Swing State Project criticized the SurveyUSA poll of NY-01 that was sponsored by Firedoglake with regard to the weighting of 18-34 voters in the 2010 election, which was subsequently promoted by Nate Silver. Chris Cillizza of the Washington Post simultaneously had the exact same criticism of the same data point. For the record, no one at Firedoglake had anything to do with the actual manner in which the poll was conducted. We had zero say in the methodology, determining the sample make up, or the weighting. Firedoglake only requested it be a poll of likely voters with a sample size of 600 in specific Congressional districts with regard to the individual mandate.
Below is a statement from SurveyUSA:
Decisions about question wording, question order, methodology and data weighting are made by SurveyUSA. The client has not attempted to tell SurveyUSA how to conduct the research. Concerns about process, weighting and/or results should be addressed to SurveyUSA.
SurveyUSA re-weighted the figures in the poll to address the concerns of SwingState. These were the original figures:
| 600 Likely Voters | All | Gender | Age | <50 / 50+ | Race | Age | ||||||||||
| Margin of Sampling Error: + 4.1% | Male | Female | 18-34 | 35-49 | 50-64 | 65+ | 18-49 | 50+ | White | Black | Hispanic | Asian/Ot | Republic | Democrat | Independ | |
| Tim Bishop (D) | 47% | 42% | 52% | ** | 42% | 46% | 50% | 43% | 48% | 48% | ** | ** | ** | 23% | 75% | 40% |
| Randy Altschuler (R) | 45% | 52% | 40% | ** | 52% | 49% | 40% | 51% | 44% | 44% | ** | ** | ** | 66% | 19% | 55% |
| Undecided | 8% | 6% | 9 | ** | 6% | 6% | 10% | 6% | 8% | 7% | ** | ** | ** | 11 | 6% | 5% |
| Total | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100%t | 100% | 100% | 100% |
| Composition of Likely Voters | 100% | 48% | 52% | 1% | 19% | 38% | 41% | 21% | 79% | 91% | 3% | 3% | 3% | 27% | 33% | 39% |
These are the figures they got when they doubled the size of the age (18 to 34) respondent population:
| 600 Likely Voters | All | Gender | Age | <50 / 50+ | Race | Age | ||||||||||
| Margin of Sampling Error: + 4.1% | Male | Female | 18-34 | 35-49 | 50-64 | 65+ | 18-49 | 50+ | White | Black | Hispanic | Asian/Ot | Republic | Democrat | Independ | |
| Tim Bishop (D) | 47% | 42% | 51% | ** | 42% | 46% | 50% | 43% | 48% | 48% | ** | ** | ** | 23% | 74% | 40% |
| Randy Altschuler (R) | 46% | 52% | 40% | ** | 52% | 49% | 39% | 51% | 44% | 45% | ** | ** | ** | 67% | 20% | 54% |
| Undecided | 8% | 6% | 9 | ** | 6% | 6% | 10% | 6% | 8% | 8% | ** | ** | ** | 10 | 6% | 6% |
| Total | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100%t | 100% | 100% | 100% |
| Composition of Likely Voters | 100% | 48% | 52% | 3% | 24% | 38% | 36% | 26% | 74% | 90% | 3% | 3% | 3% | 27% | 33% | 39% |
The results go from 47-45 to 46-45. Jay Leve of SurveyUSA says:
The outcome is no different. . . . [T]he one point that the Republican picks up is not significant. Yes, this MAY to some be counter-intuitive; the youngest voters do slightly favor the Democrat. But so do the oldest voters, and if you manipulate the data to shift the electorate younger, you are, in effect, reducing the significance of the older voters. And in this set of data (not in every set of data, but in this set of data), that’s effectively an offsetting correction.
He also responds to SwingState:
The sample is not “pretty weird.” It is a sample of likely mid-term voters. There is no one “right” way to draw a sample in a congressional district 10 months from a midterm election, but this sample was drawn carefully and defensibly. To be included in the sample (SurveyUSA’s criteria, not the client’s): the voter had to be registered with the secretary of state; had to have a telephone; had to have voted in 2008 and had to have voted in either [2006 or 2002], and had to confirm that he/she resided in the district being surveyed. By design, this was not a survey of registered voters (which would have resulted in a younger sample).
Reasonable people can disagree about exactly what percentage of the electorate in 2010 will be age 18 to 34 , and I am not defending any specific turnout target. But most would agree that midterm voters are older. That’s what these results show. When SurveyUSA re-weights the respondents in NY-01 to be younger, the survey results do not meaningfully change. This may seem to some counter-intuitive; it is not.
When, for internal analysis, SurveyUSA re-weights the respondents to be younger in AR-02, OH-01, and IN-09, the survey results do not change.
Firedoglake selected SurveyUSA because of their very long track record of accurate polling. The Wall Street Journal found them to be one of the most accurate pollsters in the 2004 election and Nate Silver ranked them the second most accurate pollster in May of 2008. We have fully confidence in SurveyUSA and selected specifically because they are one of the best in the field. We fully trust them to conduct the poll in a manner that will produce the most accurate data possible about the potential electorate.



17 Comments
Thanks Jon!
What are the turnout demographics for MA? As a total percentage, it was just as high as 2006.
it’s a shame that my comments regarding the Green Party in Rep Grayson’s thread have been scrubbed and I can no longer post there. I was polite and used no blue language and he hardly seems like a shrinking violet. Strange though, coming from Jane Hamsher who was against the HCR Senate bill, then kinda for it….then again Natural Born Killers was a mess too.
Thats the right wing for ya! challenge empirical data and industry standard methods with distortion and emotional, relativist, un-scientific ad homenim attacks.
I have never been for the Senate bill. If you’re habitually spreading disinformation in the comments you should expect to moderated, the mods frown on that.
Citizen Jane Hamsher:
I had an “experience” with one of your “hall monitors” in the Grayson post that was pretty illuminating…I don’t wanna associate myself with Citizen YngwieMalmsteen but I had a question addressed respectfully to the good Congressman about his early clerkships and his take on the SCOTUS decision released today and it was scrubbed untill an hour later I repeated the question, which went unanswered, after about a bazillion folks had asked the same question withOUT the reference to his clerkin’ for lunatic libertarian judges. What’s goin’ on Sister Jane, I have been very concerned about your thin skinned responses to lomg time FDLers who, if they don’t always adhere to your party line, at least take you seriously enough to try and keep you honest. Whats’ goin on Dear Heart, I’m wonderin’ about whether there are others objectives for FDL than providing a community-like forum for people to communicate with each other?
Citizen Jon Walker:
The corporate hounds have been sicced on FDL because this forum more than KOS presents a direct threat to any credibility that these suede shoe salespeople have with the general public. Now that SurveyUSA has responded, maybe an even wider survey with the same methodology and specific target might not just nail the coffin shut on the bastards.
Hell bro, I got a nasty attack from a front pager today as well. It’ll take more than that to run me off.
Take your meds and settle down.
I think they are turning anti-veteran myself.
If the “pretty weird” comment was silver’s ( it was tweeted by 538) then i have 0 respect left for him. I felt sorry for him, watching him trying to rebut Jane Hamshire on rattigans show. even with the expensive suits and the manhattan quaff i guess its still hard for him to talk to girls.
some of the front pagers go all completely kiss ass when theres a “grown up” on line here at FDL. i have to attribute it to age and inexperience because no one could possibly hope to gain anything else from it here.
Yea, and some people here just don’t like each other.
there’s nothing wrong with being liked, or not liked, until it gets in the way of more important shit.
I’m flattered you think I have such control over space and time, but I don’t do the moderation Norske.
And the idea that we don’t let people express a variety of opinions here is probably the funniest thing I’ve heard all day. But since the moderation staff spend the past few days exhausting themselves trying to protect the right of people to say what they want during a very heated period by interacting with them rather than banning anyone, they might not find it so funny.
I am proof of freedom of expression here. I do appreciate the opportunity to vent. I don’t post here to be liked so i don’t complain when i find im not liked. No other blog allows the amount of expression tolerated at FDL,I was banned for a month from the CPUSA blog for being to confrontational. This is not an “establishment” blog and I hope it never is. I find that all the more remarkable from the editors point of, considering how much scrutiny this blog gets. Im sure cilizsa and silver and yglesias dont spend time reading the comments, but maybe they do. their heads are fat enough. Without delvolving into ass kissing, thanks Jane Hamsher, and Marcy Wheeler, and Jon Walker and everyone else. EVERYONE(progressive) who posts here. If FDL blogger, Glen Greenwald and a few other are the standard going into the future, and The blogs will be the Newspapers of the 21st century, but much more useful because of the interactive capablity.
Indeed. I say things here I could never say over at Daily Kos, mainly because a certain group of people has worked the refs very successfully so that any terms referring to certain forms of behavior on their part (such as “Stalinist” or “cult of personality”) are excuses for troll-rating or even outright banning.
Anyway, back to the topic of this post:
This is telling. SSP, 538 and the WaPo’s Cillizza all singing in unison from the same hymnal. In harmony, too.