構造方程式モデリングによる高校生の言語学習方略使用と言語学習達成の分析 : SILLで得られたデータのより適切な分析と結果の提示方法の提案

抄録

This paper points out problems in previous studies concerning the statistical procedure of analysis and of the way results are presented based on the most popular questionnaire format in Language Learning Strategy researches, namely SILL (Strategy Inventory for Language Learning ver. 7.0 for ESL/EFL). A more appropriate statistical procedure to analyse the data, is proposed called "Structural Equation Modelling" based on the results of both "Exploratory Factor Analysis" and "Confirmatory Factor Analysis" with Oblique Rotation and Maximum Likelihood Method. Presentation with more detailed statistics is also promoted, which allows readers to reanalyse or meta-analyse. In order to validate the proposed method above of analysing and presenting the necessary statistics, a case study was conducted on strategy use of Japanese High School students using data obtained from the SILL. This case study explored which language learning strategy contributes most to the achievement of language learning.