• 作者： 蕭佳純 (通訊作者：蕭佳純)
• 作者服務機構： 國立臺南大學教育學系
• 中文摘要： 本研究以1,079名國小學童為對象，在經過四次長期追蹤調查之後，利用階層線性模式分析三個層次因素對於科學創造力成長速率的影響。研究結果發現，學生科學創造力在初階段具有個別差異且呈現先上後下的發展。參與科展的學生其科學創造力成長曲率高於未參加科展的學生；科學學習動機中的自我效能對學生科學創造力的初始狀態及成長曲率具有影響。教師創意角色認定對學生初階段的科學創造力有直接影響，也會強化成就動機與科學創造力之間的關係。若教師表現出多元的創意教學行為，也會正向調節動機對科學創造力線性成長率的影響；並且，教師若展現心胸開放、意義學習的創意教學行為，也會強化學生內在價值與科學創造力線性成長速率之間的關係。
• 英文摘要： Having 1,079 elementary pupils from grades four to six as the research subjects, longterm investigations has been proceeded four times and the growth model in Hierarchical Linear Modeling is utilized for analyzing the effects of factors in hierarchy one (time, creativity tendency), hierarchy two (gender, motivation, and experience in attending science show), and hierarchy three (teachers’ creative role identity and creative instruction) on the development of creativity. The research findings show that students’ scientific creativity presents individual differences in the preliminary stage and first-up-then-down development, and curiosity in creativity tendency appears significantly negative effects on creativity. Moreover, students participating in science shows reveal
  higher scientific creativity development curve than the ones not participating in science shows, and self-efficacy in motivation shows effects on the initial state and dynamic development of students’ scientific creativity. Teachers’ creative role identity presents direct effects on students’ scientific
  creativity in the preliminary stage and could reinforce the relations between the achievement motivation and the scientific creativity of students. Furthermore, a teacher’s multiple creative instructions could positively adjust the effects of students’ motivation on the linear development of creativity. Besides, a teacher presenting creative instructions with open mind and meaningful learning could also reinforce the relations between students’ inner value and the linear development of scientific creativity.
• 中文關鍵字： 科學創造力、階層線性模式、縱貫性分析
• 英文關鍵字： Scientific Creativity, Hierarchical Linear Modeling, Longitudinal Analysis