數位遊戲學習對學生學習成效影響之後設分析

Abstract

因為數位原民的概念，數位遊戲被認為是一種有效的學習方式。然而，根據研究顯示，結果卻是眾說紛紜。基於此，本文運用後設分析為研究工具，搜尋了78個資料庫，初步獲得21,502筆摘要，逐步隨著研究步驟篩選，最後獲得96篇符合研究目的及方法之文章，得到11,898樣本數。此96篇研究樣本依研究特性、樣本特性、研究設計特性、研究案特性、數位遊戲種類及數位遊戲特性六個面向探究，並使用29個調節變項檢驗。本研究獲致結論如下:不論是在學業成就、情意或高層次方面，DGBL 均顯著優於Non-DGBL。 學業成就達中度效果量，情意達中度效果量，高層次達高度效果量。調節變項顯著影響學習效果。因此，基於本研究結果，研究者針對現場教師提出六點建議，針對遊戲設計者提出五點建議，針對未來研究者提出五點建議。

It is assumed that Digital Game-Based Learning is effective because the students now are digital natives. However, the results from previous reviews are ambiguous. This study used meta-analysis as research tool, searched on 78 electronic databases and got 21,502 abstracts at first round search. Followed by the research procedure, 96 articles fitted in the inclusion criteria and yielded 11,898 sample sizes. These 96 articles were carefully coded into six categories: sample characteristics, research design characteristics, program characteristics, digital game categories and digital game characteristics, which 29 moderators were examined. The findings are: DGBL is significantly better than Non-DGBL in academic achievement, affective outcomes and higher level thinking. Academic achievement achieved medium effect size level; affective outcomes achieved medium level; higher level thinking achieved large effect level. Moderators significantly affected the effect of learning. Therefore, based on the research, the researcher gave 16 suggestions: six suggestions for educational field teacher; five suggestions for game designers and five suggestions for future researchers.