Development of A Scientific Creativity Assessment for Physics University Students

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

This study develops and validates a Science Creativity (SC) assessment instrument for the university-level physics domain. The instrument is designed as open-ended tasks within the mechanics domain. Item development is grounded in the cognitive aspects of physics learning, including knowledge restructuring, orchestration of multiple representations, depth of cognitive processing, as well as reasoning and thinking strategies. All tasks are positioned as a complementary package of activities: while each task may highlight specific cognitive aspects, the ensemble is directed toward building a holistic modeling experience so that SC is reflected as a unified performance. The quality of the measurement model was evaluated via Confirmatory Composite Analysis (CCA) within a PLS-SEM framework using higher-order construct modeling. CCA results indicate that the majority of indicators perform strongly. Construct reliability is generally adequate, with one dimension appearing very high due to highly homogeneous content. In general, construct coherence is stronger in dimensions based on technical products, whereas dimensions requiring cross-representational shifts tend to exhibit greater variation. Collinearity diagnostics reveal strong correlations in several blocks, which are interpreted as a consequence of integrated cognitive processes in physics. Discriminant validity was confirmed, forming a coherent SC construct. This instrument has a direct impact by providing assessments aligned with physics practice, which can be used to map SC profiles, provide diagnostic feedback, and evaluate physics learning. The feasibility of the instrument as a measure of SC linked to the cognitive aspects of physics learning serves as a basis for further refinement and cross-context validation.

Article activity feed