Three days ago we said Barack Obama was up four in Colorado. Today Rasmussen says John McCain is up a point. I don't know who's right but I can tell you one reason for the difference in our polls.
Rasmussen's numbers are based on 500 interviews conducted Wednesday night.
We did 500 interviews last Tuesday night in Colorado, and based on those McCain was up by a point as well. But we also called everyone in the sample who we didn't reach Tuesday night again Wednesday morning, Wednesday afternoon, Wednesday night, and Thursday afternoon. We got almost another 500 respondents and they went for Obama by a good amount, pushing him to a four point lead.
As far as I can tell Rasmussen just makes calls until they get 500 respondents. When we buy a sample we attempt to reach every number in it at least four times over a period of several days and go with whatever number of respondents we get based on that. And we almost always find that the further into conducting the poll we go, the better Obama does. I don't know if it's that his supporters are harder to get on the line or what, but we're finding that trend in nearly every state we poll.
I'm not saying that one way or another is better- Rasmussen has an outstanding track record- just pointing out that differences in the way we conduct our polls may help explain some of the differences in the numbers we produce. It's not as simple as just saying Rasmussen is Republican and PPP is Democratic.
The fact is than Rasmussen is leaning republican. Scott Rasmussen is a republican who goes every week on Fox News for speaking about the polls who look good for McCain.
ReplyDeleteAlso, Rasmussen is not transparency. The crosstabs are not available publicly.
The fact is that PPP is Democractic.
ReplyDeleteMaybe it's that PPP tries to find votes for Obama over the few days of the poll. Since Obama wasn't up on first night of polling, they decided to add more until Obama got in the lead.
Also the RAS poll had 700 likely voters which is a pretty healthy sample.
Brandon,
ReplyDeleteThat's absurd. Making callbacks to increase the response rate is a pretty standard good practice in polling and we do it regardless of who is winning the poll.
PPP is a very good pollster. For the primaries, without the PA "mini drama", they had a very good results.
ReplyDeleteWisconsin(the only who had Obama by more than 10%), NC, Oregon, Ohio and others.
SUSA and Rasmussen has generally underestimate Obama and at this moment for the general election it's the same
The turnout will be historical and the new voters will make the difference in much states.
Would it be worth releasing cross-tabs for the time of day the call was completed or the number of tries it took to get the respondent on the phone? That would be immensely valuable and would certainly help show how strong Rasmussen's methodological biases can be, especially in statewide polling.
ReplyDeleteThough, just FYI, while Scott Rasmussen might be a Republican, Rasmussen Reports is not a Republican polling firm. Otherwise, this would make Gallup, Pew, and Mitofsky all partisan pollsters, which is obviously not the case.
Anyway, keep up the great work and the great blog! And please start polling more congressional races.
This is interesting .. so according to this, most McCains supporters are home alot .. whereas most Obama supporters are not home much... Sounds like Rasmussen may be missing the younger crowd.
ReplyDeleteGiven that McCain does particularly well with older voters it is not surprising that calls to those that can be reached on the first call would tend to be pro-McCain. Of course, perhaps some of the difference between PPP and Ras is compounded by the fact that Ras weights its polls, and I believe has tended to have Reps make up larger shares of the respondents than non-weighing polls, like PPP.
ReplyDeleteThe Ohio poll from Rasmussen had Mccain up by 10 points. How is it that Rasmussen is the only polling brand that had Mccain up? let alone by 10. Second Rasmussen choose only 500 as a sample for a state like ohio, no offense thats a joke! how is it that the only times Mccain has substantial leads in all the ohio polls is when the sample is small; around 500.
ReplyDeletei dont think any polling brands play favorites because it would discredit their ability to predict accurately. I am just shocked that this race is so close. i am a obama supporter but starting to feel like hillary clinton had a point. this 50-state strategy is just nonsense. its gonna come down to ohio michigan and florida. which hillary could carry! To be honest i am having doubts if Barack can carry ohio and florida. Mccain has spent nothing on florida while obama has spent million and mccain is still up. however i think
ReplyDeleteobama's biggest headache right now is michigan...hillary was right!
obamawill you sound more like a PUMA than an Obama supporter. Funny how you excluded VA from your list of battleground states, or CO and NV for that matter. Seriously, Clinton never polled well in CO or VA for the general as compared to Obama.
ReplyDeletePPP in their party id in their two Ohio polls had HUGE Party Id's for the Democrats, and I brought this up on here before, but they did not post my comment because they know if it's posted it shows their the true partisan pollster. PPP you have the [D] next to your name in newspapers and on many polling sites.
ReplyDeleteFACT IS THIS:
In June: Party ID of Ohio Poll
Dem-55 to GOP-30.
In July: Party ID of Ohio Poll
Dem-46 to GOP-33.
In 2004 the GOP had a 5 point edge in party and now the Dems blow it out to 13-25% margins. The partisan pollster is PPP, not Rasmussen, who nailed the 2004 election correctly!
Spread the word I suggest you visit the election board, secretary of state sites for the battleground states and look at the voter registration numbers for the past 8 months, this is not going to be 2004 election where Reps made up the same share of the electorate as Dems who voted in November. Sorry to tell you but Dem registration numbers have shot up during the past 8 months while Reps voter registration numbers have stagnated or even slipped in some states.
ReplyDeleteCongratulations. You are registering African Americans and kids my age who rarely go out to vote. Yes you may have the edge in party id this time around, hell I'll spot you guys 5 points in party id.
ReplyDeletePPP do a poll with Dems-45 GOP-40 and see where it is.
There is no way in hell the Democrats own a 13-25 point edge in a state they trailed by 5 points in 2004. If you look at the crosstabs of the PPP polls, you even notice this state is a strong conservative state. Conservative voting for Obama are few and far between.
PPP, my local newspaper even took a shot at you for a huge party id for the Dems when even they, a liberal paper, realize its 3-5 points at the very most!
I just dont get it. why is obama running neck-and-neck with Mccain in the polls. Fox did a survey that had Obama/clinton up 6 against Mccain/romney. Obama has spent 5 million on ad alone in florida compared to mccain's 0, and he's losing in polling! ohio is a toss up, but it definitely shouldn't be. Ohio has been through alot of economic down-turns, and the democratic brand polls way higher! and colorado and virginia is going to be tough for obama- to win he would have to reverse a major historic trend. I am a die-hard obama supporter, and i think he should put hillary on the ticket, she's a hawk! and they love the clinton brand in florida, ohio and pensylvania. also to "anonymous" hillary still would poll better in ohio, pensylvania and ohio!- i think obama would be the BEST president we have ever had- better then kennedy, regan
ReplyDeleteyou guys are so dumb if you think rasmussen or ppp are partisan. don't you think they wanna get it right? so that they are more trustworthy, and are referenced more in national news networks. the best way to look at these polls is the average of polls. realclearpolitics.com
ReplyDeleteActually, PPP is a Democratic polling film, but they do no one any favors by producing incorrect results. A common fallacy is to assume partisan pollsters produce cooked numbers. If anything, the campaign pollsters are more trustworthy since they have to be accurate.
ReplyDeleteRegardless, realclearpolitics is quite biased. They are not only openly promoting a Republican agenda, but they randomly choose to include or exclude pollsters for no reasons. Seriously, if you aren't accustomed to the regression lines at pollster.com then you are missing out.
pollster.com is a joke, i cant believe the include Zogby internet polls!
ReplyDeleteNote: http://publicpolicypolling.blogspot.com/2008/08/mccain-pulls-even-in-ohio.html
ReplyDeleteRead my comments here and now read those poll numbers.. Hmmm, living in Ohio gives me an idea of how this race will shape up!