Effects of Vocabulary Learning on Listening of English Sentences for Japanese Learners with Different Proficiency: A Behavioral Analysis Approach

本项目研究了学习新词汇和训练识别新闻句子中目标的效果。在对外英语研究中，由于要采用行为分析领域的单案例设计，因此在实验开始前，需要先找出多项听力测试的效度和信度。在确认评估工具的效度和信度后，采用计算机训练程序（Ki-Ki-Mi-Mi-chan）进行多基线研究设计。在实验1中，33名日本大学生参加了16次15分钟的英语测试。在33名学生中，有8名学生同意每周四次在课外使用Ki-Ki-M-Mi-chan。他们被训练在听包含目标的句子时学习新单词并识别它们。他们的词汇识别分数和听力分数与其余25名没有接受培训的学生的表现进行了标准化。结果表明，干预对词汇识别和听力得分均有显著影响；然而，对听力的影响仅限于训练句子。实验2考察了部分训练的效果，即词汇学习的第一阶段是在没有语境（即句子训练）的情况下进行训练。研究发现，干预效果仅在词汇识别（尽管学习量比实验1中的少）上明显，而在听力分数上不明显。然后用不熟练的学习者测试训练效果。对词汇识别的影响显著；然而，5名学生中有2名学生的听力成绩没有显示出训练的积极效果。这表明我们需要区别对待这两类效果不大的听众，并需要进一步的研究来确定外语听力的效果方法。根据目前的研究项目，已发表论文3篇。包括这三个出版物的最终报告作为Toya的博士论文，于2007年10月提交给日本大学，并已被接受。少

This project investigated the effect of learning new vocabulary and being trained to recognize the targets in news sentences. As the single case design used in the field of Behavioral Analysis was to be employed to the English as a Foreign Language research, the validity and the reliability of multiple listening tests was to be sought out before the experiments. After the validity and reliability of evaluating tools were confirmed, three experiments with a computer training program (Ki-Ki-Mi-Mi-chan) in multi-baseline research design. In Experiment 1, 33 Japanese University students took 15-minute English tests for a total of 16 sessions. Of the thirty-three, eight students agreed to use Ki-Ki-M-Mi-chan four times a week outside the class. They were trained to learn new words and recognize them while listening to the sentences containing the targets. Their vocabulary recognition scores and listening scores were standardized in relation to the performance of remaining 25 students who to … More ok the tests without the training. The results showed that the effects of intervention were significant for both vocabulary recognition and listening scores; however, the effects on listening was found to be limited to the training sentences. Experiment 2 investigated the effect of partial training, that is, the first phase in which the vocabulary learning was training without the contexts (i. e., training sentences). The intervention effect was found to be apparent only for vocabulary recognition (though the amount of learning was less than that in Experiment 1) and not in listening scores. Then the training effect was tested with less proficient learners. The effects on vocabulary recognition were significant for the 5 participants; however, the listening scores of 2 students out of five did not show the positive effects of training. This indicated that we need to treat the listeners categorized as those two with little effects differently and further research is necessary in order to determine what would be the effect method of foreign language listening. Three papers have been published based on the current research projects. The final report including these three publications served as a PhD dissertation of Toya, submitted to Nihon University in Oct. 2007 and has been accepted. Less