mating net savings generally requires the use of a
comparison group as a proxy for what the participants would have done absent the
program, or self-reported information on what would have happened in the absence of the
program when comparison groups cannot be reasonably identified.