Quinnipiac finds the Democrat leading in Virginia which means...get ready for the 2013 edition of party ID wars. Their omission of the Libertarian candidate prompts a pollster conversation. And we solved the Slate gerrymandering puzzle in 7'54"; what's your time? This is HuffPollster for Wednesday, August 21, 2013.
MCAULIFFE LEADS VIRGINIA GOVERNOR’S RACE POLL - HuffPost: “Democrat Terry McAuliffe has a 6-point lead over Republican state Attorney General Ken Cuccinelli, according to a Quinnipiac University poll released Wednesday, with McAuliffe taking 48 percent to Cuccinelli's 42 percent among Virginians likely to vote in this year's election….Voters were torn on McAuliffe, with 34 percent viewing him favorably, 33 percent unfavorably, and 31 percent saying they hadn't heard enough. Cuccinelli had more detractors, with 35 percent rating him favorably, 41 percent unfavorably, and 22 percent undecided….Although this survey is the first released on the race since mid-July, polling has consistently shown a close race between McAuliffe and Cuccinelli. Quinnipiac's previous polls this May and July, surveying a wider pool of registered voters, gave McAuliffe a 4-point and 5-point lead, respectively. HuffPost Pollster's estimate, which includes all publicly available polling, currently puts McAuliffe ahead.” [HuffPost, Quinnipiac]
Democratic 'turnout advantage?' The Washington Post's Scott Clement highlights the Democratic partisan advantage in Quinnipiac's sampling of likely voters: "Democrats appear poised to maintain their turnout advantage from 2012, when the exit poll showed 39 percent of voters identified as Democrats but just 32 percent as Republicans. Democrats outnumber Republicans by an identical seven-points among likely voters in the Quinnipiac survey, and both candidates win at least 90 percent of their fellow partisans...What’s most striking is that Democrats’ five-point party identification advantage from a July Quinnipiac poll was not diminished by the firm’s switch to likely voters this month. This indicates that Cuccinelli is not benefitting from a narrower likely voter electorate, a factor that may make a victory hard to pull off." [WaPost]
Republican pollster appears dubious - GOP pollster Glen Bolger, via Twitter: "The [Quinnipiac] Poll says turnout will favor the Dems as much in '13 as in '12. It was 32% GOP/39% Dem in '12 and 37% GOP/33% Dem in '09." [@posglen]
But Quinnipiac's likely electorate looks different than 2012 - The New Republic's Nate Cohn compares the demographics: "Compare the racial composition of today’s poll to Quinnipiac’s final survey of the 2012 presidential election. In their final poll last October, President Obama held a 5 point lead, about the same as his eventual 3.9 point victory. The electorate was 68 percent white, 19 percent black, and 4 percent Hispanic, with the balance saying “other” or not responding. Today, McAuliffe holds a 6 point lead, but the electorate is distinctly less diverse: 72 percent white, 16 percent black, 2 percent Hispanic. The same thing is true for age. Quinnipiac didn’t release the weighted share of people over age 65 in last October’s poll, but the Census and exit polls put seniors at 13 and 14 percent of the electorate in 2012. Today’s Quinnipiac poll shows people over age 65 representing 21 percent of the electorate...No. Turnout does not “favor” Democrats in the Quinnipiac poll. Party ID is an attitude, not a characteristic. We don’t have a great idea of what it would look like if Democrats got a “good” or “bad” turnout in 2013. If you want to see whether Democrats are getting their 2012-esque turnout advantage, check the verifiable stuff, like age and race. By those reliable metrics, it’s completely obvious that this is not the 2012 electorate. And yet, McAuliffe still has a distinct advantage." [New Republic]
And Quinnipiac found less partisanship - While the margin separating Democrats and Republicans is similar to what exit polls found in 2012, the absolute percentages for the Quinnipiac poll composition are very different: 30 percent identify as Democrats, 23 percent as Republicans, 39 percent as independents, leaving 8 percent unclassified. Thus, a total of 53 percent identify with one of the two parties, nearly 20 points lower than the partisan self-ID on Virginia exit polls in 2012 (71 percent) and 2009 (70 percent). That difference is typical, as voters are usually in a more partisan mood just after casting their ballots as they are months before. It also allows for the Democratic advantage to narrow as independents are pressed on how they "lean" or feel more partisan closer to election day. [Quinnipiac poll demographics]
WHY NO LIBERTARIAN? - The Quinnipiac poll's vote preference question offered only McAuliffe and Cuccinelli as choices and omitted libertarian candidate Robert Sarvis. In summarizing the results on Twitter, campaign handicapper Larry Sabato labeled that exclusion a "questionable practice," noting that just "2% volunteered 'someone else'". Previous polls conducted by Roanoke College and Public Policy Polling (PPP) have found support for Sarvis at 5 and 7 percent respectively. [Roanoke, PPP]
Sabato's tweet prompted @researchrants to ask, "Don't REAL 3rd party voters just angrily choose "other" & those who really won't will just pick a major candidate, like on real life ballot?" He later asked us (or at least the one of us known as @mysterypollster on Twitter): "Any evidence re naming 3rd party candidates?... my HUNCH is it's better not to name."
Our response recalled a five-year-old National Journal column noting the way vote preference questions that prompted for third party candidates in the summers of 2000 and 2004 tended overstate their support as compared to later polls and the election results (and the same thing happened again in 2008). [National Journal]
Exit pollster Joe Lenski responded that such overstatement is "usually true but there are cases where it isn't especially where major party candidates have high negatives." He offered examples: "from 2012 - Libertarian candidates polled 6% & received 6% in IN & MO Sen, thanks to Mourdock & Akin...2006 example - Green Party candidate polled 12% & received 10% in IL Gov thanks to Blagojevich."
Massachusetts pollster Steve Koczela noted that the decision of whether to include a third party candidate is "always a bit of a balancing act. If no poll names 3d party, could miss support they do have....And opens polls up to accusations of self-fulfilling prophecy, fair or not."
Sabato closed the discussion by pointing out that "there are ONLY 3 candidates for VA GOV. So the usual problem of crowding," that sometimes makes pollster reluctant to read all of the third party candidate names in a telephone interview "isn't present" in Virginia this year. [All tweets compiled on Storify, some quotations edited slightly for clarity]
CHRISTIE’S MASSIVE LEAD OVER CHALLENGER SLIGHTLY NARROWS - Monmouth University: “Garden State voters likely to participate in the November election give Christie a 56% to 36% lead over Buono. Two months ago, he had a 30 point advantage (61% to 31%). Republican (90%) and independent (64%) voters remain firmly in the governor’s column, but Democrats are starting to return to their partisan inclinations. Christie currently garners 21% of the Democratic vote to 71% for Buono. In June, the GOP governor enjoyed the support of 36% of Democrats compared to 59% who backed their own party’s nominee.” Buono’s 20-point deficit is the closest she’s come to Christie in a poll since before Sandy hit in 2012. [Monmouth, HuffPollster chart]
HUFFPOLLSTER VIA EMAIL! - You can receive this daily update every weekday via email! Just enter your email address in the box on the upper right corner of this page, and click "sign up." That's all there is to it (and you can unsubscribe anytime).
WEDNESDAY'S 'OUTLIERS' - Links to more news at the intersection of polling, politics and political data:
-Young adults have little awareness of Obamacare. [HuffPost]
-Americans paid more attention to embassy closures and baseball suspensions than to the sale of the Washington Post or the first lab-grown hamburger. [Pew Research]
-A poll sponsored by Marjorie Margolies Mezvinsky shows that Democrats in her old district still know and like her. [Inquirer]
-Harry Enten takes on a leading question in a poll on Lamar Alexander. [Guardian]
-Jim Henson and Joshua Blank examine the mixed and often countervailing attitudes about NSA surveillance. [Texas Tribune]
-Chris Cillizza defends 2016 speculation. [WaPost]
-David Lublin compiles election datasets from around the world. [Election Passport]
-Slate turns gerrymandered districts into jigsaw puzzles. [Slate]