Trump: still a birther

Sopan Deb

Any chance the media will call him to account for this?  Your ball Lester Holt.

Another look at NC polls

PPP is out with their NC poll today and these results make a lot more sense to me.  Cooper is up over McCrory 46-41, Trump leads Clinton 45-43, and Burr and Ross are tied at 41.  I found this part particularly notable and very-much expected for McCrory:

The story in this race is Cooper’s strength with independent voters. He leads McCrory 44-33 with them in the full field, and 50-34 with them in the head to head. By contrast, McCrory defeated Walter Dalton by a 2:1 margin with independents in 2012. But they have largely soured on him with now only 36% approving of him to 49% who disapprove.

Some other good tidbits:

By a 30/23 spread, Trump voters in North Carolina say they have a higher opinion of David Duke than they do of Hillary Clinton.

Lovely people, Trump voters.  At least David Duke stands up for white people, damnit!

And this:

63% of voters in the state think Trump needs to release his tax returns, to only 24% who don’t think it’s necessary for him to do that. That includes a 62/23 sentiment among independents that he needs to do so.

I can only assume that the Clinton campaign is holding their fire for the moment on the tax return issue because this strike me as quite potent.  It’s very simple and anybody can understand the relevance and Trump clearly has something to hide.  And, the public clearly agrees the returns need to be released.

Trump voters

I know, I know, you talk to voters in a Hillary Clinton line and I’m sure you can get some ridiculous comments.  But I swear they cannot be this bad.  Good, good stuff from the Daily Show.


Polling: science and  art

Wow.  So loved this feature in the Upshot yesterday that so clearly demonstrates how much of polling is not science, but also art.  Yes, there’s plenty of science when it comes to accurately obtaining a truly random sample and using statistical inference based on the science of that sample, but the art can be just as, if not more important, in determining the final results.

Basically, your random sample is never going to perfectly match the demographics of the population about which you are drawing inferences (e.g., you only care about 700 NC voters insofar as they can help you infer the opinions of the millions of actual NC voters).  Now, if you just want “public opinion” this is easy.  If your sample is only 12% Black but the Census data says 15% Black, you just weight accordingly.  But elections are different.  Nobody actually knows what the percentage of Hispanic voters or under-3o voters will be on election day.  All we can do is make informed guesses and weight to those guesses.  But, they are guesses.  And there’s often quite defensible reasons for choosing 14% Black or 17% Black or 45% over 65 or 50% over 65 or whatever.

Not to mention, when polls try and determine “likely voters” that’s another educated guess where pollsters use a variety of defensible methods to predict who will actually vote or not.  In close elections, these choices make a huge difference in who is winning or losing in the polls.  And as the Upshot feature points out, ultimately a bigger difference than the simple margin-of-error calculations based on sample size.

Anyway, the Upshot gave raw Florida polling data to four different pollsters and asked them to come up with an estimate for a final poll.  They all showed a close race, but the results diverged not inconsiderably.  Here’s some charts that show the different results, and importantly, the assumptions behind them:

poll poll1

Who’s right?  Well insofar as this is a snapshot in time, we’ll never know.  Then again, come the November exit polls, we can see whose assumptions about the composition of the electorate got it right and who got it wrong.

I so love that the NYT did this so that people can understand how polls really work.  This is definitely going into multiple future syllabi.

When is a poll just a poll?

I got a call from a reporter yesterday who wanted to talk about the new Elon poll that shows McCrory up three in the NC Governor’s race.  I didn’t’ give her what she wanted as I said that until I saw more polls like this, I was still sticking with the RCP average which shows McCrory clearly down. (Down an average of 3.6 in the latest, and even more if you don’t include the Civitas poll– and I’m always skeptical of their polls).  Elon does a solid poll (uses humans to call landlines and cell phones), but sometimes results just don’t pass the smell test.

#1) Is there anything in the real world that would actually explain McCrory gaining significantly in the polls in the past couple weeks?!  Especially with the NCAA and ACC pulling out of NC?  Ummm, nope.  So, maybe the older polls are systematically under-estimating McCrory, but I sure as hell doubt there’s been actual movement towards him among the NC electorate.

#2) Even more notably, McCrory is way outperforming the Republican Senator, Richard Burr, who is down 1 in the same poll to his Democratic challenger.  Given my knowledge of NC politics and Political Science, I just have a hard time envisioning a reality where McCrory is actually up 3 and Burr down 1.  Going with my instincts and knowledge over a poll of <700 voters on this one.  I asked my class yesterday if they could create a plausible explanation for that pattern of results.  None even tried.  Nate Cohn noticed this as well.

So, again, not to pick on Elon.  Sometimes you do everything right and the results are still funny.  But there’s times when we need to put a poll in broader context and use some common sense and I really think this is one of those times.

%d bloggers like this: