1. Research Questions: 

RQ1: What evidence supports the validity and reliability of the Computational Thinking Anxiety Scale (CTAS) for assessing computational thinking anxiety (CTA) across late childhood and adolescence?

RQ2: How is CTA related to computational thinking performance (CTP)?

RQ3: How does this relationship diverge from general trait anxiety and vary across individual characteristics such as age, gender, and programming experience?

2.Theoretical Framework:
The study is grounded in an integrative, multi-theoretical framework that bridges affective science, cognitive development, and computational thinking theory. It posits CTA as a dynamic, multidimensional construct emerging from cognitive-affective-behavioral interactions during engagement with core computational thinking (CT) processes.

First, the Integrated Model of Emotion and Cognition (Lemerise & Arsenio, 2000) provides the foundational premise that affective states like anxiety are not mere epiphenomena but are fundamentally intertwined with cognitive processing, jointly shaping competence development. This is critical for understanding how anxiety can specifically impair components of the CT process.

Second, to operationalize the phenomenology of anxiety, we integrated the American Psychological Association’s tripartite model of emotion (APA, 2025b) with the classic ABC Model of Attitude (Eagly & Chaiken, 1998). This yielded a triple-ABC framework defining CTA across synchronized manifestations: Affective (Physiological Symptoms: e.g., arousal), Behavioral (Stressful Behaviors: e.g., avoidance), and Cognitive (Cognitive Impairment: e.g., worry, mental blocking).

Third, to anchor this affective-behavioral-cognitive superstructure specifically within the CT domain, we populated it with a refined, domain-general CT competency model (Shute et al., 2017; Wing, 2006). This model conceptualizes CT as a systematic sequence involving decomposition, abstraction, algorithmic design, debugging, iteration, generalization, pattern recognition and parallelism. This creates a precise matrix for locating anxiety (e.g., cognitive impairment in pattern-recognition).

Finally, a developmental psychometric perspective, informed by Piagetian theory (Piaget, 1972), underpins the investigation. We hypothesized that the architecture of CTA is not static but evolves in tandem with cognitive maturation, particularly during the transition from concrete to formal operational thought, leading to predictable reorganizations in the salience of specific anxiety dimensions.

This robust, layered framework guided the item generation for the CTAS, ensuring it captures the full, developmentally-sensitive phenomenology of anxiety within the unique cognitive architecture of CT, while rigorously distinguishing it from related constructs like programming anxiety or general trait anxiety.

The research employed a comprehensive, multi-phase mixed-methods design for scale development and validation, adhering to contemporary psychometric standards (Boateng et al., 2018). Scale Development: The CTAS was developed through a rigorous, theory-driven process. An initial item pool was generated by aligning the triple-ABC × CT competency matrix with phrasing conventions from established anxiety scales. A bilingual (Chinese/English) Delphi expert review (5 experts in CT, psychometrics, educational psychology) ensured content validity, conceptual clarity, and cultural appropriateness. This was followed by cognitive pretesting with the target population (n=25, aged 9-10) using think-aloud protocols to optimize comprehension and ecological validity. Measures: The primary instrument was the final 30-item (later refined to 25-item) CTAS, a self-report measure using a 5-point Likert scale. CTP was assessed via an adapted short-form of the Australian Bebras Computational Thinking Challenge, culturally validated for Chinese students and stratified by age. Trait anxiety was controlled using the Spence Children’s Anxiety Scale–Short Version (SCAS-S). Programming experience was quantified as months of formal/informal instruction. Participants and Procedure: A large sample of Chinese adolescents (N=697, aged 9-14) was recruited via stratified random sampling. The study integrated cross-sectional and longitudinal components. The total sample was randomly split into independent calibration (n=299) and validation (n=398) subsamples for Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), respectively. A longitudinal subsample (n=387) was reassessed after three months to establish test-retest reliability. Analyses: A multi-stage analytical plan was executed: 1) Item analysis and reliability assessment; 2) EFA to explore latent structure; 3) CFA to confirm the factor model and test for measurement invariance across age groups; 4) Assessment of convergent, discriminant (using HTMT ratio, Fornell-Larcker criterion), and criterion validity; 5) Network analysis to visualize CT anxiety (CTA) factor relationships; 6) Hierarchical regression to test incremental predictive validity over trait anxiety; 7) Correlation Matrix among age cohorts, CTP, programming experience, trait anxiety, and gender difference; and 7) Longitudinal analysis using Intraclass Correlation Coefficients (ICC).

This study advances CT research by making visible a dimension that has often been treated as peripheral: learners’ anxiety when engaging in core CT processes. We developed and validated the Computational Thinking Anxiety Scale (CTAS), a theory-informed, developmentally sensitive instrument for adolescents. Across 697 Chinese students aged 9–14, the final 25-item CTAS yielded a coherent eight-factor structure—Algorithms, Pattern Recognition, Parallelism, Decomposition, Generalization, Debugging, Abstraction, and Iteration—and showed excellent model fit (CFI = .974, RMSEA = .041), reliability, convergent and discriminant validity, and three-month temporal stability (ICC = .983; r = .986). Measurement invariance across age groups supports meaningful developmental comparison. Substantively, the findings show that CTA is not merely a proxy for general anxious disposition or programming exposure. CTAS scores were essentially unrelated to trait anxiety and only weakly associated with programming experience, but were substantially and negatively related to CTP. This indicates that CTA is a domain-specific affective barrier with direct relevance for learning, assessment, and instructional design. Developmental analyses further suggest that CTA has both stable and dynamic features. Algorithm-related anxiety remained comparatively robust across ages, whereas some facets varied more strongly, with the 11–12 group showing the greatest variability. This points to early adolescence as a potentially critical window for affect-sensitive CT support. Overall, the study reframes CT as a cognitive–affective learning system rather than a purely cognitive competence. The CTAS offers a transferable framework for examining how affect shapes digital competence, participation, and equity in computing education.

Ahlen, J., Vigerland, S., & Ghaderi, A. (2018). Development of the spence children’s anxiety scale-short version (SCAS-S). Journal of psychopathology and behavioral assessment, 40(2), 288-304. American Psychological Association (APA). (2025a). Anxiety. American Psychological Association. Retrieved November 19, 2025a from https://dictionary.apa.org/anxiety American Psychological Association (APA). (2025b). Emotion. Retrieved November 19, 2025b from https://dictionary.apa.org/emotion Brennan, K., & Resnick, M. (2012, April). New frameworks for studying and assessing the development of computational thinking. In Proceedings of the 2012 annual meeting of the American educational research association, Vancouver, Canada (Vol. 1, p. 25). CSIRO Digital Careers. (2023a). Bebras Australia Computational Thinking Challenge: 2023 Solutions Guide Round 1 (Middle School Years 7-8). Bebras Australia. CSIRO Digital Careers. (2023b). Bebras Australia Computational Thinking Challenge: 2023 Solutions Guide Round 2 (Primary School Years 3-6). Bebras Australia. https://bebras.edu.au/ DeVellis, R. F. (2003). Scale development: Theory and applications. Thousand Oaks. Eagly, A. H., & Chaiken, S. (1998). Attitude structure and function. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed., pp. 269–322). McGraw-Hill. Lemerise, E. A., & Arsenio, W. F. (2000). An integrated model of emotion processes and cognition in social information processing. Child development, 71(1), 107-118. Piaget, J. (1972). Intellectual evolution from adolescence to adulthood. Human Development, 15(1), 1–12. Shute, V. J., Sun, C., & Asbell-Clarke, J. (2017). Demystifying computational thinking. Educational research review, 22, 142-158. Wing, J. M. (2006). Computational thinking. Communications of the ACM, 49(3), 33-35.