<p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">看了论坛对<span lang="EN-US">SPSS</span>的讨论很受益，但对于主成分分析和因子分析的异同（参考文献１），还有以下疑惑之处：<span lang="EN-US">
<p></p></span></span></p><p></p><p></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">变量A已分为四种类型，每种类型下设有５道题，希望从调查问卷<span lang="EN-US">(likert 5</span>级量表<span lang="EN-US">)</span>的分析中能够检验此四种类型分类的合理性。那么，</span></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;"><span lang="EN-US"><p></p></span></span></p><p></p><p></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span lang="EN-US" style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">(1)</span><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">虽然两者都是对协差阵的逼近来达到数据变量的精简的<span lang="EN-US">(</span>最后都能得到<span lang="EN-US">4</span>种因子<span lang="EN-US">)</span>，但究竟用哪种方法更可取？</span></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;"><span lang="EN-US"><p></p></span></span></p><p></p><p></p><p class="MsoNormal" align="left" style="MARGIN: 0cm 0cm 0pt; TEXT-ALIGN: left; mso-layout-grid-align: none;"><span lang="EN-US" style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">(2)</span><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">大多数教材上都是如此说，“单击<span lang="EN-US">Analyze-Data Reduction-Factor</span>命令，设置因素分析主对话框，单击<span lang="EN-US">Extraction</span>按钮，按照系统默认值抽取特征值</span><span lang="EN-US" style="COLOR: black; mso-bidi-font-size: 10.5pt; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt; mso-ascii-font-family: 宋体;"><font face="Times New Roman">≥</font></span><span lang="EN-US" style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">1</span><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">的因素”；但在参考文献２，通过对标准化后的变量进行抽取特征值</span><span lang="EN-US" style="COLOR: black; mso-bidi-font-size: 10.5pt; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt; mso-ascii-font-family: 宋体;"><font face="Times New Roman">≥</font></span><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">０的因素。用这两种方法分别抽取变量标准化前及后的数据，得出来的结果完全不同，是怎么回事？后种方法甚至出现了错误？</span></p><p class="MsoNormal" align="left" style="MARGIN: 0cm 0cm 0pt; TEXT-ALIGN: left; mso-layout-grid-align: none;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;"><span lang="EN-US"><p></p></span></span></p><p></p><p></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span lang="EN-US" style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">(3)KMO</span><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">值越大时，表示变量间的共同因素越多，越适合进行因素分析。它与哪些值有关？与<span lang="EN-US">Correlation Matrix</span>的有关吗？</span></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;"><span lang="EN-US"><p></p></span></span></p><p></p><p></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">新手发文，说得比较啰嗦，请各位高手指教呵。</span></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;"><span lang="EN-US"><p></p></span></span></p><p></p><p></p><p class="MsoNormal" style="MARGIN: 0cm 0cm 0pt;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">参考文献１：林海明<span lang="EN-US">,</span>张文霖<span lang="EN-US">.</span>主成分分析与因子分析的异同和<span lang="EN-US">SPSS</span>软件<span lang="EN-US">--</span>兼与刘玉玫、卢纹岱等同志商榷<span lang="EN-US">[J].</span>统计研究<span lang="EN-US">,2005.(3):65-69 <p></p></span></span></p><p></p><p></p><p class="MsoNormal" align="left" style="MARGIN: 0cm 0cm 0pt; TEXT-ALIGN: left; mso-layout-grid-align: none;"><span style="COLOR: black; FONT-FAMILY: 宋体; mso-bidi-font-size: 10.5pt; mso-hansi-font-family: 'Times New Roman'; mso-bidi-font-family: 宋体; mso-font-kerning: 0pt;">参考文献<span lang="EN-US">2</span>：<span lang="EN-US">Xiaowenzi22 </span>与<span lang="EN-US">pinksss </span>共同制作<span lang="EN-US">-SPSS </span>中主成分分析的基本操作<span lang="EN-US">
<p></p></span></span></p><p></p><p></p><br/><br/>

[此贴子已经被作者于2007-10-22 1:59:38编辑过]