimming performance. It is recommended that
swimmers and swimming coaches make use of specific intervention strategies that eradicate the
detrimental effects of somatic anxiety immediately before competition.

Keywords
Competitive State Anxiety Inventory-2, performance, precompetitive state anxiety, stress,
swimming

Anxiety and arousal are two related psychological constructs that have been shown to affect ath-
letic performance. Arousal refers to a blend of physiological processes in a person that are related
to the intensity dimensions of motivations at a particular moment (Weinberg & Gould, 2011). On

University of the Western Cape, South Africa

Corresponding author:
Barry S Andrews, University of the Western Cape, Private Bag X17, Bellville, Cape Town 7335, South Africa.
Email: bandrews@uwc.ac.za

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

2 South African Journal of Psychology 

the other hand, precompetitive state anxiety may be regarded as anxiety prior to the onset of com-
petition and is reflective of the objective and perceived requirements of the individual or team
participating in the competition (Raglin & Morris, 1994). Novice athletes seem to experience a
steady increase in anxiety in the immediate period leading up to and during competition, while
experienced athletes may have increased anxiety during precompetition periods further away from
competition and a decrease in the immediate period leading up to and during competition (Bridges
& Knight, 2005). Precompetitive anxiety is usually caused by the psychological pressure induced
by both the presence and expectations of significant others such as coaches and parents. Weinberg
and Gould (2011) have stated that the incapability to cope with the pressure in competitive sport
can lead to poor performance and at times even deteriorate to physical illness and mental distress.
Martens, Vealey, and Burton (1990b) provided a multidimensional approach to the inquiry of
anxiety. Their multidimensional theory proposes that anxiety is sub-composed of cognitive and
somatic anxiety. They defined cognitive anxiety as “the mental component of anxiety” (Martens
et al., 1990b, p. 6). Cognitive anxiety was said to have a negative linear relationship with perfor-
mance. Somatic anxiety was defined as “the physiological and affective elements of the anxiety
experience that develop directly from autonomic arousal” (Martens et al., 1990b, p. 6). The authors
proposed that somatic anxiety affects performance in a curvilinear fashion, with both lower and
higher levels of somatic anxiety being debilitative to performance. It was also posited that somatic
anxiety reaches its maximum levels at the beginning of competition and begins to fade once the
contest starts. As such, somatic anxiety was said to have a minimal impact on performance com-
pared to cognitive anxiety (Martens, Burton, Vealey, Bump, & Smith, 1990a). However, this
depends on the duration of the event. Specifically for this study, due to the short time taken to
complete the 50-m swim, the impact of somatic anxiety might be greater than that of cognitive
anxiety. For example, Burton (1988) found that short duration events demonstrate stronger rela-
tionships between somatic anxiety and performance than longer events. To further differentiate the
two constructs, cognitive anxiety, which affects the mental element, is expressed during self-eval-
uation and self-talk, while somatic anxiety, which is the physiological element, is expressed as
autonomic arousal (Parnabas, Mahamood, & Parnabas, 2013).
The relationship between cognitive and somatic state anxiety as predictors of performance in
different sports is a function of the specific motoric and physiological requirements of the athletic
activity (Taylor, 2006). Thus, sports that involve strength, such as weightlifting, require elevated
levels of state anxiety and physiological arousal, whereas sports that are more skill-oriented, such
as archery, require low arousal and anxiety levels for peak performance (Taylor, 2006). As such,
swimming strokes that are more strenuous will require elevated arousal and anxiety levels, whereas
strokes that are less demanding in terms of strength will require less arousal and anxiety levels. In
other activities, such as performance in music, the results have been somewhat different. Yoshie,
Shigemasu, Kudo, and Ohtsuki (2009) investigated the effects of state anxiety on music perfor-
mance. The results showed that the magnitude of cognitive anxiety negatively predicted technical
accuracy, while cognitive anxiety positively predicted global performance. The authors concluded
that reducing pre-performance cognitive anxiety and understanding the symptoms of cognitive
anxiety as being facilitative to performance could enhance performance. This finding, although it
relates to music performance, may also apply to swimming performance; in other words, swim-
mers must be able to minimize pre-performance cognitive anxiety in order to enhance their swim-
ming performance. Taylor (2006) further notes that extreme levels of state anxiety impede motor
coordination and the effective use of the cardiovascular system.
Currently, the literature is equivocal about the effect of anxiety on performance (Polman,
Rowcliffe, Borkoles, & Levy, 2007). Some studies (Burton, 1988; Jones et al., 1994) have con-
cluded that state anxiety has a debilitating effect on swimming performance, whereas other studies

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

Mabweazara et al. 3

(Furst & Tenebaum, 1986) have found state anxiety to be facilitative of swimming performance.
Polman et al. (2007) concluded that state anxiety is both debilitative and facilitative to swimming
performance. Nordell and Sime (1993) have also reported that other studies have found no associa-
tion between swimming performance and state anxiety. Lundqvist, Kentta, and Raglin (2011)
examined the differences in anxiety ratings of elite swimmers on the direction modified Competitive
State Anxiety Inventory-2R. They showed that elite athletes rated more items on the scale as facili-
tative to performance, whereas sub-elite athletes rated more items on the scale as debilitative.
Females are reported to display significantly elevated levels of predicted precompetition anxiety
than males (Wilson, Raglin, & Pritchard, 2002). However, the authors point out that gender is a poor
predictor of precompetition anxiety. Although this is the case, the significance of Wilson et al.’s
(2002) findings is that state anxiety research about one gender should not overlap into that of the other
gender. As such, research about the state anxiety should focus on one gender as is the case in this
study. Wilson et al. (2002) also reported that optimistic college athletes exhibit significantly lower
levels of precompetition anxiety compared to pessimists. A sound explanation of this finding could
be that optimistic athletes are more self-confident compared to pessimistic athletes. As a result,
numerous sport psychology interventions for performance enhancement typically include techniques
designed to enhance self-confidence (Weinberg & Gould, 2011) and reduce anxiety (LeUness &
Nation, 2002). The predictability of swimming performance from state anxiety measures remains
unclear. Furthermore, it is not clear which type of anxiety best predicts swimming performance.
Much more recently, using the same participants and research instruments as in this study,
Mabweazara, Andrews, and Leach (2014) assessed the temporal changes in state anxiety among
high school swimmers prior to competition. The Competitive State Anxiety Inventory-2 (CSAI-2)
was used to measure state anxiety 7 days (baseline measures) before a major competition and 1 hr
(precompetition measures) before the competition, under the supervision of a registered psycholo-
gist. Significant differences between baseline and precompetitive cognitive state anxieties
(p = .000) and baseline and precompetitive somatic anxieties (p = .000) were reported. The authors
concluded that state anxiety significantly increases prior to competition among high school swim-
mers. The study clearly showed the importance of assessing state anxiety among swimmers, espe-
cially before a major competition by making use of a real competitive situation as compared to
creating a pseudo-competitive environment, which might not be perceived as competitive by the
young swimmer. It is in line with this, that this study assesses the predictability of performance
using state anxiety among swimmers within the same competitive context. Coaches and swimmers
should thus make use of specific psychological intervention strategies in order to be able to regu-
late state anxiety and change potentially debilitative anxiety to facilitative state anxiety. Given the
foregoing, this study extends and builds on the work of Mabweazara et al.’s (2014) study and shifts
focus from assessing the temporal changes in state anxiety prior to competition in determining the
component of state anxiety that best predicts swimming performance using precompetitive state
anxiety measures. It was hypothesized that somatic rather than cognitive state anxiety significantly
predicts swimming performance.

Method
The study utilized a quantitative, descriptive design.

Participants
Male high school swimmers comprised the study population. Purposive sampling was used to
identify 61 male swimmers from three private High Schools in Zimbabwe. The following inclusion

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

4 South African Journal of Psychology 

criteria were used: swimmers registered as High School learners at selected private schools in
Zimbabwe; swimmers whose ages ranged from 15 to 19 years; swimmers who competed in the
Falcon College Schools Swimming Gala (16 October 2012); swimmers who competed in the indi-
vidual 50 m front crawl, back stroke, breast stroke, or butterfly swimming events; swimmers who
competed at provincial level; swimmers with at least 4 years’ experience in competitive swimming.
One school was selected from the Matabeleland South Province, while the other two schools were
selected from Bulawayo Province. Private schools were specifically chosen for this study because
in Zimbabwe, specifically in the Province where the study was undertaken, most public schools
cannot afford the luxury of a swimming pool because installing and maintaining one is expensive.
As such, it is mostly private schools that can afford having a swimming pool and hiring coaching
staff. The three High Schools included in this study were chosen on the basis that they are the most
competitive.

Instruments
Precompetitive state anxiety was measured using the CSAI-2, a questionnaire developed by
Martens et al. (1990b). The CSAI-2 is a 27-item self-report measure of anxiety with three 9-item
subscales that measure cognitive state anxiety, somatic state anxiety, and state self-confidence
(Duda, 1998). In this study, the state self-confidence subscale of the CSAI-2 was excluded. The
rationale for using the CSAI-2 was that the instrument has been used in a landmark study by Burton
(1988), which utilized elite swimmers to show the reliability of the CSAI-2 in terms of stability and
consistency. The CSAI-2 was administered 1 hr prior to competition. Burton’s (1998) landmark
study also provided confirmation for the theoretical underpinnings of the multidimensional theory.
As in Burton’s (1988) study, this study utilized the CSAI-2 to measure state anxiety just prior to
competition. Initial psychometric studies on the CSAI-2 inventory have demonstrated an internal
consistency for each of the three subscales (Martens et al., 1990b). Separate alpha coefficients for
the cognitive, somatic, and self-confidence subscales ranged from .79 to .90 thereby revealing a
high degree of internal consistency (Martens et al., 1990b). Other studies employing the CSAI-2 to
measure anxiety in sport have reported alpha reliability coefficients that range from .76 to .91,
thereby confirming its internal consistency (Duda, 1998, p. 136). Similarly, in an African setting,
Adeniyi and Samson (2004) have established the reliability of this multidimensional measure of
state anxiety with Nigerian athletes and have documented an r of .80. The reliability coefficients
for the scales in this study were satisfactory (Cronbach alpha scores of .68 and .80) for the cogni-
tive and somatic anxiety scales, respectively.
In this research, any scores that were less than or equal to 12 points were considered to be low
for state anxiety, whereas scores ranging from 13 to 24 points were considered to be medium and
scores ranging from 25 to 36 points were considered to be high anxiety scores. This was done for
convenience and in line with Martens et al.’s (1990a) instructions that on the CSAI-2 inventory,
state anxiety scores range from a minimum of 9 to a maximum of 36.

Procedure
The swimmers were assessed at an Inter-Schools Swimming Gala (2012) in Zimbabwe. With the
assistance of the swimming coaches from each of the participating schools, a premier inter-schools
swimming competition was identified. This swimming competition was assumed to bring out the
competitive character of the swimmers. Such an environment would likely lead to the swimmers
experiencing precompetitive state anxiety (Gill, 1986). A brief explanation of the study was then
provided by the investigator to the school authorities, coaches, and participants, and participants’

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

Mabweazara et al. 5

parents. It was important in this study to collect anxiety measures twice in order to observe whether
there was a change in the state anxiety of swimmers from a non-competitive context to a competi-
tive context. This was done so that the precompetitive state anxiety levels observed an hour before
the competition could be attributed to the competitive nature of the environment. Therefore, the
precompetitive state anxiety of the swimmers was first assessed 7 days prior to the competition in
a non-competitive context during training at their respective schools using the CSAI-2. This assess-
ment was used as a baseline for comparison purposes. Each swimmer then completed the CSAI-2
inventory on the day of competition, an hour before competing in an individual 50 m swimming
event, that is, the front crawl, butterfly, breast, and/or back stroke. On each occasion, before the
swimmers completed the questionnaire, the CSAI-2 instructions were read aloud to them. The final
results of the swimmers’ performances were e-mailed by the head coach of the host school to the
investigator 3 days after the competition.
Race finishing position was used to evaluate performance. The finishing positions in the swim-
ming competitions were categorized by convenience in order to facilitate data comparisons.
Therefore, Positions 1 and 2 were categorized as being above average performance, with Positions
3 and 4 as average performance and Positions 5 and 6 as below average performance. The swim-
mers were then categorized as being of above average, average, or below average performance.
This was done for convenience.

Ethical considerations
Informed consent for participation in the study was obtained from the participants. Consent was
also obtained from the parents of the participants. Participation in the study was voluntary and the
participants could withdraw from the study at any point without coercion or any negative conse-
quences. Permission to conduct the study and ethical approval for the study was obtained from the
Ethics Committee of the University of the Western Cape, the Zimbabwean Ministry of Education,
Sport, Arts and Culture, and the three private schools selected in the Bulawayo and Matabeleland
South Provinces (ethics clearance number 12-7-9).

Data analysis
Descriptive and inferential statistics were used on SPSS (version 21) to analyze the data. Means
and standard deviations were calculated for precompetitive state anxiety intensity levels, swim-
ming performance, and the ages of the participants. Spearman correlations were computed to test
for significant associations between precompetitive state anxiety and swimming performance.
Hierarchical regression analyses were performed to show the strength of association of each of the
subcomponents of state anxiety with swimming performance and to show the component of
precompetitive state anxiety that best predicts swimming performance. Statistical significance was
set at p < .05. Data are presented using tables.

Results
The participants’ (n = 61) ages ranged from 14 to 19 years (mean = 16.16, standard deviation
[SD] = 1.66 years). Performance placement fell into six categories, that is, from positions 1 to 6. In
all, 10 participants (16.39%) finished in first position, 9 (14.75%) in the second position, 12
(19.67%) in the third position, 10 (16.39%) in the fourth position, 11 (18.03%) in the fifth position,
and 9 (14.75%) in the sixth position. Table 1 shows the performance placement in terms of finish-
ing position of the participants in frequencies and percentages.

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

6 South African Journal of Psychology 

Table 1. Performance placement of the participants expressed as frequencies and percentages.

Placement Frequency Percentage

1 10 16.39
2 9 14.75
3 12 19.67
4 10 16.39
5 11 18.03
6 9 14.75
Total 61 100

Table 2. Precompetitive cognitive and somatic state anxiety of swimmers based on finishing position.

Ranking N Cognitive anxiety Somatic anxiety

Finishing position (mean ± SD) (mean ± SD)
Above average 19 13.05 ± 2.39 18.05 ± 1.99
(Positions 1 and 2)
Average 22 24.77 ± 2.31 24.32 ± 2.00
(Positions 3 and 4)
Below average 20 27.10 ± 4.05 28.15 ± 2.85
(Positions 5 and 6)

SD: standard deviation.

Table 3. State anxiety scores at precompetition.

Period High Medium Low Total

n % n % n % n %
Precompetitive cognitive anxiety 27 44.26 26 42.62 8 13.11 61 100
Precompetitive somatic anxiety 31 50.82 30 49.18 0 0.00 61 100

Table 2 shows the precompetitive cognitive and somatic state anxiety mean scores for the three
performance groups, namely, swimmers with above average (Positions 1 and 2), average (Positions
3 and 4), and below average (Positions 5 and 6) finishing performance. The mean ± SD scores for
the above average group for cognitive and somatic anxiety were 13.05 ± 2.39 and 18.05 ± 1.99,
respectively. The mean ± SD scores for the average group for cognitive and somatic anxiety were
24.77 ± 2.31 and 24.32 ± 2.00, respectively. The mean ± SD scores for the below average group for
cognitive and somatic anxiety were 27.10 ± 4.05 and 28.15 ± 2.85, respectively.
Table 3 shows the state anxiety scores at precompetition (i.e., 1 hr before competition) catego-
rized into high, medium, and low frequencies and percentages. In total, 27 participants (44.26%)
experienced high cognitive state anxiety, 26 (42.62%) experienced medium, and 8 (13.11%) par-
ticipants experienced low cognitive state anxiety. In all, 31 participants (50.82%) experienced high
somatic state anxiety, 30 (49.18%) experienced medium, and none of the participants scored low
on the somatic state anxiety scale 1 hr before competition.
Table 4 shows the correlation matrix for performance (finishing placement) and precompetitive
state anxiety. Specifically, there was a significant correlation (p < .001) between precompetitive

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

Mabweazara et al. 7

Table 4. Correlation matrix for placement and precompetitive state anxiety.

Variables Variables

1 2 3 4
1. Age –
2. Cognitive anxiety .181 –
3. Somatic anxiety .058 .825** –
4. Placement .028 .787** .873** –

**p < .001.

Table 5. Model summary of the hierarchical regression in five steps.

Step Criterion Outcome b Partial Excluded R2 F

1 Cognitive anxiety Somatic anxiety .850** .721 152.768**
2 Somatic anxiety Cognitive anxiety .850** .721 152.768**
3 Somatic anxiety Placement .840** .710 144.434**
4 Cognitive anxiety Placement .787** .619 95.896**
5 Cognitive anxiety Placement .255 .250 .728 77.679**
Somatic anxiety Placement .626** .535** Cognitive anxiety

b (standardized regression coefficients) is a measure of how strongly each predictor (outcome) variable influences the
criterion variable.
**p < .001.

cognitive state anxiety and performance (r = .787). The results also indicate a significant correla-
tion (p < .001) between precompetitive somatic state anxiety and swimming performance (r = .873).
The results also indicate a significant correlation (p < .001) between precompetitive cognitive state
anxiety and somatic state anxiety (r = .825). Performance or finishing position was coded so that
faster finishing order denoted lower rankings. Both correlations were large and positive, that is, as
anxiety increased, performance or finishing position increased. In other words, the positive corre-
lation coefficients actually indicates an inverse relationship between state anxiety and performance.
There was a significant negative relationship between precompetitive state anxiety and perfor-
mance, that is, as anxiety increased, performance decreased.
Hierarchical regression analysis was performed in order to establish the component of state
anxiety that best predicts swimming performance. The hierarchical regression analysis results are
shown in Table 5.
When somatic state anxiety was regressed onto cognitive state anxiety and vice versa, identical
results were found. Both models were significant (p < .001) and explained 72% (R2) of the variance
in cognitive state anxiety.
When somatic anxiety was regressed onto placement, the result was significant (p < .001) and
explained 71% (R2) of the variance in finishing position. Somatic state anxiety was a significant
predictor of placement (p < .001), that is, for every 1-unit change in somatic state anxiety, there was
a corresponding change of 0.84 units in finishing position.
When cognitive state anxiety was regressed onto placement, the result was significant (p < .001)
and explained 62% (R2) of the variance on finishing position. Cognitive state anxiety was a signifi-
cant predictor of placement (p < .001), that is, for every 1-unit change in cognitive state anxiety,
there was a corresponding change of 0.78 units in finishing position.

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

8 South African Journal of Psychology 

When both cognitive state anxiety and somatic state anxiety were regressed onto placement, the
result was significant (p < .001) and explained 73% (R2) of the variance on finishing position. In
this final step, cognitive state anxiety did not significantly predict performance. Consequently,
somatic state anxiety became the dominant predictor of performance.
Three important observations emerge from this analysis (see Table 5):

1. Somatic state anxiety and cognitive state anxiety are in a linear relationship and can predict
each other significantly (Steps 1 and 2).
2. Somatic state anxiety and cognitive state anxiety both independently predict placement
significantly (Steps 3 and 4).
3. Somatic state anxiety dominates cognitive state anxiety in the mixed model (Step 5).

Thus, in the final step of the analysis, cognitive state anxiety was excluded from the hierarchical
regression in which case somatic state anxiety became the significant predictor of performance. It
therefore emerged that somatic state anxiety is a manifestation of cognitive state anxiety or acts in
place for cognitive state anxiety (proxy relationship).

Discussion
In this study, it was hypothesized that cognitive state anxiety rather than somatic state anxiety
would be the best predictor of swimming performance. When somatic state anxiety was regressed
onto performance, it showed to be a significant predictor of performance. Cognitive anxiety was
also a significant predictor of performance; however, in the last step of the hierarchical regression,
when both cognitive and somatic state anxieties were regressed onto performance, somatic state
anxiety became the only significant predictor of swimming performance. It was thus concluded
that somatic state anxiety acts in place of cognitive anxiety (a proxy relationship) or is manifesta-
tion of cognitive state anxiety. This is reflected in the results showing that both cognitive and
somatic state anxieties significantly predict each other.
Studies conducted by Edwards and Hardy (1996) and Jerome and Williams (2000) have also
found that competitive anxiety intensity is a significant predictor of performance variance.
However, Martens et al. (1990b) hypothesized that cognitive state anxiety would be a stronger
predictor of performance than somatic state anxiety. However, it is not clear whether these conclu-
sions about somatic anxiety and cognitive anxiety relationship with performance, were made with
the duration of events taken into consideration. The findings of this study are contradictory to this
assertion. Several other research findings, which contradict the present findings, are in agreement
with the prediction that cognitive anxiety is a stronger predictor of performance than somatic anxi-
ety (Lane, Terry, & Karageorghis, 1995a, 1995b). It is, however, worth noting that the study by
Lane et al. (1995b) employed male triathlete participants, and as such, their findings might have
been influenced by the long duration of the triathlon event.
In a study utilizing intra-individual measures, which was rooted in multidimensional theory,
Burton (1988) predicted performance to decline linearly with surges in cognitive state anxiety. In
general, research concurs with the assertion that cognitive anxiety is linked to factors in the envi-
ronment which impact perceptions of success or failure (Gould, Petlichkoff, & Weinberg, 1984;
Hanton & Jones, 1995; Jones, Swain, & Cale, 1990; Lane et al., 1995a, 1995b). A number of expla-
nations have been given for these contradictions, including “the use of poor performance measures,
failure to control for individual differences in ability, response sensitivity, anxiety tolerance, and
failure to consider subjects’ interpretations of the meaning of their anxiety symptoms” (Hardy,
Jones, & Gould, 1996, p. 149).

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

Mabweazara et al. 9

Craft, Magyar, Becker, and Feltz (2003) specifically observe that the use of the CSAI-2 has
not managed to yield stable results in terms of predicting performance. This study used an objec-
tive outcome (finishing position) as a performance measure. However, it has been noted that the
most dominant methodological limitations in examining the anxiety–performance relationship
have been complications with the working definitions of performance (Krane, 1992). Performance
may be evaluated in terms of objective results, such as finishing position, or subjective out-
comes, such as evaluation by a coach, or even based on self-perceptions of performance (Craft
et al., 2003). These different methods of evaluating performance may yield different results
(Craft et al., 2003).
This study attempted to determine the strongest predictor of swimming performance in the
50-m events. Martens et al. (1990a) probed the impact of task characteristics on the relationship
between cognitive and somatic anxiety and swimming performance. For the 50-m freestyle
event, the results of Martens et al.’s (1990a) study were similar to those of this study. Somatic
state anxiety was significantly related to performance. Martens et al. (1990a) recommend that
somatic state anxiety is a central mediator of sprint freestyle performance. As in this study,
somatic state anxiety played a significant role in predicting performance for short or sprint
events in swimming.
In this study, the results suggested that somatic anxiety is merely a manifestation of cognitive
state anxiety or that it acts as a proxy for cognitive state anxiety. Although these two components
of anxiety are theoretically autonomous, Morris, Davies, and Hutchings (1981) have recommended
that they co-vary in demanding situations such as competition since these situations contain ele-
ments related to the manifestation of both somatic and cognitive reactions. A weak reliance of
cognitive anxiety on somatic anxiety has been shown in a number of studies (Deffenbacher, 1980;
Morris & Liebert, 1973). Borkovec (1976) suggests that each constituent of anxiety may serve a
provisional or discriminative function for the other constituent.
If strong somatic reactions (e.g., heavy sweating) by an athlete are linked to a certain provoca-
tion or stimulus present in the competition arena (e.g., a highly competitive sport setting), these
reactions may indicate to a person that there is reason to worry (Martens et al., 1990a). A swimmer,
for example, may acquire conditioned somatic anxiety reactions, because of precompetitive activi-
ties, such as the precompetition team talk or the mere presence of noisy spectators. These somatic
anxiety reactions may then prompt the swimmer to start worrying. In other words, the somatic
anxiety reactions incite cognitive anxiety. This may also happen the other way round where cogni-
tions, manifested through undesirable pictures of failure, may generate a specific form of somatic
response. In this study, somatic state anxiety became a proxy for cognitive state anxiety, probably
due to the presence of a highly evaluative crowd that is made up of coaches, provincial team scouts,
parents, spectators, and teachers.
Knowing that anxiety will have an effect on performance is widely accepted in sport, but know-
ing that somatic anxiety (the physiological arousal levels) has the greatest impact on performance
can affect the coaching of swimmers. Coaches can plan and instruct their swimmers on ways to
cope with somatic anxiety, and also to be on the lookout for it so as to negate its effects on the
athlete’s performance, thereby assisting their swimmers to be more successful.
Transferability of the findings of this study is limited because of the non-randomized recruit-
ment of subjects. Also, this study did not assess whether the swimmers interpreted the experienced
state anxiety as either facilitative or debilitative to swimming; rather it focused on the influence of
anxiety intensity on swimming performance. It has been recommended that researchers consider
examining the direction dimension of state anxiety on sport performance (Kais & Raudsepp, 2005).
This study quantified performance as a function of finishing positions. Researchers are encouraged
to use world age group swimming record for specific events.

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

10 South African Journal of Psychology 

Conclusion
This study reveals that both cognitive and somatic anxiety can independently predict swimming
performance. However, somatic state anxiety was the stronger predictor of performance. We thus
conclude that somatic anxiety has the greatest impact on sprint performance (shorter swimming
events). It is thus recommended that coaches and swimmers use specific intervention strategies
that eradicate the detrimental effects of somatic anxiety on swimming performance in order to
optimize performance.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.

References
Adeniyi, A. M., & Samson, A. B. (2004). Predictors of pre-game anxiety dysphoria among teenage soccer
players. Nigerian Journal of Emotional Psychology, 6, 15–18.
Borkovec, T. D. (1976). Physiological and cognitive process in the regulation of anxiety. In G. E. Schwartz &
D. Shapiro (Eds.), Consciousness and self-regulation (Advances in Research and Theory, pp. 216–312).
New York, NY: Plenum.
Bridges, A., & Knight, B. (2005). The role of cognitive and somatic anxiety in athletic performance. Retrieved
from http://psych.hanover.edu/research/Thesis05/BridgesKnight.pdf
Burton, D. (1988). Do anxious swimmers swim slower? Re-examining the elusive anxiety performance rela-
tionship. Journal of Sport and Exercise Psychology, 10, 45–61.
Craft, L. L., Magyar, T. M., Becker, B. J., & Feltz, D. (2003). The relationship between the Competitive State
Anxiety Inventory-2 and sport performance: A meta-analysis. Journal of Sport and Exercise Psychology,
25, 44–65.
Deffenbacher, J. L. (1980). Worry and emotionality in test anxiety. In I. G. Sarason (Ed.), Test anxiety:
Theory, research and applications (pp. 111–124). Hillsdale, NJ: Lawrence Erlbaum.
Duda, J. (1998). Advances in sport and exercise psychology measurement. Morgantown, WV: Fitness
Information Technology.
Edwards, T., & Hardy, L. (1996). The interactive effects of intensity and direction of cognitive and somatic
anxiety and self-confidence upon performance. Journal of Sport and Exercise Psychology, 18, 296–312.
Furst, D. M., & Tenebaum, G. (1986). The relationship between worry, emotionality and sport performance.
In D. M. Landers (Ed.), Sport and elite performers (pp. 89–96). Champaign, IL: Human Kinetics.
Gill, D. L. (1986). Psychological dynamics of sport. Champaign IL: Human Kinetics.
Gould, D., Petlichkoff, L., & Weinberg, R. S. (1984). Antecedents of temporal changes in and relationships
between CSAI-2 subcomponents. Journal of Sport and Exercise Psychology, 6, 289–304.
Hanton, S., & Jones, G. (1995). Antecedents of anxiety and confidence in elite competitive swimmers.
International Journal of Sport Psychology, 26, 512–523.
Hardy, L., Jones, J. G., & Gould, D. (1996). Understanding psychological preparation for sport. Chichester,
UK: John Wiley & Sons.
Jerome, G. J., & Williams, J. M. (2000). Intensity and interpretation of competitive state anxiety: Relationships
to performance and repressive coping. Journal of Applied Sport Psychology, 12, 236–250.
Jones, G., Swain, A., & Cale, A. (1990). Antecedents of multidimensional competitive state anxiety and self-
confidence in elite intercollegiate middle-distance runners. The Sport Psychologist, 4, 107–118.
Jones, G., Hanton, S., & Swain, A.B.J. (1994). Intensity and interpretation of anxiety symptoms in elite and
non-elite sports performers. Personal and individual Differences, 17, 657-663.
Kais, K., & Raudsepp, L. (2005). Intensity and direction of competitive state anxiety, self-confidence, and
athletic performance. Kinesiology, 37, 13–20.
Krane, V. (1992). Conceptual and methodological considerations in sport anxiety research: From the inverted-
U hypothesis to catastrophe theory. Quest, 44, 72–87.

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016

Mabweazara et al. 11

Lane, A. M., Terry, P. C., & Karageorghis, C. I. (1995a). The antecedents of multidimensional state anxiety
and self-confidence in duathletes. Perceptual and Motor Skills, 80, 911–919.
Lane, A. M., Terry, P. C., & Karageorghis, C. I. (1995b). Path analysis examining relationships among ante-
cedents of anxiety, multidimensional state anxiety, and triathlon performance. Perceptual and Motor
Skills, 81, 1255–1266.
LeUness, A., & Nation, J. R. (2002). Sport psychology. Louisville, KY: Wordsworth Group.
Lundqvist, C., Kentta, G., & Raglin, J. S. (2011). Directional anxiety responses in elite and sub-elite young
athletes: Intensity of anxiety symptoms matters. Scandinavian Journal of Medicine and Science in
Sports, 21, 853–862.
Mabweazara, S. Z., Andrews, B. S., & Leach, L. L. (2014). Changes in state anxiety prior to competition.
African Journal for Physical, Health Education, Recreation and Dance, 20, 492–499.
Martens, R., Burton, D., Vealey, R. S., Bump, L. A., & Smith, D. E. (1990a). Development and validation of
the Competitive State Anxiety Inventory-2 (CSAI-2). In R. Martens, R. S. Vealey, & D. Burton (Eds.),
Competitive anxiety in sport (pp. 193–208). Champaign, IL: Human Kinetics.
Martens, R., Vealey, R. S., & Burton, D. (1990b). Competitive anxiety in sport. Champaign, IL: Human
Kinetics.
Morris, L. W., Davies, D., & Hutchings, C. (1981). Cognitive and emotional components of anxiety: Literature
review and revised worry-emotionality scale. Journal of Educational Psychology, 73, 541–555.
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 Counseling Psychology, 20, 321–326.
Nordell, K. A., & Sime, W. (1993). Competitive trait anxiety, state anxiety, and perceptions of anxiety, state
anxiety, and perceptions of anxiety: Interrelationships in practice and in competition. The Journal of
Swimming Research, 9, 19–24.
Parnabas, V. A., Mahamood, Y., & Parnabas, J. (2013). The relationship between cognitive and somatic anxi-
ety on performance of student-athletes of Universiti Malaysia Perlis (UNIMAP). Sport and Art, 1(3),
61–66.
Polman, R., Rowcliffe, N., Borkoles, E., & Levy, A. (2007). Precompetitive state anxiety, objective and
subjective performance and causal attributions in competitive swimmers. Pediatric Exercise Science,
19, 39–50.
Raglin, J. S., & Morris, M. J. (1994). Precompetition anxiety in women volleyball players: A test of ZOF
theory in a team sport. British Journal of Sports Medicine, 28, 47–51.
Taylor, J. (2006). Predicting athletic performance with self-confidence and somatic and cognitive anxiety as
a function of motor and physiological requirements in six sports. Journal of Personality, 55, 139–153.
Weinberg, R., & Gould, D. (2011). Foundations of sport and exercise psychology. Champaign, IL: Human
Kinetics.
Wilson, G. S., Raglin, J. S., & Pritchard, M. E. (2002). Optimism, pessimism, and precompetitive anxiety in
college athletes. Personality and Individual Differences, 32, 893–902.
Yoshie, M., Shigemasu, K., Kudo, K., & Ohtsuki, T. (2009). Effects of state anxiety on music performance:
Relationship between the Revised Competitive State Anxiety Inventory-2 Subscales and Piano perfor-
mance. Musicae Scientiae, 13, 55–84.

Downloaded from sap.sagepub.com at TEXAS SOUTHERN UNIVERSITY on May 4, 2016