Programlamanın ötesinde bilgi işlemsel düşünmenin değerlendirilmesi: 5. ve 6. sınıflar için bir test geliştirme çalışması

This study aimed to develop a valid and reliable assessment tool to measure Computational Thinking (CT) skills among 5th and 6th-grade students, based on a four-component theoretical framework: abstraction, algorithm, pattern recognition, and decomposition. An initial pool of 41 items was developed and subjected to expert review for content validity, resulting in a 26-item version. After a small-scale pilot study, the final version of the test was administered to 429 students. Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were conducted to investigate the construct validity. The reliability analysis revealed low internal consistency for each subscale (Cronbach's α = 0.30 - 0.50), suggesting the need for more items per component. Additionally, Multidimensional Item Response Theory (MIRT) analysis under the 2-Parameter Logistic Model provided detailed information on item discrimination and difficulty. While the revised model showed acceptable fit, the low internal consistency across subscales limits the tool's current utility for diagnosing individual CT components. Nevertheless, the study contributes a promising prototype for holistic CT assessment, and the findings offer critical insights into the empirical challenges of translating theoretical CT constructs into psychometrically sound tools in unplugged contexts. These findings highlight the need for further refinement of both the theoretical model and item pool to support the development of valid and reliable instruments for CT assessments in educational settings.