Statistics in the Presidential Race

I'm sure we've all heard about Donald Trump, Ben Carson, Hillary Clinton, and Bernie Sanders in the media lately.

Yeah, I know - it can get old pretty fast.

The economy, women's rights, the war in Iraq... there are a LOT of things that are hot topics right now for these candidates, and it's their responses and thoughts on these topics that drive the polls.

But... wait. The election isn't for a whole year yet. Why are we already polling? What difference does it make a year from now?


The first thing to keep in mind is that polls are not predictive.

Let me say that one more time.  
POLLS ARE NOT PREDICTIVE. 

They can't and aren't supposed to tell you who's going to be elected next year. In fact, there's some evidence to show that the further away a poll is from the actual election, the less accurate voters are likely to be about who will win.

Okay, so polls won't tell me who's going to win. What's the point, then?

The point is that they tell you how people are feeling about the candidates today. They are used by the candidates to decide what they need to do in order to improve by the time the election rolls around, and they let people know - roughly - where everyone stands.

Remember: Two polls (or six or ten or twenty!) are better than one. It's simple math - the bigger your sample size, the more accurate your results will be.

Some of the WORST offenders for uselessness include the following kinds of polls:

• "Head-to-head" match-ups - this might give you an inkling of who would be elected if people were voting today, but the fact is, they aren't. I'll say it once more: 

P O L L S  =  N O T   P R E D I C T I V E

• Polls conducted by organizations with a bias or special interest - who's to say they aren't fudging the numbers? 

• Polls conducted by phone. Only the people who really care one way or the other will bother to participate, which isn't representative of the entire voting population.

•  Polls with small sample sizes. There are nearly 150 million people in the U.S. who are registered to vote. How accurate do you think a 100-person poll would be? How about 500? Would even 1,000 people be a good representation out of 150,000,000?

Here are a few more reasons as to why polls are increasingly inaccurate.

If the rest of this was too long and you just skimmed, PLEASE READ THIS SECTION:
Polls are not predictive. Be selective about your sources of information, but as long as they're all reliable, the more sources you have, the better. If you aren't sure about whether your poll's source is biased, see if you can find it here.

Which issues are the most important to you in this election? Why? Which candidate best represents your interests?

Do you think your candidate will win in 2016? 

Why or why not? 

If not, who do you think will win?