ANOVA of a cell array

I'm looking to do a one-way ANOVA of a cell array, except when I use cellfun (as follows) I get an analysis of each individual cell as opposed to one analysis of all the cells (for example, if i have 8 cells, all of which have a different number of elements, I get 8 outputs when I want one output that compares the elements of all 8 cells).

p=cellfun(@anova1,master_maxt{1,1})

For reference, master_maxt is a 1x54 cell array, where {1,1} is a 1x8 cell array. Each of those 8 contains 57-120 elements. I am looking to do the same thing to all 54 of the arrays in master_maxt.

Thanks in advance for any help!

Answer 1

[Vedant Shah]

Hi @car,

To perform a one-way ANOVA on all the data contained within the cells of a cell array and obtain a single output that compares all of them, you first need to combine the data from all the cells into a single vector. Subsequently, you can apply the “anova1” function. For more information, please refer to the documentation using the following command in the MATLAB command line:

web(fullfile(docroot, "/stats/anova1.html")) 

To execute the one-way ANOVA, following code can be used:

% Initialize variables to store combined data and group labels 
combinedData = []; 
groupLabels = []; 

% Loop through each cell in master_maxt{1,1} 
for i = 1:length(master_maxt{1,1}) 
    % Extract data from each cell 
    data = master_maxt{1,1}{i}; 
    
    % Append the data to the combinedData array 
    combinedData = [combinedData; data(:)]; 
    
    % Create a group label array for this cell and append it to groupLabels 
    groupLabels = [groupLabels; i * ones(length(data), 1)]; 
end 

% Perform one-way ANOVA 
p = anova1(combinedData, groupLabels); 

In this approach, the data from all cells is collected, and each data point is assigned a group label corresponding to its original cell. This process allows for a single output that compares all the cells as intended.