Exploring Clusters of Novice Programmers' Anxiety-Induced Behaviors During Block- and Text-Based Coding: A Predictive and Moderation Analysis of Programming Quality and Error Debugging Skills.

The article investigates how clusters of anxiety-induced behaviors among novice programmers predict programming performance in block-based and text-based coding environments. Using physiological (Apple Watch, ECG), behavioral observation, and self-report measures, the study identified three distinct anxiety clusters in each environment—"stay calm," "stay hesitant," and "to-calm" in block-based programming, and "to-hesitant," "stay hesitant," and "stay anxious" in text-based programming. Results indicate that novice programmers experience higher anxiety levels in text-based environments, which correlates with lower program quality and error-debugging skills, while block-based environments are associated with calmer states and better performance. The study found no significant moderating effects of personality traits, gender, or computer experience on the relationship between anxiety clusters and programming outcomes. These findings suggest the importance of tailored instructional strategies and policy considerations favoring block-based programming as an introductory tool to reduce anxiety and improve learning outcomes in programming education.