Clint here crosses the boundary between ordinal & interval
scales (strictly he also needs a value "uncomfortable", say 3 * mild).
It is now appropriate to work out means, totals, even standard
deviations; everything being in units of "mild". To legitimately
carry out out a t-test or regression, he needs only to declare that
his sample is large enough for the Normal approximation to work
(n>50, say). Otherwise, he can use bootstrapping for his
confidence intervals & p-values.
Another possibility, if the Normal approximation does not work, is for
Clint to define his "discomfort" variable and then to define confidence
intervals for median differences between two groups using the -cendif-
program, which is part of the -somersd- package. The -somersd- package can
be installed from SSC using Web-aware Stata, and includes .pdf manuals. Type
