2005-08-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/704746摘要：平行分析乃因素數取決方法之一，其因素數目正確決定率在以連續資料進行之正交因素的模擬研究中，均優於其他因素數目決定方法。然由於心理學研究的構念間常有關聯，且因素分析多以李克式量表測量之題目進行，故本研究擬就三方面探討平行分析之表現：（1）平行分析於以李克式量尺測量之正交多因素模型之表現，（2）平行分析於連續變項斜交多因素模型之表現，（3）平行分析於以李克式量尺測量之斜交多因素模型之表現。<br> Abstract: Parallel analysis has been shown to be superior to other methods for determination of number of factors in exploratory factor analysis. Past simulation research, however, was limited to factor models with orthogonal factors and continuous measures. Because many constructs studied in psychological research or social sciences are intercorrelated, and variables measured on Likert-type scales have been frequently submitted to factor analysis, the present study is proposed to investigate the performance of parallel analysis in a series of conditions including: (1) performance of parallel analysis on orthogonal factors with indicators measured on Likert-type scales, (2) performance of parallel analysis on oblique factors with continuous variables, and (3) performance of parallel analysis on oblique factors with indicators measured on Likert-type scales.因素數目平行分析李克式量尺Number of factorsparallel analysisLikert-type scales