# How Do I Get a Pollster to Notice Me?

**URL:** <https://boards.straightdope.com/t/how-do-i-get-a-pollster-to-notice-me/782534>\
**Category:** Politics & Elections\
**Created:** [March 19, 2017, 6:17am UTC](https://boards.straightdope.com/t/how-do-i-get-a-pollster-to-notice-me/782534 "2017-03-19T06:17:43Z")\
**Posts on this page:** 1\
**Showing post:** 20

<div class="post-metadata">

**Author:** ![Silver\_lining](https://avatars.discourse-cdn.com/v4/letter/s/aca169/32.png) [@Silver\_lining](https://boards.straightdope.com/u/Silver_lining)\
**Post date:** [April 4, 2017, 2:01pm UTC](https://boards.straightdope.com/t/how-do-i-get-a-pollster-to-notice-me/782534/20 "2017-04-04T14:01:10Z")

</div>

> [@kenobi\_65](#):
>
> I’m a professional market researcher, and this is the real answer.
> 
> A true random poll needs to be done from a truly random sample, including (ideally) some people who really don’t want to participate, since their opinions might differ from those who would eagerly participate.
> 
> And, most political polls have sample sizes that aren’t terribly large – a few hundred to a few thousand people, at most, within a state. Thanks to statistical modeling, you don’t need a much bigger sample than that to get poll numbers that are representative of what you’d get from a census (that is, surveying _eveyrone_), assuming that you’ve done a good job of building your sample.
> 
> Polls that are done among volunteers (or “convenience samples”) are notoriously unreliable – remember the “polls” that one of Trump’s spokesmen pointed to after the debates, claiming them as evidence that Trump had won the debates?

Well several on line polls said Trump won the debate. Hillary Clinton was given the questions in advance, a highly unfair advantage.

---

_[View the full topic](https://boards.straightdope.com/t/how-do-i-get-a-pollster-to-notice-me/782534)._