Research On The Characteristics Of "Idea Improvement" Forming Deep Learning Under Knowledge Construction

Against the background of educational reform in China,learning science has become the theoretical basis of education reform,and the knowledge construction theory is also an important research direction in the field of learning science.As an innovative teaching method,knowledge construction teaching has attracted more and more attention from educators.At present,research on knowledge construction in China has mainly focused on the study of "idea,dialogue,and roles" in knowledge construction communities.Building on previous research,this study explores the characteristics of deep learning formed by "idea improvement" under knowledge construction from the perspective of " idea improvement ".The main focus of this article is to explore how students form deep learning outcomes through the "idea improvement" path,and to summarize the path characteristics that lead to different deep learning outcomes.The study uses a design-based research method and content analysis to investigate a undergraduate course and two graduate courses in educational technology at a certain 211 normal university and a provincial normal university,all of which use applied knowledge construction teaching method and have instructors with over ten years of experience in knowledge construction teaching.The purpose of this study is to improve the evaluation tool and enhance the ability to analyze learning outcomes.All ideas data from the three courses were sorted in chronological order according to the ideas expressed by each student.Using the SOLO classification evaluation theory as a basis,the Biggs SOLO classification evaluation scale was improved through three rounds of iteration.In the first round,the evaluation tool was used to evaluate data from the graduate course "Learning and Teaching Theory" in 2021,and the tool was adjusted based on content analysis to form the first version of the SOLO classification evaluation tool.The second round of iteration applied the first version of the tool to evaluate data from the graduate course "Research Methods in Educational Technology" in 2020,resulting in further adjustments to the evaluation principles.The third round of iteration applied the second version of the tool to evaluate data from the undergraduate course "Learning Science" in 2018,resulting in the final version of the SOLO classification evaluation tool consisting of four structural levels and eleven evaluation principles.This tool can be used to evaluate the deep learning outcomes of learners in knowledge construction courses.After sorting the student ideas,content analysis was used to track the "idea improvement" path of students with different deep learning outcomes.The idea paths were classified according to the SOLO structural level to obtain the proportion of each idea category in the class,the idea states of each structural level,and the idea improvement situation,thereby summarizing the 13 "idea improvement" characteristics that contribute to deep learning outcomes.Finally,the differences in "idea improvement" characteristics between different structural levels were compared,revealing that the differences between single-point and multi-point "idea improvement" characteristics mainly lie in the spirit of exploration,concentration,and the ability to introduce authoritative materials.The differences between relational and multi-point levels relate to the number of ideas,interest in the problem,and use of other ideas.For extended abstract level ideas,the differences lie in responding to ideas,applying scaffolding,and using keywords.The study identifies areas for improvement and provides a prospect for future research.