If you work with longitudinal data, you may see the data arranged in a tall and thin format or in a short and fat format. The tall and thin format is useful for some data analyses, such as mixed model analysis, and the short and fat format is useful for other data analyses, such as graphical display. You often need to convert from one format to the other. The UCLA Statistics website shows how this is done.
The UCLA Statistics website shows data arranged in a tall and thin format
famid year faminc
1 96 40000
1 97 40500
1 98 41000
2 96 45000
2 97 45400
2 98 45800
3 96 75000
3 97 76000
3 98 77000
which they call the long format. To convert this data to a short and fat format (which they call the wide format)
famid faminc96 faminc97 faminc98
1 40000 40500 41000
2 45000 45400 45800
3 75000 76000 77000
they use the following code
proc transpose
data=long1
out=wide1
prefix=faminc;
by famid ;
id year;
var faminc;
run;
I took the liberty of laying the code out slightly differently, but it’s the same code.
If you have data in the short and fat format, you can also use proc transpose to convert to a tall and thin format.
proc transpose
data=wide1
out=long1;
by famid;
run;
The UCLA site also shows some more complex applications of proc transpose on