PA PrimD: Public Policy Polling: Obama Leads PA by 4%

[Alcon]

Last poll 4/15

Obama 47% (+2)
Clinton 43% (+1)

[Alcon]

PPP was right on in WI, and wrong nearly every place else.  I'm not counting on it.

You're confusing them with some other pollster.

PPP's lone mistake so far was underestimating Obama in Georgia.  I'm betting this will be the second one, but they've been stellar so far.

[Alcon]

The concern that exists with them regarding PA is more the wild swings seen in their polls of late rather than their past record.

Yes.  They swung a good deal beforehand too, but this is very worrying.  But they were not "wrong nearly every place else."

It's also possible that they got one totally whacked sample and then corrected it.

[Alcon]

Statistically, SurveyUSA and PPP are ranked first and second on my favorite pollster ratings.  Who shall triumph?!

[Alcon]

Numbers-wise, polling companies with large swings typically translates into lack of weighting (see e.g. Gallup, SUSA).  But then again, I really don't know much about PPP's methodology - have they made it specific to you, Alcon-fascist?

They use a tighter voter screen than SUSA.  That may suggest that a lot of the voters SUSA is picking up for Clinton are what I'll call "potential voters" -- ones that PPP eliminates from their sample.  I think PPP has a similar screen to Rasmussen though.

[Alcon]

Numbers-wise, polling companies with large swings typically translates into lack of weighting (see e.g. Gallup, SUSA).  But then again, I really don't know much about PPP's methodology - have they made it specific to you, Alcon-fascist?

They use a tighter voter screen than SUSA.  That may suggest that a lot of the voters SUSA is picking up for Clinton are what I'll call "potential voters" -- ones that PPP eliminates from their sample.  I think PPP has a similar screen to Rasmussen though.

Thanks - helps a bit.  I think I see what is going on here now.

Want to fill me in, then?  'Cause I can see numerous scenarios going on here (although I think a lot will involve turnout of the rural working-class, who I'm guessing are not excited right now).

In fact, I think turning this campaign negative may have turned those voters off and lost Clinton a decent number of votes.  Double-edged sword.  Only a guess.

[Alcon]

My only real support for that (it's more theorizing ) is that the undecideds have not gone down in the last few days as they tend to.  But that could simply be projection, or coincidence, or just general tomfoolery.

Shouldn't Zogby be crapping a new poll soon?

[Alcon]

I don't recall where I read this, Pollster.com I believe, but the PA primary polls are finding:

1. More undecideds than previous polls
2. A lesser rate of deciding in the last few days

They adjusted for different methodologies and all.

But damned if I can find the reference.  I'm reasonably sure I saw it though.

[Alcon]

Sorry, I'm really sick right now and it's reflecting in my posts.  I'm not actively trying to be vague.

I don't see anything on the RCP index relating to undecideds.  Not even their chart includes undecideds.  Would you mind a link to whatever you're seeing?

[Alcon]

PPP has some 'splaning to do, but they still have a solid record.  I guess they're entitled to their 1-in-20, even if this was more like a 1-in-40.  I still think they're promising--we should consider that this is their first year of being a big-league pollster and they're ranking #2.  That's impressive and they don't deserve to be lumped with ARG.

[Alcon]

First off, please do not call us or e-mail us and tell us we suck! We are well aware

Haha.

It's pretty easy, based on the exit polls, to see where we went wrong. We had the black vote at 18% when it turned out to be 14-15%, and we had the under 45 vote at 41% when it turned out to be 31%.

I reweighted the results from our final Pennsylvania poll to those figures for race and age, and the result of our poll was flipped- Hillary leading 49-46. Assume she pulled 60% of the undecideds and that gives her the 52-48 lead that the extrapolation from the original exit polls does.

So it's pretty easy to see where we went wrong. I'm glad that it was misweighting rather than some sort of systematic issue with our polls. Obviously we'll take some heat in the next few weeks for being so far off, and we certainly deserve it, but at the same time I don't think this should completely discount our credibility as a company- we've been pretty darn good a lot of other times this cycle.

I'm not really satisfied with that explanation, but in fairness they made it when they were just looking at initial exit polls.  I hope they're going to (transparently) look more into what they did wrong, because they did something seriously wrong.  I do like their attitude, though .