Hi Julie:
The dSPM and sLORETA values are statistical scores akin to z-scores and F-scores; therefore they are unit-less and can be interpreted at deviation from “standard fluctuations” as defined from noise or baseline variance stats for instance. If you want to compare two such measures in two different conditions or subject groups, I suggest you run statistical tests between samples (e.g., a t-test). You can still appreciate empirically a difference between two measures is the baseline used to compute the dSPM or sLORETA scores is the same. Hope this clarifies a bit.