Traditionally, big media outlets (AP, CNN) conduct exit polling on election day – they ask who did you vote for, and why, and get demographic info on the voter. As I understand it, they also combine exit polls results with actual results to make early calls while only 2% of the actual votes have been counted.
Have they accounted for the fact that not everyone votes on Election Day at the polls?
To take it to an extreme: if all Democrats vote by mail and all Republicans vote in person…a lot of media will be calling a Trump landslide and saying that “violence in the cities” was the most important issue this year.
Back to the OP:
An “exit poll” is really an informal poll of voters. When everyone voted in person on one day they were simple to implement. In many states early in-person voting starts 2 or 4 weeks before Election Day. It’s more labor intensive for the media to include those voters in exit polling starting from day 1, but it absolutely needs to be done or the noise in their projections will be bigger than the signal.
As to mail-in votes, they ned to be conducting telephone polling or similar polling to take the termparature of who’s voting by mail and for which candidates.
Normally the mainstream media are torn between their desire for a scoop and for eyeballs versus the desire to be rigorous. Mindful of “Dewey Wins!” they dont’ want too much egg on their faces.
This time around one hopes they’ll be extra rigorous as they’re the only thing standing between continued democracy and the political Heist of the Century.
Broadcasting polls based on only Election Day in-person voters and leaving any caveats to footnotes delivered sotto voce just at the cut to commercial will be the last large mistake those media empires make. Putin knows what to do with media empires.
In the United States, much of the media has been voluntarily following that code of ethics in recent years, waiting until after all the polls close, even in the Westernmost states, to call a winner. Thus, no matter what they say, it won’t affect how any subsequent voters vote because there will be no subsequent voters.
That’s very different from calling an election based on known totals at the close of the polls, but without knowing vast numbers of mail-in votes yet. This STILL can’t cause anyone to vote differently than they would have, but it CAN result in an election being called very badly wrongly.
The mainstream media’s restraint has been sorely tested in recent years by both RW propaganda media (Fox) and by pure-play online media (Breitbart, HuffPost, etc.) It will be further tested by the state-sponsored online trolls this time.
If they want to remain relevant to the unfolding story they’ll have to say something other than “We don’t know and anyone else who tells you otherwise is lying / confused.” Even though they’d be telling the objective truth by saying that.
I’m not sure if this is what you meant, but they only wait until each state closes before they’re willing make a call on that state.
But anchors on the pundit shows and their “map analysts” start dropping hints in the late afternoon by zeroing in on certain bellweather areas and demographics and talking about how the exit polling compares to past results.
Mail-in is probably going to throw a monkey wrench into that because exit polling is usually done in the morning.
I’m way too lazy to look up how each state tallies its mail ins. As LSLGuy said, they are gonna have to somehow figure out a way to do telephone polls on mail in voters or they are screwed as far as “exit polling.”
And when you look at pre-election phone polls by state, they’re usually done over a number of days and I don’t think that’s because they WANT to spread it out, it’s just because it takes that long to do it.
The Gallup Poll really does exist; I was chosen to participate a while back. I got a questionnaire booklet about the Postal Service, and that, I did fill in and send back, because they’re legitimate.
