The PPP Poll uses an awfully weird voter screen.

[Zynx]

Obama only has a 44% approval rating with voters for the MA special election. Unless Democratic turnout is absolutely abysmal, and I mean really fucking bad, that won't be the case in the election. Obama likely has close to a 57-60% job approval rating in MA.

That being said, we have to do a much better job at getting our voters engaged not only in MA, but nationally or we will see some pretty scary poll results as we near November.

[Bitwit1234]

1. Since there are more registered democratic voters

than republicans, if the voters turn out the republicans won't stand a chance. Why do democrats stay away and the republicans always turn up.

[DCBob]

3. They are a few red meat issues that just drive the conservatives crazy..

and compel them to vote: gays, guns, God, taxes, military and big government. Democrats issues are more complex and nuanced which requires an inspiring, charismatic, intelligent leader to explain them and get voters to the polls -- e.g. Kennedy, Clinton, Obama.

[CreekDog]

2. Didn't PPP have Clinton over Obama by 15 points in North Carolina in 2008?

it was a startling number that didn't pan out.

Obama trounced Clinton on North Carolina shortly after.

they've made some big headlines --that turned out to be big mistakes.

[backscatter712]

4. Both PPP and Rasmussen cook their polls one way or another.

If they don't get numbers that make the GOP look good, they'll used biased framing and questions, questionable voter screens, odd sampling, etc. until they get polls that make the GOP look good.

[DrToast]

5. PPP is a Democratic poling firm

[ShadowLiberal]

6. Keep in mind, it's VERY hard to predict turnout for special elections

Also keep in mind it seems that a lot of the polls conducted in Massachusetts have an unusually small sample size of a bit over 500, instead of the usual 1000, which increases the margin of error farther.

[Zynx]

7. The extra accuracy gained by going from 500 to 1000 is not as huge as people make it out to be.

500 is generally well within the range of a good sample.

[quiller4]

8. PPP has done very poorly predicting special elections.

They were way off on NY 23 and showed Owens losing badly. They do a very good job predicting turnout in regular elections. It is all about choosing the right screen for likely voters.

I suspect the Globe poll is more reliable and it show Coakly winning handily.