Competitive Sport Anxiety Inventory-2 (CSAI-2)

1. Abstract

The Competitive Sport Anxiety Inventory-2 (CSAI-2), developed by Rainer Martens, Robin S. Vealey, and Damon Burton (1990), stands as one of the most foundational and extensively investigated psychometric instruments in sport psychology. Designed to measure competitive state anxiety as a multidimensional construct, the CSAI-2 emerged out of theoretical advancements that partitioned situational competitive tension into separate cognitive and physiological components, while concurrently measuring state self-confidence. The scale comprises 27 self-report items evenly distributed across three distinct 9-item subscales: Cognitive State Anxiety (measuring negative performance expectations, cognitive preoccupation, and worries concerning failure), Somatic State Anxiety (measuring the athlete's subjective perception of autonomic physiological arousal, muscular tension, and visceral discomfort), and State Self-Confidence (measuring affirmative beliefs concerning personal capability, mental fortitude, and likelihood of competitive success).

Respondents evaluate each item using a 4-point Likert-type scale ranging from 1 (Not at all) to 4 (Very much so), yielding subscale scores between 9 and 36 without a global composite score. Over four decades of psychometric evaluation have demonstrated satisfactory internal consistency for its subscales, with Cronbach's alpha coefficients typically ranging from .79 to .90 across diverse athletic populations, competitive skill levels, and cultural environments. Confirmatory factor analyses have confirmed the three-factor baseline structure, though psychometric evaluations by Cox, Martens, and Russell (2003) subsequently led to a revised 17-item variant (CSAI-2R) to address cross-loadings and enhance construct validity. The CSAI-2 continues to serve as an indispensable measurement paradigm for investigating temporal anxiety dynamics prior to competition, examining directional perceptions of anxiety symptoms, assessing the efficacy of psychological skills training, and testing theoretical models of human performance under psychological pressure.

2. Keywords

Competitive Sport Anxiety Inventory-2, CSAI-2, competitive state anxiety, cognitive anxiety, somatic anxiety, sport self-confidence, multidimensional anxiety theory, sport psychometrics, athletic performance, pre-competitive arousal, Rainer Martens, psychological assessment in sport

3. Authors

The original conceptualization and empirical development of the Competitive Sport Anxiety Inventory-2 was executed by a collaborative group of prominent American sport psychologists led by Rainer Martens, Ph.D., formerly of the Department of Kinesiology at the University of Illinois at Urbana-Champaign and founder of Human Kinetics Publishers. Martens pioneered the study of competitive anxiety in sports, having previously developed the unidimensional Sport Competition Anxiety Test (SCAT) in 1977.

Martens was joined in this instrument's creation and validation by Robin S. Vealey, Ph.D., Professor of Sport Psychology in the Department of Kinesiology, Nutrition, and Health at Miami University (Ohio), whose seminal theoretical work on sport confidence provided the conceptual foundation for integrating self-confidence into the inventory; and Damon Burton, Ph.D., Professor Emeritus of Movement Sciences at the University of Idaho, renowned for his empirical investigations into goal setting, cognitive stress, and multidimensional competitive anxiety models.

The psychometric revision of the instrument, culminating in the Revised Competitive State Anxiety Inventory-2 (CSAI-2R), was spearheaded by Richard H. Cox, Ph.D. (University of Missouri-Columbia), Matthew P. Martens, Ph.D. (University of Missouri), and William D. Russell, Ph.D. (Missouri Western State University), published in 2003.

4. Purpose

The primary purpose of the CSAI-2 is to provide a standardized, psychometrically rigorous, and multidimensional measure of immediate, state-dependent anxiety and confidence responses triggered by competitive athletic stress. Prior to the development of the CSAI-2, competitive anxiety was predominantly assessed as a unidimensional trait or general emotional state using non-sport-specific scales, such as Charles Spielberger's State-Trait Anxiety Inventory (STAI), or unidimensional athletic instruments like Martens' original Competitive State Anxiety Inventory (CSAI; Martens et al., 1980). These early tools were limited in their capacity to differentiate physiological activation from cognitive worry or to capture the moderating influence of an athlete's self-confidence.

Recognizing that an athlete may experience profound physiological arousal (e.g., accelerated heart rate, muscle tension) without necessarily harboring catastrophic worries about performance, Martens and colleagues recognized the necessity for an instrument that independently assesses these experiential domains. The CSAI-2 fulfills several major applied and theoretical functions:

• Temporal Tracking of Pre-Competitive Stress: The inventory enables researchers and applied practitioners to track the distinct temporal trajectories of cognitive anxiety, somatic anxiety, and self-confidence across varying time intervals leading up to competition (e.g., one week, 24 hours, 2 hours, and 15 minutes pre-event).
• Diagnostic Assessment for Performance Optimization: In applied sport psychology, the instrument serves as a diagnostic battery to identify the primary modality of an athlete's competitive vulnerability. Athletes exhibiting elevated somatic anxiety benefit most from physiological interventions (autogenic training, progressive muscle relaxation, diaphragmatic breathing), whereas those suffering from elevated cognitive anxiety require cognitive restructuring, rational-emotive approaches, or attentional refocusing.
• Empirical Testing of Performance Prediction Models: The CSAI-2 provides operationalized variables required to evaluate theoretical models of motor performance under stress, including Multidimensional Anxiety Theory, Hardy's Catastrophe Model, and Hanin's Individual Zones of Optimal Functioning (IZOF).
• Outcome Evaluation of Applied Interventions: Applied practitioners use the CSAI-2 in clinical trials and field interventions to quantify pre-to-post changes in emotional states following psychological skills training programs such as mental imagery, self-talk training, and biofeedback.

5. Psychological Construct

The CSAI-2 operationalizes competitive state anxiety as an individualized, transient, multidimensional psychological state that is activated in evaluative athletic contexts. Rather than treating anxiety as an undifferentiated general drive, the instrument measures three distinct, interacting psychological subconstructs:

Cognitive State Anxiety

Cognitive state anxiety represents the mental component of the stress response. It is defined as the athlete's conscious awareness of negative expectations, cognitive concerns, subjective distress, and potential failure regarding performance outcomes or social evaluation. In competitive environments, this construct manifests through self-doubts, intrusive catastrophic thoughts, perceived inadequacies relative to opponents, and worry over external judgment from coaches, teammates, or spectators.

Items measuring cognitive anxiety (e.g., "I am concerned that I may not do as well in this competition as I could do," "I am concerned about choking under pressure," and "I have self-doubts") isolate pure cognitive worry from visceral physical activation. Research confirms that cognitive state anxiety is highly responsive to changes in objective outcome expectations, perceived probability of defeat, and the personal significance attached to the contest.

Somatic State Anxiety

Somatic state anxiety represents the physical, physiological, and affective manifestation of the competitive stress response. It reflects the individual's subjective perception and cognitive interpretation of autonomic sympathetic neuroendocrine activation. Rather than directly recording physiological indices such as galvanic skin response, electromyography (EMG), or plasma catecholamines, somatic anxiety captures the athlete's phenomenological awareness of these visceral changes.

Items within this subscale (e.g., "My body feels tense," "I feel tense in my stomach," "My heart is racing," and "My hands are clammy") capture peripheral arousal, localized gastrointestinal distress, and heightened muscular tension. Somatic anxiety operates as a rapid, conditioned response to environmental cues immediately preceding the athletic event, characteristically dissipating rapidly once the event commences.

State Self-Confidence

State self-confidence in the CSAI-2 framework reflects an athlete's situational conviction, belief, and positive certainty regarding their competence to perform successfully, cope with task demands, and execute requisite motor actions under pressure. Grounded in Bandura's self-efficacy theory and Vealey's sport-specific confidence framework, state self-confidence is conceptually treated not merely as the simple mathematical absence of anxiety, but as an independent, active psychological construct that functions in dynamic opposition to cognitive worry.

Items on this subscale (e.g., "I feel self-confident," "I'm confident I can meet the challenge," and "I'm confident because I mentally picture myself reaching my goal") measure subjective feelings of mental composure, perceived control, and optimism. High state self-confidence can protect an athlete from the disruptive effects of somatic arousal, buffering motor execution against intrusive worries.

6. Theoretical Framework

The CSAI-2 is theoretically anchored in Multidimensional Anxiety Theory (MAT), which synthesized foundational paradigms from cognitive psychology, clinical neuropsychology, and sport behavior models (Martens et al., 1990). The framework integrated three primary theoretical contributions:

Liebert and Morris' Dual-Process Model

The separation of anxiety into cognitive and somatic dimensions originated from the work of Liebert and Morris (1967) within test anxiety research in academic settings. They demonstrated that test-related anxiety consisted of "Worry" (cognitive distress regarding failure) and "Emotionality" (perceived physiological arousal). Martens and colleagues applied this paradigm to competitive athletic settings, hypothesizing that cognitive and somatic state anxiety represent structurally and functionally distinct mechanisms triggered by different competitive antecedents.

The Tripartite Structure and Temporal Dynamics Hypothesis

Multidimensional Anxiety Theory posits specific, testable hypotheses regarding the temporal emergence and performance implications of each dimension:

• Cognitive Anxiety: Hypothesized to maintain a negative linear relationship with athletic performance. Because cognitive worry depletes central executive working memory resources and disrupts visual-motor attentional focus, increases in cognitive anxiety are predicted to progressively impair performance. Temporally, cognitive anxiety can fluctuate days or weeks before competition as expectations of success or failure shift, remaining stable right up to the start of the event unless situational expectations change.
• Somatic Anxiety: Hypothesized to demonstrate an inverted-U relationship with athletic performance, consistent with the Yerkes-Dodson Law. Moderate somatic arousal facilitates physiological readiness and reaction speed, whereas excessive or insufficient arousal leads to motor coordination errors or sluggishness. Somatically, arousal symptoms remain low until approximately 24 to 48 hours before competition, increase sharply in the final hours and minutes before the start, and drop rapidly once competition begins.
• State Self-Confidence: Hypothesized to maintain a positive linear relationship with performance. Higher confidence fosters task persistence, optimal effort expenditure, and constructive coping when facing competitive setbacks.

Theoretical Refinements and Directional Perceptions

Subsequent sport psychology scholarship expanded the original theoretical architecture of the CSAI-2. Notably, Graham Jones and colleagues (Jones & Swain, 1992; Jones, 1995) introduced the Control Model of Competitive State Anxiety, noting that the CSAI-2 historically measured only symptom intensity (i.e., how much anxiety is felt). Jones argued that athletes also evaluate anxiety along a directional continuum—perceiving their cognitive and somatic sensations as either debilitative (harmful) or facilitative (helpful) to performance, dependent on their perceived degree of situational control and self-confidence.

7. Validity

The validity of the CSAI-2 has been subjected to extensive empirical verification across diverse sport contexts, athletic tiers (recreational, collegiate, elite, and Olympic), and translated international adaptations:

Construct and Factorial Validity

During its initial development, Martens et al. (1990) conducted iterative item analysis and exploratory factor analyses on data collected from large cohorts of intercollegiate and high school athletes. More than 70 preliminary candidate items were refined down to 27 items that loaded cleanly (> .40) onto their designated factors (Cognitive Anxiety, Somatic Anxiety, and Self-Confidence) without significant cross-loadings. While subsequent confirmatory factor analyses by Cox, Martens, and Russell (2003) revealed some psychometric vulnerability in several original CSAI-2 items (e.g., cross-loadings of reverse-coded Item 14 and cognitive item ambiguities), the basic three-factor model consistently shows superior statistical fit compared to unidimensional or two-factor models.

Convergent and Discriminant Validity

Convergent validity is evidenced by significant correlations between CSAI-2 subscales and established clinical and sport psychometric inventories:

• The Somatic State Anxiety subscale correlates strongly with the physical arousal indices of the STAI Form Y-1 (r = .65 to .78) and physiological biomarkers, including elevated resting heart rate and electrodermal responses prior to competition.
• The Cognitive State Anxiety subscale shows strong convergence with the Sport Competition Anxiety Test (SCAT; r = .55 to .68) and the worry subscale of the Sport Anxiety Scale (SAS/SAS-2; r = .62 to .74).
• Discriminant validity is supported by the orthogonal to moderate inverse correlations observed between State Self-Confidence and Cognitive State Anxiety (typically r = -.45 to -.65) and Somatic State Anxiety (r = -.25 to -.40). Somatic and Cognitive anxiety share a moderate positive correlation (r = .50 to .62), confirming they measure distinct facets of state anxiety rather than identical processes.

Predictive and Criterion Validity

Extensive field research supports the CSAI-2's predictive utility. Burton (1988) tested the instrument with collegiate swimmers and verified that cognitive anxiety and self-confidence accounted for substantial variance in objective swimming times, whereas somatic anxiety demonstrated an inverted-U trend. Furthermore, studies across golf, rifle shooting, tennis, and gymnastics have repeatedly demonstrated that high CSAI-2 cognitive anxiety scores predict performance disruptions under intense competitive evaluation.

8. Reliability

The psychometric evaluation of reliability for state instruments requires distinct criteria compared to trait inventories, because state anxiety is, by definition, dynamic and sensitive to fluctuating situational variables. Consequently, internal consistency represents the primary index of reliability for the CSAI-2, while test-retest reliability is evaluated only across neutral baseline periods where competitive conditions remain constant.

Internal Consistency

Extensive empirical studies consistently demonstrate robust internal consistency across all three CSAI-2 subscales:

• Cognitive State Anxiety: Cronbach's alpha coefficients typically range between .79 and .89 across varied sport cohorts (Martens et al., 1990; Burton, 1988; Krane, 1994).
• Somatic State Anxiety: Cronbach's alpha coefficients consistently fall between .81 and .88, demonstrating strong item homogeneity in capturing visceral and muscular arousal perceptions.
• State Self-Confidence: Cronbach's alpha coefficients routinely demonstrate high reliability, frequently falling between .86 and .92.

Cross-cultural adaptations across Spanish, French, Greek, Turkish, Portuguese, and Mandarin populations have replicated these internal consistency metrics, yielding alpha values generally exceeding the recommended psychometric threshold of .70 for research and applied diagnostics.

Test-Retest Stability (Under Invariant Baseline Conditions)

When administered across non-stressful classroom or non-competitive practice conditions separated by brief intervals (e.g., 2 to 7 days), stability coefficients for the CSAI-2 subscales range from .60 to .75. Conversely, as an athlete moves closer to competitive events, test-retest stability coefficients predictably decline. Rather than reflecting psychometric measurement error, this variance reflects the instrument's sensitivity to the real-time temporal elevation of competitive stress.

9. Factor Analysis

The structural dimensionality of the CSAI-2 has been the subject of extensive psychometric investigation, shifting from early exploratory factor methods to rigorous contemporary confirmatory factor analysis (CFA) frameworks.

Exploratory Factor Analyses (EFA)

During the initial development of the CSAI-2, Rainer Martens and colleagues carried out principal component and principal axis factoring procedures utilizing varimax and promax rotations across multiple samples totaling more than 1,200 high school and collegiate athletes. The analyses revealed a clean three-factor structure accounting for over 45% of the total variance:

• Factor 1 (Cognitive Anxiety) aggregated items reflecting performance worry, fear of failure, and concentration concerns.
• Factor 2 (Somatic Anxiety) captured visceral, autonomic, and muscular tension sensations.
• Factor 3 (State Self-Confidence) gathered statements of personal capability, mental resilience, and positive expectations.

Confirmatory Factor Analyses (CFA) and Model Fit

Later evaluations using structural equation modeling and CFA subjected the CSAI-2's 27-item measurement model to strict goodness-of-fit benchmarks. Seminal investigations by Lane, Sewell, Terry, Bartram, and Nesti (1999) indicated that while the three-factor model was conceptually supported, the empirical fit indices of the full 27-item inventory frequently failed to meet modern psychometric standards (e.g., Comparative Fit Index [CFI] < .90; Root Mean Square Error of Approximation [RMSEA] > .08).

Psychometric Critique and the CSAI-2R Revision

In a comprehensive re-evaluation, Cox, Martens, and Russell (2003) tested the CSAI-2 on a diverse sample of 3,310 athletes across 34 sports. Their multi-sample CFA revealed that 10 of the original 27 items were compromised by low factor loadings, cross-loadings, or ambiguous semantic content. Notably, reverse-coded items (such as Item 14: "My body feels relaxed") displayed method effects and loaded inconsistently across factors.

By removing problematic items, Cox and colleagues developed the 17-item Revised Competitive State Anxiety Inventory-2 (CSAI-2R), consisting of:

• Somatic Anxiety: 7 items (Items 1, 4, 6, 9, 12, 15, 17)
• Cognitive Anxiety: 5 items (Items 2, 5, 8, 11, 14)
• Self-Confidence: 5 items (Items 3, 7, 10, 13, 16)

The revised model established strong construct representation, measurement invariance across gender, and superior statistical fit indices (CFI = .95, RMSEA = .05) while retaining the original three-factor theoretical foundation.

10. Instrument / Measurement Tool

• Full Instrument Name: Competitive Sport Anxiety Inventory-2 (CSAI-2)
• Alternative Title: Competitive State Anxiety Inventory-2
• Developer / Creators: Rainer Martens, Ph.D., Robin S. Vealey, Ph.D., and Damon Burton, Ph.D.
• Constructs Measured: Multidimensional Competitive State Anxiety (Cognitive Anxiety, Somatic Anxiety) and State Self-Confidence.
• Target Population: Competitive athletes across all developmental stages (adolescent, collegiate, adult, elite/Olympic).
• Administration Format: Self-administered paper-and-pencil or digital psychometric questionnaire.
• Completion Time: Approximately 4 to 6 minutes for the full 27-item scale; 2 to 3 minutes for the revised 17-item CSAI-2R.
• Item Count: 27 total items (9 items per subscale).
• Response Scale: 4-point Likert-type scale formatted as:
  • 1 = Not at all
  • 2 = Somewhat
  • 3 = Moderately so
  • 4 = Very much so
• Scoring Procedures:
  • Cognitive State Anxiety: Sum items 1, 4, 7, 10, 13, 16, 19, 22, and 25 (Score range: 9 to 36).
  • Somatic State Anxiety: Sum items 2, 5, 8, 11, 14, 17, 20, 23, and 26 (Score range: 9 to 36). Note: In standard original scoring, item 14 is reverse-coded (1=4, 2=3, 3=2, 4=1), although in revised variants it was removed entirely due to method variance.
  • State Self-Confidence: Sum items 3, 6, 9, 12, 15, 18, 21, 24, and 27 (Score range: 9 to 36).
  • Important Psychometric Note: There is no total or aggregate score. Subscales must be interpreted independently to preserve construct validity.

11. Permissions & Fee and Test Year

The Competitive Sport Anxiety Inventory-2 was formally finalized and published in book format in 1990 within the monograph Competitive Anxiety in Sport (Human Kinetics Publishers), building on preliminary development versions introduced in 1980 (Martens et al., 1980). The revised version (CSAI-2R) was published by Cox, Martens, and Russell in 2003 in the Journal of Sport & Exercise Psychology.

Licensing and Research Availability: The CSAI-2 and CSAI-2R items have been made accessible for academic, clinical, and scholarly research. The original publishers and authors generally permit the reproduction, adaptation, and translation of the inventory for non-commercial educational, scientific, and athletic assessment purposes, provided that appropriate scholarly attribution is cited. Researchers and practitioners planning to deploy the inventory within commercial diagnostic software, fee-based sports analytics platforms, or published tests should consult Human Kinetics or the respective journal copyright holders for commercial permissions.

12. References

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191

Burton, D. (1988). Do anxious swimmers swim slower? Reexamining the elusive anxiety-performance relationship. Journal of Sport & Exercise Psychology, 10(1), 45–61. https://doi.org/10.1123/jsep.10.1.45

Cox, R. H., Martens, M. P., & Russell, W. D. (2003). Measuring anxiety in athletics: The Revised Competitive State Anxiety Inventory–2. Journal of Sport & Exercise Psychology, 25(4), 519–533. https://doi.org/10.1123/jsep.25.4.519

Hardy, L. (1996). A test of catastrophe models of anxiety and sports performance against multidimensional anxiety theory models. The Sport Psychologist, 10(2), 140–156. https://doi.org/10.1123/tsp.10.2.140

Jones, G. (1995). More than just a game: Research developments and issues in competitive state anxiety in sport. British Journal of Psychology, 86(4), 449–478. https://doi.org/10.1111/j.2044-8295.1995.tb02565.x

Jones, G., & Swain, A. (1992). Intensity and direction dimensions of competitive state anxiety and relationships with competitiveness. Perceptual and Motor Skills, 74(2), 467–472. https://doi.org/10.2466/pms.1992.74.2.467

Lane, A. M., Sewell, D. F., Terry, P. C., Bartram, D., & Nesti, M. S. (1999). Confirmatory factor analysis of the Competitive State Anxiety Inventory-2. Journal of Sports Sciences, 17(6), 505–512. https://doi.org/10.1080/026404199365852

Liebert, R. M., & Morris, L. W. (1967). Cognitive and emotional components of test anxiety: A distinction and some initial data. Psychological Reports, 20(3), 975–978. https://doi.org/10.2466/pr0.1967.20.3.975

Martens, R. (1977). Sport Competition Anxiety Test. Human Kinetics Publishers.

Martens, R., Burton, D., Rivkin, F., & Simon, J. (1980). Reliability and validity of the Competitive State Anxiety Inventory (CSAI). In C. H. Nadeau, W. C. Halliwell, K. M. Newell, & G. C. Roberts (Eds.), Psychology of Motor Behavior and Sport—1979 (pp. 91–99). Human Kinetics.

Martens, R., Vealey, R. S., & Burton, D. (1990). Competitive Anxiety in Sport. Human Kinetics Publishers.

Spielberger, C. D. (1966). Theory and research on anxiety. In C. D. Spielberger (Ed.), Anxiety and Behavior (pp. 3–20). Academic Press. https://doi.org/10.1016/B978-1-4832-3131-0.50006-8

Vealey, R. S. (1986). Conceptualization of sport-confidence and competitive orientation: Preliminary investigation and instrument development. Journal of Sport Psychology, 8(3), 221–246. https://doi.org/10.1123/jsp.8.3.221

13. Items of the Scale