Manipulating dataset: (from column variables to row observations)

[mark_ph]

Dear All,

I have the following dataset:

data have;

     input ID Jan Feb;

datalines;

100     1     2  

200     3     4    

;

I would like to put the values of the two columns Jan and Feb in a unique column, named value, and add a column named month which should indicate whether the value comes from the column Jan or Feb of the dataset have. That is, I would like to obtain the following dataset:

data want;

     input ID month value;

datalines;

100     Jan     1    

100     Feb     2    

200     Jan     3

200     Feb     4

;

Any help would be highly appreciated.

[Reeza · Accepted solution]

Proc Transpose:

proc transpose data=have out=want

  (rename=(_name_=Month COL1= Value));

by ID;

run;