Creating new variable set from existing data

[RCPenguin]

Hello all,

I have a data set with test scores and I am trying to compare scores of subject a to subject b (for only only gender)

I attempted to do a Wilcoxon-Rank sum test but I cannot use my class as gender when I proc npar1way because I need it to look at the two subjects not the two different genders as all genders for my data being used will be used one particular gender.  My question is am I able to create an overarching variable that I can put my subjects in.  Something along the lines of subject: a=0, b=1, c=2, d=3 and so on so that I can eventually do

proc npar1way wilcoxon data = ...

class subject;

var diff;

You get the idea.  Very dirty code at the moment but this has been racking my brain and I am not sure if I should just use a different nonparametric test (like Wilcoxon Signed Rank) or maybe a parametric test.  Thank you for your time.