Good point on "the source of the data is 'registered voters', not people who have jobs." In a round-about way, polling registered voters might make sense, as t

Good point on "the source of the data is 'registered voters', not people who have jobs." In a round-about way, polling registered voters might make sense, as they have -- however indirectly -- to actually change the system.