典型相关分析及其在外语教育研究中的应用

定量实证研究在我国外语教育研究领域取得了长足的进步,但也暴露了方法论方面的一些问题,因而加强外语教育研究方法论的训练仍是一项长期而艰巨的基础性工作。通过介绍典型相关分析(Canonical Correlation Analysis)并结合实例详细讨论其在外语教育研究中的具体应用表明,作为研究两组变量相关关系的多元统计分析方法,典型相关分析在处理两组变量的相关性问题上明显优于简单相关分析、多元回归分析和因素分析。而且,由于外语学习过程的多因素特征,典型相关分析在外语教育研究领域具有广阔的应用前景。

Conducting and Interpreting Canonical Correlation Analysis in Foreign Language Education Research

This article introduces the statistic principles,SPSS operations and an applied example of Canonical Correlation Analysis(CCA).CCA is a multivariate statistical model that facilitates the study of interrelationships among multiple independent variables and multiple dependent variables.In CCA,there are three layers of analysis:(1) to explore whether pairs of canonical variates are significantly related;(2) to find out how the predictor variables relate to their respective canonical variates,and how the criterion variables relate to their respective canonical variates;and(3) to see how the variables on each side relate to the canonical variates on the other side.The article concludes by pointing out the implications and applicability of CCA for foreign language education research.