<![CDATA[Computational thinking (CT) has become increasingly important in physics education, where prospective physics teachers are expected not only to develop computational solutions but also to interpret, justify, and communicate computational reasoning. However, CT assessment in higher education remains diverse, and communication is often embedded within broader CT constructs rather than examined as an explicitly measured domain. This study aimed to develop and preliminarily validate a communication-explicit CT efficacy scale for undergraduate physics education students. We employed an instrument development and preliminary validation design, integrating expert content review, iterative Rasch-based item refinement, and partial least squares structural equation modeling (PLS-SEM). Five experts evaluated the initial 64-item Indonesian-language self-report instrument, and we administered it to 100 third-semester physics education undergraduates from two universities in Indonesia. Rasch analysis using a polytomous rating-scale model refined the instrument into a 52-item scale representing six domains: creativity, algorithmic thinking, cooperativity, critical thinking, problem solving, and communication. The results indicated satisfactory preliminary measurement evidence, with person reliability of 0.91, item reliability of 0.93, and mean item Infit/Outfit MNSQ values of 0.98/1.01. PLS-SEM further showed acceptable domain-level reliability and convergent validity, with composite reliability values ranging from 0.834 to 0.935 and AVE values from 0.544 to 0.677. The communication domain demonstrated acceptable reliability (CR = 0.843; AVE = 0.642) and a statistically significant but modest relationship with the exploratory higher-order CT efficacy representation (β = 0.299, t = 5.622, p < 0.001). This study's main contribution is the empirical examination of communication as an explicitly assessed CT-related efficacy domain within physics education, without assuming it as a universal component of CT. The developed scale provides a multidimensional diagnostic profile that may help physics educators identify students’ perceived readiness to engage in computational modeling, problem solving, and scientific communication. Further validation with larger, more diverse samples is recommended to establish the instrument's broader applicability.]]>
