医学统计作业及答案分析.docVIP

结果分析结论 刘 欢 临床八年0901班 2204090116 该资料为重复设计资料，故考虑采用重复设计资料的方差分析。 进行“球对称”检查，结果（见图1），=5.173，Mauchly’s W=0.625，P=0.076＞0.01,不拒绝“球对称”假设，采用方差分析的结果。 Mauchlys Test of Sphericity(b) Measure: MEASURE_1 Within Subjects Effect Mauchlys W Approx. Chi-Square Df Sig. Epsilon(a) Greenhouse-Geisser Huynh-Feldt Lower-bound factor1 .625 5.163 2 .076 .727 .871 .500 Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to an identity matrix. a May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table. b Design: Intercept+Group Within Subjects Design: factor1 图1 Tests of Within-Subjects Contrasts Measure: MEASURE_1 Source factor1 Type III Sum of Squares df Mean Square F Sig. factor1 Linear 2.344 1 2.344 13.870 .003 Quadratic .518 1 .518 7.313 .019 factor1 * Group Linear 2.351 1 2.351 13.918 .003 Quadratic .183 1 .183 2.578 .134 Error(factor1) Linear 2.028 12 .169 Quadratic .851 12 .071 图2 由组内差别方差分析表（图2）可得，组内前后测量时间对家兔血清胆固醇浓度影响的检验结果：F=13.918，P=0.003＜0.01，即组内前后测量时间对家兔血清胆固醇浓度影响有差别，差别有统计学意义。 Tests of Between-Subjects Effects Measure: MEASURE_1 Transformed Variable: Average Source Type III Sum of Squares df Mean Square F Sig. Intercept 948.460 1 948.460 4119.613 .000 Group 7.296 1 7.296 31.692 .000 Error 2.763 12 .230 图3 由组间差别分析表（图3）可得，两组组间处理因素对家兔血清胆固醇浓度影响的检验结果：F=31.692，P＜0.01，即组间处理对家兔血清胆固醇浓度影响有差别，差别有统计学意义。 两组数据处理因素与前后测量时间的交互总用检验结果（图4）：F=10.566，P=0.001＜0.01，即两组处理因素与前后测量时间有交互作用。 Tests of Within-Subjects Effects Measure: MEASURE_1 Source Type III Sum of Squares df Mean Square F Sig. factor1 Sphericity Assumed 2.862 2 1.431 11.932 .000 Greenhouse-Geisser 2.862 1.455 1.967 11.932 .001 Huynh-Feldt 2.862 1.742 1.643 11.932 .001 Lower-bound 2.862 1.000 2.862 11.932 .005 factor1 * Group Sphericity Assumed 2.534 2 1.267 10.566 .001 Greenhouse-Geisser 2.534 1.455 1.742 10.566 .002 Huynh-Feldt 2.534 1.7