New user of RStudio here, beginner at statistics, grateful for any hints : )
I have data from a psychology experiment in which there is a control and an experimental group. Both groups are tested on reaction times (RTs) for locating where a dot appears on a screen (left/right). Before the dot appears, the participants are presented with two images, one on the left and one on the right. The images can be happy, sad, or neutral. The measures are repeated at three different points in time (T0, T1 and T2) :
Emotion Location Time Group RT Score.D
1 Sad Incongruent T0 XP 625 22.83693
2 Happy Congruent T1 CTRL 783 22.83693
3 Sad Congruent T2 XP 578 22.83693
4 Happy Incongruent T0 XP 780 22.83693
5 Sad Incongruent T2 CTRL 566 22.83693
6 Happy Congruent T1 XP 535 22.83693
Here is a subset of my data that can be directly imported in R :
structure(list(Emotion = structure(c(2L, 2L, 2L, 2L, 2L, 2L), .Label = c("Happy",
"Sad"), class = "factor"), Location = structure(c(2L, 2L, 2L,
2L, 2L, 2L), .Label = c("Congruent", "Incongruent"), class = "factor"),
Time = structure(c(1L, 1L, 1L, 1L, 1L, 1L), .Label = c("T0",
"T1", "T2"), class = "factor"), Group = structure(c(2L, 2L,
2L, 2L, 2L, 2L), .Label = c("CTRL", "XP"), class = "factor"),
RT = c(625L, 783L, 578L, 780L, 566L, 535L), Score.D = c(22.83693,
22.83693, 22.83693, 22.83693, 22.83693, 22.83693)), row.names = c(NA,
6L), class = "data.frame")
My supervisor told me to conduct a repeated measures ANOVA with the treatment type (control/experimental) as between-group factor, the D score as an independent variable and the three points in time as the within-group variable.
I am confused as to how I should achieve this in RStudio. Here is my attempt :
Incong.Cong.XP.CTRL.T0T1T2.formatted$Group <- factor(Incong.Cong.XP.CTRL.T0T1T2.formatted$Group)
Incong.Cong.XP.CTRL.T0T1T2.formatted$Time <- factor(Incong.Cong.XP.CTRL.T0T1T2.formatted$Time)
Incong.Cong.XP.CTRL.T0T1T2.formatted$Emotion <- factor(Incong.Cong.XP.CTRL.T0T1T2.formatted$Emotion)
rANOVA.BA.ech1 <- with(data = Incong.Cong.XP.CTRL.T0T1T2.formatted, aov(Score.D ~ Time*Emotion + Error(Group/(Time*Emotion))))
summary(rANOVA.BA.ech1)
This works, but I cannot compute Mauchly's sphericity test on this, which I would very much like to do.
The package ez
allows for a rANOVA together with Mauchly's test. Here is the code I tried :
ezANOVA(Incong.Cong.XP.CTRL.T0T1T2.formatted, dv = .(Score.D), wid = .(generated_uid), within = .(Time,Emotion), between = .(Group), type = 1)
But I get this error message :
One or more cells is missing data. Try using ezDesign() to check your data.
Yet there are no NA cells in the data...
What's wrong here ?