Levene's Test of equality of Error Variances 方差齐性检验 Dependent variable:红细胞数 P=1.000,按α=0.05检验 df1 df2 000 1.000 水准,接受H。,可认为各 Tests the null hypothesis that the emor variance of the dependent variable is equal across groups. 格子方差齐同 a. Design: Intercept+A+B+A * B 方差分析表 Tests of Between-Subjects Effects 今因素A Dependent variable:红细胞数 Type lll sum P<0.0005,可{ra of Squares Mean Square Eta squared cted model 2.963 98.750 000 974 认为使用和不 19.507 195071950.750 000 996 1.688 1688168 000 955 使用A药的红 908 000 919 821 细胞增加总体|mo 8.000E02 81.000E-02 ot al 22.550 均数不等。 Corrected total 3.043 a. R Squared = 974 (Adjusted R Squared =.964) ◇同上道理推出:使用和不使用B药的红细胞增加总体均数不 等。A药和B药存在交互作用Levene's Test of Equality of Error Variancesa Dependent Variable: 红细胞 数 .000 3 8 1.000 F df1 df2 Sig. Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept+A+B+A * B Tests of Between-Subjects Effects Dependent Variable: 红细胞数 2.963a 3 .988 98.750 .000 .974 19.507 1 19.507 1950.750 .000 .996 1.688 1 1.688 168.750 .000 .955 .908 1 .908 90.750 .000 .919 .368 1 .368 36.750 .000 .821 8.000E-02 8 1.000E-02 22.550 1 2 3.043 1 1 Source Corrected Model Intercept A B A * B Error Total Corrected Total Type III Sum of Squares d f Mean Square F Sig. Eta Squared a. R Squared = .974 (Adjusted R Squared = .964) 方差齐性检验 方差分析表 P=1.000 ，按=0.05检验 水准，接受H0，可认为各 格子方差齐同。 ❖同上道理推出：使用和不使用B药的红细胞增加总体均数不 等。A药和B药存在交互作用。 ❖因素A P<0.0005，可 认为使用和不 使用A药的红 细胞增加总体 均数不等