Implementing a Formative feedback Framework in Kit-Build Concept Mapping: Evidence from Three Classroom Trials

Abstract

Kit-Build Concept Map (KBCM) has been established as an effective tool for formative assessment, with previous research demonstrating its capability to provide automated assessment, visualize student understanding, and support instructional feedback. However, while formative assessment using KBCM has been achieved, a systematic method for implementing pedagogically grounded feedback based on its analytics has not been well established. This study proposes a formative feedback framework and reports its classroom implementation and evaluation. In this framework, teachers prepare proposition-to-text dictionaries and tiered feedback questions targeting misconceptions identified through missing links in student maps. The feedback is delivered in progressive stages, encouraging students to revisit source text, correct single-proposition errors, and integrate multiple propositions for deeper understanding. The framework was implemented in three classroom trials with undergraduate informatics students (n=16, n=10, n=10). Although the effect of feedback alone varied depending on content complexity, large overall learning gains were observed from the reading phase to the final feedback stage in all trials (effect sizes: 0.895, 0.878, 0.886, p<0.05). These results indicate that the implementation of the proposed framework can effectively enhance student understanding by systematically addressing misconceptions through targeted, supportive feedback.