Nonclustered vs Clustered PK indexes - Performance

I know that the @ssn is beeing generated with rand



The thing is, that with cluster key or with noncluster key, the time I get in the end (timedif) are +/- the same.

and it shouldnt be.