(1) What is the meaning of the numbers after the plus signs in the registration figures for each state? (In other words, what is the base to which the new number of GOP registrations is being compared?)
(2) I know it may vary from poll to poll, but in general, would newly registered voters be counted by pollsters claiming to be polling and reporting on LIKELY VOTERS? (In other words, if you are a newly registered voter who has never voted before in your location or at all, are you thrown out of the poll results because you don't have a voting history, and therefore thought to be unlikely to vote by the poll's definition?)
Usually the + means more Republicans than Democrats (as in the FL numbers).
Sometimes it means an increase over 2012 numbers so you’d have to look at context.
But normally if someone says, when discussing absentee ballots, Rs +30,000, that means so far the Republicans have 30,000 more absentee requests than Democrats.
As to your second question, absolutely no insight as to how a pollster would describe these people. Actually, I doubt they’d ever ask “when did you register?” They would likely say, “are you a registered Democrat or Republican or independent?”
>>(1) What is the meaning of the numbers after the plus signs in the registration figures for each state? (In other words, what is the base to which the new number of GOP registrations is being compared?)<<
The article uses the word “Net,” so I would say the “+” means net gain (including D losses) versus 2012 registered voters. I would ask the author but I couldn’t find any way to contact him.
>>(2) I know it may vary from poll to poll, but in general, would newly registered voters be counted by pollsters claiming to be polling and reporting on LIKELY VOTERS? (In other words, if you are a newly registered voter who has never voted before in your location or at all, are you thrown out of the poll results because you don’t have a voting history, and therefore thought to be unlikely to vote by the poll’s definition?)<<
Typically a newly registered voter wouldn’t be considered a likely voter. Likely presidential election voters are usually ones that have voted in the last 1, 2, sometimes 3 presidential elections. Some pollsters will simply ask something like “How likely are you to vote in the upcoming election? Not likely, somewhat likely, likely, very likely?”
Good questions. I’m sure you all know since the MSM have written about it from time to time, that pollsters do have concerns that they are not able to reach people who may be first time voters, newly energized voters who have stayed out of many past elections because they feel disenfranchised by the system itself regardless of who they voted for in the past. I gotta think that if there is a grouping of newly energized voters that the group would favor Trump, only because Hillary represents the establishment, and she is not new, and voters have already had a chance to vote for her in 2008.
As we get closer to November at what point to pollsters start considering absentee/mail ballots? It’s becoming big enough to matter, you can’t just count registered voters and likely voters, you have to start counting “already voted” (AVs). An interesting thought experiment at the least, but would you blend LVs with AVs, or run a poll of just AVs entirely. This would present a set of new problems, particularly trying to figure out if AVs are representative of the body politic or if they are imbalanced in some way that they can’t even try to correct for in the weighting. Additionally, in what way is this edge for Rs among AVs suggestive of voter enthusiasm, and does it project likely turnout to the polls in November. Lots to consider for those of us who enjoy watching and dissecting these polls.