# Exit polls: voters who don't say

**URL:** <https://boards.straightdope.com/t/exit-polls-voters-who-dont-say/273283>\
**Category:** Factual Questions\
**Created:** [November 5, 2004, 11:16pm UTC](https://boards.straightdope.com/t/exit-polls-voters-who-dont-say/273283 "2004-11-05T23:16:18Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Quartz](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/quartz/32/267_2.png) [@Quartz](https://boards.straightdope.com/u/Quartz)\
**Post date:** [November 5, 2004, 11:16pm UTC](https://boards.straightdope.com/t/exit-polls-voters-who-dont-say/273283/1 "2004-11-05T23:16:18Z")

</div>

The exit polls in the last election got it wrong. When they counted voters, did they include those who don’t state how they voted?

‘We sampled 1000 voters. 700 told us how they voted …’

If they didn’t - and I don’t think they did - what bias does this introduce?

---

<div class="post-metadata">

**Author:** ![Freddy\_the\_Pig](https://avatars.discourse-cdn.com/v4/letter/f/a587f6/32.png) [@Freddy\_the\_Pig](https://boards.straightdope.com/u/Freddy_the_Pig)\
**Post date:** [November 6, 2004, 1:02am UTC](https://boards.straightdope.com/t/exit-polls-voters-who-dont-say/273283/2 "2004-11-06T01:02:25Z")

</div>

Exit polls don’t generally report the number of refusals. Whether or not this introduces bias depends, obviously, on whether the people who refuse tend to vote for the candidates in the same proportion as the people who answer.

In past years, it appears that they did, because we never saw a consistent pattern of error. (An individual exit poll could always be wrong, since all polls have a margin of error, but it was equally likely to err in favor of Republicans or Democrats.)

In this election, a larger percentage of Republicans apparently refused to answer. The logical reason would be ever-growing Republican mistrust of the media.

The unpredictability of this variation in refusal rates will probably make exit polls useless as a forecasting tool, so we’ll go back to waiting for the actual votes. Bravo, I say!

---

<div class="post-metadata">

**Author:** ![SiouxChief](https://avatars.discourse-cdn.com/v4/letter/s/d6d6ee/32.png) [@SiouxChief](https://boards.straightdope.com/u/SiouxChief)\
**Post date:** [November 6, 2004, 2:17am UTC](https://boards.straightdope.com/t/exit-polls-voters-who-dont-say/273283/3 "2004-11-06T02:17:29Z")

</div>

There is also an ever-growing movement to purposefully lie to or mislead exit pollsters. People are getting quite creative with it, too, I might add.

The end result, hopefully, will be the end of the exit poll.

I agree with Freddy. Bravo!

---

<div class="post-metadata">

**Author:** ![aruvqan](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/aruvqan/32/2891_2.png) [@aruvqan](https://boards.straightdope.com/u/aruvqan)\
**Post date:** [November 6, 2004, 3:13am UTC](https://boards.straightdope.com/t/exit-polls-voters-who-dont-say/273283/4 "2004-11-06T03:13:21Z")

</div>

> [@Quartz](#):
>
> The exit polls in the last election got it wrong. When they counted voters, did they include those who don’t state how they voted?
> 
> ‘We sampled 1000 voters. 700 told us how they voted …’
> 
> If they didn’t - and I don’t think they did - what bias does this introduce?

FWIW, I _categorically_ refuse to say how I voted. mrAru doesn’t even know who I vote for. If I am not even telling my husband, best friend in the whole world and lover who I voted for, why should I tell anybody else. Secret ballots are a right, and a protection.