Gingrich Surging in South Carolina

Nate Silver unequivocally answers a question he posed yesterday--does Newt Gingrich have momentum in South Carolina?--in his FiveThirtyEight blog post today: "Yes, Mr. Gingrich does have momentum — and a lot of it." He writes, "Six different South Carolina polls have been released so far today, and they show a split in their results." Three of them, automated polls by Rasmussen Reports, Public Policy Polling and InsiderAdvantage, "have Mr. Gingrich in the lead...by margins ranging from 2 points to 6 points."

The polls strongly indicate that Mr. Gingrich's performance at the Monday debate has contributed to his rise in the polls. Marist for NBC News, one of the more traditional polls conducted by live interviewers, "showed Mitt Romney leading by 10 percentage points." But when Marist "split its results," it appeared that the debate did indeed give Mr. Gingrich some momentum with South Carolina voters. Mr. Silver writes, "Mr. Romney’s lead was smaller — 5 points — in the post-debate sample." Further, the aforementioned automated polls that show Mr. Gingrich in the lead "were conducted on Wednesday, after the debate and the qualified endorsement Mr. Gingrich received from Sarah Palin."

The FiveThirtyEight forecast model suggests that Mr. Gingrich is, in fact, gaining significant momentum. According to Mr. Silver, "The model now shows a virtual tie in the race, with Mr. Gingrich projected to get 34 percent of the vote and Mr. Romney 33.6 percent and each candidate having about a 50 percent chance of winning."

Most of Mr. Gingrich's gain, however, does not appear to be connected to a drop in the polls for Mr. Romney. Mr. Silver notes, "Mr. Gingrich has gained ground in the polls more than Mr. Romney has lost it. Two days ago, our forecast had Mr. Gingrich with 22.6 percent of the vote, so he has gained about 11 points since then. Meanwhile, Mr. Romney’s projection has declined less than 3 points — to 33.6 percent from 36.1 percent — over the same interval."

So where is Mr. Gingrich getting his new projected votes? According to Mr. Silver, "Mr. Gingrich’s gains have come from Rick Santorum and Rick Perry (who dropped out of the race today and endorsed Mr. Gingrich), as well as from undecided voters." Mr. Silver does go on to acknowledge that automated polls, because they usually do not call cell phones and "also generally have lower response rates than traditional polls," may not be the best source.

For now, Mr. Silver anticipates that the Saturday primary could yield a number of different results. He writes: "Our research shows that high levels of volatility and disagreement in the polls make polling aggregation methods less reliable. In other words, the margin of error on the forecast is especially high, enough so that either Mr. Romney or Mr. Gingrich could emerge with a clear victory by the time that the vote takes place on Saturday."

Early Polls Show Romney's Lead in South Carolina Shrinking

Nate Silver of the FiveThirtyEight blog writes that the first South Carolina GOP primary poll contradicts his earlier prediction that Mitt Romney would increase "his modest lead in the South Carolina polls once his victory in New Hampshire was reflected in the surveys there." Instead, the poll, an automated survey from InsiderAdvantage, "shows Mr. Romney’s lead contracting in South Carolina," though Silver "urge[s] a lot of caution in interpreting it."

The poll "was conducted on Wednesday and shows Mr. Romney with 23 percent of the vote, just 2 points ahead of Newt Gingrich at 21 percent. The poll has Rick Santorum and Ron Paul tied for third place at 13 percent, with Jon M. Huntsman Jr. having moved ahead of Rick Perry for fifth place."

But Silver is not completely sold on InsiderAdvantage. He writes, "InsiderAdvantage has a mixed track record and rates fairly poorly in the FiveThrityEight pollster ratings, which is one reason to interpret these numbers with some care. In addition, Matt Towery, the head of InsiderAdvantage, formerly served as the head of Mr. Gingrich’s political organization from 1992 until Mr. Gingrich left Congress."

Silver does note, however, that "another polling firm that rates more highly, Public Policy Polling, said in a series of Twitter messages that it has also detected a tight race between Mr. Romney and Mr. Gingrich in South Carolina, with little evidence of favorable momentum for Mr. Romney following his New Hampshire victory."

Negative early news for Romney in South Carolina is partially offset by early polling in Florida, however. Silver writes: "Confusing matters further is that a poll of Florida, from Rasmussen Reports, did show some gains for Mr. Romney. The poll, conducted on Wednesday, showed Mr. Romney with 41 percent of the vote there, better than he has done in any prior survey of the state. The poll also showed Mr. Gingrich’s numbers on the decline." Silver predicts that "if the initial results from South Carolina and Florida are confirmed by other pollsters, they would suggest that Florida could play a backstop role for Mr. Romney, in somewhat the same way that New Hampshire did for him relative to Iowa."

However, at this stage Silver suggests that we wait before attaching much weight to the early poll results from either South Carolina or Florida. 

Mormon Republicans Loyal to Romney; Atheist Republicans Disproportionately Go For Paul

According to Gallup, Republican and Republican-leaners of voting age who identify as Mormon are disproportionately likely to support Mitt Romney, while Ron Paul garners more support among atheist Republicans than he does among Republicans of all religious denominations. Romney collects 71 percent of the Mormon Republican vote, exceeding the 23 percent he scores among all Republicans, and Paul garners 19 percent of the atheist vote, more than the 10 percent of all Republicans who support him. Gallup surmises that Paul's support among atheist Republicans at least partly reflects "the youthful skew in Paul's support and the tendency for young Americans to have no formal religious identity."

But Republican voters do not always support the candidate who shares their religious denomination: "Catholic Republicans were no more likely than average to support the two Catholics in the GOP race, Newt Gingrich and Rick Santorum; Protestant Republicans did not disproportionately support any particular candidate." 26 percent of Republicans who identify as Catholic support Gingrich, less than the 28 percent of all Republicans who support him. And four percent support Santorum, only slightly more than the three percent of all Republicans who support him.

Similar to Catholic Republicans, Protestant Republicans "reflect the overall sample average in their candidate support." The three major Protestant candidates are Ron Paul, Rick Perry, and Michele Bachmann (who has just suspended her campaign). Paul only gets 10 percent of the Protestant vote, which matches his score among all Republicans; Perry gets eight percent of the Protestant vote, only one more percentage point than his support among all Republicans; and Bachmann scores seven percent of the Protestant vote, just surpassing the six percent of all Republicans who support her.

Although Romney attracts the vast majority of the Mormon Republican vote, Mormons represent a relatively small percentage of the voting population. Gallup writes: "Given that Mormons constitute about 2% of American adults and 4% of Republicans nationally, the more important factor may be any possible effect that Romney's faith has on the vote of highly religious non-Mormons, a much larger voting bloc. The current data show that highly religious Protestants do give Romney slightly lower support than he gets among all Republicans, although the five-point difference is not huge."

Gingrich Joins Romney at Top of GOP Race

Gallup reported on recent polling data that shows Mitt Romney and Newt Gingrich as the top choices among Republicans for their party's 2012 presidential nominee. 20% of Republicans and Republican-leaning independents nationwide say Romney is their top choice, while 19% say they would most likely support Gingrich; Herman Cain comes in at third with 16%. Among Republicans and Republican-leaners who are currently registered voters, however, Gingrich jumps into first place with 22%, followed closely by Romney (21%) and Cain (16%).

The USA Today/Gallup poll was conducted Nov. 13-17, close to two weeks after a similar Nov. 3-6 poll. In that short time, Gingrich's support has increased from 12% to its current figure of 19% among Republicans nationwide. During this same time, Cain appears to have lost support--perhaps the result of recent allegations of sexual harassment levied against him. While 21% of Republicans supported him in the Nov. 3-6 poll, a smaller 16% favor him today. Rick Perry also continued to slip, from 11% in the last poll to 8% in this most recent one.

While most candidates appear to receive roughly equal support from conservative and moderate/liberal Republicans and Republican-leaners, Gingrich and Cain seem to hang their hat on the more conservative wing of Republican voters. Conservative Republicans and Republican-leaders are most likely to favor Gingrich (23%), followed by Romney (20%) and Cain (18%). But among moderate/liberal Republicans and Republican-leaners, Romney leads the race, checking in with 20% of the votes, while Gingrich and Cain each only garner 12% of the group's vote. According to Gallup, however, "conservatives outnumber moderates and liberals by better than 2-to-1 in the Republican rank-and-file." Gallup writes: "Gingrich and Cain appear to have benefited most from the decline in Perry's support."

Age appears to be a fairly large determinant of respondents' preferences. The GOP race would be a two-man battle between Gingrich (34%) and Romney (28%) if it were up to voters over 65 years of age. In fact, "Gingrich's support is heavily concentrated among Republicans who are at least 50." Part of this, Gallup surmises, may be due to the fact that Gingrich "has been out of public office for more than a decade, and [is] thus a less familiar figure to younger Republicans." Cain and Ron Paul, on the other hand, get most of their support from younger respondents. This would appear to favor Gingrich and Romney, as older Americans are typically more likely to head to the polls than younger Americans.

Gallup concludes: "With the first official nominating contest, the Iowa caucuses, now just six weeks away, there is no clear national front-runner for the Republican nomination...the current contest stands to be the most competitive and perhaps most unpredictable for the Republican nomination since 1972, when the parties shifted the power to choose their presidential nominees away from party leaders at the national convention to the rank-and-file voters in state primaries and caucuses."

Elections and Rural Bias

Every four years, America's chief politicians make a pilgrimage to Iowa and New Hampshire, two states that in to a large extent influence who will be America's two presidential nominees. Iowa and New Hampshire, of course, don't have the most delegates of any state, but the media coverage, money and momentum that candidates receive from high finishes in these first two states often makes them unbeatable (the 2008 election was unique in this regard, with both primaries competitive well past Super Tuesday). In 2007, Brown economists Brian Knight and Nathan Schiff set out to quantify exactly how much more influence voters in Iowa and New Hampshire had in the 2004 Democratic primary than voters in other states. Knight and Schiff noted that anecdotal correlations between candidates who win Iowa or New Hampshire and those who win the nomination are too simplistic because those correlations could reflect underlying candidate strength and examined changes in polling data in late states as a response to the outcomes of primaries in early states. In short, they found primary voters and caucus goers in the earliest states had eight times the influence of those in the latest states.

It's not just the nomination system that biases American policy toward rural areas. States with small populations (which tend, of course, to be rural) are overrepresented in the Senate and Electoral College. To take the most extreme case, a citizen of Wyoming has 69 times the influence in the United States Senate as a citizen of California (that is to say, Wyoming's 544000 residents have the same representation as California's 37 million) and 3.7 times the influence in the Presidential election (three electoral votes per 544000 residents versus 55 per 37 million residents).

David Leonhardt in the New York Times argues that these systems of unequal representation have profound policy consequences. He notes that early-voting states are more likely to receive pork than late-voting ones if they supported the winning candidate, and that the Iowa caucuses have turned ethanol subsidies into an untouchable third rail despite little evidence they contribute much to solving the climate crisis. Most of all, he argues that America has an "anti-urban policy bias," focusing on the issues important to rural and small-town voters even though he believes that "national prosperity depends on urban prosperity."

ZM cf

Down Huckabee, Race for GOP Presidential Nomination Without a Clear Leader

According to Gallup, following Mike Huckabee's decision to withdraw his name from consideration, the race for the 2012 GOP presidential nomination lacks a clear leader. A March-April 2011 trial heat ballot among Republicans and Republican-leaning independents, adjusting for the withdrawals of Huckabee and Donald Trump, suggests the lead is shared by three politicians: Mitt Romney (20%), Sarah Palin (18%), and Newt Gingrich (11%).
But current voter preferences are based more in recognition than anything else. Those three politicians also lead potential GOP candidates in name recognition among the polled population, with 96% recognizing Palin; 84% Gingrich; and 83% of those polled recognizing Romney. Writes Gallup: "The most well-known candidates lead in nomination support at this point, while those who are not as well-known lag behind."

Evidence for this view--that recent, post-Huckabee trial ballot results can be attributed primarily to candidate recognition and not real, strongly rooted voter preference--was found not only in the three most well-known candidates leading the results, but also from the relatively low level of excitement voters expressed for those three candidates. Positive Intensity scores, meant to measure the intensity of a voter's support for his/her candidate, showed that voters are less excited about these three potential candidates than they had been about Huckabee. While Palin scored only 16, Romney 14, and Gingrich 13, Huckabee's final Positive Intensity score for May 2-15 polling reached 25. The Positive Intensity score is calculated by taking the percentage of voters who recognize and have a strongly favorable opinion of a candidate and subtracting from that the percentage of voters who recognize and have a strongly unfavorable opinion of the candidate.Among those candidates who registered lower name recognition, Herman Cain (29% recognition) and Michele Bachmann (21%) received the highest Positive Intensity scores with 24 and 21, respectively. For those two and others "in the Republican field beyond Palin, Gingrich, and Romney," the greatest challenge "right now is increasing their name recognition." For Palin, Gingrich and Romney, the goal must be to increase their Positive Intensity scores among voters, assuming all three participate--or remain--in the race.According to Gallup, only one thing is clear following Huckabee's departure: "There is no clear front-runner in the race for the 2012 Republican presidential nomination."

NH cf