A new poll shows the governor's race in Virginia is still close, and the New Jersey Senate primary still isn't. A Republican pollster rants about PPP's methods. And in less predictable news, George Will defends the ACS. This is the HuffPost Pollster update for Tuesday, July 16, 2013.
MCAULIFFE EDGES OUT CUCCINELLI IN VIRGINIA POLL - Democratic firm Public Policy Polling: "PPP's most recent poll of Virginia voters shows a tight race for the
governorship brewing in November. Democratic nominee and former chairman of the
Democratic Party Terry McAuliffe leads Republican Attorney General Ken Cuccinelli
41% to 37%, with Libertarian nominee Robert Sarvis receiving 7% of the vote.
Cuccinelli is viewed in a mostly negative light by the state, as 32% of the state has a
favorable opinion but 47% has an unfavorable opinion, including 57% of independents." [PPP]
Cohn: 'Things Are Looking Up For McAuliffe - The New Republic's Nate Cohn is bullish on McAuliffe's candidacy, but not because of the new PPP results. The head to head numbers haven't moved, he writes, "because persuadable voters just aren't paying attention yet. According to a PPP poll out today, 57 percent of voters don't have an opinion of [GOP lieutenant governor hopeful E.W.] Jackson; 30 percent don't have an opinion of McAuliffe, and 22 percent don't have an opinion of Cuccinelli. As a result, there are plenty of undecided voters, with McAuliffe and Cuccinelli only combining to hold 78 percent of the vote." Rather, Cohn thinks Jackson's "extremism" and a scandal embroiling Republican Governor Bob McDonnell "raise the real possibility of widespread revulsion toward the Republican ticket. McAuliffe will have the resources to exploit the GOP's weakness... In a polarized state like Virginia, the race will probably stay close. But as the campaign gets under way and the ads turn negative, it's easy to envision McAuliffe solidifying a modest lead. At this point, you'd rather be McAuliffe." [New Republic]
More results tomorrow - In an email, Quinnipiac says it's releasing “results of a poll of Virginia voters, asking their opinions about Gov. Bob McDonnell and the issue regarding his ties to businessman Jonnie Williams Sr." New results on the governor's race will likely follow later in the week.
BOOKER MONOPOLIZES NJ SENATE PRIMARY FIELD - Monmouth, applying the first likely voter screen in the race: “Newark Mayor Cory Booker maintains a daunting lead in the Democratic nomination contest to fill the late Frank Lautenberg's U.S. Senate seat. The Monmouth University Poll finds Booker doing well among all Democratic groups, including in his opponents' own backyards....Booker currently claims support of nearly half (49%) of likely voters in the Democratic primary to be held four weeks from today. Significantly fewer say they intend to vote for Congressman Frank Pallone (12%), Congressman Rush Holt (8%), or General Assembly Speaker Sheila Oliver (3%). Another 28% are undecided." [Monmouth .pdf]
GOP POLLSTER UNLOADS ON PPP - B.J. Martino, a senior vice president at the Republican polling firm The Tarrance Group, went on an 30-minute Twitter rant on Tuesday questioning the unorthodox method used by PPP to select samples and weight data: "Looking at @ppppolls new VA SW. Wondering how many interviews they discarded to get down to 601 completes? Because @ppppolls discards a LOT of interviews. Of 64,811 conducted for @DailyKos /SEIU in 2012, they discarded almost 23K. Sure, a handful of the @ppppolls discards were not valid interview responses. Most appear valid completes. @ppppolls says discarding interviews is a kind of retroactive quota on race, gender and age. Why just not weight the data? [@BJMartino tweets aggregated on Storify]
PPP's explanation of how they weight data - PPP's explanation of their method appears on their "About Us" page: "Our first step in weighting is to survey more than enough people. This allows us to then be able to randomly reject individual surveys from demographics that are overrepresented. For example, in our polling more women answer relative to men, and not enough African-Americans answer our surveys. Our random selection eliminates any potential bias from the rejections, plus it functions like a quota, only after the fact. PPP also employs a mathematical weighting scheme that assigns a weight based on each demographic." PPP confirms that for the DailyKos/SEIU national surveys Martino examined from 2012, it weighted by only one variable, age. PPP has not conducted a survey sponsored by Daily Kos since November 2012. [PPP]
Martino sees partisan shifts when controlling for age & gender: Martino tweet critique continues, reporting on numbers he tabulated using raw data published by DailyKos throughout 2012: "I looked a little deeper into that last pre-election poll @ppppolls did for Daily Kos/SEIU. The discards, @ppppolls claims, controls for older white women. So I controlled for them. The Party ID of @ppppolls discarded age 45+ white women was R+4.5. Of those remaining? R+10.2. Interestingly, the difference on the ballot from @ppppolls discarded data to remaining unweighted: Romney +7.6, and Romney +17...Independent older white women in @ppppolls final poll, the discarded were Romney +8.7, the remaining Romney +19.4." [ibid]
So discarding interviews made one subgroup in one PPP poll more supportive of Romney? What's the gripe? Via email, Martino clarifies: "The random process they use to discard older white female interviews in this case, changes the reported composition of and the opinions of the older white females who remain. The discard process can be (can be, not saying is) manipulated to produce desired results. Even random discards within a selected sub-group can be the result of choices the pollster made. Ultimately, why discard at all when you are already weighting after the fact?"
Response from PPP - In response to an email query, PPP's Tom Jensen defended his company but not the specific methodology challenged by Martino: "I'm sure there are as many methods for weighting polls as there are polling companies. We've been doing things the way we do them for over a decade and it's served us well. Our polls don't have a record of bias no matter how much Republicans want them to and our new VA poll is very similar in its findings to the last ones from Rasmussen and Quinnipiac. Beyond that I don't really feel the need to defend or explain our methods- since we had a much better handle on the election than Republican polling companies last year I'm sure not going to tell them our secrets."
GEORGE WILL DEFENDS CENSUS ACS - Columnist George Will takes fellow conservatives to task for targeting the American Community Survey (ACS): "If the survey were voluntary, compliance would plummet and the cost of gathering the information would soar. The data, paid for by taxpayers and available to them at no charge, serve what the nation needs most — economic growth. Target, Wal-Mart and other large retailers tailor their inventories to regional, even neighborhood, differences revealed in the ACS's granular data. Home builders locate markets rich in people age 25 to 34 and renters. Information improves the efficiency of markets — and of governments, too. There are systemic reasons why democratic governments frequently behave foolishly...The ACS cannot cure systemic problems, but abolishing it would require government to be unnecessarily ignorant...Clearly, conservatives should favor the nation applying to itself the injunction “Know thyself." Besides, if conservatives do not think information about society — the more the merrier — strengthens their case, why are they conservatives?" [WaPost]
So does AEI - Michael R. Strain, a former Census economist and New York Census administrator now a resident scholar at the American Enterprise Institute, also makes the case for the ACS as an economic tool: "The American Community Survey (ACS) is one of the most important sources of data in the country. Its enormous sample size of 2.9 million housing units and its annual publication enable it to provide extremely detailed and useful statistics down to the neighborhood level, and permit the precise estimation of social and economic trends. Why can't the private sector take over the surveys? Without mandatory compliance, the cost of generating accurate and useful data would soar - it is unlikely that the private sector would invest the resources necessary to produce high-quality social and economic statistics. Even if the private sector were willing to invest the resources, it is not clear that households and businesses would be willing to disclose sensitive information to a consortium of private firms. The government has a strong need for accurate data, and if the private sector took over the surveys then the government would almost surely find itself purchasing a vastly inferior product. Since our nation's founding, the government has conducted the national census every decade - the ACS is designed to replace the long-form component of the decennial census - as is specifically authorized by the U.S. Constitution. The founding fathers didn't leave this function to the private sector, and neither should we." [AEI, h/t @JoshdelaRosa]
IS WINNING WHITE AND LATINO VOTERS A ZERO SUM GAME? - Democratic pollster Margie Omero: “To listen to Republicans and Republican-watchers talk about immigration, one would think the debate is about balancing Latinos and whites in a zero sum game of electoral politics. Work too closely with the president and risk a primary challenge in Republicans' increasingly white Congressional Districts. Kill the legislation and risk Republican chances to take back the White House in 2016. There is nothing in the polling to back this up. First, Latinos don't just care about immigration. In the runup to the 2012 election, Gallup showed Latino voters prioritized immigration higher than did white voters, but still lower than health care or unemployment....Second, just like immigration doesn't automatically move all Latinos, reforming immigration doesn't immediately alienate "everyone else." In the 2012 exit polls, about two-thirds of voters said they supported a pathway to citizenship." [HuffPost]
TUESDAY'S 'OUTLIERS' - Links to more news at the intersection of polling, politics and political data:
--24 percent of black men under age 35 say the police have treated them unfairly during the last 30 days. [Gallup]
-Drew DeSilver notes that the filibuster is obscure and mysterious to most Americans. [Pew Research]
-Harry Enten thinks that Anthony Weiner can win and Elliot Spitzer's early lead will fade. [Guardian]
-Nate Cohn finds that Rick Perry had the same problems with Hispanic voters that plagued the GOP in 2012. [New Republic]
-Only 32 percent of Americans tell pollsters they keep a monthly household budget, and Jenn Agiesta thinks that overstates the reality. [AP]
-Kevin Collins records an "interesting survey sampling strategy" in use by the Washington Metro Transit Authority. [@KWCollins]
-Republican pollster Matthew Jason says americans see improved border security as a "key component" to immigration reform. [POS]
-Jonathan Alter reports that the Obama campaign's Project Narwhal failed — "'a huge disaster,' as one senior staffer put it." [Wired]
-The Bureau of Labor Statistics explains why jobs numbers revise (spoiler: initial reports are based on incomplete samples). [BLS via @FactTank]
-Jiffy Lube finds the vaunted "net promoter score" had almost no correlation to customer return visits. [AdAge via @RobertMoran]
-A timeline of Crayola's crayons makes for a colorful infographic. [@justinwolfers]
POLLSTER UPDATE 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).