I was hoping someone gravitated towards this poll... It really goes to show how little people understand about polling, recognized and unrecognized.
Allen West has been pushing the USC/LA Times poll these last couple of weeks as a SIGN that DRUMPF is rebounding, but in reality, DRUMPF isnt rebounding at all.
What the USC/LA Times does is what makes them an UNRECOGNIZED national poll.
Unlike all the other polls which are conducted by contacting either REGISTERED VOTERS, or LIKELY VOTERS and uses different people each week, the USC/LA Times poll uses the same 3200 people each week and these are people are considered "ELIGIBLE" voters.
1600 democrats and 1600 republicans are used and week after week, the same people are polled. Depending on participation, the USC/LA Times poll can sample between 1800 and 3200 of its original 3200 people week after week, it just depends who they contact.
Since the source of polled people are the same, there is no scientific basis behind it, nor is there a margin of error included in this poll. Depending on the week, the needle moves +2 to +4 depending on how many people they sampled.
Most of the time, the poll shows TRUMPvCLINTON TIED, and that stands to reason, given the original make up of the polling base.(1600 D to 1600 R)
This "experiement" by USC is really silly at best, and to use it to demonstrate DRUMPFS movement forward is just goofy.
We could start a weekly poll on this board, and the results would come out the same week after week, because nobody here is going to flip on there choice for president, so the poll would be a giant waste of time.
TOS.