Repeated-measures ANOVA with no between-factor

Hi all!

I am trying to perform a two-way repeated measures ANOVA by using the function fitrm and ranova. I have two within-factors (e.g. Condition & Side) and no between-subject factors. I tried the code below, based on the different topics on the mathworks help page. Cause I have no between-subjects factor I thought i should use ~1 as constant. However, I keep getting the following error:

Error using RepeatedMeasuresModel.fit (line 1347)

The between-subjects design must have full column rank.

Error in fitrm (line 77)

s = RepeatedMeasuresModel.fit(ds,model,varargin{:});

Error in Statistics_JvdH_versie5 (line 7778)

rm_APMoS_MSZP = fitrm(t_APMoS_MSZP,'CWSaff-CWS2laff~1','WithinDesign',Condition)

Does anyone know how to solve this problem? Or know another way to perform this analysis in matlab.

See code below:

t_APMoS_MSZP = table(MoS_AP_side1_CWS_MSZP,MoS_AP_side2_CWS_MSZP,MoS_AP_side1_FWS_MSZP,...
    MoS_AP_side2_FWS_MSZP,MoS_AP_side1_CWS2_MSZP,MoS_AP_side2_CWS2_MSZP...
    ,'VariableNames',{'CWSside1','CWSside2','FWSside1','FWSside2','CWS2side1','CWS2side2'});

Condition = table([1 1 2 2 3 3]',[1 2 1 2 1 2]','VariableNames',{'Condition' 'Side'});

rm_APMoS_MSZP = fitrm(t_APMoS_MSZP,'CWSside1-CWS2side2~1','WithinDesign',Condition)
[b1_APMoS,A_APMoS,C_APMoS,D_APMoS] = ranova(rm_APMoS_MSZP,'WithinModel','Condition*Side');

Answer 1

[Scott MacKenzie]

You don't need a between-subjects factor to use ranova. Your design has two within-subjects factors, and that's just fine.

I see two minor issues in your code. First, you've used "Condition" both as the name of the within-subjects design and as the name of one of the variables in the design. Probably not a good idea. Below, I changed the name of the design to withinDesign. Second, you need to define the Condition and Side variables as categorical. I added two lines to do this, and got a clean run of your code with some random test data. See below.

% test data
MoS_AP_side1_CWS_MSZP = rand(20,1);
MoS_AP_side2_CWS_MSZP = rand(20,1);
MoS_AP_side1_FWS_MSZP = rand(20,1);
MoS_AP_side2_FWS_MSZP = rand(20,1);
MoS_AP_side1_CWS2_MSZP = rand(20,1);
MoS_AP_side2_CWS2_MSZP = rand(20,1);

t_APMoS_MSZP = table(MoS_AP_side1_CWS_MSZP,MoS_AP_side2_CWS_MSZP,MoS_AP_side1_FWS_MSZP,...
    MoS_AP_side2_FWS_MSZP,MoS_AP_side1_CWS2_MSZP,MoS_AP_side2_CWS2_MSZP...
    ,'VariableNames',{'CWSside1','CWSside2','FWSside1','FWSside2','CWS2side1','CWS2side2'});

withinDesign = table([1 1 2 2 3 3]',[1 2 1 2 1 2]','VariableNames',{'Condition' 'Side'});
withinDesign.Condition = categorical(withinDesign.Condition);
withinDesign.Side = categorical(withinDesign.Side);

rm_APMoS_MSZP = fitrm(t_APMoS_MSZP,'CWSside1-CWS2side2 ~ 1','WithinDesign',withinDesign);
[b1_APMoS,A_APMoS,C_APMoS,D_APMoS] = ranova(rm_APMoS_MSZP,'WithinModel','Condition*Side');

[Jaimy van den Hout]

Thanks for your answer. I changed my script according to your suggestions. It did not work at first, but I found out that my data in t_APMoS_MSZP was in the wrong format (due to changes I made earlier ....).

Thanks for the help, it works fine now!