假定方差齐同时的两两比较方法选项 Univariate: Post Hoc Multiple Comparisons: for Observed Means Factor(s) Post Hoc Tests for: Continue 因素 厂别 别一 性别 Cancel Help Equal Variances Assumed 厂LSD 厂SNK 厂 Waller-Duncan 两两比较检验因素 M Bonferroni 厂 Tukey Type IType ll Error Ratio: 100 厂 Sidak 厂 Tukey,s-b 厂 Dunnett 厂 Scheffe 厂 Duncan Control Category R-E-G-YF Hochberg s gT2 Test 厂 R-E-G-Y Q Gabriel C 2-sided C< Control C)Control Equal Variances Not Assumed y Tamhane's T2 Dunnett's T3 t Games-Howell Dunnett's C 假定方差不齐时的两两比较方法选项
假定方差齐同时的两两比较方法选项 假定方差不齐时的两两比较方法选项 因 素 两 两 比 较 检 验 因 素
估计边缘均数 显示边缘均数 Univariate: Options Estimated Marginal Means Factor[s] and Factor Interactions: Display Means for: 因素和因素交互效应栏 LOVERALL 厂别*性别 厂别 性别 厂别性别 厂 Compare main effects Confidence interval adjustment LSD (none) Display Descriptive statistics V Homogeneity tests Estimates of effect size v Spread vS level plot observed power 厂 Residual plot 观察 厂 Parameter estimates Lack of fit Contrast coefficient matrix General estimable function 效能 Significance leyel:1.05 Confidence intervals are 95% Continue Cancel Help 设定检验水准
估计边缘均数 显示边缘均数 设定检验水准 观察 效能 因 素 和 因 素 交 互 效 应 栏
Between-Subjects Factors 因素别和因素性及其各水平值和 Value Label N 厂别1.00绣品厂 12 例数 200电视机厂 9 3.00蓄电池厂 性别1.00男 13 2.00女 因素r和因素性的描述性统计 Descriptive statistics Dependent variable:血原卟啉 厂别 性别 Mean Std. Deviation 绣品厂男 31.425 16.1372 女 31.6375 10.1383 Tota315667 11.6808 「电视机厂男 24.6800 10.0380 女 28.7500 9.8578 Total264889 9.5616 蓄电池厂男167.3000 27.9205 48254949 女 136.6000 32.9515 Tota1460462 33.7084 13 Total 男 70.6385 69.2871 13 76.0714 58.0616 21 Total739941 61.6118 34
Between-Subjects Factors 绣品厂 1 2 电视机厂 9 蓄电池厂 1 3 男 1 3 女 2 1 1.00 2.00 3.00 厂 别 1.00 2.00 性 别 Value Label N Descriptive Statistics Dependent Variable: 血原卟 啉 31.4250 16.1372 4 31.6375 10.1383 8 31.5667 11.6808 1 2 24.6800 10.0380 5 28.7500 9.8578 4 26.4889 9.5616 9 167.3000 27.9205 4 136.6000 32.9515 9 146.0462 33.7084 1 3 70.6385 69.2871 1 3 76.0714 58.0616 2 1 73.9941 61.6118 3 4 性 别 男 女 Total 男 女 Total 男 女 Total 男 女 Total 厂 别 绣品厂 电视机 厂 蓄电池 厂 Total Mean Std. Deviation N 因素厂别和因素性别及其各水平值和 例数 因素厂别和因素性别的描述性统计
Levene's Test of Equality of Error Variances 方差齐性检验 Dependent variable:血原卟啉 F df1 df2 P<0.05,可认为各格子总体方 3646 280121 Tests the null hypothesis that the error variance of 差不齐同 the dependent variable is equal across groups. a. Design: ntercept+厂别+性别+厂别*性别 方差分析表 Tests of Between-Subjects Effects Dependent variable:血原卟啉 Source of squares Mean square Parameter Power Corected Model 112048.135 522409.627 47462 000237.311 1.000 Intercept 148999409 1148999.409315572 000315572 1.000 厂别 106415.205 253207.603112691 000225381 1.000 性别 588.380 588.380 1.246 274 1.246 厂别*性别 1904.449 952.225 2.017 152 4.034 381 检验效能很大 Error 13220.384 28 472.157 Total 311422.920 Corrected Total.519 a. Computed using alpha =.0 b R Squared=. 894(Adjusted R Squared = 876) ◇因素rP<0.0005,可认为不同厂别之间职工的血原卟啉全不等或不全等 ◇因素性P>0.05,可认为不同性别之间职工的血原卟啉相等 心因素别性P>0.05,可认为厂别与性别之间不存在交互作用
Levene's Test of Equality of Error Variancesa Depe ndent Variable: 血原卟啉 3.646 5 2 8 .012 F df1 df2 Sig. Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept+厂 别+性 别+厂 别 * 性 别 Tests of Between-Subjects Effects Dependent Variable: 血原卟啉 112048.135b 5 22409.627 47.462 .000 237.311 1.000 148999.409 1 148999.409 315.572 .000 315.572 1.000 106415.205 2 53207.603 112.691 .000 225.381 1.000 588.380 1 588.380 1.246 .274 1.246 .190 1904.449 2 952.225 2.017 .152 4.034 .381 13220.384 2 8 472.157 311422.920 3 4 125268.519 3 3 Source Corrected Model Intercept 厂 别 性 别 厂 别 * 性 别 Error Total Corrected Total Type III Sum of Squares d f Mean Square F Sig. Noncent. Parameter Observed Powera a. Computed using alpha = .05 b. R Squared = .894 (Adjusted R Squared = .876) 方差齐性检验 方差分析表 P<0.05，可认为各格子总体方 差不齐同 ❖因素厂别P<0.0005，可认为不同厂别之间职工的血原卟啉全不等或不全等 ❖因素性别P>0.05，可认为不同性别之间职工的血原卟啉相等 ❖因素厂别*性别P>0.05，可认为厂别与性别之间不存在交互作用 检 验 效 能 很 大
边缘均数表 厂别*性别 Dependent variable:血原卟啉 95% Confidence interval 厂别 性别 Mean Std. Error Lower Bound Upper Bound 31.425 10.865 9.170 53.680 7.682 15.901 电视机厂 24.680 28.75010.865 6.495 51.005 蓄电池厂 女男女男女 16730010.865 145.045 89555 136.600 7.243 121.763 151437
厂 别 * 性 别 Dependent Variable: 血原卟 啉 31.425 10.865 9.170 53.680 31.637 7.682 15.901 47.374 24.680 9.718 4.774 44.586 28.750 10.865 6.495 51.005 167.300 10.865 145.045 189.555 136.600 7.243 121.763 151.437 性 别 男 女 男 女 男 女 厂 别 绣品厂 电视机 厂 蓄电池 厂 Mean Std. Error Lower Bound Upper Bound 95% Confidence Interval 边缘均数表