The problem is that the line endings in that file are \r\r\n, i.e., two carriage return characters followed by a line feed character.
One way to get around this is to replace those line endings with one that textscan can handle. Here I replace them with a single line feed character:
% you won't need to unzip the zip file, this is just to run it here:
unzip('ConcatenatedCsv.zip')
% open, read, and close the file:
fileID = fopen('ConcatenatedCsv.csv');
data = char(fread(fileID).');
fclose(fileID);
% replace \r\r\n with \n in data:
data = strrep(data,char([13 13 10]),newline());
% now do textscan as before, but on the text
% itself (data) rather than the file handle fileID:
delimiter = ',';
startRow = 4;
formatSpec = '%f%f%f%f%f%f%f%f%f%f%[^\r\n]';
dataArray = textscan(data, formatSpec, 'Delimiter', delimiter, 'TextType', 'string', 'EmptyValue', NaN, 'HeaderLines' ,startRow-1, 'ReturnOnError', false, 'EndOfLine', '\r\n');
ConcatenatedCsv = [dataArray{1:end-1}];
% check the size:
size(ConcatenatedCsv)
