國小學生體育前瞻性與反應性後設認知量表之建構與驗證

Abstract

在體育學習中，後設認知歷程透過特定的學習行為來激發自我調節，進而影響學習者在運動中的後設認知表現。本研究旨在建構並驗證一份適用於國小高年級學童之體育後設認知量表，並以Luke (1998) 的後設認知能力模型 (metacognitive ability) 和Chevalier等人 (2015) 的雙重控制機制 (Dual Mechanisms of Control) 為理論架構。研究一旨在建構體育學習情境中的後設認知模型，綜整既有後設認知理論，並參考Luke與Hardy (1999) 援引Flavell (1979) 所提出的理論觀點，將體育課中的後設認知界定為在動態環境中，個體調節內在思維與外在行為表現的綜合性機制。經由專家效度檢核後，確認模型核心內涵，並凸顯前瞻性與反應性後設認知的構面，以描繪學習者在體育課中展現的相應行為。研究二旨在發展具有良好信度與效度的兩種控制機制體育後設認知量表。經因素分析後，前瞻性後設認知量表可解釋總變異量為43.64%，反應性後設認知量表則為52.10%。前瞻性後設認知量表的Cronbach's α為 .93，反應性後設認知量表則為 .95；組合信度 (CR) 介於 .76至 .85，平均變異萃取量 (AVE) 介於 .42至 .52，顯示本量表具備良好的內部一致性與可接受的建構效度。研究三旨在運用結構方程模式分析兩種控制策略的後設認知與內在動機和終身運動傾向的關聯證據，同時瞭解不同背景變項學生之差異。綜合上述三個系列研究，描繪出體育後設認知之內涵並提供量化證據，為體育教學研究與後設認知概念提出新的思考方向。

In physical education (PE), metacognitive processes are activated through specific learning behaviors, fostering self-regulation and influencing students’ metacognitive performance during physical activities. This study aimed to develop and validate a metacognition scale for upper elementary students within PE contexts, grounded in Luke’s (1998) model of metacognitive ability and Chevalier et al.’s (2015) Dual Mechanisms of Control theory. Study 1 constructed a theoretical model of metacognition in PE by synthesizing existing metacognitive theories. Drawing on Luke and Hardy’s (1999) interpretation of Flavell’s (1979) framework, PE metacognition was defined as an integrated mechanism for regulating internal thoughts and external behaviors in dynamic environments. Expert validation confirmed the model’s core structure and emphasized both proactive and reactive metacognitive dimensions, thereby illustrating learners’ corresponding behaviors in PE classes. Study 2 focused on the development of two reliable and valid subscales aligned with the dual control framework. Exploratory factor analysis revealed that the proactive metacognition scale explained 43.64% of the total variance, while the reactive metacognition scale accounted for 52.10%. The Cronbach’s α was .93 for the proactive subscale and .95 for the reactive subscale; composite reliability (CR) ranged from .76 to .85, and average variance extracted (AVE) ranged from .42 to .52, indicating good internal consistency and acceptable construct validity. Study 3 used structural equation modeling to examine the relationships among metacognitive control strategies, intrinsic motivation, and lifelong physical activity tendencies, while also exploring differences based on student background variables. Together, the three studies outline the structure of PE metacognition and offer empirical evidence, contributing new insights for both metacognition research and PE instructional practice.