本研究運用『後設分析法』(meta-analysis)，探討電腦網路學習相對於傳統教學，對學生學習成效之影響，並進一步探討不同文件性質、受試者性質、研究方法、與研究案性質之變項是否會造成學習成效之差異，以及學習後態度之影響。 本研究資料主要源自五大類。蒐集之研究案共22篇。各篇研究案之量化數據皆轉化為效果大小(Effect Size)，各變項則進一步進行單因子變異數分析。各研究案數據資料經統計分析後，結果歸納如下： 一、整體成效： 1.網路教學應用於學生認知學習上可帶來中等的學習成效，顯示電腦網路學習確實優於傳統教學(ES=0.4857)。 二、各變項之分析： 1.僅有『學校位置』一項變項達顯著性差異；網路教學在南部與東部地區有較佳的學習成效，愈靠近北部，網路學習效果愈不明顯。 三、學習者對網路教學的態度： 研究結果肯定資訊融入教學確實能增進學生的學習成效，並增進學生透過網路學習的信心，產生正向學習的學習態度。 最後根據本研究結果，提出相關實驗研究之建議。
A meta-analysis was performed to synthesize existing research comparing the effects of web-based learning with traditional instruction on students' achievement, and attitudes toward computer networked learning after experiments. The studies were located from five sources. Totally, 22 studies were collected. The quantitative data of the collected studies were transformed into effect Size (ES). Each studied variables were analyzed by one-way ANOVA. The main findings of this study can be summarized as follows: 1. The whole effects suggest that computer networked learning more effective than traditional instruction on both cognitive and affective learning. Their ES are 0.4857 and 0.3448, respectively. 2. For the studied variables, the results show that only the variable of school area had a statistically significant impact on the mean ES. 3. Finally, students held more positive attitude toward web-based learning after experiments. Based on these findings, implication for future studies are suggested.
