What's the diff. between likely voters and registered voters?

[napi21]

I hear this diff. made all the time in polls and there's always a diff in the results.

[kiahzero]

1. Registered voters is everyone who has paperwork on file

Likely voters are the people who will probably get themselves to the polls.

[gristy]

2. Likely voters are likely to vote. Registered voters are registered to vote

Kinda self-explanatory.

[troublemaker]

3. I think what the poster wants to know

is how pollsters distinguish them. They use things like 'did you vote in the last election' and whether the pollee knows where her local polling place is... stuff like that. Everyone uses their own formula.

[gristy]

4. can't you just let me be snide in peace?

troublemaker... :D

[Feanorcurufinwe]

5. Highly subjective distinction.

In theory, polls of 'likely voters' are more accurate -- but only if the pollster can really figure out who is likely to vote -- something of which I am very sceptical. In practice, this becomes a highly subjective criteria.

[napi21]

6. Thanks everybody.

I suspect the subjectivity is at least one of the reasons poll results vary so much. The quesstion of did you vote in the last election isn't a very good indicator because some people only vote if they are really enamoured with a particular candidate, and some won't vote at all if they don't like their parties candidate but won't cross lines to vote for the opposition.

[Piperay]

8. I think they ask them if

they have been following the campaign, know the issues and are planning on voting in the coming election but I agree it is hard to distinguish realistically who is actually going to vote.

[tritsofme]

7. It depends on who conducts the polls,

Sometimes they will outright ask if you plan on voting in the next election, and how likely it is you will do so.

They also might ask if you voted in the last election.

It is an important difference because only about 50% of the voting age population votes, and only somewhere around 80% of registered voters will vote in a presidential election.

This is why polls that have solid methods of picking out the voters most likely to go to the polls on election day are the most reliable.

[caffefwee]

9. There is a difference

But it's of no relevance. Bush's thugs will find a way to alter the outcome again. Karl Rove has something cooking I'm sure.

[Josh]

10. Conventional wisdom is that likely voters are more accurate in a race -

but I read an opinion piece that this election will be decided by voters who DIDN'T vote at the last election (particularly on the left, as Bush hadn't yet done the things that they now want to vote heavily against) rather than those who did. Thus in the polling we see, it may be more prescient to look at the "registered voters" and for whom they will vote.

[mobuto]

14. Well, its complicated

Because no two people can agree on what constitutes a likely voter. When you see, for example, a Zogby poll predicting the voting of "likely voters," Zogby ISN'T looking at 2000 voters, he's making a guess as to who will be likely to vote this year. Is he right? Is he wrong? Is Gallup's formula better? That's what's very hard to say. But in theory, likely voter polls should be more accurate than registered polls. Whether that theory reflects reality is another question.

[caffefwee]

11. Rob Reiner was on C Span this morning

He said you can't think of it as a 50-50 race. You have to think landslide. Bush will win or lose in a landslide which means he's more likely to lose because there's no way the whole country is going to vote for him, according to Reiner's thinking.

[Chango]

12. My explanation

"Likely" voters are those who have been identified by some means as having the highest propensity to show up to vote - on average. This is biased towards Republicans because of class distinctions that make it easier for them to show up. If lower class constiutents of Democrats are highly motivated to vote - as it appears they are this year - the "likely" classification is out the window.

[caffefwee]

13. Don't forget

They make it harder for others to show up. Florida 2000 was a perfect example.