Category Archives: NJ Senate 2013 Special Election

Additional Analysis from the Report on 2013 Rutgers-Eagleton Final Election Polls

Last week we released a report from Langer Research Associates commissioned by the Eagleton Institute of Politics examining the reasons for our mis-estimates of the U.S. Senate race in October and the gubernatorial race in November 2013. In addition to examining question order priming effects, which were found to be the primary cause of the mis-estimates, Langer Research Associates examined some general operational aspects of the Rutgers-Eagleton Poll to assess the degree to which any of them influenced the estimates. None of these items were found to be causes of the mis-estimates, but Langer’s assessment provides useful guidance for the Poll.

This week we provide a brief summary of the operational portions of the report, along with our responses, which in some cases include changes in some of our processes going forward. We do this as part of our commitment to transparency and our educational mission.


Likely Voter Modeling

Issue: Pre-election polls have to estimate who will vote, which is done through likely voter modelling. This means identifying through a series of questions which respondents are likely to turn out to vote and which are not.

Summary of Langer Analysis: The Rutgers-Eagleton Poll employed a series of questions including self-reported likelihood of voting, awareness of the election date (for the Special Senate election), reported last time voting, following news of the election, and attention to debates. The calculation of likely voter was applied independently to the Senate and Governor elections. The Poll employed a scoring methodology, which assigned points to each response. The likely voter modeling was conceptually sound. However the cutoff point used greatly overestimated actual turnout. Likely voter models that overstate turnout include non-voters in their vote-preference estimates, which can compromise the accuracy of these estimates.

Likely voter modeling, however, was not the culprit in the discrepancies in the 2013 Rutgers-Eagleton estimates. Constructing three tighter likely voter models with turnout estimates as low as 32 percent in the Senate race and 38 percent in the gubernatorial contest made no substantive difference in vote-preference estimates.

Rutgers-Eagleton Poll Response: The LV screens were relatively loose in these reports, due to the size of the original sample, which simply did not allow tightening screens since that would result in too small samples. The Rutgers-Eagleton Poll staff did test varying screens and found few differences, as also determined by the Langer analysis. So the looser screens were chosen for reporting. One potential revision for future LV screens would be to use propensity scoring rather than a cutoff approach, which would allow all cases where the likelihood of voting was greater than zero to remain in the sample, weighted to reflect their relative propensity to vote. One attempt to improve the estimates that was employed was an adjustment to reflect the greater likelihood of Republicans turning out, beyond what was appearing in the LV screens. Such an approach is not industry standard and should not be employed in the future. As it turned out, this adjustment made no significant difference in the estimates. But even if it improves an estimate, we agree with the Langer report that this approach departs from best practices and should not be employed.

The Rutgers-Eagleton Poll is currently involved in a broad research project to reassess the weighting process that we use, and we anticipate the results of that project will begin to be used in polls beginning in the 2014-2015 academic year.


Weighting

Issue: Non-response generally results in variation between the sample that is completed and target population norms which are based on U.S. Census data. One potential problem could be incorrect weighting of the sample prior to reporting the results. The Rutgers-Eagleton Poll is a random digit dial (RDD) survey, requiring that respondents be asked if they are registered voters in order to determine if they should be included in the sample for the purposes of asking election-related questions. For both the October and November 2013 polls, those who responded that they were not registered to vote were immediately terminated, meaning no additional questions were asked. Thus the samples are of registered voters only and must be weighted to norms for registered voters.

Summary of Langer Analysis: The Langer Report suggests that it is standard practice to weight to demographic variables for the full population, not to the registered voter population. To do so would require not terminating non-registered voters and at least asking them a series of demographic questions. Since the Rutgers-Eagleton Polls analyzed here were of registered voters only, this option was not available. The registered voter sample was weighted to the Census Bureau Current Population Survey (CPS) from March 2012 using age, gender, race and ethnicity as target demographics. Had the Poll sampled all adults, the all adult sample would have been weighted to the Census’s American Community Survey (ACS), the standard source for weighting population surveys. It is important to note that the CPS registered voter norms were from the 2012 election, making them 11 months old in 2013, which means they do not capture voter registration changes that might occur during the election season.

Rutgers-Eagleton Poll Response: The Rutgers-Eagleton Poll will continue to generally weight samples to age, gender, race, and ethnicity targets for the population from which the sample is drawn. We will make sure that we are using the most recent available norms at all times. We may do more “all adult” samples which will include subsamples of registered voters, and will allow the full sample to be weighted to current ACS norms. However, in doing so we will necessarily have smaller samples of registered voters since to increase the overall sample size would require additional financial resources not currently available.


Sampling, including cell phones

Issue: Surveys of NJ residents must be based on a probability sample of landline and cell phone respondents in New Jersey. In this survey, respondents were asked if they were registered voters and were terminated if they were not. A second area of investigation is the relative share of cell phone calls placed as part of the sample. NJ has one of the lowest cell phone-only penetration levels, but nonetheless a significant number of residents cannot be reached without dialing cell phones.

Summary of Langer Analysis: The sampling process appears appropriate for both cell phones and landlines. However, the termination of non-registered voters means the sample must be weighted to norms for registered voters, which are generally less current than adult population norms. As noted above, the Langer report suggests that non-registered respondents should be retained for the collection of demographic data before termination.

Given the increasing use of cell phones and in particular the increasing proportion of cell-phone-only households in the United States, the inclusion of a robust sample of cell phones is a necessary practice. The Rutgers-Eagleton Poll uses an overlapping dual frame sample that includes a sample of cell phone respondents regardless of whether or not they have landlines. Estimates from the federal National Health Interview Survey (NHIS) indicate that, as of December 2011, 16.5 percent of New Jersey adults used cell phones only, and an additional 24.7 percent relied “mostly” on cell phones. These proportions surely have increased since then. The October and November Rutgers-Eagleton polls, using an overlapping dual frame design, included 17 percent and 23 percent cell phone interviews (weighted to 22 and 26 percent, respectively), with 4 and 7 percent cell phone-only respondents (weighted to 6 and 8 percent, respectively), well below available NHIS estimates.

Rutgers-Eagleton Poll Response: The Rutgers-Eagleton Poll has increased the cell phone target to 30 percent of the sample and will monitor whether this provides a reasonable share of cell phone-only households. We are also collecting additional information from respondents including the number of adults in the household (for landlines) and the number of adults sharing a cell phone (for cell respondents). These data will help with improving weighting calculations.


Question Wording & Field Dates

Issue: Field dates and question wording are other potential causes of differences in survey estimates.

Summary of Langer Analysis: The review finds no indication that either field dates or wording influenced Senate or gubernatorial vote preference estimates in these surveys. Question wording, while different in each survey, in all cases was balanced and neutral.

Rutgers-Eagleton Poll Response: Question wording was slightly different between the Senate and gubernatorial head-to-head questions. The biggest difference was that voters who responded don’t know in the gubernatorial question in October we not asked about which way they leaned. They were asked this in November, and the Senate vote asked about leaners in October.

 

Advertisements

Leave a comment

Filed under 2013 NJ Election, NJ Senate 2013 Special Election

Analysis of Rutgers-Eagleton 2013 Pre-election Polls Released

Following inaccurate results for final pre-election polls in October 2013 (NJ Special Senate) and November 2013 (NJ Governor), the Eagleton Institute of Politics commissioned an outside study by Gary Langer of Langer Research Associates of New York to identify reasons for the outcomes of these polls. Today, The Eagleton Institute of Politics and Eagleton Center for Public Interest Polling are releasing this analysis to the public as part of a commitment to transparency and education.

The Rutgers-Eagleton Poll reported a final pre-election poll for the special Senate election between then-Newark Mayor Cory Booker, a Democrat, and Republican former Bogota Mayor Steve Lonegan in which Booker held a 22-point lead. Booker ultimately won by 11 points. In the final November gubernatorial pre-election poll, Rutgers-Eagleton had Republican Governor Chris Christie ahead of his Democratic challenger state Senator Barbara Buono by 36 points: Christie won by 22.

The Langer report identifies the primary reason for the inaccurate results as the failure to put the “head-to-head” questions, which asked respondents for their vote intention, at or near the beginning of the questionnaire. Because these questions were asked after a series of other questions, it appears that respondents were “primed” to think positively about Governor Chris Christie in the November survey, which then may have led Democrats and independents in particular to over-report their likelihood of voting for the Governor. A similar process occurred with the October Senate poll, where voters were first reminded of how little they knew about Lonegan and how much they liked Booker before being asked the vote question.

Ruth B. Mandel, director of the Eagleton Institute of Politics stated that, “In response to these results, Eagleton chose to contract with an independent, highly respected, outside survey research firm to review its recent work and offer suggestions for improvement.” She added, “The Institute is committed to contributing to political knowledge in New Jersey and nationally with credible, impartial data. When we saw we had a problem, we knew we had to learn why and what to do about it.”

“Gary Langer and his colleagues spent many hours examining multiple aspects of our polling to understand what went wrong,” said David Redlawsk, director of Eagleton’s Center on Public Interest Polling (ECPIP) and professor of political science at Rutgers. “We are grateful for the efforts they put in and the advice they have provided, both in terms of this specific issue and general operations of the Rutgers-Eagleton Poll. The results of this report will make what we do even better.”

The Rutgers-Eagleton Poll has been a valued source of information about the views of New Jersey residents for over 40 years. As an academic-based survey research organization, ECPIP strives to be transparent and accessible. “We have a special obligation to take our educational mission seriously, which includes informing the public as well as learning from our own errors.” Redlawsk notes that survey research results released by the Rutgers-Eagleton Poll, for example, aim to meet the transparency standards set by the American Association of Public Opinion Research (AAPOR). Further, in recent years, the Rutgers-Eagleton Poll has been providing open informal insights and perspectives about survey research from Redlawsk and members of his staff through its blog at https://eagletonpollblog.wordpress.com. And for many years full data from the Poll has been freely available, generally after a one-year period, at http://eagleton.libraries.rutgers.edu/.

Langer’s major finding is that the order in which the head-to-head ballot test questions were asked most likely added inadvertent bias to the results in both the October and November Polls, although the results came out in opposite partisan directions in the two polls. Decisions made by ECPIP to maintain the standard set of questions about political figures including Cory Booker and ratings of Chris Christie at the beginning of the questionnaire worked to particularly prime Democrats in the November poll and Republicans in the October poll to support the candidate from the other party – Christie or Booker.

Redlawsk noted that the cause was a decision to maintain an ongoing four-year series of questions about Governor Christie that have been asked at the very beginning of a Rutgers-Eagleton NJ Poll since the governor’s inauguration. “We made this decision purposefully to maintain the integrity of our time series,” said Redlawsk. “This long-term research has greatly informed our understanding of public opinion about Governor Christie, and we had concerns that moving these questions after a head-to-head vote question would bias those results for the same reason we ended up biasing the vote questions.”

Most pre-election head-to-head polls focus only on the election and do not include long batteries of additional questions. The Rutgers-Eagleton Poll was unable to field separate pre-election surveys and thus combined the head-to-head polls with the regular surveys of New Jersey public opinion. “In retrospect, this was the wrong choice when one goal was to be as accurate as possible with pre-election numbers,” noted Redlawsk. “We should have either fielded a separate poll or just focused on our long-term work, rather than trying to do both at the same time.”

The Langer report on the cause of the pre-election poll mis-estimates is available to the public now on the Rutgers-Eagleton Poll website at http://eagletonpoll.rutgers.edu (PDF).

 

Leave a comment

Filed under 2013 NJ Election, Buono, Chris Christie, Cory Booker, NJ Senate 2013 Special Election, Steve Lonegan

#2 Senator Cory Booker – Rutgers-Eagleton Poll’s 2013 Top 5 Countdown

2.) Cory Booker wins the Special Senate Election

This special election was not supposed to happen – at least not this year. Newark Mayor Cory Booker had been on a different path, with speculations rising over a possible challenge to Christie for the governorship and then an early (and much-criticized bid) to unseat fellow Democrat Senator Frank Lautenberg (D-NJ) in 2014. But all of that changed when Senator Frank Lautenberg passed away on June 3, 2013, leaving behind a vacant Senate seat. Christie called for a special election, and a race between Booker and former Bogota, New Jersey Mayor Steve Lonegan ensued. The special Senate race in New Jersey became major national news as both candidates stood behind their stark ideological differences and as Booker’s large lead – once thought inevitable – took a surprise turn and began to narrow in the final weeks, making the race more of a contest than anyone had expected. Polls widely varied throughout the shortened campaign, with who exactly would turn out on the specially set date extremely difficult to predict. Booker ultimately won by 10 points – suggesting that the race tightened up as time went on, but also reinforcing that uncertainties that are always part of special elections. The Rutgers-Eagleton Poll recapped the twists and turns of the campaign for the London School of Economics and Political Science’s American Politics and Policy blog.

Booker

And on to #1 for 2013…

Leave a comment

Filed under Cory Booker, NJ Senate 2013 Special Election

Well, Booker won, but we were pretty far off…

While not all the final numbers are in as this gets written, it looks like Cory Booker has won the Senate race by between 11 and 12 points. Our final poll, which ended last Sunday, had him up 22 points in our likely voter sample. That’s pretty far off. Monmouth and Quinnipiac were essentially right on the mark.

We’ll try to figure out why we missed it this far over the next weeks. After all, while we’d like to get it “right” we always stand by two points: 1. Polling is NOT a prediction, no matter how much we sometimes act and talk as if it is. It provides an estimate at the point in which the poll is done. It does not foretell the future. 2. We are an academic operation, and even when we get what seem to be out-of-the-norm results, we look for the learning opportunities for ourselves and for others in what we find.

What’s interesting to us is that our other numbers in the same poll make “sense” – for example, our favorability ratings for Booker are in the same range as other polls that have reported them.  And as our releases over the next week will show, other numbers from the same sample also seem reasonable. So something is odd specifically in what we recorded for the Booker-Lonegan vote question. It doesn’t look like a technical problem; we’ve checked that. We recorded what people told us. And the sample doesn’t seem demographically odd as a whole. Given these two points, no matter how we thought about likely voters in our modeling, we kept showing Booker +20 or so.

On obvious problem, then, is that we failed to get a good “likely voter” screen as much as we tried.  Our likely voter calculation did screen out more unlikely Booker supporters than it did unlikely Lonegan supporters, which suggested Lonegan’s supporters would turn out at a higher ratio than would Booker’s.

And in a very quick look, that is be exactly what happened, but even more than we expected.

Last year Republican Joe Kyrillos lost to Sen. Bob Menendez by a 17-point margin. Kyrillos won 7 counties, Menendez won 14 counties. This time around, Lonegan appears to have won the same 7 counties plus another two. More interestingly, Lonegan overperformed Kyrillos in 18 of 21 counties – that is, he did better as a percentage of the vote than did last year’s GOP candidate throughout the state.

Booker overperformed in exactly zero counties as this is written, instead getting the same percentage in 3 counties that Menendez did last year. So it does look like Lonegan supporters were much more likely to turn out than we were seeing in our data.

Another indicator is that in the 7 counties Kyrillos won last year, turnout was just under half of what it was in 2012 (about 46% of last year’s total at the moment).

But in the counties Menendez won last year (and Booker won all but two this year) turnout appears to only be about 40% of the number that voted last year.  Again, Booker not only underperformed Menendez, but also had lower turnout in his winning counties.

Finally,  there are real regional differences: Lonegan’s “best” over-performance came in:

Cumberland +13 points over Kyrillos’ results
Salem +11
Warren +9
Atlantic +8
Ocean +8
Gloucester +8
Cape May +7
Sussex +7
Camden +5

Except for Sussex and Warren, these are southern and shore counties.
Booker’s least worst counties were:

Essex +0 points versus Menendez’s results
Bergen +0
Monmouth +0
Hudson -1
Mercer -1
Morris -1

Nothing South Jersey in this list, though Booker held even in a couple Republican counties.

This leads us to wonder if the fact that Booker is African-American played any role. If it did for us, why not for other pollsters? We have some thoughts on this point we will be investigating, and we’ll come back and talk about that once we’ve done so.

These are not final numbers at this point, so there may be a percentage point shift here and there. But one part of the story seems clear. We assumed better turnout in Democratic counties than actually occurred and we underestimated Republican turnout. And,  Booker underperformed compared to Lonegan all over the state, if we use 2012 as a baseline, which we did not see coming in what people were telling us last week.

It might be worth noting that in most places a 11-12 point win is pretty much a landslide. But given the low initial expectations for Steve Lonegan, it probably seems disappointing to Booker forces. But give Lonegan lots of credit. he ran an impressive attention-getting campaign, for better or worse, and probably can count this as a moral victory if nothing else.

As for us, well we do have another election on which to focus!

Leave a comment

Filed under 2013 NJ Election, Cory Booker, NJ Senate 2013 Special Election

Rutgers-Eagleton Poll: Booker Lead Greater than Expected

CLICK HERE FOR THE TEXT OF THE 10/14 BOOKER-LONEGAN RELEASE FROM THE RUTGERS-EAGLETON POLL. Alternatively, the full text of the release follows a somewhat lengthy speculation on why this poll differs from a brand new Monmouth poll showing a much smaller margin for Booker.

The following analysis is by Rutgers-Eagleton Poll Director David Redlawsk  [EDITS as noted 6:40pm EDT, 10/14/13].

Well, these are the times that pollsters hate. I think I know how the Gallup people must have felt when their 2012 vote models showed Romney ahead in the waning days of last year’s election, something that was clearly not true.

Why, you ask? Because today’s Rutgers-Eagleton Poll puts Cory Booker up 22-points on Steve Lonegan, a far larger margin than any other recent poll, and, in particular well above Monmouth’s 10-point lead for Booker, also out today. Monmouth’s polling dates fall within the dates we polled as well – though we were in the field longer. However, looking at our daily numbers we don’t think the longer fielding time is the issue, though we do see movement more toward than away from Booker over the last week, especially on our last day of polling, Oct 13, the day after Monmouth ended their poll. In fact, of our week of polling, the two days with the best Lonegan performance are the 10th and 12th, two of the three days Monmouth polled.

Still, I am puzzled, to be quite honest. I understood our large margin for Booker right after the primary – at that time Lonegan had gotten nothing but bad press and Booker had just come off a huge primary victory. But here it is harder to see what’s happening.

What’s especially odd from our perspective is that numbers other than the vote – for example, favorability – are very similar between the two polls.

So, anyway, which is it: Booker up 10, or Booker up 22?

We’ve looked at our data over and over. We’ve looked at other indicators besides the Booker-Lonegan race and those numbers seem quite reasonable, as we’ll report later. We looked at our weighted partisan breakdown for likely voters, and that’s not substantially different either. What is different between us and Monmouth today? We see a few key areas:

Gender questions

No, we don’t mean the Lonegan campaign’s attacks on Booker’s masculinity. That’s irrelevant.  But the difference between how men and women plan to vote does matter.

Monmouth shows essentially NO gender gap – in their poll 53% of men and 51% of women support Booker. We show a huge gender gap: 51% of men support Booker (basically the same as Monmouth) but we have 67% of women on his side (versus only 27% voting for Lonegan.) This alone could account for much of the difference. Interestingly, other polls have also shown a gender gap, most notably a recent Quinnipiac poll which had women going for Booker 62%-31%, although they showed men supporting Lonegan 51%-44% .

Independence

We have independent voters supporting Booker over Lonegan by a margin of 49% to 41%. But Monmouth’s poll has it the opposite: independents are 48% Lonegan and 43% Booker. we also have more Democrats (96%) supporting Booker and fewer Republicans (74%) supporting Lonegan. For Monmouth the number are 90% and 86% respectively.

Is it the Old Folks?

In our poll, voters 65 and over favor Booker 53%-43% over Lonegan.  Monmouth does not report an age crosstabulation so we do not know how different we are. However, we do know that Monmouth has a lot more senior citizens in its Likely Voter sample, at 36% of all voters, where we have just 30% of likely voters 65 and over. Since older people are more likely to support Lonegan, that could be a factor.

METHODOLOGY!  [EDIT, 6:40PM, 10/13: Huffpost Pollster talks about this point this evening…]

So here’s an interesting issue – Monmouth is using IVR for part of their polling. IVR stands for Interactive Voice Response – or in other words, robo-polling. In this method, computers dial landline numbers and respondents answer computer generated questions by keying responses on their phone keypad. With IVR the pollster has no idea who they are actually talking to – it could be anyone who picks up the phone. There is no human interaction. For the most part people who respond to IVR polls are often quite different from those who respond to live callers. Monmouth knows this so they supplement the IVR with live caller cellphones and some live landlines.  But more than half their sample is from IVR. It would be very interesting to see what the numbers say just for the IVR sample, compared to their live callers. However, they do not report this. Maybe they see no difference. [EDIT: Patrick Murray at Monmouth tells Huffpost that they did in fact see no difference. But since that was NOT disclosed in their press release or their methodology statement, it was impossible to know that when writing this post. I will note we don’t actually know what “no difference” means. Is in no difference in the marginals? Or no difference in the makeup of the samples? Pre or post-weighting?]

My guess is that IVR brings the older sample, and also results in the lack of a gender gap. But that’s only a guess right now. [EDIT, 6:40PM, 10/14: And apparently Patrick Murray suggests the same gender result in both IVR and live landlines, but again he isn’t completely clear on this.]

Race

This leads to one other interesting possibility. With IVR respondents don’t have to talk to anyone. This lessens what we call “social desireabilty” which has been shown to matter when race and ethnicity are a factor in preferences. People know it is “wrong” to express overtly racist attitudes. Likewise, there has been evidence in the past that voters may over report support for a non-white candidate in a live-caller poll. IVR does not have this problem – people can be honest without anyone knowing except the computer. Is it a factor here? I’m really not sure how deeply race comes into this election. If it does, then we may well get more supporters for Booker in our live calling than he will get in the privacy of the voting booth.

Finally, it’s also about likely voters and sampling

Both Rutgers-Eagleton and Monmouth try to identify likely voters. From what Monmouth says in their release, they used a listed registered voter sample and consider people who voted in two of the last four general elections to be likely voters. If they asked any other questions to determine this, they did not say in their release so we assume they did not. [EDIT 6:40PM, 10/14: Apparently, according to Huffpost, Monmouth did ask two additional screening questions. But again this was not made clear in the Monmouth release or methodology statement.]

We use Random Digit Dialing (RDD), not a listed sample. The disadvantage is we do not know voting history, and we have to screen for registered voters – and people may lie about this. The advantage is that we can hear people talk, and can get to the right person. We also theoretically have better coverage in that many people do not provide phone numbers when they register or provide incorrect numbers. With RDD theoretically anyone can get a call.

We ask a series of questions to determine likely voters – including awareness of the election date, attention to the campaign, a direct “will you vote question”, and a vote history question. Combining these gives us a model for likely voters that has fewer older people, and slightly more white people, but about the same share of each party as Monmouth. So the likely voter screen does not seem to be a huge source of the difference, at least in terms of  partisan breakdown.

[EDIT 6:40PM, 10/14: We also checked different levels of screening that were more or less strict, based on different cutoffs on our questions. The results did not vary by more than a couple points.]

So…

The upshot is that we report numbers quite different from other recent polls. Are we certain about them? Of course not – there are many reasons we could be way off. But we could also be within the ballpark.  In the end, every poll is an estimate and some will be on the mark and some will be off. Take each one with a grain of salt. Looking across all polls, if I had to guess (as opposed to poll), I think Booker will win with a margin in the mid-teens.

Still, a lot depends on who remembers to vote on Wednesday, October 16 and how effective the campaigns are a getting out their supporters.  If we’re wrong, we’ll try to figure out what happened. If we’re right, well, you heard it here first!

Finally, we could have simply not released this poll, but we might as well put it out there and see if others can see a problem that we are missing.

RELEASE TEXT FOLLOWS:

RUTGERS-EAGLETON POLL: BOOKER HOLDS LARGER THAN EXPECTED LEAD OVER LONEGAN ON EVE OF SPECIAL SENATE ELECTION

NEW BRUNSWICK, N.J – With just two days to go until the Oct. 16 special U.S. Senate election, Newark Mayor Cory Booker holds a 58 percent to 36 percent lead over former Republican Bogota Mayor Steve Lonegan among likely voters, a new Rutgers-Eagleton Poll finds. Another 3 percent prefer someone else, and about 3 percent remain undecided. While the Democrat’s lead is less than he enjoyed following the August primary, compared to other polls this latest Rutgers-Eagleton poll suggests Booker has recently gained in his bid to succeed the late Frank Lautenberg.

The debates between Booker and Lonegan, which highlighted the stark policy differences between them, may have played an important role in Booker’s current performance. While 43 percent of all likely voters paid some attention to a debate, independent voters who followed the debates strongly favor Booker, 59 percent to 37 percent. But independents ignoring the debates prefer Lonegan by a 45 percent to 42 percent margin.

For partisans, attention to the debate reinforced support for their own party’s candidate. At this point, Booker does much better than Lonegan within his party: 96 percent of likely Democrats are in Booker’s corner, while Lonegan wins only 74 percent of Republicans. GOP backers are less likely to have followed the debates: 36 percent compared to 49 percent of Democrats and 43 percent of likely independent voters.

“Other recent polls showed a narrower lead for Booker, but voters we talked to seem to have moved back in his direction,” said David Redlawsk, director of the Rutgers-Eagleton Poll and professor of political science at Rutgers. “The debates presented a stark picture of the differences between the candidates, which appears to have led independents to prefer Booker. While both candidates came out swinging, Lonegan’s attacks bordered on the taboo, with his ‘floating dead bodies’ and ‘big black hole’ comments about Newark – reminiscent of the rhetoric that got him into hot water early in the campaign. Independents in particular dislike these kinds of attack.”

Results are from a sample of 513 likely voters with a margin of error of +/- 4.3 percentage points. These likely voters are part of a sample of 798 registered voters polled statewide using live callers to both landline and cell phone households from Oct. 7-13.

Voters paying some attention

The uniqueness of the special Senate election makes estimates of turnout difficult and suggests that the final result will be unusually dependent on the candidates’ abilities to motivate their supporters. There is some evidence that voters are paying attention. More than half of registered voters know there is an election in October, and 45 percent know it will be held Oct. 16. Just over half are following the election at least fairly closely, with about one in five paying very close attention. Sixty percent of registered voters claim they are very likely to vote Wednesday.

Booker holds a commanding lead with minority voters, urbanites and voters in the South Jersey/Philadelphia region. Lonegan leads among voters in shore counties, 46 percent to 44 percent, while losing northwestern exurban voters, 51 percent to 41 percent, a better performance than his statewide margin.

“Here’s the key: if the Booker campaign can turn out urban voters and Democrats who say they will vote, he will be in the range we estimate,” said Redlawsk. “If Lonegan’s turnout operation is stronger and Booker’s base stays home thinking it is all wrapped up, then all bets are off.”

Booker ratings down slightly on Lonegan attacks

Lonegan’s attacks throughout the campaign appear to have had some impact. While Booker receives positive ratings from 54 percent of likely voters, this is a nine-point drop from an early September Rutgers-Eagleton Poll. Concurrently, his unfavorability has almost doubled to 32 percent.  Still, asked to rate the Democrat on a scale of 0 to 100 degrees, voters remain warm toward Booker with an average score of 56 degrees. The overriding emotional response to Booker is “hope,” with 48 percent saying his campaign makes them feel hopeful and 33 percent proud. In contrast, only 12 percent are angered and 19 percent worried by Booker.

Lonegan, however, continues to leave either a negative impression or none at all on most likely voters. While favorable impressions of the Republican have risen eight points to 30 percent, the share of voters who hold unfavorable impressions climbed to 34 percent. Over one third have no opinion on the GOP candidate. In general, likely voters are quite cool toward Lonegan; he averages only 41 degrees on the poll’s thermometer. Voters feel more negative about Lonegan than Booker: 37 percent are worried, and 30 percent angered by his campaign. Thirty percent are hopeful and 16 percent are proud of the candidate.

Many voters see Booker as a self-promoter

Booker is not without weaknesses. A third of likely voters say Booker is more about self-promotion and that life in Newark has improved little under this watch. While Lonegan has gained some support though this line of attack, a plurality (47 percent) continues to believe Booker has made real improvement in Newark.

Booker’s position on issues is seen as “about right” for 42 percent of likely voters, but 44 percent think he is too liberal, even for blue New Jersey. Lonegan, however, is thought too conservative by a large majority of likely voters; 60 percent say he is further to the right than most of the state, while only 18 percent say his views are right on par with the preferences of New Jerseyans.

“By all measures, voters see Booker as somewhat more in tune with New Jersey and see Lonegan as less so,” said Redlawsk. “Although Lonegan has tarnished Booker’s image, Booker remains the clear preference of voters, generating warmer feelings and a strongly positive impression, and taking positions more in line with what voters want.”

The party base, women and independents support Booker

Booker’s advantage over Lonegan is driven by overwhelming support from his own party base and a winning margin with independents. Ninety-six percent of likely Democratic voters back Booker, compared to only 74 percent of likely Republican voters who support Lonegan.  Independents also are in Booker’s corner, 49 percent to 41 percent for Lonegan. Newark’s mayor captures 16 percent of the Republican vote, while Lonegan peels off only 2 percent of Democrats.

Booker leads across virtually all demographic groups. While both men and women go for Booker, women are 13 points more likely than men to say they will vote for the Democrat and 18 points less likely than men to say they will vote for his opponent, evidencing a wide gender gap.

Booker also gets support from almost 9 in 10 likely voters who say he has made a real difference in Newark, but among those who think Booker is too much about his own self-promotion, only one in five say they will vote for him.

4 Comments

Filed under 2013 NJ Election, Cory Booker, NJ Senate 2013 Special Election, Steve Lonegan

LIKELY VOTERS GIVE BOOKER LARGE LEAD, MOST EXPECT HIM TO WIN; LONEGAN WIDELY UNKNOWN

We start this next round of polling results with the upcoming special U.S. Senate election, to be held on October 16. The candidates are Newark Mayor Cory Booker and former Bogota Mayor Steve Lonegan. It has been less than a month since the primary election that made both their party’s nominee, though neither had much trouble winning their respective primaries. Turnout in the Democratic primary surprised most observers, with more than 20% of Democrats going to the polls, much higher than expected. Republican turnout was quite low, however. Perhaps this reflected a less competitive Republican primary, or maybe a lack of enthusiasm for Lonegan. Hard to tell. But in our first post-primary head-to-head test of the two candidates, less than strong Republican numbers suggest it might have been the latter.

Booker holds on to 93% of Democrats, with only 3% of Democrats crossing over to support Lonegan. But nearly one in five Republicans says they will vote for Booker. Given this, along with strong registered voter advantage Democrats hold and a solid showing among independents, Booker now holds a large 64% – 29% lead among the likely Senate voters in our sample. Is the race really that lopsided? Or are we reflecting a couple bad media weeks for Lonegan including the Tweet by one of his staff of a racially insensitive map of Newark and Lonegan’s recent questions about Booker’s masculinity. Neither has created a positive image for the Republican. In fact, his core problem could be he has no image at all among a majority of likely voters, and among those with an impression, half are positive and half are negative. Meanwhile, ratings of Booker are overwhelmingly positive among likely voters.

One thing of note in this likely voter sample – it is very Democratic, in fact 23 points more Democrat than Republican. We don’t weight likely voter samples to party registration, so this is what we actually found in our polling. This matters, given that nearly all Democrats say they will vote for Booker. Obviously the more Democrats who make it through the likely voter screen the better the result for Booker. So we took a look to see what would happen if the sample were less Democratic, say 40% Democrat and 22% Republican, that is, 18 points more Democratic than Republican instead of 23 points. Assuming nothing else changed, we’d find a 30 point lead for Booker. No matter how we slice the sample Booker is doing very well. It would take a massive Republican turnout assumption to change this right now.

A note about likely voters in this poll. It is difficult to figure out who will show up for a special election on a Wednesday in mid-October. In the end we decided on three simple screens: when the respondent last voted (this year, last year, somewhere in the past before that), how closely they are following the senate election, and self-stated likelihood of voting in the senate election. These three questions result in a 0-10 scale; we have somewhat arbitrarily decided that those scoring 7 or higher are likely voters. This gives us 462 LVs in a sample of 814 registered voters. As it turns out it doesn’t make a lot of difference where we set the cutoff, so this seems reasonable.

The text of the release follows. Click here for a PDF of the release text with questions and tables.

LIKELY VOTERS GIVE BOOKER LARGE LEAD, MOST EXPECT HIM TO WIN; LONEGAN WIDELY UNKNOWN

NEW BRUNSWICK, N.J. – With just over five weeks until the Oct. 16 special U.S. Senate election, Newark Mayor Cory Booker has opened a large lead over former Bogota Mayor Steve Lonegan, according to a new Rutgers-Eagleton Poll. Almost two-thirds (64 percent) of likely voters support Booker, 29 percent plan to vote for Lonegan and 6 percent are undecided.

Most likely voters think Booker, a Democrat, will win – including Republicans and Lonegan voters. And in a campaign where the candidates disagree on almost everything, solid majorities in both camps say candidate’s issue positions are more important than leadership style.

As in the primary, Booker benefits from name recognition supported by his positive impression on most voters; 63 percent (versus 19 percent unfavorable), have a favorable impression of Booker, while 17 percent are neutral or simply do not know him. Lonegan, however, leaves no impression with more than half of likely voters. Among those with an impression, 22 percent are favorable and 22 percent are unfavorable.

The special election is drawing modest attention among registered voters: just over half claim they are following the election at least fairly closely with about a quarter giving it close attention. Just under 60 percent of registered voters say they are very likely to vote in the October election.

“Booker appears to be building an insurmountable lead,” said David Redlawsk, director of the Rutgers-Eagleton Poll and professor of political science at Rutgers. “While special elections are notoriously hard to predict, given uncertainties about turnout, Booker’s name recognition, celebrity-type status and stances on issues that align more with New Jersey’s ‘blue’ political climate seem to be driving momentum toward him and away from Lonegan. On top of that the Republican’s most recent news highlights attacking Booker’s masculinity have been quite unflattering.”

Results come from a sample of 462 likely voters with a margin of error of +/- 4.5 percentage points. All totaled, 925 New Jersey adults were polled statewide among both landline and cell phone households from Sept. 3-9. Within this adult sample are 814 registered voters, with a margin of error of +/-3.4 percentage points, from which the likely voter sample is taken.

Booker’s lead wide across the board

Booker’s double-digit advantage over Lonegan is driven by overwhelming support from his own party base and independents. More than 90 percent of likely Democratic voters back Booker, compared to only three-quarters of likely Republican voters who support Lonegan. Among independents, Booker holds a 52 percent to 38 percent lead. Booker even captures 19 percent of the Republican vote while Lonegan peels off only 3 percent of Democrats.

“The nearly unanimous party support is a key for Booker,” said Redlawsk. “Democrats seem motivated in this election, and may even be more likely to turn out than Republicans, who are much more split on their candidate.”

Booker handily leads across virtually all demographic groups. He especially wins over likely women voters by a huge margin, 73 percent to 21 percent. Likely male voters show a tighter race – 55 percent for Booker to 37 percent for Lonegan. Booker takes a commanding lead with minority voters, as well as younger voters and urbanites.

Booker also gets favorable ratings from some who do not plan to vote for him, but most with a favorable impression support him. Lonegan also gets majority support from those with a favorable impression of him. The problem is there are many fewer of these voters. Moreover, a quarter of those who like Lonegan still plan to vote for Booker, and Booker overwhelmingly wins the large number who have no opinion of Lonegan.

Voters paying some attention

A majority of all registered voters are paying some attention to the Senate race; 24 percent are watching it very closely and 32 percent fairly closely. But 44 percent are paying little attention and are not likely to vote. Democrats and independents are more likely than Republicans to be following the campaign very closely. Booker wins two-thirds of registered voters paying very close attention but only leads 52 to 27 percent among those paying just some attention. Lonegan’s battle will be uphill even if most registered voters show up. Among those with any chance of voting, Booker still holds an almost insurmountable 59 percent to 26 percent lead. Increasing turnout does not seem likely to pay off for Lonegan, at least right now, noted Redlawsk.

Most voters say issues are key

More than 60 percent of all likely voters say that a candidate’s stance on issues is more important than leadership style. Men are nine points more likely than women to prefer issues over style, though a solid majority of both genders call issues more important. Regardless, Booker wins among both groups: those who favor issues and those who vote for style.

About 80% percent of likely voters expect Booker to win, no matter their personal preference.  Sixty-four percent of Republicans, 60 percent of conservatives, and 69 percent of those favoring Lonegan believe Booker will win the special Senate election. Every demographic group believes the odds are against Lonegan.

4 Comments

Filed under 2013 NJ Election, Cory Booker, NJ Senate 2013 Special Election, NJ Voters, Steve Lonegan

On to the NJ Special Senate General Election!

OK, so let me start by admitting I was wrong. About what, you ask? Well, along with virtually every other pundit in the state, we here at the Rutgers-Eagleton Poll expected turnout in the special Senate primary to be abysmal, if even that. The primary was being held with little lead time, in the middle of August, when everyone it seems is down the shore. Moreover, both Republicans and Democrats appeared to have essentially uncompetitive races. And of course, turnout in primaries is generally low anyway. All of this combined to assure us that we’d see record low turnout.

Well, mea culpa. Though I suppose we were half right. Republican turnout looks to have been about 130,000 voters out of around 1.085 million registered Republicans, or about 12%. That’s pretty bad, by any standard. Democrats, though, nearly doubled this, with more than 350,000 showing up to vote, out of just under 1.8 million, or right around 20%. Now 20% might not seem like much, but it is a far cry from the anticipated “no one will vote” story.

The winners were Steve Lonegan, with 79% of the vote (with 98% of precincts reporting,) and Cory Booker with 59%. In both cases, the winners out performed the few polls that were out, but to be fair, Quinnipiac’s last poll was pretty close on both sides. For the Democrats, whether it’s that the 9% “undecideds” in that poll went more for Booker than any other candidate or that (more likely in my mind) undecideds didn’t actually show up and Booker had a better turnout operation than the others, Quinnipiac did a good job of what we thought would be a very difficult task: identifying likely voters. Also, the Monmouth poll done earlier suggested in their likely voter screen that turnout could range between 200,000 and 400,000 Democrats – with that high end seeming pretty unlikely. Yet, in fact, Democratic turnout was closer to the high than the low end.

We had decided not to poll because we were very uncertain about our likely voter model in this context; apparently the primary voters who turned out looked a lot like any likely primary voters, despite the oddity of an August primary. Most likely this is due as much to Cory Booker’s ground operation as the support for Frank Pallone and Rush Holt within their own Congressional Districts (both appear to have basically won their district.) [correction: the Star Ledger reported on Aug. 15 that Booker appears to have narrowly won in Holt’s district while Pallone did indeed win his.]

This morning, I did a quick and dirty look at Democratic turnout by county. Some interesting results:  (All Senate election and turnout numbers from http://project.wnyc.org/election2013. Registered voter numbers come from the state Board of Elections.)

Top 5 counties:  Mercer, 26%; Monmouth, 25%; Essex, 23%; Hunterdon, 23; Somerset, 23%.

Of these, Holt won 2 (Mercer and Hunterdon) and Pallone won one (Monmouth). Booker handily won Essex with 67% and got 53% in Somerset. The problem for both Pallone and Holt is that they won (or did best in) counties with many fewer Democrats than Booker, who, not surprisingly won overwhelmingly in the Democrat-rich counties of Essex, Bergen, and Hudson, even though turnout in the latter two was at or below the state average. Booker didn’t need a massive turnout percentage there to win, just lots of votes. Essex itself, for example, has more Democrats than Mercer and Monmouth combined.

Some other interesting things that reinforce the local aspect of politics. First, Middlesex, with about the same number of Democrats as Bergen (both about 50,000 Democrats behind Essex) had 22% turnout, but a true three-way race, with Booker at 38%, Pallone at 33% and Holt at 25%. Of course both Pallone and Holt both represent parts of Middlesex, accounting for their much stronger showing than in most of the rest of the state.

Booker got less than 50% in only 6 counties: Warren, Ocean, Mercer, Middlesex, Monmouth, and Hunterdon. Of these Pallone represents parts of Monmouth and Middlesex, and Holt has parts of Mercer, Middlesex and Hunterdon. Where they knew them, voters preferred Pallone or Holt, otherwise Booker’s wider recognition and celebrity status no doubt played a role in burying both Congressmen.

Also, for all the early news reports of “vote by mail” efforts in Camden county supposedly orchestrated by the Booker campaign, Camden turnout was the lowest in the state, at 11.3% with just over 15,000 reported Democratic votes. However, as of this writing, voters are reported from only 82% of precincts (compared to 98% of the state) so the numbers in Camden are no doubt higher, but will still most likely put it in the bottom 5 of turnout. But what Camden really has in common with other southern counties is that all of the candidates came from central or north Jersey. And for the most part, the turnout percentages reflect that.

Bottom 5 counties:  Camden, 11%; Gloucester, 12%; Salem, 13%; Cumberland, 13%; Warren, 14%/

Where the candidates live, and are better known, voters were simply much more likely to show up. No great surprise, perhaps, but interesting none the less.

So on to the special Senate general election on Wednesday October 16. Given the primary turnout, perhaps we should be more upbeat about turnout for that election. On the other hand, if the initial polling is right (and we suspect it is) then this will be a snoozer of a race, with Booker handily beating Lonegan. Doesn’t mean the campaign won’t be interesting; with Lonegan in it, it is likely to have at least some fireworks. But turnout for October? Who knows. But regardless, Rutgers-Eagleton will be polling this one, with out first poll on the race due out in early September. Watch for it!

2 Comments

Filed under 2013 NJ Election, Cory Booker, Frank Pallone, NJ Senate 2013 Special Election, NJ Voters, Rush Holt, Sheila Oliver