4
$\begingroup$

This question already has an answer here:

I have data consisting of two measures per subject (variable score, measured at Day 0, and Day1 per each subject). There are two groups, and each subject belongs to one of them (variable Group). I would like to know if there are differences in the score values between these two groups of subjects. I tried aov and lme in R, and I got completely different results.

Why is the Group:Day effect significant using aov, and it is not significant using lme? Here is the code I used:

> demo1.aov <- aov(score ~ (Group*Day) + Error(Subject/Day), data = demo1)
> summary(demo1.aov)


Error: Subject
       Df Sum Sq Mean Sq F value Pr(>F)
       Group      1    0.2   0.198    0.15    0.7
       Residuals 50   65.9   1.318               

Error: Subject:Day
       Df Sum Sq Mean Sq F value Pr(>F)  
Day        1    0.3    0.34    0.45  0.507  
Group:Day  1    4.0    4.02    5.29  0.026 *
Residuals 50   38.0    0.76                 
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> demo1.lme<-lme(score ~ (Group*Day), random = ~ 1|Subject/Day, data = demo1)
> anova(demo1.lme)

    numDF denDF F-value p-value
    (Intercept)     1    50   0.394   0.533
    Group           1    50   0.078   0.781
    Day             1    50   1.225   0.274
    Group:Day       1    50   2.064   0.157
$\endgroup$

marked as duplicate by mkt, mdewey, Michael Chernick, kjetil b halvorsen, usεr11852 Jan 16 at 23:32

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.