05/09/2013 04:29 pm ET

POLLSTER UPDATE: Census Turnout Data Shows Record Black Turnout


Political data geeks rejoiced over the release of the biennial Census report on voter turnout, a report which shows black turnout exceeding white turnout for the first time ever -- or perhaps for the second time. Ed Markey's lead in Massachusetts looks bigger than it did last week. And, without the links, could you guess which Outlier item is the Onion headline and which is from Business Insider? This is the HuffPost Pollster update for Thursday, May 9, 2013.

CPS SURVEY FINDS SURVEY FINDS HIGHER TURNOUT FOR BLACKS THAN WHITES - NY Times: "The turnout rate of black voters surpassed the rate for whites for the first time on record in 2012, as more black voters went to the polls than in 2008 and fewer whites did, according to a Census Bureau report released Wednesday...According to the Census report, 66.2 percent of eligible blacks voted in the 2012 election, compared with 64.1 percent of eligible non-Hispanic whites. An estimated two million fewer white Americans voted in 2012 than in 2008, just as about 1.8 million more blacks went to the polls, more than 90 percent of them voting to re-elect President Obama, exit polls showed." [NY Times]

A drop among younger voters - George Mason University Professor Michael McDonald: "Republicans may feel encouraged for their future election prospects, in that fewer non-Hispanic Whites voted in 2012 than 2008, since Whites tend to vote Republican. However...The good news for Democrats is that the decline in non-Hispanic White turnout was primarily due to younger people abstaining. Turnout among all 18 to 24 year olds dropped 7.3 percentage points between 2008 and 2012. Indeed, this decline is evident across all races and ethnicities...These young people also present an opportunity for Republicans because they are the most persuadable potential voters, in that they have not yet become as fixed in their partisanship as people tend to do as they age." [HuffPost]

Demographic change - TNR's Nate Cohn: "The big problem for Republicans is that the primary cause of the declining white share of the electorate is demographic change, not high black turnout... If the exit poll survey results are plugged into the CPS, and adjusted for the discrepancies between the exit poll's electorate, the CPS, and the actual results, then Obama would have won the election by 2.6 points with '04 minority turnout rates and 1.9 points if white and minority turnout returned to 2004." [TNR]

A PRIMER: Reasons to love the CPS... - Is this report based on just another survey? No. The voting statistics come from the Current Population Survey (CPS), a survey administered by the Census Bureau for the Bureau of Labor Statistics (BLS) for the purpose of measuring the monthly employment rate and other labor force characteristics. Because it is designed to accurately measure these stats at the national and state level, the CPS has an enormous sample size, roughly 60,000 households per month. Initial interviews are conducted in person and, because participation is compulsory, the overall CPS response rate is over 90 percent -- a rate conventional surveys come nowhere close to matching. [Census.gov]

In November of even-numbered years, CPS includes two "supplemental" questions on voting, specifically whether each member of the household is registered to vote and whether each voted in the general election (the standard methodology of CPS is for one person in each household to answer questions on behalf of everyone in that household). [CPS Voting and Registration]

So the reasons to value the CPS voter survey is that it features a true random sample, involves a huge number of interviews (allowing for analysis down to the state level) and achieves exceptionally high coverage and response rates.

...and reasons for caution - But as with any survey, some caveats are in order. First, the vote questions are self-reported, and as is well known, some respondents say they voted when they haven't ("for the purpose of appearing to behave in a socially desirable way," the Census explains). [CPS Report]

Second, as Michael McDonald explains at length, the CPS voting supplement has an extra layer of "non-response." Respondents agree to participate in the monthly CPS survey eight times over the course a year. In any given month, some respondents fail to complete the questionnaire. In November, roughly in seven respondents failed to respond to the voting question, either because he or she skipped the question or didn't complete the supplemental survey that month. [HuffPost]

Why the non-response matters - In calculating the turnout percentage, the Census treats missing data as a "no" response, which makes the over-reporting of turnout appear smaller. Also, the percentage with no answer to the voting question declined slightly from 13.8 percent in 2008 to 12.8 percent in 2012, but because those who didn't answer had been interviewed previously, their demographics can be taken into account. McDonald: "The decline in non-response was mostly centered among African-Americans, of which 18.0% did not respond in 2008 compared with 15.0% in 2012. In comparison, non-Hispanic White non-response declined from 12.8% in 2008 to 12.2% in 2012....When non-respondents are removed from the turnout rate calculations, African-Americans first exceeded non-Hispanic White turnout in 2008 -- 78.9% to 75.5%. They did so again in 2012, 77.9% to 72.8%." [ibid.]

More on over-reporting - Pew Research's Paul Taylor and Mark Hugo Lopez: "The gap between the voter turnout reported by the Census Bureau and the actual national vote tallied by election officials widened in 2012. According to the Census Bureau's 2012 Current Population Survey November Supplement on Voting and Registration, an estimated 133 million U.S. citizens voted in 2012. That estimate is higher—by 4 million votes—than the national tally of 129 million votes cast for president." [Pew Research]

MASSACHUSETTS: TWO NEW POLLS SHOW MARKEY LEADING GOMEZ - Two news Massachusetts surveys show Democratic Rep. Ed Markey leading Republican nominee Gabriel Gomez by comfortable margins. A MassInc/WBUR poll has Markey leading by 8 points (46 to 38 percent), while a Suffolk University poll has Markey ahead by 17 (52 to 35 percent). Two automated polls released last week showed a closer race between the two candidates [MassInc, Suffolk, Pollster chart]

Suffolk "house effect?" - The very large Markey lead on the Suffolk poll is consistent with two Suffolk surveys conducted in the fall of 2012, which both overstated Barack Obama's victory over Mitt Romney. The final Suffolk polls had Obama leading Romney by 33 and 32 percentage points respectively; Obama defeated Romney in his home state by a 23 point margin (61 to 38 percent). Suffolk's polls on the 2012 Senate race were generally in line with other surveys. Their final poll had Warren ahead by 7; she won by an 8 point margin. [2012 presidential chart, 2012 Senate chart, 2012 results]

THURSDAY'S 'OUTLIERS' - Links to more news at the intersection of polling, politics and political data:

-Two thirds of Americans support Jason Collins' decision to come out. [WaPost, ABC News]

-75 percent say say immigration policy needs at least major changes, with 35 percent saying it needs to be "completely rebuilt." [Pew Research]

-Rahm Emanuel's job approval rating slips, especially among black voters. [Tribune]

-Americans still down on Congress, still like their own representative better. [Gallup]

-Both Israeli's and Palestinians want President Obama to play a larger role in resolving the Israeli-Palestinian conflict. [Pew Research]

-Charles Franklin will join the Marquette University Law School full time to run its poll. [Marquette]

-A "push poll" allegedly targets mayoral candidate Wendy Greuel. [LA Times]

-Sides and Vavreck report on the how much the air war and ground game mattered in campaign 2012. [Monkey Cage]

-Open and machine readable now the standard for all government data. [White House, via @PaulBlu]

-What the Obama campaign's chief data scientist is up to now. [Atlantic]

-"20 percent of Democrats say they have 'been to the club.' Only 10 percent of Republicans have 'been to the club.' [Business Insider]

-"Malicious Focus Group Convinces Marketers Cinnamon Mountain Dew Is The Next Big Thing." [The Onion]