Margin of Σrror

Margin of Σrror -

Anthony Weiner climbs as NYC mayor race heads for tightest finish since 1977

When the candidates are polling this poorly and closely to each other, anything can happen

The New York City mayoral race is a zoo, and we’re all witnesses to it. Over the past two months, I said Anthony Weiner was stronger than initial polling suggested, Christine Quinn might not make the runoff (held if no candidate reaches 40% in the first round) and Bill Thompson probably would end up in the second round. The polling that came out this week makes me doubt none of these beliefs, yet the race remains very unsettled.

This week Marist and Quinnipiac released polls with slightly different, though, mostly consistent results. Marist had Weiner at 25%, Quinn at 20%, Thompson at 13%, Bill de Blasio at 10% and John Liu at 8%. Quinnipiac put Quinn at 19%, Weiner at 17%, Thompson at 16%, de Blasio at 10% and Liu at 7%. You’ll note that both surveys paint Quinn, Thompson, de blasio and Liu in nearly the same place and any differences are within the margin of error. Weiner’s higher percentage in the Marist poll may, as Mark Blumenthal noted on Wednesday, be because he’s in the news a lot and is benefitting from Marist pushing undecideds harder.

That should not be mistaken for Quinn having any sort of strong support. The Marist poll finds that among those who strongly support a candidate Weiner’s lead over Quinn’s extends to 17pt – 40% to 23%. Quinnipiac has her unfavorable rating among Democrats climbing to 31% – 12pt higher than in May and by far her highest of the year. What is going on?

While many analysts were stuck claiming that Weiner was a product of name recognition, they failed to recognize the same and perhaps to a greater extent was true of Quinn. I wrote in March that Quinn didn’t have a record that some Democrats would like once they got to know it. Anti-Quinn advertisements have been running in New York, and they clearly have had an effect.

Quinn had also been polling particularly strongly among African-Americans and Latinos in prior Marist surveys, which didn’t make much sense. Quinn is seen as a kind of heir apparent to Mayor Michael Bloomberg who polls worst among minorities. The latest Marist poll has her falling back to third among blacks at 19% and second among Latinos at 16%. Those numbers may fall further.

Weiner, meanwhile, has seen his numbers climb. In the ballot test, he’s at his high point in both the Marist and Quinnpiac polls. He’s cut his deficit in a potential runoff against Quinn from 15pt down to 2pt per Marist. Marist has his net favorable among Democrats rising from 0 to +16pt in the last month alone, while Quinnpiac show the percentage of Democrats thinking he should run for mayor up to 52% from 41% last month.

It’s easy to say that Weiner’s rise is because more people are hearing his name, but I don’t think that’s necessarily true. Weiner’s rise in the last month has occurred, despite the percentage having an opinion of him staying the same. That means he’s changing minds like I suggested he might in April. This transformation has occurred even as a number of stories hit the press about Weiner’s lack of a solid congressional record, past racially tinged campaigns, and of the damage his sexual escapades had on the women he had conversations with.

The biggest story, however, from the polling is the predictable rise of Bill Thompson. It’s been my contention since day one that Thompson, an African-American, would pick up the lion share of votes from African-Americans, who will make up about 30% of the electorate. That would put him in prime contention for a runoff spot that will probably be earned with just a little greater than 20% of the vote. Both the Marist and Quinnipiac polls have Thompson rising from the mid teens to 21% among African-Americans in the past month. That should go up even more.

The underlying voter sentiment gives Thompson even more hope. Thompson has the best net favorables in both the Marist and Quinnipiac polls of any of candidate among all Democratic voters – including whites. We can see this at work in potential runoffs, which he is in statistical dead-heats with Quinn and Weiner per Marist, even as Marist has him trailing both of them in the initial round.

Thompson may also benefit from the fact that pollsters in New York City seem to have a difficult time surveying minority candidates and voters. The leading minority candidate has over-performed his final polling significantly in every Democratic primary since 1989. Pollsters underestimated the percentage of minority voters going for the minority candidate in the past two general elections – including Thompson in 2009.

At the end of the day, though, any of the top three candidates can advance. I went through the polling that I could find since 1989, and I can’t find a single poll this late in the mayoral primary campaign when the leading candidate had less than 26% and certainly not less than 20%. There simply is no precedence for this in the past 30 years.

Indeed, the only race I can ever remember that shares the slightest resemblance to this one is 1977. That race featured Democrats Bella Abzug, Herman Badillo, Abe Beame, Mario Cuomo, Ed Koch and Percy Sutton. Abzug was thought of as a favorite with Beame close behind. Polling in that race had Abzug leading with right around 20% until mid-August. Then Koch “surged” forward to win the first round with less than 20% with Cuomo close behind, while none of the six earned less than 10%.

The lesson from that campaign that should be applied to this one is that when the candidates are polling so poorly and close to each other anything can happen. I wouldn’t even count out Bill de Blasio who is lurking with 10%. If you buy the Quinnipiac poll, he’s less than 10pt back. With two months to go and most voters not tuned into the race yet, it could be 1977 with someone we wouldn’t think of coming from behind. I don’t expect it, but in this race expect the unexpected.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

The Republican party is not ‘electorally damaged’ | Harry J Enten

The GOP might be feeling the pressure to adjust its political platforms to win national elections, but it doesn’t need to

Just how screwed is the Republican party electorally? We all know Mitt Romney lost by nearly 4 points, and the Republican party has won the national vote for president only once since 1992. That’s likely why there’s a lot of pressure for Republicans to pass immigration reform to “get with the times” and not lose the growing Latino vote for a generation. I, however, continue to think the idea that Republicans have to adjust to win national elections is hogwash.

Why do I say such a thing?

1. Voters thought Mitt Romney was actually closer to them ideologically

John Sides put together this graph from a bunch of surveys taken during the 2012 campaign.

Voters were asked where they put themselves ideologically and then to place each of the candidates on the same ideological scale. Throughout the campaign, the average voter was closer ideologically to Mitt Romney than to President Obama.

That, of course, doesn’t match the hype these days. You could, of course, argue that Mitt Romney was closer to the center than the average Republican politician, but the facts don’t really bare it out. His donor base, which can be used to measure ideology since people most often donate to those they line up with on the issues, ended up being right in the middle of those of the last Congress. Same goes with the congressional endorsements he racked up during the primary: they were in the center of the caucus.

Then how the heck did Obama win? Again, it goes back to the fact that ideology doesn’t matter much in presidential elections. The economy was doing well enough, so voters rewarded President Obama with another term.

2. In a 2016 presidential matchup, Republicans are performing just fine

There’s been a lot of hype around Hillary Clinton’s potential 2016 run and rightly so. That said, most understand that her current general election polling numbers won’t stand up over time. She’s still coming off a height of popularity since serving in the non-partisan position of Secretary of State. As she has reentered the fray, her favorables have fallen off a little and should continue to as they have in the past when she pursues a partisan agenda.

At this point, a better feel of how a Democrat might do if the election were held today is polling involving Vice President Biden. Biden is a well-known standard bearer of the Democratic label. He’s been serving in a partisan office for as long as many have been alive. His favorable rating generally matches up with President Obama’s approval rating in the polls.

Last month, Biden was either neck-and-neck or trailed leading Republicans. He was 1pt ahead of Marco Rubio and 2pt ahead of Rand Paul in an early May Public Policy Polling survey. He was behind by 4pt against Paul and 6pt behind Jeb Bush in a Quinnipiac poll.

All of them are also less well-known than Biden, yet they are polling at or ahead of him. You could argue that Bush is more moderate than the mainstream Republican candidates, though, Rubio and Paul are almost certainly to the right of it. They incidentally are three of the top four leading contenders for the nomination right now. You wouldn’t expect mainstream Republican candidates to be polling this well if the party were too out of the mainstream to win.

3. Republicans have been winning a lot of national elections since 1994

House elections, for better or worse, have turned into national affairs. That’s why the presidential vote in each district is so predictive of the outcome in the House races. The good news is the nationalization of the races allow us to get a pretty idea how competitive the parties are.

I don’t just mean that we see how many Republicans and Democrats there are at the end of each election; I mean how many votes were won by each party’s candidates. This helps to mostly eliminate the effects of urban packing and redistricting.

Since 1994, Republicans have won the national House vote seven times. The Democrats have won it only three times. Republicans have won it in presidential years such as 1996, 2000, and 2004. They won it in 2010 by a large enough margin that even with a presidential year turnout, they still would have taken the House vote by more than 3pt.

If the existence of more extreme Republican candidates was hurting Republicans anywhere, you’d expect it to be in the House given they are furthest to the right. It might not hurt them in seats won, but you’d at least expect it to make a dent in the national vote. Yet, Republicans in the House actually outperform Mitt Romney, winning a higher percentage of the vote.

Overall, the idea that Republicans are somehow too extreme to win elections just doesn’t hold much water upon scrutiny. That doesn’t mean Republicans can’t make moves to update its get-out-the-vote techniques or try and shore up portions of the electorate. Just keep in mind, as Sean Trende notes, moves can be made to win many different groups, not just the ones most mainstream commentators think.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Obama’s approval takes a hit over NSA leaks | Harry J Enten

Declining public trust in his administration over the surveillance revelations should worry the president, but not based on one poll

President Obama’s approval rating is down in the aftermath of the NSA leaks. The median of five surveys taken before and after the leaks has his net approval off 4pt. The current HuffPollster aggregate has his net approval standing at -1.3pt, which is its lowest point (in this aggregation) since last November’s election. This decline coincides with falling trust both in government and in Obama himself.

Indeed, the biggest sign of trouble for Obama is not that his approval has fallen, but that it’s been accompanied by a similar collapse in trust – even as consumer sentiment is rising. That’s unusual because, as I noted here, a very good predictor of changes in a president’s approval and trust in government is changes in consumer sentiment. Historically, the only times consumer sentiment hasn’t correlated with these two factors is during war or scandal. That suggests that the NSA surveillance revelations are regarded, by some Americans at least, as “a scandal”.

It’s important, however, to note exactly how much damage has been done. A 4pt drop in net approval is significant, but it’s not as if the dam has broken. It’s nowhere near the 17pt drop in net approval than CNN/ORC showed in their poll released on Monday. That survey showed Obama’s worst net approval rating in two and a half months. It’s fair to say, as Mark Blumenthal did, that the CNN poll “exaggerates” Obama’s decline.

Responsible commentators like Blumenthal and Nate Cohn both mentioned what was likely an outlier result, but cited the average as the more reliable guide. Not all pundits followed their judicious example.

At CNN, one would be hard-pressed to find any mention of another poll. Instead, readers just have “political analyst” Gloria Borger declaring that Obama needs to execute a twist on the “Green Lantern” political play to get the public back on side. Despite what Aaron Sorkin or Drew Westen would have you believe, presidents rarely have some magical power to sell the public on a policy that people don’t agree with.

CNN, of course, is simply doing what all news organizations do when they sponsor a poll: they want to sell their “exclusive” findings for all they’re worth. That’s not much of an excuse, though: political analysts on the network should be telling the people how it is, using all the datapoints available, not solely citing the company’s own poll. This is what the Huffington Post did after their initial survey on the NSA controversy conflicted with a Pew/Washington Post poll.

Some of the polls being debated in the media also do not meet reputable polling firms’ technical criteria, though both the Gallup and Time magazine polls conducted during the same period certainly do. These called cellphones and used live interviewers. Both polls gave Obama positive net approvals. Why wouldn’t those polls be cited?

The same goes for news organizations that didn’t sponsor any poll. Take, for example, this National Journal article on Obama’s trust problem. I understand that when polls conflict with each other, there’s a belief that citing more than one poll can distract from an argument. The thing that’s so frustrating in this circumstance is that a wider look at the polling does support the belief that Obama has paid an approvals price over the NSA leaks – even if it isn’t to the same degree as CNN discovered. But if the one poll doesn’t agree with the majority, perhaps you shouldn’t be building a thesis around it.

This is especially the case when you’re dealing with crosstabs (the more detailed breakdowns within individual polls). I saw a lot of articles citing CNN’s finding that young voters’ support had “plummeted” to 48%. The margin of error on that crosstab was 7.5%. Meaning that Obama’s approval might well have been 55.5% – or equal to Gallup’s relatively stable weekly number, which has a far smaller margin of error. Add on the fact that younger voters are especially hard to reach thanks to low response rates, and you see the need to be careful.

Also the case is that we could already see a small drop in Obama’s approval rating late last week, when I wrote about that in the context of many polls. So we knew that the president had a trust problem developing before the CNN/ORC poll came out.

Finally, the advantages of sticking to a polling average, median, or some form an aggregate, have already proved their worth – notably, during the 2012 campaign. Anyone relying solely on Gallup (which has subsequently taken solid steps to improve accuracy) and Rasmussen to predict the 2012 result would have been embarrassed. Whether it be the CNN/ORC survey now or an outlier Quinnipiac poll last month, the lesson is: go to the polling average, even if the story isn’t as sexy.

That said, Obama should be concerned about his approval rating – especially the trust factor that appears to underpin its slide.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Marco Rubio’s immigration reform push wins GOP establishment loyalty | Harry J Enten

Immigration reform has enabled Senator Rubio, once a Tea Party poster-boy, to reposition himself deftly for a 2016 presidential run

On Monday, Politico reported that “With immigration, Marco Rubio risks DC insider label.” Rubio, of course, has put a lot of chips on the table in trying to push for immigration reform for undocumented immigrants. And I’ve got one piece of advice for Senator Marco Rubio on this issue: run to the establishment. Run to it as much as you can.

I have not been particularly on board the “Marco Rubio for president, 2016″ train. In fact, I wrote a whole article devoted to the five reasons why I wouldn’t bet on him. Two of those reasons, however, look a whole lot less applicable in the wake of Rubio’s leadership on the immigration issue.

Previously, I had worried that Rubio would not appeal to the establishment wing of his party. Rubio had won his Senate seat in 2010 by challenging the establishment candidate, and he has run up an anti-establishment record as a senator. This was perhaps my main worry about a possible run by him for the presidency: see, contrary sometimes to appearances, the grassroots is not where the power is in Republican politics.

Cohen, Karol et al’s The Party Decides tells us that the presidential candidate with the most establishment backing almost always wins the nomination in the modern era. Jerry Ford won over Ronald Reagan in 1976; Reagan over George HW Bush in 1980; Bush over Bob Dole and Jack Kemp in 1988; Dole over all in 1996; and George W Bush over John McCain in 2000. All of these winners had clear support from the party establishment.

Mitt Romney followed that tradition in 2012. He had by far the most endorsements from party leaders in the lead-up to and during the primary season. That’s why I was fairly certain he was going to win the nomination, even as he faced challenges from the “very conservative” wing in his party.

The only Republican candidate who has won the nomination in the past 40 years without clear party support was McCain in 2008. In that year, however, there was no favorite, and McCain clearly tried to make good with the GOP grandees, after his 2000 bid. In the end, it worked. After the first two primaries, McCain became the establishment candidate.

Rubio’s push on immigration reform is going to get him some insider credit. Faith and Freedom Coalition President Ralph Reed looks favorably upon Rubio’s push. Most donors like Rubio’s new role. He’s also bound to win plaudits from the GOP establishment in the Senate, which is pushing for immigration reform. Overall, this is definitely the correct move for a person who might otherwise be seen as too “outsidery”.

Immigration reform has the additional advantage of being seen as a moderating force. It’s backed by most Americans and is generally supported by the party establishment because it’s seen (rightly or wrongly) as an electoral winner.

Rubio’s Senate record paints him as one of the most conservative senators. He was the seventh most conservative senator in the 112th Congress, sandwiched between Jim Inhofe and Ron Johnson. As I wrote before, it’s unlikely the Republican party will nominate a very conservative candidate in 2016. When it liked Barry Goldwater in 1964, and Ronald Reagan in 1980, the party had recently controlled the presidency. But when the party hasn’t been in the White House for eight years or more, it goes for a more centrist pick in order to win.

This isn’t to say that Rubio should become a moderate. Far from it. It’s neither the moderate nor the very conservative vote that dictates who wins Republican primaries for president. Rubio’s key is to win the “somewhat conservative” vote that tends to take into account what the establishment says, as well as wanting to pick the most electable. By the Buckley rule, the “somewhat conservative” bloc chooses the most electable conservative.

That’s the reason that McCain was able to win in 2008 when he lost in 2000. He lost the very conservative vote and won the moderate vote in both years. The switch was among the voters who considered themselves “somewhat conservative”. These are the voters that Rubio must win, if he wants the nomination.

So, I think Rubio is making a smart political play by supporting immigration reform so openly. It’s the type of issue that will garner him plaudits from the party establishment – which generally gets to pick Republican nominees. It’ll help to reassure Republicans that he can win, which will likely be a chief concern for primary voters in 2016, as it was in 2012.

At the same time, Rubio can point to other issues where he is an outsider conservative, such as the debt ceiling. Rubio is proving, perhaps, that he knows how to balance the wings of his party correctly.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

NSA surveillance fears give Republicans a great in with young Americans | Harry J Enten

Young people are more Democratic and pro-Obama – but they’re even more pro-liberty. The GOP should seize this chance

You don’t have to be a rocket scientist to know that the Republican party has a problem with the millennial generation. Americans born after 1980 voted by some 25pt for President Obama over Mitt Romney, while the rest of the electorate narrowly supported the Republican candidate. It’s my view that Republicans can do little to attract most Millennials – but privacy may be the exception.

Our Guardian poll conducted by Public Policy Polling indicated that Democrats were more supportive of President Obama’s position on the National Security Agency (NSA) leaks than other parts of the electorate:

• 58% of Democrats were OK with the government collecting data on internet or phone data, compared with 44% of voters overall

• By a 4pt margin, Democrats were more fearful of putting national security at risk than of infringing civil liberties. That compares with a 22pt margin going the other way among the overall sample

• Only 28% of Democrats said that the information learned made them less likely to support President Obama against 48% of the electorate overall

These results aren’t too surprising given that a political base usually support its leader. That’s why you saw far more Democrats supporting Obama’s actions on collection and access of phone and internet records than for a somewhat similar policy under President Bush, while Republicans were far more receptive of the policies under Bush than under Obama.

You’d expect, therefore, that young voters – who are the bedrock of the Democratic base – to be receptive to President Obama’s actions. You’d be wrong. Young voters are overwhelmingly against the NSA’s access to internet and phone records.

• A meager 27% said the government should be accessing internet or phone records – 31pt lower than Democrats overall

• 73% were more fearful of an infringement of their civil liberties, compared to 15% who were more afraid of putting national security at risk. This 58pt margin is a reversal of the 4pt margin of Democrats who were more scared of a terrorist attack than abridgement of civil liberties

• 69% of young voters said that what they learned from the NSA leaks made them less likely to support President Obama – 41pt greater than Democrats. Exactly zero young voters in our sample said what they learned made them more likely to support Obama

To me, this indicates a tremendous opening for Republicans: 69% of young voters think that this country needs a fresh conversation about privacy versus security – and they are not with President Obama or the Democrats on this issue. If Republicans don’t at least discuss it, it will be an opportunity lost.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Polls show Obama’s real worry: NSA leaks erode trust in government | Harry J Enten

With the NSA revelations doubling down on the IRS and AP scandals, the president’s approval rating is hitting new lows

Thursday, the Guardian released a poll conducted on Monday and Tuesday nights by Public Policy Polling looking at America’s reaction to the National Security Agency (NSA) controversy. The public appears to be reacting negatively to the revelations – and it seems to be hurting President Obama.

We found 50% of American voters believe the NSA should not be collecting telephone or internet records, compared to the 44% who think they should. The results hold even when respondents were told that the data the government is collecting is “metadata” (and not necessarily actual content of communications).

These results are consistent with a CBS News poll, Fox News poll, and YouGov survey that showed only 38%, 32%, and 35% of Americans respectively approved of phone record collection in order to reduce the chance of a terrorist attack. A Gallup poll was consistent with these, showing only 37% approved monitoring of Americans’ phone and internet use.

The results conflict with a Pew Research/Washington Post survey, which showed 56% of Americans found the NSA’s tracking of phone records to be acceptable. Why the difference?

As Mark Blumenthal pointed out Wednesday, the difference could well have to do with the Pew Research/Washington Post poll pointing out that the government had a “court order”. A court order would, to most, probably imply something less sinister; other pollsters had not made this distinction.

That said, the Guardian survey confirms the Pew survey in another important way. Nate Cohn recognized Wednesday that only 45% of Americans approved of the government monitoring of Americans’ emails and computer information. The Guardian survey discovered a very similar 41% of Americans who feel this way.

It’s fair to say the majority of Americans are, at the very least, unhappy with one or more aspects of the NSA data-mining revelations. The question is, though, whether or not President Obama will suffer political fallout from the leaks. The answer seems to be that he very well may.

First, Americans are increasingly catching up with the details of the controversy. Only 27% of Americans Pew/Washington Post said they were paying attention close attention to the NSA leaks. In fact, a full third said they weren’t following the news closely at all. That number, however, was only through 9 June. The YouGov poll conducted through the 10 June has the percentage who had not heard any news at all about the leaks at 10%. The Guardian survey, taken through 11 June, has it down to 8%. So, by now, most Americans have heard about the NSA leaks.

Second, Americans claim they are less likely to back Obama because of the information released to the public: 48% of American voters in our poll say they are less likely to support him because of the recent disclosures, while only 17% says they are more likely to support him. Fifty-two percent of voters who identify as independent said they were less likely to support the president, versus 8% who were more likely. Even Democrats, who are the most likely to think the leaks aren’t a big deal, are by a margin of 3pt less likely, rather than more, to back Obama because of the controversy.

Third, while it can be misleading to compare between pollsters to detect a trend, Obama’s approval rating does seem to be sliding (when controlling for pollster). Obama’s rating among registered voters in the YouGov survey (pdf) is only 46% (only 44% among adults). These are two-month lows.

Among all adults, the 44% number is Obama’s lowest since the November election. Obama’s net approval of -2pt in the Gallup daily tracker is also the lowest since his re-election. The same is true in the Fox poll, where his 44% approval is the lowest in over a year.

More worrisome for Obama is that his approval may be going down even as Americans think the economy is approving. Consumer sentiment rose to its highest level in six years in May at 84.5 in that index. Research by political scientists Robert Erikson, Michael Mackuen and James Stimson in Macro Polity shows that the No 1 predictor of changes in presidential approval is changes in consumer sentiment. In fact, it predicts over 90% of the changes.

Check out this chart from Republican strategist Adrian Gray published before we learned about the NSA leaks:

Obama’s three-month average approval rating had been tracking nicely with consumer sentiment through most of his presidency. It did rise above where one would expect, during his re-election honeymoon, but it’s continued to fall even further below where we would expect, post-honeymoon.

The overall Huffpollster aggregate, even without the trackers, shows his approval dropping to an all-time low level since re-election. Some surveys such as the NBC/Wall Street Journal poll show no movement in Obama’s approval. Even if there were no movement, this still isn’t a good sign for the president.

The point is, Obama’s approval should be rising given higher consumer sentiment. But it isn’t.

When the IRS and Associated Press scandals first broke, I pointed out that the one factor that predicts election results better than consumer sentiment is trust in government. Consumer sentiment and trust usually track together – except in times of government controversy. It’s one of the reasons the Democrats lost so many seats in 1994 during the first Clinton administration, even as consumer sentiment was decently high.

Trust in government after these scandals has been falling. In the recent NBC/Wall Street Journal poll, 55% of Americans said the IRS targeting made them doubt the “overall honesty and integrity” of the Obama administration. Only 48% of voters in Fox News poll taken after the release of the NSA information said Obama was “honest and trustworthy” – the lowest level the poll ever recorded. More than a third (35%) of voters believe the administration has been less open than previous administrations – a record high.

In light of the public’s negative reaction to the NSA leaks, trust in government could fall further. This would likely lead to a drop in the president’s approval. At this point, these controversies about government overreach, including the NSA revelations, look to be hurting Obama, even if his approval ratings are just sliding rather than crashing. The question going forward is whether fallout from the NSA revelations accelerates that decline.

• Editor’s note: this article was updated with new material at 10.15am on 13 June

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Can the Republicans expect a midterm election bump in 2014? | Harry J Enten

Who turns out to vote is believed to favor the GOP in midterms, but changes in the electorate may not make so much difference

Barack Obama won the 2008 presidency by 7.3pt. Two years later, his Democratic party lost 63 seats and control of the House of Representatives. Two years after that, Obama won re-election by 3.9pt. The incongruity of the middle result has led people to wonder why Democrats did better in the presidential years but did worse in the midterm election.

Did Democrats do so poorly in 2010 because of poor turnout among their core constituencies: minorities and young voters? After all, the percentage of eligible voters that turns out drops by about 20pt between midterm and presidential years. Are most of those minority and young voters? If so, Republicans would have fared far worse in 2010 with presidential year turnout and will do well in 2014.

If you look at the exit polls, the minority share of the electorate went down from 26% in 2008 to 23% in 2010. The 18- to 29-year-old percentage of the electorate fell from 18% to 12%, while the 65-year-old and older percentage of the electorate rose from 16% to 21%. But are these data correct, and what type of effect do they have?

I believe the minority share of the vote did not drop by 3pt. The Current Population Survey (CPS), which most would agree is far better at picking up changes in the electorate and matching pre-election surveys, shows a drop of only 1.2pt, from 23.7% in 2008 to 22.5% to 2010. Such a drop is not unusual and is consistent with the 1.2pt drop from 2004 to 2006, even as the Democrats romped in the 2006 midterms with an 8pt victory in the national House vote.

We can figure how much of a difference this change in the racial make-up of the electorate makes by multiplying the percentage of whites won by the Democrats by the percentage of the vote they made up of the electorate in 2008 and 2010, and doing the same for non-whites. Applying the racial turnout from the 2008 election to the 2010 election would have slimmed the Republican margin by less than 1pt from the 6.6pt victory they took in 2010. This would have been more than enough to take 50-plus seats and recapture the House.

Even if you apply the drop-off seen in the exit polls, the Republicans would have won by more than 4pt. Again, this would have been more than enough to take back control of the House.

What about the youth effect? The CPS showed a drop in the 18- to 29-year-old vote from 17.1% in 2008 to 11.3% in 2010. This 11.3% is virtually identical to the 11.2% that 18- to 29-year-olds made up in the Democratic romp of 2006. Meanwhile, the share of the 65-year-old plus vote climbed from 19.5% in 2008 to 23.9% in 2010. That may seem a lot, but keep in mind that the voting differences between age groups is far less than the differences in voting between racial groups. Non-whites voted 38pts more Republican than whites in 2010, while 18- to 29-year-olds voted only 17pt more Democratic than voters 65 years and older. So, it isn’t surprising that the total age impact of 2010 versus 2008 is about 2pt on the margin, despite the comparatively wide turnout differences between age groups.

Combining the impact of race and age, you’d be looking at a 3pt Republican gain between the 2008 and 2010 electorate. That still would have been enough for the Republicans to win more than 45 seats and win back the House. Keep in mind, though, the largest portion of the electorate that is “minority” are youth. Putting together the age and race effects, therefore, may make the presidential and midterm electorates seem more different than they are.

Perhaps the best way to check whether the electorate changes between midterms and presidential years is to look at data from Pew Research. Pew, a pre-eminent pollster that has led the way in developing sampling techniques that get full coverage of the American population, publishes registered and likely voter data just before each election. Unlike other pollsters whose likely voter electorate were a less accurate estimate of the 2012 result, Pew’s likely voter data were closer than its registered data. In fact, its likely voter data has been more accurate in every election except one since 2002. This suggests that they are picking up something very real.

We can compare the likely voter to registered voter gap in midterms and presidential elections in Pew polls to see if Republicans really do have a turnout advantage in midterm elections. The likely voter to registered voter gap in 2010 was 7pt. That is, Republicans led by 6pt in the likely voter result, while trailing by 1pt among registered voters. That may seem like a big gap, but bear in mind that the difference was 4pt in both 2008 and 2012. In the 2006 midterms, it was also only 4pt. In fact, the average difference between Pew’s likely voter and registered voter result in midterm elections over the past decade, as the Democratic-leaning millennials have joined the electorate, has been 4.3pt. The average difference in presidential years has been 4pt.

The Pew data suggest that Republicans are actually not gaining an advantage; or if there is one, it’s minimal. My own estimate is that it’s probably about 2pt when looking at all the data. That’s consistent with the idea that the current party affiliation differences between age groups has disadvantaged Democrats in midterm elections, while the differences in turnout by different racial groups likely don’t have much of an effect. This gap between midterm and presidential year electorates has developed recently, as Democrats used to be quite popular with the older voters of the “greatest generation”, though they are significantly less liked by the older voters of the “silent generation”.

Two points is certainly not nothing, yet it’s not a ton either. President Obama won by nearly 4pt in 2012. Republicans would definitely have won the House and comfortably so in 2010, even with a 2008 electorate. If Republicans win big in 2014, the difference in the electorate will only be a small part of it. But if they were to barely hold onto the House, then demographics may have made the difference.

Then again, other factors will have played a bigger difference, such as the usual “penaltythe president’s party suffers in midterm elections or a possible backlash against Obamacare.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Gallup’s 2012 election polling errors were only part of the problem | Henry Enten

Gallup was caught out badly, but other national pollsters were off, too. It’s time to look at different methods and new technologies

We all know that Gallup screwed the pooch in the 2012 presidential election. It had Mitt Romney leading through most of October and in its final poll by a point – a 5pt error. Gallup sought to prove to the polling world that it was seriously investigating its 2012 polling errors by issuing a report on Tuesday. In the write-up, Gallup noted that although there was no single cause, a likely faulty voter screen and too few Hispanics were among the problems. This comes as no surprise to others including Mark Blumenthal and myself.

It’s worth the time, though, to point out, as I have and Gallup did on Tuesday, that the Gallup effect was only half the problem.

The average of polls done in the final week, excluding Gallup and Rasmussen, had Obama’s lead over Romney more than 2pt too low. I might be willing to look the other way, except the polling average in 2000 had George W Bush winning and had a margin error of again more than 2pt. The error in margin in 1996 was off by 3pt. The 1980 average saw an error of more than 5pt. The years in between 1980 and 1996 were not much better. In other words, the “high” error in national polling even when taking an average isn’t new; in fact, it seems to be rather consistent over the years.

Worse than the error in the final polls was how the national polls took the consumer for a ride in October 2012 before finally settling in the final week. Anyone remember when Pew Research published a poll after the first debate in 2012 that had Mitt Romney up by 4pt among likely voters? I don’t mean to single out Pew, but because of Pew’s sterling reputation, this poll got an outsized amount of attention even as most of us suspected that it probably didn’t reflect the truth. Other pollsters, too, showed a bounce for Romney that propelled him into the lead after the first debate, though not all to the same extent.

The state polling, meanwhile, did not show an analogous large bounce. It consistently had Obama leading in the states he needed to be leading in. Moreover, it showed Obama holding very similar positions to those he did prior to the first debate in the non-battleground states.

Look at YouGov, for example, which polled before and after the first debate. In Florida, Obama was ahead by 2pt before the debate and 1pt after it. In blue New York, Obama was ahead by 22pt before the first debate and 24pt after. In red Georgia, Romney was up by 7pt before the debate and 8pt afterward. Pollsters like ABC/Washington Post, CNN/ORC,and Public Policy Polling (PPP) all did better on the state level throughout October than they did at the national level.

It’s not the first time the state polling beat the national surveys. Back in 2000, for instance, state poll followers knew that Al Gore had a really good shot at winning. National survey followers, though, were surprised when Gore won the national vote. That’s why smart poll aggregators like Drew Linzer, Nate Silver and Sam Wang barely looked at national polling in 2012 when trying to project the winner. It’s also why the Obama campaign didn’t conduct national surveys.

I asked Gallup about state polling on Tuesday, and why it didn’t try to do individual state polls and/or then sum up, as Silver did, to calculate the national vote. After all, besides polling accuracy, the ball game of presidential elections for pollsters is state elections. Gallup’s response was telling. First, it said that polling 50 individual states via live interviewer to come up with a national estimate would be too time-consuming and cost too much money. That’s fair. Second, Gallup said that it didn’t just poll the swing states because it was interested in knowing what all Americans thought, not just swing-state voters. (I agree and made the same point in an earlier column.)

But for those of us who are interested in knowing who is going to win, Gallup’s answer is not satisfying. Other live pollsters like CNN/ORC, Marist, Quinnipiac and the Washington Post did very good statewide polling in 2012. Gallup hasn’t conducted a statewide general election poll since 2006 and hasn’t done so in a general presidential election since 2004. (Those 2004 polls, by the way, weren’t very good.)

Moreover, the option now exists for pollsters to use other technologies to poll most states, if not all 50. We have interactive voice response (IVR) or robo-polls that are relatively cheap and can survey many people quickly. As long as you properly weight in younger voters, as does PPP and SurveyUSA, these polls work quite well in predicting who is going to win the national election. We also have the somewhat less expensive, randomly selected internet surveys such as Knowledge Networks, and the cheaper volunteer internet polling, which YouGov and Ipsos have implemented successfully. These volunteer surveys hold a lot of promise in the future as more and more people get rid of landlines and have computers.

The point is that there are proven ways to poll that produce more consistently accurate portrayals of the election than doing a single live telephone interviews of a randomly selected population in a national poll. In fact, it’s already being done. That’s not to say that good-quality probabilistic national surveys don’t have a place. No one has proven to me that IVR or non-random internet surveys are as good as probabilistic telephones surveys on issue questions beyond the ballot test. The problem, again, is that I’d look to other sources in preference to a survey that interviews some number of respondents in one survey, a different set of respondents in the next survey, and so on.

One of the biggest takeaways from the American Association for Public Opinion Research (AAPOR) conference is on the usefulness of panel research. That is, you have a set number of respondents, weighted to the correct population parameters, who get interviewed over and over again. This leads to less volatility, and you can actually see how different respondents are reacting to the campaign. Panels can be difficult to do by phone, yet are rather easily obtained in randomly-selected internet samples that pretty much everyone, including those against volunteer internet samples, agrees do just as good a job at finding the true public opinion on issue questions. You can actually see how well panels worked with the Rand American Life Panel. It was the only national tracking survey in 2012 that had both convention bounces and Obama leading throughout the month of October.

None of this is to say that live telephone surveying is bad or useless, by any stretch. Most of the national telephone polls in 2012 were better than Gallup’s. It just seems to me that we shouldn’t only be examining Gallup for 2012′s polling failings. It might be time for even the most ardent defenders of live telephone national interviews to look at other methods in greater depth. Whether it be for the presidential horserace or more in-depth issue questions, different and (in some cases) less expensive survey styles have shown a trend to do better or at least as well.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Cory Booker is the big winner from Christie’s call on New Jersey elections | Harry J Enten

Chris Christie’s decision to hold separate polls for senator and governor may suit him but won’t please the Republican party

Chris Christie has made his decision: New Jersey will have a special Senate election on 16 October 2013. The selection comes as a surprise given that the smart money was on either November 2013 or November 2014 (given conflicting state statutes). No other election is scheduled for that date, and Christie’s move will cost taxpayers a cool $12m. Additionally, a primary has been scheduled in August. That’s another $12m.

The reason Christie chose these dates is simple enough. He was fearful of having the Senate election on 5 November 2013, when he will be running for re-election as governor. That date could have led to higher turnout – higher, that is, than the usual lower, off-year turnout, which tends to benefit Republican candidates. Christie was also likely fearful of being linked with the national Republican brand in a predominantly Democratic state.

But why not November 2014, in that case? The answer is that Christie didn’t want to push the election to November 2014 because that postponement would have led to a court battle with Democrats, who want the election in 2013. Such a date would have allowed a Republican incumbent, whom Christie will have appointed, to have time to build statewide name-recognition. Now, the seat-filler will not be able to raise his or her name-recognition by a prolonged period in office, so Democrats are unlikely to challenge Christie’s 2013 date.

Put another way, Christie is trying to have it both ways: he doesn’t want an open legal battle with Democrats over the date of the election; at the same time, he doesn’t want another Democrat on the ballot for his election. I’m not sure, though, that this is the wisest move on his part.

Christie said he “doesn’t care” about the cost of a special election. That is not exactly the type of wording fiscally-conscious national Republican primary voters want to hear. It’s also a line likely to be used by Democrats against Christie given that he had previously made a move to ensure that state elections were held on the same today precisely in order to save money.

National Republicans won’t approve the idea of an incumbent Republican senator not having time to build name-recognition either. Democrats, of course, aren’t happy that there won’t be a Senate race to help boost turnout in November, which would help their New Jersey gubernatorial candidate, Barbara Buono. In addition, vulnerable Democratic state legislators may be hurt by depressed turnout because of the prior election a mere month before the state elections in November 2013.

It’s possible therefore that Christie is being too cute by half – storing up problems for himself with a national GOP that already has its doubts about the New Jersey governor’s conservative credentials.

Another question to be answered is whom Christie will now pick to fill the Senate seat before the October election. If he can persuade the relatively well-known former Republican Governor Tom Kean Sr, who had a +37pt net favorable in 2010, to run, then all might be forgiven by national Republicans. Kean would have a real chance of winning the seat and may actually benefit from a compressed election schedule. Aside from Kean Sr, it’s unlikely that other Republican candidates (who may include state Senator Tom Kean Jr, Lt Governor Kim Guadagno, or state Senator Joe Kyrillos) would stand much of a chance.

Any Republican nominee would likely have to face off against Newark’s popular Democratic mayor, Cory Booker. Booker is likely to be the real winner from Christie’s decision. He has the money and name-recognition to win both a primary and general election in a short campaign. There’s no real way that the state Democratic party can choose a different candidate.

Finally, Booker can now dismiss the possibility that a popular Christie on the same ballot might have helped a Republican Senate candidate’s odds. Barring a Kean Sr candidacy, Booker is almost certainly going to be the next senator from New Jersey. Even if Kean Sr was on the ballot, Booker would stand a solid chance in a blue state like New Jersey.

So, the election for New Jersey’s next United States senator is well under way. Chris Christie has eliminated worry about interference from a Senate race appearing on the ballot with him, yet he’s likely to face harsh questions about the cost of running two separate elections from Democrats. National Republicans, also, will be critical of his spending decision and may hold against him the fact that he has deprived a GOP Senate incumbent of a November 2014 election.

Christie is still a heavy favorite for re-election as governor, but expect his enemies to try to use this decision over when to hold the gubernatorial election against him. Cory Booker, meanwhile, gets his primary and is standing pretty.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Lautenberg’s legacy for New Jersey rivals Chris Christie and Cory Booker | Harry J Enten

The death of Senator Frank Lautenberg creates both pitfalls and opportunities for the Republican Christie and Democrat Booker

The death of New Jersey Senator Frank Lautenberg has upended the political landscape in the Garden State. New Jersey’s Republican Governor Chris Christie now must select someone to fill the vacancy in the upcoming days. New Jersey voters will then vote on a replacement in an election at some point. When that election is could have major political ramifications for well-known political figures Christie and Newark Mayor Cory Booker. Why?

For Christie, 2013 was supposed to be the year of his re-election. Christie had benefitted greatly from voters’ positive views of his handling of Hurricane Sandy. No Democratic heavyweights were willing to run against him. Only the politically weak New Jersey state Senator Barbara Buono, who Christie leads by 30+ points, put her hat into the ring. Christie was also going to benefit from the fact that no regularly scheduled federal elections were to take place this year in New Jersey. The off-year electorate is typically older than one would see in a presidential election year and would pad Christie’s margin by about 2-4pt in the event that the polls tightened.

For Booker, the path from mayoralty to United States Senate was fairly clear. He was going to raise a ton of money, beat off any competition in a Democratic primary in which he already led by 40pt, and crush any Republican competition in a state that hasn’t elected a Republican senator in 40 years. It wouldn’t matter that Booker inelegantly pushed Lautenberg out of running for re-election 2014. It wouldn’t matter that he left the Democratic party without a politically strong standard-bearer in 2013 and may have cost his party control of the state legislature. The public loves Cory Booker.

Now, neither Christie nor Booker are in a comfortable position. Christie has to select a replacement. He’ll probably pick a Republican, but what type of Republican? Will he pick an elder statesmen like former Governor Tom Kean who wouldn’t run for re-election? In doing so, he’ll give away any advantage that an incumbent might have and almost certainly needs to win a Senate seat in this very Democratic state. Yet, a more moderate selection won’t help him among the Republican base if he decides to run for president in 2016. Will he make a more polarizing pick like State Senator Tom Kean Jr or Lt Governor Kim Guadagno, who almost certainly would run for re-election? Doing so might alienate Christie from the heavily Democratic electorate in New Jersey.

The greatest question for Christie though is when to hold the election. Without confusing anyone too much detail here, suffice to say that conflicting state statutes mandate that the election needs to be held in November 2013 or November 2014. Which statute is right is likely to be determined in court. Most state Democrats no doubt want November 2013. Such a close election gives little time for any Republican senatorial candidate to build the necessary statewide name recognition to win. Further, if the Democrats somehow lose in 2013, they get another shot in the regularly scheduled 2014 election. More important, though, is the impact an earlier election will have on the gubernatorial race.

A Senate election in 2013 could help to nationalize the gubernatorial election taking place on the same day. Christie leads the gubernatorial race by 30pt because he takes 30% of the Democratic vote and 65% of the independent vote that split evenly between Obama and Romney in 2012. The last thing Christie needs is for these voters to have a reminder on the ballot that he’s a member of the Republican party that is not well liked in New Jersey. In addition, Christie doesn’t want Democratic voters to come out as strongly as they do in a presidential election year atmosphere, as they might if Booker is the Democratic nominee. Christie doesn’t need to take chances given his lead has already been cut from 40+ to 30+ pt; and most agree that it’s likely to be cut further.

Of course, Christie probably doesn’t want to end up with a prolonged court battle over the date of an election. The mere action of going to court to fight with Democrats over the date of the election is going to put Christie in the press for being a Republican. He might think that his 30+ pt lead is wide enough, and the only thing that could screw it up is a publicized battle with state Democrats. Not choosing November 2014, however, will definitely perturb national Republicans who know that a 2014 election is the only chance they have to win the seat.

If the picture is murky for Christie, it’s pretty clear for Booker. He also wants 2014. How could it be that Christie and Booker want the same thing? If the election is held in 2013, the state Democratic party picks the Democratic candidate. If the election is held in 2014, the Democratic nominee will be selected in a primary process. While there is the smart thought that the state Democratic party wouldn’t dare pass over the popular Booker, who polls significantly better in a primary and general election than other Democrats, it’s difficult to predict the actions of a few people.

State Democrats may dislike Booker’s prior actions so much that they are willing to pass him over. They could make the following argument: Booker may lead his opponents by 40pt in a primary match-up, yet he’s only at 50% in a hypothetical primary contest. You could say that half of New Jersey Democrats prefer someone else. They also know that any Democrat selected is likely to win in 2013, and a close election may, if anything, boost turnout that could help Buono in the gubernatorial race.

The fact is that Booker is sporting a +70 favorable rating among New Jersey Democrats. Yes, a 2014 election could give a Republican incumbent a slightly greater chance at defeating Booker, but that’s unlikely to make much of a difference given Booker is ahead by 20+ pt against all potential Republican candidates.

Whether New Jersey voters select a United States senator in November 2013 or 2014 is unknown. My guess is 2014 will be the date because that’s the date Christie’s probably going to select. The court is more likely to confirm than overturn the decision of the head of the state, especially given that a recent clarifier by that State Office of Legislative Services says 2014 is the proper date. That’s good news for Booker and Christie, who will say that 2014 is what state law tells him to do. I’d still bet on both to win in 2013, if the Senate election is this year.

Frank Lautenberg’s death, however, has definitely added a wrinkle to both Chris Christie and Cory Booker’s political planning.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds