資源描述:
《經(jīng)典spss因子分析.pdf》由會員上傳分享,免費(fèi)在線閱讀,更多相關(guān)內(nèi)容在教育資源-天天文庫。
1、[例11-1]下表資料為25名健康人的7項(xiàng)生化檢驗(yàn)結(jié)果,7項(xiàng)生化檢驗(yàn)指標(biāo)依次命名為X1至X7,請對該資料進(jìn)行因子分析。X1X2X3X4X5X6X73.763.660.545.289.7713.744.788.594.991.3410.027.5010.162.136.226.144.529.842.172.731.097.577.287.0712.661.792.100.829.037.082.5911.764.546.221.285.513.981.306.925.337.302.403.270.620.443.367.638.848.398.747.003
2、.3111.683.534.761.129.649.491.0313.5713.1318.522.359.731.331.009.879.8711.063.708.592.981.179.177.859.912.627.125.493.689.722.643.431.194.693.012.175.982.763.552.015.511.341.275.814.575.383.431.661.611.572.801.782.093.725.905.761.558.845.407.501.979.849.271.5113.609.0212.671.758.394
3、.922.5410.053.965.241.434.944.381.036.686.499.062.817.232.301.777.794.395.372.279.467.311.0412.0011.5816.182.429.555.354.2511.742.773.511.054.944.524.508.071.792.101.298.213.082.429.103.754.661.729.416.445.1112.502.453.100.9111.2.1數(shù)據(jù)準(zhǔn)備激活數(shù)據(jù)管理窗口,定義變量名:分別為X1、X2、X3、X4、X5、X6、X7,按順序輸入相應(yīng)數(shù)值
4、,建立數(shù)據(jù)庫,結(jié)果見圖11.1。圖11.1原始數(shù)據(jù)的輸入11.2.2統(tǒng)計(jì)分析激活Statistics菜單選DataReduction的Factor...命令項(xiàng),彈出FactorAnalysis對話框(圖11.2)。在對話框左側(cè)的變量列表中選變量X1至X7,點(diǎn)擊?鈕使之進(jìn)入Variables框。圖11.2因子分析對話框點(diǎn)擊Descriptives...鈕,彈出FactorAnalysis:Descriptives對話框(圖11.3),在Statistics中選Univariatedescriptives項(xiàng)要求輸出各變量的均數(shù)與標(biāo)準(zhǔn)差,在CorrelationMa
5、trix欄內(nèi)選Coefficients項(xiàng)要求計(jì)算相關(guān)系數(shù)矩陣,并選KMOandBartlett’stestofsphericity項(xiàng),要求對相關(guān)系數(shù)矩陣進(jìn)行統(tǒng)計(jì)學(xué)檢驗(yàn)。點(diǎn)擊Continue鈕返回FactorAnalysis對話框。圖11.3描述性指標(biāo)選擇對話框點(diǎn)擊Extraction...鈕,彈出FactorAnalysis:Extraction對話框(圖11.4),系統(tǒng)提供如下因子提取方法:圖11.4因子提取方法選擇對話框Principalcomponents:主成分分析法;Unweightedleastsquares:未加權(quán)最小平方法;Generalize
6、dleastsquares:綜合最小平方法;Maximumlikelihood:極大似然估計(jì)法;Principalaxisfactoring:主軸因子法;Alphafactoring:α因子法;Imagefactoring:多元回歸法。本例選用Principalcomponents方法,之后點(diǎn)擊Continue鈕返回FactorAnalysis對話框。點(diǎn)擊Rotation...鈕,彈出FactorAnalysis:Rotation對話框(圖11.5),系統(tǒng)有5種因子旋轉(zhuǎn)方法可選:圖11.5因子旋轉(zhuǎn)方法選擇對話框None:不作因子旋轉(zhuǎn);Varimax:正交旋轉(zhuǎn);
7、Equamax:全體旋轉(zhuǎn),對變量和因子均作旋轉(zhuǎn);Quartimax:四分旋轉(zhuǎn),對變量作旋轉(zhuǎn);DirectOblimin:斜交旋轉(zhuǎn)。旋轉(zhuǎn)的目的是為了獲得簡單結(jié)構(gòu),以幫助我們解釋因子。本例選正交旋轉(zhuǎn)法,之后點(diǎn)擊Continue鈕返回FactorAnalysis對話框。點(diǎn)擊Scores...鈕,彈出彈出FactorAnalysis:Scores對話框(圖11.6),系統(tǒng)提供3種估計(jì)因子得分系數(shù)的方法,本例選Regression(回歸因子得分),之后點(diǎn)擊Continue鈕返回FactorAnalysis對話框,再點(diǎn)擊OK鈕即完成分析。圖11.6估計(jì)因子分方法對話框11
8、.2.3結(jié)果解釋在輸出結(jié)果窗口中將看到