The Competitive State Anxiety Inventory\u20132 (CSAI-2)<\/strong> is the preeminent psychometric instrument designed to assess multidimensional pre-competitive state anxiety<\/a> and state self-confidence in athletic populations. Developed by Rainer Martens, Damon Burton, Robin S. Vealey, Linda A. Bump, and Damon E. Smith in 1990, the CSAI-2 emerged from the necessity to separate the cognitive and physiological components of competitive stress, diverging from historical unidimensional constructs. The scale consists of 27 self-report items evenly distributed across three distinct subscales (9 items each): Cognitive State Anxiety<\/em>, which assesses subjective worries, concerns regarding failure, and negative performance expectations; Somatic State Anxiety<\/em>, which evaluates the perception of autonomic physiological arousal, muscular tension, and visceral reactions; and State Self-Confidence<\/em>, which captures positive expectations of capability, perceived competence, and psychological resilience in competitive contexts.<\/p>\n Respondents evaluate each item using a 4-point Likert scale<\/a> ranging from 1 (“Not at all”) to 4 (“Very much so”), yielding subscale scores ranging from 9 to 36 without composite total scoring. Extensive psychometric testing across collegiate, elite, and recreational sports reveals solid internal consistency, with Cronbach’s alpha<\/a> coefficients typically spanning 0.79 to 0.83 for Cognitive State Anxiety, 0.82 to 0.88 for Somatic State Anxiety, and 0.85 to 0.90 for State Self-Confidence. Factor analytic investigations have confirmed a robust three-factor orthogonal or modestly correlated structure, while temporal tracking studies demonstrate differential pre-competition trajectories: cognitive anxiety and self-confidence remain relatively stable in the days preceding competition until event-specific expectations shift, whereas somatic anxiety spikes precipitously immediately prior to the contest and dissipates rapidly upon event initiation. The instrument serves as a cornerstone in sport psychology research, performance diagnostics, and applied psychological skills interventions.<\/p>\n Competitive State Anxiety Inventory-2, CSAI-2, competitive state anxiety, cognitive anxiety, somatic anxiety, state self-confidence, sport psychology, athletic performance, multidimensional anxiety theory, pre-competition arousal, psychometrics, Martens.<\/p>\n The Competitive State Anxiety Inventory\u20132 was developed by a team of sport psychologists and researchers headed by Rainer Martens, Ph.D.:<\/p>\n The fundamental purpose of the Competitive State Anxiety Inventory\u20132 (CSAI-2) is to provide an empirically valid, multidimensional operationalization of pre-competitive state anxiety. Prior to its construction, competitive stress was predominantly conceptualized and measured as a unidimensional construct using instruments such as the State-Trait Anxiety Inventory (STAI Form Y-1<\/a>; Spielberger, 1983) or the original Competitive State Anxiety Inventory (CSAI<\/a>; Martens, 1977). While the original CSAI situated anxiety assessment within competitive physical environments, it conflated cognitive apprehension with autonomic physical sensations, treating subjective worry and somatic symptoms as interchangeable indicators of general emotional arousal. Consequently, early studies reported contradictory, weak, or non-significant relationships between pre-game anxiety and athletic motor performance.<\/p>\n To overcome these theoretical and empirical shortcomings, Martens and colleagues drew upon broader psychological conceptualizations by Liebert and Morris (1967) and Davidson and Schwartz (1976), who demonstrated that anxiety consists of distinct cognitive and somatic modalities. The CSAI-2 was engineered specifically to differentiate subjective cognitive concerns from physiological reactivity, while simultaneously recognizing state self-confidence as an independent psychological construct rather than merely the negative pole of cognitive anxiety. Thus, the inventory aims to isolate these three components to examine their unique antecendents, trajectories, and performance ramifications.<\/p>\n In research settings, the CSAI-2 enables sport scientists to examine complex performance models, including Martens et al.’s (1990) Multidimensional Anxiety Theory<\/a>, Hardy’s (1990) Catastrophe Model of anxiety and performance, and Hanin’s (1997) Individual Zones of Optimal Functioning (IZOF<\/a>) framework. It provides a standardized mechanism to quantify baseline state anxiety across various time-to-event intervals (e.g., 48 hours, 24 hours, 2 hours, and 15 minutes pre-event), allowing researchers to track temporal fluctuations and evaluate psychological readiness.<\/p>\n In clinical and applied sport psychology settings, the CSAI-2 serves as a diagnostic profiling tool. Applied practitioners use subscale profiles to tailor psychological skills training (PST). For instance, an athlete exhibiting elevated somatic anxiety accompanied by moderate cognitive worry and high self-confidence benefits most from autonomic down-regulation techniques such as progressive muscle relaxation<\/a>, diaphragmatic breath control, or biofeedback. Conversely, an athlete characterized by debilitating cognitive anxiety and eroded self-confidence requires cognitive interventions, such as cognitive restructuring, rational emotive behavioral analysis, self-talk modification, or mental imagery. The inventory thereby bridges theoretical psychometrics and customized performance optimization.<\/p>\n The CSAI-2 operationalizes competitive state anxiety across three conceptually autonomous yet interactive subscales. Unlike trait anxiety, which represents an enduring personality disposition to perceive competitive scenarios as threatening, state anxiety is defined as a transient, situation-specific emotional condition marked by consciously perceived feelings of tension, apprehension, and heightened activation of the autonomic nervous system.<\/p>\n Cognitive state anxiety represents the mental component of the anxiety response. It is defined as subjective, negative expectations, evaluations, and self-doubts regarding competitive performance and its consequences. Athletes experiencing elevated cognitive state anxiety harbor pervasive worries about competitive failure, potential embarrassment, falling short of personal standards or external expectations, and the inability to maintain attentional focus under pressure. Exemplary items from the CSAI-2 include: “I have self-doubts”<\/em> (Item 4), “I am concerned about losing”<\/em> (Item 10), and “I’m concerned that others will be disappointed with my performance”<\/em> (Item 22). Psychologically, cognitive state anxiety is elicited by evaluations of threat to self-esteem, perceived inadequacy relative to an opponent, or excessive importance attributed to outcome goals. From an information-processing perspective, high cognitive anxiety consumes working memory capacity, disrupts visual search behaviors, and provokes conscious reinvestment in automated motor control, leading to impaired performance execution.<\/p>\n Somatic state anxiety represents the physiological and affective elements of the anxiety experience that stem directly from autonomic arousal. It reflects the individual’s subjective perception and interpretation of visceral, neurovegetative, and muscular bodily sensations triggered by the sympathetic fight-or-flight response. Somatic anxiety encompasses muscular tension, tremors, gastrointestinal discomfort, elevated cardiovascular output, and vasomotor responses. In the CSAI-2, somatic anxiety is measured through items such as: “My body feels tense”<\/em> (Item 8), “I feel tense in my stomach”<\/em> (Item 11), “My heart is racing”<\/em> (Item 17), and “My hands are clammy”<\/em> (Item 23). Unlike cognitive anxiety, which is sustained by cognitive appraisal, somatic anxiety is an automatic physiological reaction that typically remains dormant until the physical context of the event is approached, peaking during the minutes immediately prior to execution.<\/p>\n State self-confidence refers to an athlete’s temporary belief or certainty regarding their capability to achieve success and execute physical skills effectively in an upcoming competitive challenge. Grounded in Albert Bandura<\/a>‘s (1977) self-efficacy theory and Robin Vealey’s (1986) sport-confidence conceptualization, self-confidence was intentionally incorporated into the CSAI-2 as an independent psychological construct. Rather than serving as the mere mathematical inverse of cognitive anxiety, self-confidence operates as a distinct psychological buffer against debilitating anxiety symptoms. Representative CSAI-2 items include: “I feel self-confident”<\/em> (Item 9), “I’m confident I can meet the challenge”<\/em> (Item 15), and “I’m confident of coming through under pressure”<\/em> (Item 27). Athletes with high state self-confidence display heightened perseverance, resilient attentional control, adaptive task-focused coping, and an inclination to interpret physiological arousal as facilitative rather than threatening.<\/p>\n The CSAI-2 is grounded primarily in Multidimensional Anxiety Theory (MAT)<\/strong> formulated by Martens, Burton, Vealey, Bump, and Smith (1990). MAT integrates foundational insights from general cognitive psychology, neurobiology, and clinical psychology to explain how different facets of competitive state anxiety develop and exert differential effects on motor performance.<\/p>\n Prior to MAT, the dominant conceptual paradigm in sport psychology was the Yerkes-Dodson Law<\/a> (1908) or Inverted-U Hypothesis, which postulated that athletic performance increases with physiological arousal up to an optimal midpoint, beyond which progressive increases in arousal cause systematic performance deterioration. However, empirical tests of the Inverted-U regularly yielded ambiguous results because researchers treated arousal, cognitive worry, and somatic distress as a unified entity. Building upon the two-component models of anxiety pioneered by Liebert and Morris<\/a> (1967) and expanded by Davidson and Schwartz (1976), Martens et al. posited that cognitive and somatic anxiety represent qualitatively distinct dimensions characterized by disparate antecedents, temporal patterns, and performance relationships.<\/p>\n Multidimensional Anxiety Theory advances three fundamental hypotheses:<\/p>\n Beyond performance predictions, MAT outlines distinct temporal patterns for each subscale prior to competitive engagement. Cognitive state anxiety is hypothesized to remain relatively stable in the days and weeks preceding a competition, fluctuating only when an athlete’s subjective probability of success changes (e.g., changes in opponent status or adverse external circumstances). Conversely, somatic state anxiety is hypothesized to remain at basal resting levels until approximately 24 to 48 hours before the contest, experiencing an exponential increase as the exact moment of competition draws near, followed by an immediate decrease once physical action commences. State self-confidence mirrors cognitive anxiety in its stability, exhibiting fluctuations only when perceived capability is reassessed. Decades of subsequent investigations (Burton, 1988; Craft et al., 2003; Woodman & Hardy, 2003) have corroborated the temporal divergence between cognitive and somatic states, while refining our understanding of how cognitive anxiety and confidence interact under extreme physiological pressure.<\/p>\n The validity of the CSAI-2 has been investigated extensively using construct, convergent, discriminant, and predictive validation protocols across diverse sports, age cohorts, and skill tiers.<\/p>\n The construct validity of the CSAI-2 was established through multi-stage instrument development involving item generation, content expert panels, exploratory factor analyses, and subsequent confirmatory factor modeling (Martens et al., 1990). During initial scale development, an item pool of 79 candidate items derived from existing anxiety metrics and clinical interviews was reviewed by elite athletes and sport psychologists. Items displaying low factor loadings, dual-subscale loadings, or weak corrected item-total correlations were systematically removed, yielding the refined 27-item structure. Construct validity is supported by experimental studies showing that targeted stressors manipulate subscale scores independently; for example, competitive stress inductions involving outcome uncertainty selectively elevate cognitive anxiety, whereas physical stressors or impending physical execution selectively amplify somatic anxiety scores.<\/p>\n Convergent validity has been established by correlating CSAI-2 subscales with validated trait anxiety and general state anxiety instruments. The CSAI-2 Somatic State Anxiety subscale correlates strongly with the physical anxiety subscale of the Sport Anxiety Scale (SAS<\/a>; Smith et al., 1990; $r = 0.65$<\/span> to $0.78$) and physiological indices of autonomic arousal, such as galvanic skin response (GSR) and heart rate variability (HRV). The Cognitive State Anxiety subscale shows strong convergence with the SAS Worry subscale ($r = 0.68$<\/span> to $0.81$) and the State Trait Anxiety Inventory (STAI-S; $r = 0.60$<\/span> to $0.74$).<\/p>\n Discriminant validity is supported by the factor correlations among the CSAI-2 subscales themselves. Martens et al. (1990) reported weak to moderate correlations between Cognitive and Somatic anxiety ($r \\approx 0.36$<\/span> to $0.54$), indicating that they share less than 30% of their variance. Furthermore, the correlation between Cognitive State Anxiety and State Self-Confidence is negative and moderate ($r \\approx -0.45$<\/span> to $-0.62$<\/span>), confirming that confidence operates as an autonomous psychometric dimension rather than merely the direct inverse of cognitive worry.<\/p>\n Predictive validity investigations have yielded nuanced findings across athletic disciplines. In a seminal meta-analysis of 29 studies, Craft, Magyar, Becker, and Feltz (2003) evaluated the predictive utility of the CSAI-2 for competitive athletic performance. The authors calculated mean weighted effect sizes ($z_r$<\/span>), reporting that state self-confidence exhibited the strongest positive correlation with performance ($z_r = 0.25$<\/span>), cognitive state anxiety had a weak negative correlation ($z_r = -0.05$<\/span>), and somatic state anxiety displayed an effect size approximating zero ($z_r = -0.03$<\/span>). Similarly, Woodman and Hardy (2003) conducted an extensive meta-analysis showing that cognitive anxiety significantly impairs performance when self-confidence is low, but has a negligible effect when confidence is high. These empirical findings underscore that while CSAI-2 subscale scores predict performance outcomes, their relationship is moderated by task characteristics, sport categorization (individual vs. team sports), and interactive psychological variables.<\/p>\n The reliability of the CSAI-2 has been corroborated through numerous studies across athletic levels, competitive settings, and cultural demographics.<\/p>\n Internal consistency for each of the three 9-item subscales consistently satisfies standard psychometric thresholds across the published literature:<\/p>\n Because the CSAI-2 is designed specifically to capture state<\/em> fluctuations that vary as a function of environmental demands and temporal proximity to competition, conventional long-term test-retest reliability is theoretically inapplicable. A psychometrically valid state measure must be responsive to changing external contexts. However, Martens et al. (1990) assessed the stability of the instrument across resting baseline conditions (non-competitive baseline testing separated by one to two weeks), documenting test-retest correlation coefficients ranging from $r = 0.65$<\/span> to $r = 0.73$<\/span>. In contrast, when measured repeatedly within the competitive window\u2014such as 48 hours, 24 hours, 2 hours, and 15 minutes pre-competition\u2014the instrument exhibits intentional, predictable instability, capturing the dynamic escalation of somatic tension and contextual adjustments in cognitive anxiety.<\/p>\n The internal structural validity of the CSAI-2 has been evaluated through extensive Exploratory Factor Analyses<\/a> (EFA) and Confirmatory Factor Analyses<\/a> (CFA).<\/p>\n During the original development phase, Martens et al. (1990) conducted principal components and principal axes factor analyses using orthogonal (Varimax) and oblique (Promax) rotations across multiple samples totaling over 1,200 participants. The analyses consistently isolated three primary factors corresponding to Cognitive State Anxiety, Somatic State Anxiety, and State Self-Confidence, with eigenvalues exceeding 1.0. All 27 retained items exhibited primary factor loadings greater than 0.40 on their designated construct, with negligible cross-loadings (typically < 0.25) on alternative factors, indicating clear structural separation<\/span>.<\/p>\n Subsequent psychometric scrutiny employing Confirmatory Factor Analysis (CFA) within structural equation modeling frameworks provided rigorous evaluation of the proposed three-factor measurement model. Lane, Sewell, Terry, Bartram, and Nesti (1999) examined the 27-item CSAI-2 across a diverse sample of competitive athletes ($N = 1,213$<\/span>). Their initial CFA indicated that while the three-factor model was conceptually superior to alternative one-factor or two-factor models, the overall goodness-of-fit indices were marginal ($RCFI = 0.87$<\/span>; $RMSEA = 0.07$<\/span>). Item-level analyses revealed that several items, particularly Item 14 (“My body feels relaxed”<\/em>), Item 24 (“I’m confident because I mentally picture myself reaching my goal”<\/em>), and Item 25 (“I’m concerned I won’t be able to concentrate”<\/em>), exhibited high residual covariances and cross-loadings on secondary dimensions.<\/p>\n Cox, Martens, and Russell (2003) conducted a comprehensive CFA re-evaluation ($N = 503$<\/span>), which confirmed moderate fit for the original 27-item instrument ($\\chi^2\/df = 2.45$<\/span>, $CFI = 0.88$<\/span>, $NNFI = 0.87$<\/span>, $RMSEA = 0.061$<\/span>). They observed that reverse-worded somatic items (such as Item 14, which requires the respondent to assess relaxation rather than direct tension) compromised measurement model fit. This psychometric finding prompted Cox and colleagues to develop the Revised Competitive State Anxiety Inventory\u20132 (CSAI-2R<\/a>), which streamlined the instrument to 17 items. Despite the development of the revised variant, the classic 27-item CSAI-2 remains widely utilized because its original subscale structure continues to demonstrate robust discriminant validity across international samples, establishing a rich historical baseline across three decades of competitive anxiety research.<\/p>\n The Competitive State Anxiety Inventory\u20132 was formally published in 1990 in the seminal research volume Competitive Anxiety in Sport<\/em>, authored by Rainer Martens, Robin S. Vealey, and Damon Burton, and published by Human Kinetics Publishers (Champaign, IL). The scale was placed in the public academic domain for non-commercial research and educational applications by its developers to foster scientific investigation into athletic stress. Researchers, practitioners, and sport psychology students may administer the CSAI-2 for academic, clinical, and field investigations without royalty fees, provided full academic attribution is rendered to Martens, Burton, Vealey, Bump, and Smith (1990). Commercial exploitation, publication within fee-for-service software platforms, or redistribution for profit requires explicit written permission from Human Kinetics and the copyright holders.<\/p>\n2. Keywords<\/h2>\n
3. Authors<\/h2>\n
\n
4. Purpose<\/h2>\n
5. Psychological Construct<\/h2>\n
Cognitive State Anxiety<\/h3>\n
Somatic State Anxiety<\/h3>\n
State Self-Confidence<\/h3>\n
6. Theoretical Framework<\/h2>\n
\n
7. Validity<\/h2>\n
Construct Validity<\/h3>\n
Convergent and Discriminant Validity<\/h3>\n
Predictive and Criterion Validity<\/h3>\n
8. Reliability<\/h2>\n
Internal Consistency<\/h3>\n
\n
Test-Retest Stability<\/h3>\n
9. Factor Analysis<\/h2>\n
Exploratory Factor Structure<\/h3>\n
Confirmatory Factor Analytic Evidence<\/h3>\n
10. Instrument \/ Measurement Tool<\/h2>\n
\n
\n
\n
\n
11. Permissions & Fee and Test Year<\/h2>\n
12. References<\/h2>\n
\n
13. Items of the Scale (Questionnaire)<\/h2>\n