I have a survey that has 204 items/variables each rated from 0-4 or 99 (unlcear). I need to know what percentage of people rated each item each number (0-4, 99). Obviously I can run proc freq and get individual tables for each of the 204 variables, but I want to compile the frequency counts into a new dataset. I have tried using output, but it only uses the last variable.   Is there a way to create a dataset from the frequency tables where all of the items are observations and the Likert ratings are variables? And then each cell contains the frequency/percentage of each rating? I would also like to keep the missing values and have 0's inserted where a possible rating was not endorsed.   For example I currently have 97 participants and 204 items structured as such:   Obs     Item1      Item2     Item3    Item4 1            0              .            2           99 2            4              1            2            1 3            0              2            1            2   I want a dataset structured like below (even better if it were percentages instead of counts):   Obs      Freq0     Freq1     Freq2     Freq3    Freq4      Freq99      Missing Item1      2              0           0             0             1              0                0 Item2      0              1           1             0             0              0                1 Item3      0              1           2             0             0              0                0 Item4      0              1           1             0             0              1                0   Any ideas are appreciated!   ... View more