探討心智模型導向問題解決教學模式對國小學生問題解決能力之影響

Abstract

本研究目的在探討不同問題解決教學模式對學生科學概念理解、科學推理能力、問題解決能力的影響，及科學概念理解、科學推理能力、心智模型對問題解決能力影響的差異，並進而探討學生心智模型與其問題解決能力的關係。 本研究採用準實驗設計，研究對象為台灣北部某國小五年級四個班級的108名學生。實驗組學生（54人）接受「心智模型導向之問題解決教學」，對照組學生（54人）接受「一般問題解決教學」。教學主題為「熱的傳播與保溫」、「聲音與樂器」，每單元問題解決授課時數九節（三週），二單元共十八節（六週）。 研究結果顯示實驗組與對照組學生經過問題解決教學後，科學概念理解測驗成績、科學推理能力測驗成績並無顯著差異，但實驗組的問題解決能力測驗成績、心智模型分數顯著優於對照組。其次，實驗組不論在後測或追蹤測的解決方法成功整合人數皆高於對照組，可見實驗組學生整合能力明顯優於對照組且持久。 此外，經由逐步迴歸（stepwise regression）分析，心智模型前測、後測、追蹤測對問題解決能力前測、後測、追蹤測都具有最高的解釋力。進而由問題解決能力測驗質性分析結果發現，實驗組學生經過教學後持有較完整心智模型的百分率較高，而且持有較完整的心智模型的學生提出較正確有效的解決方法。 學習歷程分析結果顯示，二組學生在不同問題間分數的高低趨勢一致。其次，實驗組與對照組學生在單一心智模型熟悉問題、遷移問題的分數皆未達顯著差異，但問題間達顯著差異；同時在六次問題解決的分數亦無顯著差異，但問題間達顯著差異。

This study attempts to explore the impact of mental model oriented scientific problem solving on students learning of science in terms of their problem solving ability, scientific concept understanding, and scientific reasoning ability. This study further explores the relationships between students’ mental model and their problem solving ability. The design of this study is a one factorial quasi-experiment design. There were 108 fifth-grade students involved in this study from four intact classes in north part of Taiwan. There were 54 students in the experimental group who received mental model oriented scientific problem solving, while the other 54 students in the control group who received general scientific problem solving for eighteen classes period during a semester. The three tests of the science concept test, scientific reasoning test and problem solving test were administered for both experimental and control group students before, directly after, and after four weeks of the study as pre-, post- and retention-test. Both group students problem solving worksheet were also collected. Results indicated that there is no significant difference between two groups, regardless of science concept test and scientific reasoning test, while the experimental group significantly outperformed than to the control group in problem solving test. In addition, the experimental group generated a greater amout of integrating solution, regardless of post- and retention-test. Results also indicated that mental model is the best predictor for problem solving ability. Problem solving test were further analyzed through qualitative method which clearly indicate that experimental group students provided greater percentage of higher level mental model, and the students with higher level mental model can provide more higher levels of problem solving solution.