How to perform 2-way repeated-measures Manova using RepeatedMeasuresModel?

Hi,

I'm currently stuck on performing a two-way repeated-measures Manova in Matlab using the RepeatedMeasuresModel. My data set consists of six units, from each of which I measured three properties, which might be highly correlated. Each unit was tested using two within-subject factors each of which has two levels (4 conditions in total). Thus, my data table consists of six rows and twelve columns (3 properties * 4 conditions).

I created the within-table as follows:

within = table({'A';'A';'A';'A';'B';'B';'B';'B';'C';'C';'C';'C'},...
{'1';'2';'1';'2';'1';'2';'1';'2';'1';'2';'1';'2'},...
{'a';'a';'b';'b';'a';'a';'b';'b';'a';'a';'b';'b'},...
'VariableNames',{'Property','Condition1','Condition2'});

The table looks like:

    Property    Condition1    Condition2
    ________    __________    __________

    'A'         '1'           'a'       
    'A'         '2'           'a'       
    'A'         '1'           'b'       
    'A'         '2'           'b'       
    'B'         '1'           'a'       
    'B'         '2'           'a'       
    'B'         '1'           'b'       
    'B'         '2'           'b'       
    'C'         '1'           'a'       
    'C'         '2'           'a'       
    'C'         '1'           'b'       
    'C'         '2'           'b'

The repeated-measures Model I created using

rm = fitrm(t,'M1-M12~1','withinDesign',within);

where t is the data table and M1-M12 are the responses.

If I now use

manova(rm)

the output looks like this:

    Statistic     Value      F     RSquare    df1    df2    pValue
    _________    _______    ___    _______    ___    ___    ______

    Pillai       0.54237      0    0.54237    6        0    NaN   
    Wilks            NaN    NaN        NaN    6        0    NaN   
    Hotelling        NaN    NaN        NaN    6      NaN    NaN   
    Roy              Inf    NaN        NaN    6        0    NaN

I don't know, how I can specify, which of my within-subject factors are the different properties and which ones are the different conditions.

Help of any kind would be highly appreciated.

Best, Christian

Answer 1

[Aditya]

Hi Christian,

To perform a two-way repeated-measures MANOVA in MATLAB using the RepeatedMeasuresModel, you need to ensure that your within-subject factors and response variables are correctly specified. It seems like the issue might be with how the within-subject factors are defined and how the fitrm function is being used.

Here is an example code:

within = table({'A'; 'A'; 'A'; 'A'; 'B'; 'B'; 'B'; 'B'; 'C'; 'C'; 'C'; 'C'}, ...
               {'1'; '2'; '1'; '2'; '1'; '2'; '1'; '2'; '1'; '2'; '1'; '2'}, ...
               {'a'; 'a'; 'b'; 'b'; 'a'; 'a'; 'b'; 'b'; 'a'; 'a'; 'b'; 'b'}, ...
               'VariableNames', {'Property', 'Condition1', 'Condition2'});
% Assume your data table t is structured with columns M1 to M12
rm = fitrm(t, 'M1-M12 ~ 1', 'WithinDesign', within);
manovaResults = manova(rm);
disp(manovaResults);