【作 者】：
杨九民， 陈 辉， 杨文蝶， 李 妍， 喻邱晨， 皮忠玲
【关 键 词 】：
教学视频； 字幕； 字幕类型； 学习效果； 元分析
【栏 目】：
学习环境与资源
【中文摘要】：
教学视频是否需要添加字幕是许多教学实践者在视频制作后期考虑的重要问题。从目前的实证研究来看，在教学视频中呈现字幕能否促进视频学习效果，字幕对学习效果的影响存在哪些边界条件？已有研究尚未得出一致的结论。为解决上述问题，文章采用元分析方法，对44篇国内外实验和准实验研究文献进行分析。主效应检验发现：在教学视频中呈现字幕能够提升学习者的保持成绩（g=0.283）；然而，字幕对迁移成绩（g=0.069）的促进效果不显著。调节效应检验发现：在保持成绩上，教学视频中字幕的效果受到字幕类型、字幕语言类型、视频类型和视频时长的调节影响。具体而言，关键词字幕、关键词凸显字幕显著优于全字幕，汉语字幕显著优于英语字幕，教师与PPT结合式视频和课堂实录式视频的字幕效果显著优于动画式视频，中等时长视频的字幕效果显著优于长时长视频。在迁移成绩上，字幕的效果受到字幕语言类型、视频类型、知识类型和视频时长的调节影响。具体而言，汉语字幕、课堂实录式视频、陈述性知识、中等时长视频条件下的字幕效果显著优于英语字幕、动画式视频、程序性知识、短时长视频。研究对教学实践有如下启示：教师在设计教学视频时，需综合考虑视频类型、知识类型、视频时长等因素，建议添加关键词字幕或关键词凸显字幕，以提升视频教学质量。
【英文摘要】：
Whether to add subtitles to instructional videos has always been an important issue for many teachers when designing and producing instructional videos. From the current empirical studies, can subtitles in instructional videos promote video learning performance? What are the boundary conditions for the effect of subtitles on learning effectiveness? No consistent conclusions have been drawn from the existing studies. To address these questions, this paper uses a meta-analytic approach to analyze 44 domestic and international experimental and quasi-experimental research literatures. The main effect test finds that presenting subtitles in instructional videos could enhance learners' retention performance (g=0.283). However, subtitles have no significant effect on promoting transfer performance (g=0.069). The moderating effect test finds that the effect of subtitles in instructional videos is moderated by subtitle type, subtitle language type, video type and video duration in terms of retention performance. Specifically, keyword subtitles and keyword highlighting subtitles are significantly better than full subtitles, Chinese subtitles are significantly better than English subtitles. The subtitle effect of teacher-PPT combined videos and classroom recording videos is significantly better than that of animation videos. The subtitle effect of medium duration video is significantly better than that of long duration video. In terms of transfer performance, the effect of subtitling is moderated by subtitle language type, video type, knowledge type, and video duration. Specifically, the effect of Chinese subtitles, classroom-recorded video, declarative knowledge, medium-length video is significantly better than English subtitles, animation video, procedural knowledge, and short-length video. The study has the following implications for teaching practice: when designing instructional videos, teachers need to consider video type, knowledge type, and video duration, and suggest adding keyword subtitles or keyword highlighting subtitles to improve video teaching quality.