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.