You mention the file "data_1" in a comment here:
which makes me think possibly your files are called data_1, data_2, etc., rather than sheet*whatever.csv. If that's true, then try using data_*.csv in fullfile instead of sheet*.csv to get the set of file names to read:
for a = 1 : length(dirNames)
fileDir = char(dirNames(a));
S = dir(fullfile(fileDir,'data_*.csv'));
S = natsortfiles(S);
% Loop inside folder
for k = 1:length(S) % read data in specified sheet
On the other hand, if the files are called file1.csv, file2.csv, etc., as is stated in the question itself:
Then you should use file*.csv in fullfile:
for a = 1 : length(dirNames)
fileDir = char(dirNames(a));
S = dir(fullfile(fileDir,'file*.csv'));
S = natsortfiles(S);
% Loop inside folder
for k = 1:length(S) % read data in specified sheet
The point is, you have to give fullfile the wild-card pattern that actually matches your file names.
By the way, S is not 0; it is an empty struct array (of size 0-by-1).
By the way again, you can use curly braces { } to index dirNames:
fileDir = dirNames{a};
and avoid the explicit call to char, regardless of whether dirNames is a cell array of character vectors a string array.
