I would like to draw a bar chart with "unequal samples". Here is an example code
A = [16 20 15 17 22 19 17]';
B = [22 15 16 16 16 18]';
C = [23 9 15 18 13 27 17 14 16 15 21 19 17]';
group = [ ones(size(A));
2 * ones(size(B));
3 * ones(size(C))];
figure
boxplot([A; B; C],group)
set(gca,'XTickLabel',{'A','B','C'})
The output is as below:
However, I would like to have a distance between group1,2 with group 3. As same as what you see in the figure below:(this figure is just a copy paste from another source but the distance between box plot of each group is visible)
I tried to use 'factorgap' in such command
figure
boxplot([A; B; C ],group,'factorgap',[50,1])
However, because the number of samples in each group is different it did not work.
Any suggestion?
The first solution I propose you is in fact a small workaround that consists in inserting another, invisible group between the second and the third one:
A = [16 20 15 17 22 19 17]';
B = [22 15 16 16 16 18]';
C = [23 9 15 18 13 27 17 14 16 15 21 19 17]';
group = [
ones(size(A));
2 * ones(size(B));
3;
4 * ones(size(C))
];
figure();
boxplot([A; B; NaN; C],group);
set(gca,'XTickLabel',{'A','B','','C'});
Here is the output:
Now, let's build up something serious:
% Define the sample data...
A = [16 20 15 17 22 19 17]';
B = [22 15 16 16 16 18]';
C = [23 9 15 18 13 27 17 14 16 15 21 19 17]';
% Find the length of the largest vector...
A_len = numel(A);
B_len = numel(B);
C_len = numel(C);
max_len = max([A_len B_len C_len]);
% Transform vectors into fixed size vectors of length max_len...
A = [A; NaN(max_len - A_len,1)];
B = [B; NaN(max_len - B_len,1)];
C = [C; NaN(max_len - C_len,1)];
% Define labels and groups...
L1 = [repmat('A',1,numel(A)),repmat('B',1,numel(B))];
L2 = repmat('C',1,numel(C));
L = [L1 L2];
G = [repmat('1',1,numel(L1)),repmat('2',1,numel(L2))];
% Plot the boxes...
boxplot([A B C],{G';L'},'FactorGap',50);
Here is the output: