Is that part of the reason we can't quantify whether these programs are always really, truly helping? I know we have some anecdotal support, but is that one of the reasons the Auditor General couldn't measure whether these programs were working or not?
We have different definitions, depending on who you're talking to, I guess.