頁籤選單縮合
題 名 | 典型相關分析簡介=An Introduction to Canonical Correlation Analysis |
---|---|
作 者 | 傅粹馨; | 書刊名 | 教育研究 |
卷 期 | 6 1998.06[民87.06] |
頁 次 | 頁25-40 |
分類號 | 520.28 |
關鍵詞 | 母數統計法; 典型相關係數; Parametric method; Canonical correlation; |
語 文 | 中文(Chinese) |
中文摘要 | 典型相關分析乃是處理兩組變項間的相關,它包含了母數統計方法,換言之,母 數統計法皆是典型相關分析的特例。典型相關分析可藉著雙變項的觀念來表示,簡言之,典 型相關亦即兩個組合分數的相關。與典型相關分析之解釋有關的係數如下:典型相關係數、 典型相關係數平方、加權係數、結構係數、index 係數、 adequacy 係數、重疊係數和共同 性係數。本文以實際研究之資料分析來說明這些係數是如何計算而得,這有助於對典型相關 分析的瞭解。 文中呈現 SAS 與 SPSS 執行結果的部分報表,SAS 比 SPSS 提供了多些的訊 息。 |
英文摘要 | Conventional canonical correlation analysis investigates the degree of relationship between two sets of variables. It subsumes all parametric methods. In other words, all parametric methods are special cases of canonical correlation analysis. Generally speaking, the canonical correlation can be presented in bivariate terms, that is, the bivariate correlation between the two composite score(synthetic score) is a canonical correlation. Several coefficients can be calculated to aid in interpretation, including canonical corelation, squared canonical correlation, function coefficient, structure coefficient, index coefficient, adequacy coefficient, redundancy coefficient, and communality coefficient. An analysis of real data set was employed to illustrate how the coefficients were caculated. The partial output from SAS and SPSS were presented too. Actually, SAS output provided more information than SPSs did. |
本系統中英文摘要資訊取自各篇刊載內容。