國內運用創造力教學模式對學生創造力影響之後設分析

Meta-Analysis of The Effect of Creativity Teaching Model on Domestic Students' Creativity

本研究旨在後設分析臺灣2007年至2018年間，38篇有關創造力教學模式（包含創造力教學／創造思考教學／創造性問題解決教學）之相關研究作一量化的統整，結合階層線性模式（HLM）的分析技術，希冀對創造力教學相關模式之推動成果獲致較明確的結論。並且進一步探討不同文件性質、樣本性質、不同研究方法、不同研究性質對創造力影響之差異。結果顯示，創造力教學的教學總效果為低度效果量，提供創造力教學融入能提升學生創造力的依據，再次驗證近十年來，以創造力為題的教學方案，對創造力產生正向立即的影響，更能客觀地解釋其成效。除此之外，出版類型、研究設計、領域科目、研究工具、教育階段是影響創造力教學成效的重要調節變項，有對照組的研究設計其效果量大於無對照組，在領域科目上，仍是以「自然與生活科技」此一領域的實驗效果量較佳；在研究工具方面，以威廉斯測驗的研究效果量較佳，表示使用該測驗可以看出創造力或創造思考教學的效果達顯著。而在教育階段，以中年級的效果較佳，低年級的效果最差，也符合創造力發展的歷程。

Purpose: The influence of a teacher's creative teaching on students and the teacher is a popular issue in educational administration and research on curriculum and instruction. Relevant studies are numerous. From the aspect of research methodology, some relevant creative teaching problems are worth thinking. For instance, following questions are rather difficult to answer. Does teachers' creative teaching present effects (influence)? If so, which subject and what type of students is it used for? What kind of creative teaching behaviors would be more effective? What is the generality? How is the variability? Accordingly, in addition to applying traditional literature review to the research, the use of quantitative meta-analysis for integrating past research results might be able to comprehensively inspect the questions. Multilevel analysis has become a popular research issue domestically and internationally. Methods: Based on the innovation and breakthrough of methodology, multilevel meta-analysis is utilized in this study for integrating relevant studies so as to discuss the effectiveness of creative teaching models. In this case, it does not simply achieve the purpose of literature integration but presents the value of statistical methodology application and promotion. As a result, a quantitative integration of 38 pieces of research on creativity teaching models (including creativity teaching / creative thinking teaching / creative problem-solving teaching) in Taiwan during 2007-2018 is preceded meta-analysis in this study. With the combination of hierarchical linear modeling (HLM), it is expected to acquire more definite conclusions about the promotion of creativity teaching models. Furthermore, the effects of various document properties, sample properties, research methods, and research properties on creativity are discussed. Results/Findings: The results reveal low total creativity teaching effect for the reference for the integration of creativity teaching into the promotion of students' creativity. It also verifies the positive and immediate effect of creativity based teaching programs on creativity in the past decade to more objectively explain the effectiveness. Moreover, type of publication, research design, area and subject, research tool, and education stage are important moderators on the effectiveness of creativity teaching. Research design with control groups is comparatively stricter; the experimental effects in the area of “science and technology” are better; and, research with Williams test appears better effects, showing good reliability of the test. In terms of education stage, middle-age graders present better effects, while lower graders show the worst effects. It also conforms to the creativity development process. Conclusions/ Implications: As a result, “meta-analysis” combined with the analysis of Hierarchical Linear Modeling (HLM) is used in this study for the quantitative integration of learning outcomes with creative teaching models (including creativity teaching/creative thinking teaching/creative problem-solving teaching) domestically, expecting to acquire a definite conclusion for the learning effectiveness with creative teaching models.