我有以下数据(3x3主题内设计):
subj phase cope mean
1 A a ...
1 A b ...
1 A c ...
1 B a ...
1 B b ...
1 B c ...
1 C a ...
1 C b ...
1 C c ...
2 A a ...
2 A b ...
2 A c ...
2 B a ...
2 B b ...
2 B c ...
2 C a ...
2 C b ...
2 C c ...
...
阶段和应付都是分类的,并且DV(平均值)是连续的。
如何使用此数据在主题ANOVA / lme中运行3-by-3(ABCxabc),我该怎么做事后测试?
在网上搜索后,我发现了以下3种可能的选择:
1. baselinemodel <- lme(m ~ 1, random = ~1 | subj/phase/cope, data = csv_temp, method = "ML")
copemodel <- lme(m ~ cope, random = ~1 | subj/phase/cope, data = csv_temp, method = "ML")
phasemodel <- lme(m ~ phase, random = ~1 | subj/phase/cope, data = csv_temp, method = "ML")
fullmodel <- lme(m ~ phase + cope, random = ~1 | subj/phase/cope, data = csv_temp, method = "ML")
interactionmodel <- lme(m ~ phase * cope, random = ~1 | subj/phase/cope, data = csv_temp, method = "ML")
anova(baselinemodel, copemodel, phasemodel, fullmodel, interactionmodel)
2. m.lme <- lme(m ~ cope * phase, data = csv_temp, random = ~1 | subj)
3. m.lme <- lme(m ~ cope * phase, random = list(subj = pdBlocked(list(~1, pdIdent(~ cope - 1), pdIdent(~ phase - 1)))), data = csv_temp)
anova(m.lme)
他们产生了不同的结果。哪种方法是正确的?以及如何跟进事后测试(目前,这两种方法都无法产生显着的相互作用)?