« BACK  |  PRINT

RS

FRONT PAGE CONTRIBUTOR

Is Rasmussen biased toward the Republicans? Not in California.

By: Neil Stevens (Diary)  |  July 14th, 2010 at 03:30 PM  | 

RESIZE: AAA

Fiorina Boxer

Certain critics either say or imply that Rasmussen Reports is skewed toward Republicans, just because this cycle he predicted early that the 2010 electorate would look nothing like that of 2008.

But that’s not the same as having a partisan bias, and in fact, comparing the latest Rasmussen poll of the California Senate race with SurveyUSA hints there is no such partisan bias to be found.

I say that because SurveyUSA is looking much better than Rasmussen for the Republican. SurveyUSA has Carly Fiorina ahead of Barbara Boxer 47-45 (MoE 4), winning independents 49-40. It’s a slim lead, with a 41% chance that Boxer’s actually up, but it’s a lead.

Rasmussen’s outlook is much friendlier to Boxer, putting her up 49-42 (MoE 4.5). Interestingly enough, Rasmussen has Fiorina winning independents by even more than SurveyUSA does, 12 points.

One big difference seems to be in the Democrats. SurveyUSA gives 83% of them to Boxer, but Rasmussen jumps that to 89. Up to a 12 point swing in that segment goes a long way toward explaining the overall swing from SurveyUSA’s R+2 to Rasmussen’s D+7.

But after this poll, I don’t want to hear about how Rasmussen’s systematically skewing things toward the Republican. I just don’t want to hear it, because the evidence isn’t there. I mean, unless you want to say that the firm’s deliberately swinging things the other way sometimes to cover its tracks, at which point you’re basically accusing Rasmussen Reports of being another Research 2000, a charge that needs a lot more proof than a few good-for-the-Republicans poll here and there.

Crossposted from Unlikely Voter

Tags: , , , , , , ,

COMMENTS

POSTING RULES  |  LOGIN TO COMMENT  |  REGISTER TODAY
  • rdelbov

    Rasmussen’s numbers–you either need the premium service or in some cases he does not give them out.

    A little simple math, however, shows that in this poll at least Rasmussen is showing a turnout very similar to 2008.

    This poll also may not be big enough-only 500 likely voters-to show the angry towards Boxer in the Central Valley and other Ag areas. Somewhere near 16% of democrats -indies should be in this area and this poll certainly does not show any discontent in it. I would suggest that the +1000 voters in the recent Field poll and SUSA would be a better indication of what’s going on in CA.

  • http://www.hakubi.us/ Neil Stevens

    The MoE encapsulates any issues with the sample size, unless you’re of the opinion that there are systematic errors in the sampling method.

    What is leading you to believe Rasmussen is missing entire regions of the state? Serious question; this is big news if you see it.

  • rdelbov

    89% of democrats voting for Boxer in a poll that’s too skewed towards san Francisco and LA democrats

    Follow this logic

    1. In 2004 Boxer got 58% of the vote
    2. In this poll she has 49% of the vote
    3. In 2004 she got 92% of democrats
    4. In this poll she gets 89% of democrats with 5% undecided

    Wow we have seen poll after poll where Boxer gets 80% or so of democrats. She only got 80% in her primary race. Now we see 89% for her with 6% undecided. She is matching her 2004 numbers but this is not the same rodeo.

    We have also see numerous polls and anecdotal evidence about the Central Valley area (That’s part of 18-19-11-20-21-23 congressional districts) suffering on the water issue. Boxer has voted to shut off their water and the valley is going brown. Yet this poll shows Boxer is winning 90% of the democrat vote. They don’t seem to be be polling in Fresno or Modesto.

    Rasmussen has found some very liberal democrats–It defies logic for her to loss indie support and not lose a smidgen of Democratic support.

  • http://www.hakubi.us/ Neil Stevens

    You’re expecting Democrats to be undecided at the same rate as the DTSes?

    It’s an interesting point but I’m not sure it’s indicative of much. I’d expect registered Democrats to be more likely to be decided for her than the squishy center left DTS poll in the state.

  • Russ Martin

    I don’t know much about polling methodology. I’ve got to believe that the mid-terms will show a significantly lower turnout for the Dems, versus the ’08 election – no presidential election, the overall apathy being reported on the Dems, as opposed to the passion growing on the right.

    Do the pollsters have a way to account for this? I know some whether the person being polled is likely to vote and I seem to remember Rasmussen does this. I’m just curious.

  • earlgrey

    How many voters that haven’t voted for years will work their way out of the woodwork, because they are made at one side or another? Maybe I just thrive on the idea of change (the good kind), but CA is one state that has had some of the biggest problems. It is hard to believe that there won’t be an impact by that.

  • thelibrul

    I think most of Rasmussen’s number differ because they have been using likely voter models as opposed to registered voter models. For example, they see huge swings when a candidate wins a primary as compared to other pollsters, which tend to immedialtey come back down to earth. (See Elaine Marshall and Rand Paul). Also their presidential ratings tend to have more dramatic changes as compared to Gallup. Recently Obama’s numbers we t from -15 to -1 within 4 days, which is a huge swing, whereas Gallup is typically steady.

  • http://www.hakubi.us/ Neil Stevens

    Every pollster addresses that differently. Some just poll registered voters. Others come up with different likely voter models. These models are structured in different ways. Some just ask the voter how likely he is to vote. Some base it on other factors like 2008 voting, partisan leanings, and the like.

    And there’s no way to know who’s right until November.

  • http://www.hakubi.us/ Neil Stevens

    But their model is more aggressive than others.

  • thelibrul

    Well, I think the pres. approval polls are all either RV or automated. I think it’s good to see what likely 2010 voters think of the prez bc how they view him certainly affects the midterms, but I don’t think LV models 2.5 years from the next prez ection mean that much for his reelect.

    Honestly, I think Ras is either going to look very good on Nov 3rd or very bad. They’re results tend to be more of an outlier as compared to others, yet make sense considering the enthusiasm gap.

  • tngal

    that will be on the ballot? Just curious. California always has a controversial prop or two on a ballot.

  • rdelbov

    but when he gets results that are out of line with other pollsters.

    1. His mix between D’s-R’s-I’s–his likely voter model is still closer to 2008 then what Field & other pollsters are projecting seems out of line. Boxer a 89 for D—82% for Carly F among republicans plus a 12% lead among indies still leaves Carly 7% behind. Where is the GOP surge in 2010 among likely voters?

    2. Boxer getting 89% of democrats is not only higher then three other recent polls but seems out of line with how democrats elsewhere are doing.

    3. Carly F. at 82% among the GOP voter base is not only low compared to a few recent polls but this is the only poll that has shown her getting less party support then Barbara B.

    Great pollster even the Yankees don’t win every game. I think its also more then a MOE problem as every part of the RAS model seems off

  • IJB

    The Initiative that’s getting the most media heat is the Initiative to legalize pot in toto (which is actually losing! in *CA*!! I did not see that coming!).

    Probably the most important Initiatives are the redistricting Initiative which will give redistricting to a “citizens commission” (which I’m sure will pass, unfortunately…), the 50+1 vote to pass budgets while still requiring 2/3 votes for tax increases Initiative (which I’ll vote for), and the Initiative that suspends CA’s inane “Global Warning” regulations.

  • SteveLA

    IJB

    Just like Prop 8, the Pot Head ballot measure will bring some to the polls who are not in favor of legalizing dope. How will that help or hurt each camp? I tend to think the Anti-Pot voters will be Firoria supporters…but that’s just a guess.

  • philbo

    She beat true-blue conservative Republican Chuck DeVore only because Palin interfered and deemed her as having the best chance of beating Boxer. This pissed off a lot of us out here. Fiorina is inexperienced and ran in the primaries mainly arguing that her opponents were men and this was a woman’s seat. She is testy when asked tough questions and narcissistic about “girly” things.

    She may end up winning in the end because it is, after all, Boxer. But a Tea Party backed DeVore would have captured the national mood and delivered a meaningful victory. It’s bad enough that the GOP is running an open-borders liberal Meg Whitman for governor. Very depressing. Good job Sarah!

    At least we have Tea Party candidate John Colbert ahead in the polls here in Pasadena. That should be a big story but no one ever reports on it. This was the seat that the Hollywood crowd targeted after Republican conservative Jim Rogan led the impeachment efforts on the House floor. Pasadena should always be a Republican district but the GOP has surrendered here. Thank God TEAPAC recruited John Colbert!

  • http://www.hakubi.us/ Neil Stevens

    RiNOs… can’t live with ‘em… can’t deport ‘em.

  • disintelligentsia

    that at least in prior years, the GOP in California has been pretty disspirited. It’s hard to get really excited when year after year the party has been offering RINO’s like Arnold and we’ve been a permanent minority in the legislature with no real chance previously to knock off Babs or Fein(franken)stein. This kind of on-the-ground reality does likely suppress voter turnout for GOP voters and the RAS model just incorporates this.

    However, I think this year is not typical. We have a different climate and with the potential to take out Babs I think our GOP voters will smell the blood in the water and turn out the vote. There is the real hope that the CA GOP will see new life in it and can be a force once again – particularly if the voters are disenchanted enough with our current budget crisis in CA (not to mention the nation).

  • disintelligentsia

    He was a Devore supporter (look at his posts) and probably isn’t very excited about Fiorina. God knows I have my reservations about her, but I am ready to vote for her because the place for dissent was in the primary – where I did vote for Devore.

    Since he did support Devore, I wouldn’t be labeling him a RINO (unless you think his support was some kind of false flag op because he though Devore would be a weaker candidate in the General election). I guess you could define him as a RINO in the sense that he is too conservative for the GOP and isn’t thrilled with its leadership (as he advocated a third-party tea party concept in CA in one post). But everyone knows that’s not the typical definition of RINO – which is a liberal pretending to be a Republican.

    Based on philbo’s support of Devore I would reinstate his registration. Honest differences of opinion should be respected.

  • proudgop

    I am not sure what to make of the two races in California

    1.How does Gov Race Impact Senate Race?
    2. I have been concerned with SUSA poll showing Boxer leading with Independents when thats not been case for any Dem of lately
    3. Carly isn’t spending that much of her money so far I sorta hope she ramps it up and is able to define herself in positive light
    4. CA like NY and IL might not sample Dems correct in that many people in big cites San Fran/LA might not even have phone numbers and only cell phones and those do tend to scew to left

  • tngal

    one way or the other. So I was wondering how they played out in the polls neil was referring to – or were those questions even asked of the ones answering the survey?

  • IJB

    1. Neil thinks Whitman will help Fiorina. I look at these races as being more ‘decoupled’ (with Whitman having the much better chance), though I’m sure Whitman will help Fiorina on the margins.

    2. Any poll that shows something at variance with every other poll can probably be ignored. However, that said, it would not surprise me at all if CA Independents skew more Democrat than nationally – however, even I’m dubious that Boxer’s actually *leading* with CA Indies…

    3. Fiorina’s resources are more limited that Whitman’s – she’s probably smart to wait and not go after Boxer until after Labor Day.

    4. I’m pretty sure pollsters can normalize for that kind of stuff.

  • eburke

    in the business. That’s why he’s been so accurate over the years.

  • http://www.hakubi.us/ Neil Stevens

    I’ve said the same thing myself at UnlikelyVoter and probably xposted it here. Rasmussen will be a genius, or discredited, in the end. That firm is really out on a limb.

  • http://www.hakubi.us/ Neil Stevens

    If you’re going to come in here and bash our top of the ticket nominees, then you’re a RiNO.

    Just because you backed Chuck DeVore, it doesn’t mean you’re not a RiNO.

    If Chuck can endorse Carly and move on, then any real Rep ublican can do the same. philbo couldn’t, and I have zero tolerance for that junk.

    He can apply for reinstatement if he wants, and if the site leadership decides to bring him back, he’ll be back and that’ll be the decision made.

    But really, coming into my thread to bash Whitman and Fiorina is really dumb if you want to keep your account.

  • http://www.hakubi.us/ Neil Stevens
  • http://www.hakubi.us/ Neil Stevens

    Though general elections *Should* be easier to poll than primaries and specials, so a polling disaster like that day in PA should be less likely in a general.

  • http://www.hakubi.us/ Neil Stevens

    They’re starting to mix in portable-only people in their poll results.

  • nivlem

    convicted felons and dead people.
    They are voters again, remember?

  • cactusjack

    the Ukrainian Elvis type dude that looked like a cross between Lurch and Robert Goulet. Thanks, Neil.

  • http://www.hakubi.us/ Neil Stevens

    There are other videos floating around, interviewing him all these years later now that YouTube has made him popular again. I wish I could understand what he’s saying about it all.

  • rdelbov

    inspires a passion level of democrats equal or nearly equal to Obama in 2008? I don’t buy. 89% for Boxer with 5% undecided among democrats–Obama got 92% in 2008. Obama got 59% in 2008 and Boxer is 49% with roughly the same democratic vote.

    California’s indies (many moderates and liberals-some conservatives) go from 64-31 for Obama to a 12% lead for Carly F but the democrats from the central valley to San Diego man up for Boxer in 2010?

    I don’t buy it.

  • cactusjack

    if the exiled could only come back and see what they were replaced with, they would be so jealous.

  • tngal

    “If you?re going to come in here and bash our top of the ticket nominees, then you?re a RiNO.”

    I did not vote for Obama, neither did I put my checkmark beside McCain, our “top-of the-line nominee at the time. I supported Fred ( even though he had technically dropped out.) I’m not Rino. I’m a person who still sleeps well at night. McCain hit it right on a couple of issues, namely militarily, but wrong on a few others. His wrongs were unforgivable. I’m still not convinced McCain’s in our corner on those negative issues despite Sarah’s accolades of her former dance partner.

    Everyone has that ONE candidate they’re not comfy with, regardless of the l(R) by thieir name. Although I do agree w/you that philbo is wrong on this one. So, now you can hate me. Chocolate?

  • IJB

    In PA & OH, Rasmussen has gotten one set of results, and PPP and Quinnipiac have gotten another. Consistently.

    So, PA & OH are going to settle this issue one way or the other.

    (The other place is FL – Ras has consistently polled more to Rubio’s benefit, while others have had Crist up much higher. So FL will answer this question too…)

    (OTOH, KY is either going to make PPP look like geniuses, or it’ll leave them with a bunch of egg on their faces…)

  • http://www.hakubi.us/ Neil Stevens

    Honestly, you can vote however you want, and of course I mean that. It’s the American way.

    But most of us know not to come to RS with our deviations.

    philbo now learns that lesson.

  • IJB

    There were, IIRC, three polls released just before that election – two showed Burns (slightly ahead), and one showed Critz ahead. That’s not enough to say “the outlier was right” because the sample set basically wasn’t big enough to even declare an “outlier”.

    And the reason the two pro-Burns got it wrong is they projected much higher Indie turnout than actually happened.

    That’s why I actually think PA-12 may be a surprise GOP upset in November – if the Indies come out like the poll samples back in May had them, then Critz is not going to win by 7.5%. Far from it…

  • http://www.hakubi.us/ Neil Stevens
  • http://www.hakubi.us/ Neil Stevens

    But redistricting is going to be a huge fight, and it wouldn’t surprise me if the marijuana picked upsteam.

    These wars don’t really start until the million dollar ad campaigns are under way.

  • IJB

    6 or more is always what you’d prefer. But even 4 or 5 would be better than 3…

  • http://www.hakubi.us/ Neil Stevens

    Yes that’s a fair point.

    I thought you meant the sample sizes of the individual polls.

    Of course, the only race that ever gets a *good* sample of polls is the Presidential.

  • barleycorn

    Anyone else get the feeling we worried too much for 45 years?

  • Adjoran

    One of the strongest components of most “likely voter” tests is voting in previous elections and, for midterms, voting in previous midterms.

    My guess is Rasmussen puts heavier weight on this than upon voter enthusiasm, or perhaps doesn’t tightly qualify self-proclaimed enthusiasm.

    Compare Boxer’s electoral record and the Democratic record in California generally to the state of the state, with her job approval lower and high unemployment today, It defies belief that she will draw the same numbers of Democrats as she did when times were better for her state and Party.

    2008 in particular was an unusual year, with drastically increased participation by blacks and young voters who typically turn out in far lower percentages even in Presidential years. The history of these demographic groups is even poorer turnout in midterms.

    If Boxer can hold her Party turnout steady under such conditions, she will be Senator for Life. I don’t buy it.

  • http://www.hakubi.us/ Neil Stevens

    I wouldn’t expect people to have opinions about them until September anyway. :)

  • thelibrul

    Because Ras has been such a prolific pollster (they must have done at least 50% of the polls) they have truly driven a narrative of robust Republican domination, while at the same time having polls that tend to be more of an outlier when compared to other good pollsters, such as PPP and Quinnipiac. But if you take their numbers out of the mix, Republicans don’t do nearly as well. So, Ras’ numbers have created a certain perception of total doom, which has set expectations for Republicans very high, and if they’re wrong there will be a lot of people scratching their heads. I think PA-12 was one of those, although imperfect due to the Dem Sen primary. It’ll be interesting to watch (especially if you obsess over all things statistics, as I do).

  • thelibrul

    N/T

  • thelibrul

    They couldn’t even lip-sync songs without words correctly!