Americans generally stick to partisan positions when they’re asked about filling Justice Antonin Scalia’s seat. National polls of the GOP primary race are showing healthy variance. And it’s becoming a huge problem for pollsters to ignore cell phones. This is HuffPollster for Thursday, February 18, 2016.
AMERICANS ARE SPLIT ON NOMINATING A SUPREME COURT JUSTICE - Anthony Salvanto, Fred Backus, Jennifer De Pinto and Sarah Dutton: “With an unexpected vacancy on the Supreme Court as a result of Justice Antonin Scalia's death, it's unclear if he will be replaced before the November election. Forty-seven percent would like to see the next justice appointed by President Obama before the election in November, while nearly as many, 46 percent, would like to see the new justice appointed by whoever is elected in November. Views are highly partisan: 82 percent of Republicans would like to see the next president appoint Justice Scalia's replacement, while 77 percent of Democrats want President Obama to make that appointment.” [CBS]
Voters are equally split on whether the Senate should act on a nomination - Mark Murray: “American voters are divided — especially along party lines — whether the U.S. Senate should vote this year on President Obama's eventual nominee to succeed Antonin Scalia on the Supreme Court, according to results from a new national NBC News/Wall Street Journal poll. Among Democratic voters, 81 percent want the Senate to vote this year, while just 9 percent disagree. But those numbers are flipped among Republicans — 81 percent of them want to leave the position vacant, while 11 percent prefer to vote this year. Independents are split — 43 percent this year, 42 percent next year.” [NBC]
Knowing the history of Supreme Court nominations makes a difference - Kevin Quealy: "Americans are divided by political party about whether Mr. Obama should be the one to nominate the replacement, a new online poll shows. But there is also a fascinating wrinkle. The more people are told about the history of Supreme Court nominations, the more they tend to agree that the Senate should consider the president’s nomination, not delay it. Forty-six percent of registered voters said that the next justice should be nominated by the president this year, and 39 percent said the winner of the 2016 presidential election should be the one to put forth a nominee….When given more historical context, respondents were more likely to say the justice should be nominated this year. This new information persuaded Democrats most and independents somewhat less so." [NYT]
CRUZ SURPASSES TRUMP IN NATIONAL POLL, BUT IS HIS LEAD REAL? - HuffPollster: "The Republican primary may have just shifted in Ted Cruz's favor, according to an NBC/Wall Street Journal national poll released Wednesday. The poll, which stands contradictory to other state and national polls, shows the Texas senator 2 percentage points ahead of longtime front-runner Donald Trump. The poll...shows Cruz in first place with 28 percent of the vote to Trump's 26 percent. The results are within the poll's 4.9 percent margin of error. Sen. Marco Rubio (R-Fla.) is in third place with 17 percent…. Republican pollster Bill McInturff...explained to NBC that the results might be an indication that the poll is 'right on top of a shift in the campaign.' That shift, McInturff said, could be a 'momentary 'pause' before the numbers snap back into place,'" or it could be an enduring change." [HuffPost]
Other national polls show Trump ahead by substantial margins - Four other national polls conducted around the same time as the NBC/WSJ poll all show Trump ahead by margins ranging from 15 to 29 points. The consistency of these polls indicates that the NBC/WSJ poll showing Cruz ahead is likely a statistical outlier -- although the 29-point lead for Trump is probably also an outlier. That doesn’t mean the poll or its methods failed; every pollster will eventually produce an outlier simply because of the uncertainty involved in sampling people. Public pollsters who release these numbers anyway should be commended -- it means they aren’t suppressing information or trying to make their poll look like everyone else’s poll. Of course, the NBC/WSJ numbers could be the correct ones, but since there isn’t a national primary we won’t ever have an election to confirm or refute the numbers. The best option is to go with the averages, which show Trump ahead by a wide margin. [Morning Consult, CBS/Times, Suffolk/USA Today, Quinnipiac]
Are we seeing evidence of a ceiling for Trump’s support? - Nate Silver: “There’s a fair bit of evidence that Trump is likely to encounter some upward resistance.….We know he has a high floor. He’s proven it with votes. But he may also have a low-ish ceiling….[T]here’s been a lot of evidence from the start that Trump does underwhelmingly as a second choice….But this is also why it was a good sign for Trump that he got 35 percent of the vote in New Hampshire instead of the 24 percent he got in Iowa. If you’re at 35 percent — well, you don’t need that much more to go from a plurality to an outright majority. At 24 percent, it’s much tougher.” 
TRUMP LEADS IN SOUTH CAROLINA WITH TWO DAYS UNTIL THE PRIMARY -Republicans will vote on Saturday, and polls show Donald Trump as the most likely winner. According to HuffPost Pollster's aggregate, Trump has a solid 16-point lead over Ted Cruz, 36 percent to 20 percent, with Marco Rubio at 16 percent. All the other candidates are polling in the single digits. Fifty delegates are at stake in the Palmetto State: 26 are awarded to the overall primary winner and the rest are awarded to the candidate who wins each of the state's seven districts.
WHY POLLSTERS SHOULDN’T CALL ONLY LANDLINE PHONES - Some pollsters try to lower costs by focusing their efforts on landline telephones rather than calling cell phones, which the Federal Communications Commission requires them to dial by hand. Landlines can be dialed more cheaply, and pollsters can use automated voice technology on landlines to avoid the expense of interviewers. The FCC bans automated polls on cell phones. But pollsters who rely primarily on landline phone calls without taking alternate steps to reach cell phone users are missing larger and larger segments of the population. Updated estimates of wireless-only households finds that the majority of adults in 11 states live in households that only use cell phones. When "wireless only" and "wireless mostly” -- defined as answering most calls on a cell phone -- households s are combined, just seven states fall below the 50 percent threshold. [CDC]
HUFFPOLLSTER VIA EMAIL! - You can receive this daily update every weekday morning via email! Just click here, enter your email address, and click "sign up." That's all there is to it (and you can unsubscribe anytime).
THURSDAY'S 'OUTLIERS' - Links to the best of news at the intersection of polling, politics and political data:
-A USA Today/Suffolk University poll finds little support for Michael Bloomberg as an independent presidential candidate. [USA Today]
-Nate Silver illustrates what would have to happen for Bernie Sanders to win the Democratic nomination. 
-Jonathan Bernstein explains how endorsements can change the primary contests. [Bloomberg]
-Aaron Blake thinks Donald Trump would perform worse if the race were winnowed to two candidates. [WashPost]
-Chris Cillizza explains why South Carolina is do or die for Jeb Bush. [WashPost]
-Jeff Stein dives into the strategy behind Bernie Sanders' electoral coalition. [Vox]
-Toni Monkovic explains the importance of the black vote in the Democratic primary. [NYT]
-Illinois residents are the least confident in their state government, while North Dakota residents are the most confident. [Gallup]
-The initial findings of an inquiry into why British pollsters missed in the 2015 election indicate problems with sampling and identifying who would turn out to vote. [HuffPost]