Wednesday, November 2, 2016

How is the Recent Email Controversy Affecting the Polls?

By Nicole Sandelier-Monmouth University Polling Institute Graduate Assistant
Last Friday, FBI Director James Comey sent a letter to congressional leaders stating that the FBI had “learned of the existence of emails that appear to be pertinent to the investigation” of Hillary Clinton’s use of a personal email server.  With Election Day right around the corner, how will the new revelation impact the presidential race? 
It is important to note that even before the recent news regarding Clinton’s emails, national polls were already tightening.  According to the Real Clear Politics 4-way national average, Clintons’ lead had been on a decline.  On October 18th, Clinton led Trump 46% to 39%, and on the day that the Comey news broke, Clinton’s lead had fallen to 45% to 41%.  As of today, Clinton is hanging onto a slim 2-point lead (45% to 43%) nationally.
Although it may still be too early to tell, as of now there are scare data suggesting the recent news regarding Clinton’s emails has caused voters to rethink their vote preference. A recent national ABC News/ Washington Post poll found 63% of voters nationally saying the recent news does not affect how likely they are to support Clinton.  Recent Monmouth University polls in Indiana, Missouri, and Pennsylvania draw an even stronger conclusion. Fewer than 5% of voters in each state say Comey’s letter actually caused them to change their vote choice. Since this finding includes supporters of both candidates, the net effect of Comey’s letter is only a net 1 or 2 point gain for Trump. With all the coverage and talk focusing on Comey’s decision to re-open the investigation, there is little evidence it has been overwhelmingly detrimental to the Clinton campaign and her standing in the polls…yet. 

Tuesday, October 4, 2016

Clinton Enjoys a Post-Debate Bump as Majority Feel Trump Does Not Have Presidential Temperament

by Ashley Medina and Nicole Sandelier
Monmouth University Polling Institute graduate assistants

A Monmouth University Poll released the morning of the debate suggested that the vast majority of voters (87%) did not expect to learn anything that would change their minds based on the first presidential debate.  With the majority of voters already set on their presidential candidate selection, Trump and Clinton have shifted their attention to gaining the support of undecided voters. Presidential temperament may be one of the factors that helps sways undecided voters. 

The national Monmouth University Poll that came out on debate day found that nearly 6-in-10 voters believe Hillary Clinton has the right temperament to sit in the Oval Office, while just 35% feel the same about Donald Trump’s temperament.  A FOX poll conducted just after the event mirrors pre-debate findings on presidential temperament stating that 67% of likely voters say Clinton has a presidential temperament while only 37% say Trump has the temperament to be president.

The most recent Monmouth University Polls in the battleground states of Colorado and Pennsylvania appear to be reflective of national views concerning both Clinton’s and Trump’s temperament.  A majority of likely Colorado (61%) and Pennsylvania (64%) voters feel that Hillary Clinton has the right temperament to be president.  Meanwhile, only 31% of likely Colorado and Pennsylvania voters feel that Donald Trump has the temperament to be president.  With Election Day just around the corner, the candidate’s presidential temperament will continue to play a key role in swaying undecided voters in battleground states.

According to Nielsen, an estimated 84 million people watched the first presidential showdown between candidates.  Recent polls have expressed voters’ opinion showing Clinton as the clear winner of the first debate (ABC/ The Washington PostPolitico/ Morning Consult). 

The latest Politico/Morning Consult Poll confirms Monmouth’s pre-debate findings, with approximately 8-in-10 voters (81%) stating that the debate did not change their ballot decision. About 1-in-10 (9%) voters said that the debate has influenced their selection for president.  Nonetheless, post-debate findings are confirming what pre-debate polls suggested.  The first presidential debate reaffirmed many voters' ballot selection and did little to sway voters' minds.

Monday, August 29, 2016

Historical Presidential Nominee Favorability Ratings

A Monmouth University Poll released today (http://monmouth.edu/polling-institute/reports/) underscored the historically high level of negative attitudes toward both major party nominees for president. 
The number of voters who cannot bring themselves to voice a favorable opinion of either major party nominee is unlike anything witnessed in past elections.  Only 2% have a favorable opinion of both Hillary Clinton and Donald Trump while one-third (35%) do not have a favorable opinion of either candidate.  These results are unprecedented according to polling data going back more than 30 years.
The number of voters in elections going back to 1984 who had a favorable opinion of both candidates was never lower than 5% – in fact registering as high as 19% in 2000.  Conversely, the number of voters who did not have a favorable opinion of either nominee was never higher than 9% – a fraction of what is being seen in the current election.

Among the 1-in-3 voters in the current poll who do not have a favorable opinion of either nominee, 21% say they have an unfavorable opinion of both candidates, 7% have an unfavorable view of Clinton while expressing “no opinion” of Trump, and 8% have an unfavorable view of Trump while expressing “no opinion” of Clinton.  Even taking into account differences in question wording and methodology compared to past election polls, the number of voters who hold negative views of both candidates is indisputably a record high.
Monmouth combined the data from its four national polls conducted this summer to get a better sense of these disapproving voters.  Based on this four-poll average, those with an unfavorable opinion of both nominees are dividing their support almost evenly among Trump (24%), Clinton (21%), and Johnson (22%), with Stein at 8%.  Among those who hold a negative view of one nominee and no opinion of the other candidate, however, the vast majority are voting for the candidate of whom they have no personal opinion.  This includes 77% of the “unfavorable Clinton/no opinion Trump” group who are voting for Trump and 75% of the “unfavorable Trump/no opinion Clinton” group who are voting for Clinton.
This is not surprising because the vast majority of “no opinion on Clinton voters” lean Democrat and the vast majority of “no opinion on Trump” voters lean Republican.  It just seems that they can’t bring themselves to admitting to a favorable opinion of the person they are grudgingly supporting.
It’s also worth noting that there are more Republicans than Democrats among voters who have an unfavorable opinion of both candidates and this negative group is also much more likely to be college educated.  The demographic composition of each voter group is below.
Among those who have an unfavorable opinion of Trump but no opinion of Clinton:
·         44% describe themselves as Democrats and 33% are independents who lean Democrat
·         51% are white, 21% are black, 23% are Hispanic, and 6% are Asian or other race
·         42% are under age 35, 26% are 35-49, 21% are 50-64, and 10% are 65 and older
·         41% are men and 59% are women
·         39% have a college degree
Among those who have an unfavorable opinion of Clinton but no opinion of Trump:
·         45% describe themselves as Republicans and 29% are independents who lean Republican
·         84% are white, 3% are black, 7% are Hispanic, and 7% are Asian or other race
·         23% are under age 35, 18% are 35-49, 33% are 50-64, and 25% are 65 and older
·         58% are men and 42% are women
·         46% have a college degree
Among those who have an unfavorable opinion of both Trump and Clinton:
·         29% are Republicans and 21% lean Republican, 13% are Democrats and 20% lean Democrat, and 18% are self-described independents who do not lean toward either party.
·         80% are white, 6% are black, 10% are Hispanic, and 4% are Asian or other race
·         36% are under age 35, 24% are 35-49, 26% are 50-64, and 15% are 65 and older
·         54% are men and 46% are women
·         56% have a college degree
It’s also worth noting that nearly 1-in-4 of those voters who do not have a favorable opinion of either candidate are considered to be unlikely to turn out to vote this November.  This compares to less than 1-in-10 with a favorable opinion of one of the candidates who are considered to be unlikely voters.
For the record, among those who have a favorable opinion of Clinton only:
·         72% describe themselves as Democrats and 19% are independents who lean Democrat
·         58% are white, 24% are black, 12% are Hispanic, and 5% are Asian or other race
·         22% are under age 35, 26% are 35-49, 28% are 50-64, and 24% are 65 and older
·         35% are men and 65% are women
·         53% have a college degree
·         93% are voting for Clinton
Among those who have a favorable opinion of Trump only:
·         62% describe themselves as Republicans and 25% are independents who lean Republican
·         89% are white, 2% are black, 7% are Hispanic, and 2% are Asian or other race
·         16% are under age 35, 27% are 35-49, 31% are 50-64, and 26% are 65 and older
·         57% are men and 43% are women
·         42% have a college degree
·         94% are voting for Trump
Another historical note: the difference between the two candidates’ favorability ratings correlates extremely closely with the actual margin of victory.  For example, Barack Obama had a 6 point advantage over Mitt Romney in candidate favorability in 2012 and ended up winning the popular vote in that election by 4 points.  Ronald Reagan had a 17 point favorability advantage over Walter Mondale in 1984 and won that election by 18 points.  Even in the razor thin election of 2000, Al Gore had a one point favorability edge over George W. Bush and won the national popular vote by half a percentage point despite losing the Electoral College.  The same is true in 2004 (favor +5R; vote +3R), 1996 (favor +6D; vote +8D), 1992 (favor +5D; vote +6D), and 1988 (favor +8R; vote +7R).  According to the average of recent polls reported by HuffPost Pollster, Clinton has about a 6 point advantage on this metric.
There are also intriguing down-ballot implications.  Some pundits point to the 1996 election when the GOP tried to disconnect the Congressional races from its presidential nominee who was trailing in the polls.  In that year, however, opinion of Bob Dole was fairly positive, with 50% of voters holding a favorable opinion of him.  This year, the top of ticket nominees in both party are largely negative, with Trump doing significantly worse among his fellow Republicans than Clinton is doing among her fellow Democrats.  This suggests that the GOP could have a bigger problem holding its base in down ballot races where their nominee is seen as aligned too closely with Trump.


Monday, August 8, 2016

A Poll Sample's Party Composition

A note on party composition in polling samples.

Some commenters have noted that the Democratic advantage in the latest Monmouth University Poll is larger than in our poll taken just prior to the two parties' conventions . Specifically, voters in the current poll self-identify their party leanings as 35% Democrat, 26% Republican, and 39% independent or other.  In the July poll it was 33% Democrat, 28% Republican, and 39% independent or other.

Contrary to some misperceptions - largely by those unhappy with the overall results of the latest poll - Monmouth did not "choose" the sample to look this way.  Party identification is a self-reported attitude based on where people see themselves fitting in the current political environment.

It is not the same as party registration or partisan voting behavior (e.g. consistently voting in one party's primaries), which is a more stable metric. I wrote about these differences in more detail a few years ago (Party ID Apples and Oranges).  While the data in that analysis were drawn from New Jersey voter files and poll samples, the underlying message is the same.  Party self-identification can move with the political climate, while party registration is more stable.

Monmouth's 2016 presidential polling uses a combination of voter lists and random digit dialing. The voter list includes data on voter registration and past primary voting.  According to this metric, 34% of the Monmouth sample are registered or active Democrats, 34% are Republicans, and 32% are independents or something else.

In other words, the Monmouth sample is evenly divided between Democrats and Republicans when it comes to registration and past voting behavior.  Yet when asked how they see themselves politically, these same voters are 9 points more likely to call themselves Democrats rather than Republicans.

The question you should be asking yourself, in light of events over the past few weeks, is why that might be so.





Tuesday, May 24, 2016

The Case for Including 3rd Party Candidates in Presidential Polls

by Ashley Medina
Monmouth University Polling Institute graduate assistant


As it becomes increasingly likely that the American public is now looking at their two major party candidates for the 2016 election, pollsters will begin to test the head to head matchup between Hillary Clinton and Donald Trump with more frequency. However, what many of these pollsters may fail to account for are the number of voters who may be looking for another option come election day.

A recent NBC News/Survey Monkey poll found 16% of voters nationwide say they would vote for a generic “3rd party” candidate rather than vote for either Clinton or Trump. These numbers suggest that a substantial number of U.S. voters may be seeking another option this November. While the U.S. electorate has expressed similar sentiments in the past, a single third party candidate has received that large of a vote share only once before. 

In 1992, self-funded Reform Party candidate Ross Perot won nearly 19% of the total 20% of votes earned by independent and third party candidates. The next largest showing for a single independent or 3rd party candidate came in the 1968 presidential election when American Independent Party candidate George Wallace earned nearly all 14% of the third-party candidate votes that year. Perot ran again in 1996, but this time, earned just 8% of the 10% total vote that independent and 3rd party candidates received. The 1996 election marked just the third time since 1948 that third party and independent candidates combined received at least double digit support. 

If current polling remains consistent, the third party gains in this upcoming presidential election could reach double digits. However, there are some caveats facing third party candidates during this cycle. For one, there will likely be several candidates vying for independent and third party votes. Additionally, many of them are largely unknown to most Americans and are likely to remain unknown unless they can make it to the debate stage. In order to do so, these candidates must appear on enough state ballots to mathematically earn an Electoral College victory as well as average at least 15% in national polls. Without the opportunity to participate in presidential debates, they will struggle to increase their name recognition. 

However, only three polls to date have included individual third party candidates. The first of these, a national Monmouth University Poll taken in March, found that in a match-up between the two front runners, Hillary Clinton held a ten point lead over Donald Trump. When Libertarian third party candidate Gary Johnson was added to the mix, both Clinton's and Trump’s numbers fell as Johnson pulled in 11% of the vote. This pattern was mirrored in a similar national Public Policy Poll where Clinton held a 6 point lead over Trump, but Clinton’s lead shrunk to 4 points when two third party candidates were added to the mock ballot, with Johnson at 4% and Green Party candidate Jill Stein at 2%. In a more recent national Fox News Poll, results were consistent with these third party findings. In this poll, when respondents were asked to choose first between Clinton and Trump, Trump led Clinton by 3 points, but when given the option of choosing between Clinton, Trump, and Johnson, Trump’s and Clinton’s vote share dropped 3 points each as Gary Johnson garnered 10% of the vote.

Given high voter discontent, it is likely that the third party vote will be higher than average this year, but we will not know just how high unless other polls include third party candidates in their surveys. As the rules stand, including these third party candidates in more polls is necessary if they are to have a chance at participating in the presidential debates.

The national polling requirements for third party candidates are rather unrealistic given the fact that a third party candidate was only once able to cross the 15% margin in the past 70 years. A look at Wallace’s regional appeal in 1968 suggests that this requirement may be unfair, as Wallace was able to earn enough Electoral College votes to impact the final outcome. More recently, in 2000, it is possible Ralph Nader’s 3% share of the vote was a contributing factor in that year’s race.

With this in mind, it is clear that even five percentage points in the polls can reflect the mood and preferences of significant segments of the U.S. voting base and as such, the voices of third party supporters should be represented on the presidential debate stage. It is for this reason that more pollsters should use methodologically sound ways to include these candidates in their polls. 

WATCH: Monmouth Poll Director discusses these issueshttp://bit.ly/1OLGHS6

Thursday, May 19, 2016

Republican Disenfranchisement reaches the House: Paul Ryan Fails to Endorse Trump

by Ashley Medina
Monmouth University Polling Institute graduate assistant

Recently, the political divide within the Republican Party became even more evident when Speaker of the House Paul Ryan issued a statement expressing that he is “not ready” to endorse Republican frontrunner Donald Trump.  However, the Speaker’s unwillingness to endorse the billionaire may hurt his own political career as opposed to that of the presidential candidate.

Trump and some of his supporters have voiced strong positions concerning the issue of Ryan’s statement but what does the greater electorate think?  Results from a recent (5/6-9) YouGov/Economist Poll cite that two out of three voters who participated (or plan to participate) in the GOP primaries and caucuses believe that Ryan should endorse Trump. 

Personal differences aside, Ryan now has to measure how the general Republican base’s attitudes and allegiances will affect his standing.  Keeping in mind the tremendous popular support the billionaire has been able to cultivate, it may be in Ryan’s best interest to officially support the candidate if he would like to maintain favor among the Republican electorate.

As the political climate stands now, Republican voters are actually more likely to side with Trump who among all American voters, is viewed favorably by only 30% and unfavorably by 64%. By comparison, Ryan is viewed somewhat more positively, with 34% of US voters viewing him favorably and 38% unfavorably. 

Ryan may be better liked than Trump among all voters, but among Republicans only, two out of three actually have positive views of Trump.  Should Trump mobilize his supporters against the Speaker, Ryan is likely to face political ramifications for his recent statements.  This, along with the fact 49% of GOP voters disapprove of what Ryan has done as Speaker of the House, may motivate him to “get ready” sooner rather than later to support the presumptive presidential nominee.

Wednesday, May 11, 2016

West Virginia’s Trump Supporting Sanders Voters

What is up with West Virginia Democrats? Eight years ago, Hillary Clinton won every single county on the way to a 2-to-1 victory over Barack Obama.  This year she lost every single county and got trounced by Bernie Sanders.

Well, here’s the thing.  Many of those voters aren’t really Democrats at all – at least not by any standards we would call a Democrat in the rest of the country.  While Democrats are still competitive for statewide office there, West Virginia has been solidly red in presidential elections for more than a decade. 

In fact, the exit poll included two questions about the November election pitting Donald Trump against either Clinton or Sanders.  According to results shown on MSNBC’s primary night coverage, nearly 3-in-10 of these Democratic primary voters actually said they will vote for Trump in either match-up. 

Let that sink in.  Three-in-ten voters who just cast a ballot in the Democratic primary said they would be voting for Trump in November regardless of “their” party’s nominee.  For the record, most of these Trump supporters voted for Sanders over Clinton – 60% to 12%, with another 28% of these mischief-makers voting for one of the largely unknown other names on the ballot.

These Trump supporters who took part in the Democratic primary are more likely than others to be from coal mining households (53%), more likely to be very worried about the nation’s economy (81%), and more likely to want the next president to be less liberal than Obama (69%).  The latter question has been asked in every exit poll this season and this is the only place where that many voters in a Democratic primary said they want to move in a less liberal direction!

These voters are most likely “legacy” Democrat.  They belong to the party as it exists in West Virginia, but they disdain the Democratic brand on the national stage.  It’s not that they like Bernie Sanders, but it’s more likely that they really detest Hillary Clinton.  If these voters did not participate in the presidential primary, we would have likely seen an extremely close margin between Sanders and Clinton rather than Sanders’s 15 point win.

And this may not be the strangest West Virginia outcome in the past few cycles. Remember that four years ago, a convicted felon who was incarcerated in Texas at the time got 41% of the Democratic primary vote against Obama.

So, let’s just mark the West Virginia primary down as one strange footnote to a very strange primary season.

Wednesday, May 4, 2016

Cruz Zigged While the GOP Electorate Zagged

The Cruz campaign’s attempt to coalesce the #NeverTrump movement around their candidate #NeverHappened.  In hindsight, the attempt to position him as the establishment alternative may not have been the wisest move.

Ted Cruz entered the 2016 presidential race with a reputation as the Senate Republican conference’s enfant terrible.  He ended his campaign as the establishment’s last hope to deny Donald Trump the party’s nomination.  The problem is that GOP voters’ desire for a political outsider intensified just as he was making this pivot.

Exit polls conducted by the national media’s National Election Pool asked voters in 24 different contests this year whether the next president should have experience in politics or be from outside the political establishment.

In the first four contests held in February, Republican voters were divided – 49 percent wanted an outsider while 45 percent favored someone with political experience.  The preference for an insider fell off in early March’s Super Tuesday primaries – with 49 percent still wanting an outsider but only 41 percent looking for political experience.

By the mid-March primaries, a 52 percent majority of GOP voters preferred a political outsider compared to 41 percent who still wanted an experienced politician.  The gap widened in the April contests, with nearly 6-in-10 Republicans (59 percent) wanting an outsider and just 37 percent favoring an insider.  In yesterday’s Indiana primary, the results for this question stood at 59 percent outsider and 35 percent insider.

The Cruz recasting gambit worked to the extent that he was ultimately seen as the establishment candidate – 68 percent of Indiana Republicans who want an insider voted for him.  In the very first contest of 2016 – the Iowa caucuses – Marco Rubio was actually the preferred candidate of voters wanting someone with political experience, even though Cruz was the overall winner on the night.

However, Donald Trump has been the favored choice of GOP voters looking for a true outsider since the very beginning of the primary season.  He won 46 percent of this group’s vote in Iowa, culminating with a 78 percent showing in Indiana three months later.

“In retrospect, Cruz's pivot to being the face of the establishment was a mistake.  Cruz ceded the outsider mantle to Trump at the very same time the Republican base's desire for an outsider grew,” said Patrick Murray, director of the independent Monmouth University Polling Institute.

Wednesday, March 2, 2016

Super Tuesday Polling: How Did Monmouth Do?

To say the 2016 primary season has been surprising would be an understatement.  A presumed Democratic nominee who was supposed to coast to victory has faced a tough challenge.  The GOP race is now coalescing around a front runner who practically everyone would have laughed off less than a year ago.

This topsy-turvy situation has amplified the already significant challenges that pollsters face when trying to take the pulse of voters in presidential primaries and caucuses.  Super Tuesday presented the first large-scale polling test of the nomination contest.  Interestingly, only a few national polling organizations devoted significant resources to polling these races (more on that below).

Monmouth University conducted polls in four of the nine primary states in the weeks leading up to Super Tuesday – Alabama, Oklahoma, Texas, and Virginia.  We had a significant miss in the Oklahoma Republican primary where, like every other poll in the state, we forecast a Donald Trump victory when Ted Cruz was the actual winner.  On the other hand, Monmouth was the only poll to accurately forecast a Bernie Sanders victory in the Democratic contest in that very same state.  Monmouth’s forecasts were accurate in the other three states we polled.

Monmouth’s Democratic primary polls were closest – or second closest in the case of Texas – to the actual margin of victory of all the polls taken in those four states over the past month.  Our Democratic primary results compared to the actual margin of victory:
   TX :  Vote - Clinton +32  /  Monmouth Poll - Clinton +34
   AL :  Vote - Clinton +59  /  Monmouth Poll - Clinton +48
   VA :  Vote - Clinton +29  /  Monmouth Poll - Clinton +27
   OK :  Vote - Sanders +10  /  Monmouth Poll - Sanders +5

On the GOP side, Monmouth came closest to the actual margin of victory in Texas and Alabama:
   TX :  Vote - Cruz +17  /  Monmouth Poll - Cruz +15
   AL:  Vote - Trump +22  /  Monmouth Poll - Trump +23

Monmouth had the correct winner in the Virginia GOP primary, but the margin was tighter than our poll:  Vote - Trump +3 / Monmouth Poll - Trump +14.  It’s worth noting that our Virginia poll was conducted a week prior to Super Tuesday, specifically before the crucial debate when Rubio decided to take on Trump.  That performance appeared to help Rubio’s performance in Virginia, although not enough to change the overall outcome.

As mentioned, we had the wrong winner in Oklahoma GOP primary:  Vote - Cruz +6 / Monmouth Poll - Trump +12.  Every poll in Oklahoma had Trump ahead by a significant margin.  Examining all the state exit polls from Super Tuesday shows that Oklahoma actually had the highest number of Republican voters (33%) who said they made up their mind in the last few days.  These voters broke significantly for Cruz (37%) and Rubio (38%) over Trump (15%).

Despite the stakes involved in Tuesday’s races, only four national polling organizations polled in at least three of the nine primary states during the month of February.  In addition to Monmouth, CBS/YouGov and NBC/Wall Street Journal/Marist polled in three states and PPP conducted Democratic-only primary polls in all nine states.

In Texas, the one state where all four organizations polled the Democratic primary, Monmouth’s 32 point margin for Hillary Clinton was the closest to the actual result.  CBS/YouGov had the margin at 24 points, PPP had 23 points, and NBC/Wall Street Journal/Marist had 21 points.  In the Texas Republican primary, Monmouth’s 15 point margin for Cruz was also closer to the actual result.  NBC/Wall Street Journal had a 13 point margin and CBS/YouGov had an 11 point margin.

This primary season has been particularly tough to poll, so it is understandable why more national polling operations did not dip their toes in the water.  Despite the one missed call, Monmouth is pleased with its Super Tuesday polling results.

Full information on Super Tuesday polling can be found here:
Republicans:  http://www.realclearpolitics.com/epolls/2016/president/march1GOP.html


Monday, February 8, 2016

After the Messaging, it's Time for the Ground Game.

The final Monmouth University Poll in New Hampshire found Donald Trump with a sizable lead over his competition in the Republican race, while Bernie Sanders held a ten point lead over Hillary Clinton in the Democratic contest.

The big question on the GOP side is who will come in second place, with at least four candidates realistically vying for the spot.  On the Democratic side, the question is whether Clinton can reduce her deficit to the single digits.

In the retail-heavy political environment of New Hampshire, it may all come down to the ground game – how many voters can each campaign personally contact.  Monmouth asked its poll respondents whether they had been contacted, and if so, on behalf of whom.

Interestingly, since registered independents can – and do – vote in either party’s primary, a significant number of likely voters in each contest say they were contacted by both Republican and Democratic campaigns.  The numbers below give a relative sense of how intense that outreach has been – and which candidates are excelling in their field operations.

Candidates are listed in rank order of total voter contacts, assuming about the same number of voters will turn out in each party’s primary.  The first number in parenthesis is the percentage of likely Republican voters who report being contacted by someone promoting that candidate.  The second number is the percentage of likely Democratic voters who say the same. 

New Hampshire Voter Contacts
1. Hillary Clinton (13 / 39)
2. Bernie Sanders (13 / 35)
3. Jeb Bush (31 / 8)
4. John Kasich (26 / 8)
5. Ted Cruz (22 / 10)
6. Marco Rubio (22 / 8)
7. Donald Trump (19 / 9)
8. Chris Christie (17 / 7)
9. Carly Fiorina (16 / 4)
10. Ben Carson (11 / 5)
11. Rand Paul (10 / 5)
12. Martin O'Malley (0 / 6)

Thursday, January 7, 2016

Testimony on Proposed Changes to New Jersey legislative Reapportionment Process

Testimony of
Patrick Murray
Monmouth University Polling Institute

New Jersey Senate
Committee on State Government, Wagering, Tourism & Historic Preservation
January 7, 2016

Re SCR188 (proposed constitutional changes to legislative apportionment process)

Mr. Chairman, Madame Vice-Chairwoman and members of the committee:

I am Patrick Murray, director of the Monmouth University Polling Institute.  Although I have been known, from time to time, to comment on the efficacy and responsiveness of the internal workings of state government in New Jersey, I rarely take a public position on a piece of legislation.  When I do it is largely because the process by which the legislation was drawn up does not adhere to principles of good government.

It is for this reason that I appear before you today to express my strong opposition to SCR 188.

This proposed Constitutional amendment will not achieve its stated aim of designing a fair legislative map with at least ten competitive districts.

Furthermore, the wording of the ballot question and interpretive statement seems to be deliberately designed to fool New Jersey voters into supporting a Constitutional change against their own interests.

First, let me say that I endorse the provisions that call for the immediate appointment of the public member or the Apportionment Commission and that codify a public access process while still giving the commission the flexibility for negotiating in private.  I also agree with the wisdom of granting legislative leaders the power to appoint commission members in return for barring current legislators from serving on the commission. 

However, I must object to the entire amendment because of the language in paragraphs 2.c. and especially 2.d.

Let me start with the “fair representation” provision in paragraph 2.c.  According to the Judiciary Committee statement, this provision utilizes the standards established by Dr. Donald Stokes, who served as the commission’s public member in both 1981 and 1991. 

On close examination, it does not.

Stokes’s fairness doctrine states that the number of seats a party holds in the legislature after each election should correlate to its share of the vote in that election.  For example, if the statewide vote splits 50-50 split between the Democratic and Republican candidates for office, then the share of legislative seats should be evenly divided.  However, according to projections Stokes included in his 1993 monograph, if a party wins 60 percent of the vote, it would be reasonable to see that party take as many as 75 percent of the seats. 

Fair enough, but Stokes’s fairness test must to be applied to the map as a whole after the fact.  You cannot a priori set aside 30 evenly-divided safe districts and then work on 10 so-called competitive districts and guarantee that you will come up with a fair correlation of seats to the statewide vote share.

More importantly, though, the crucial metric used by Stokes is whether seats in the legislature correlate to the total vote for the legislature.  Not how those seats correlate to the vote for a variety of unrelated offices such as President, Governor, or U.S. Senate as set forth in that paragraph.

This linkage is truly bizarre.  As we know, voters use a different set of criteria when evaluating who to support in elections for federal offices versus state offices as well as for executive positions versus legislative ones.

If gubernatorial elections told us what type of representation New Jersey voters want in their legislature, I would be directing my remarks today to the Republican Chairman of this committee.  By the same token, if presidential elections told us what New Jersey voters want in a legislature, there would be only one Republican sitting on this committee today.

However, even if Stokes’s fairness doctrine was applied correctly, it would still be unfair in practice.  In determining the legislative vote share of the two parties, Stokes did not employ a straight tally of the statewide vote, but used a district-based vote share average.  In other words, instead of using millions of data points – i.e. individual votes – to determine the New Jersey electorate’s intent, Stokes used only 40 data points – the two-party percentage margin in each district.

Stokes claimed that, due to widely varying voter registration and turnout rates in each district, this formula would be more representative of the will of all constituents – assuming that non-voters have the same preferences as those who actually showed up to vote.

This may be true in theory, but it is not supported by the data.  I examined election results from the past five legislative cycles – which is exactly what Stokes would do.  I found 19 instances where one party or the other did not field a full slate of candidates for either the Senate or the Assembly, which represents a not insignificant 6 percent of all races during that period.  Moreover, 14 of those 19 cases – or nearly three-quarters of these uncontested races – were instances where the Republican Party did not field a full slate.  That means that 14 of the data points used in the Stokes fairness test would produce a result at or near a 100 percent vote share for the Democrats compared to only 5 data points that would produce the same result for Republicans.

This would falsely skew the overall vote share result toward the Democrats, unless you actually believe that there were no minority party voters living in any of those 19 uncontested districts.

On the one hand, using non-legislative elections to determine the legislative maps fairness relies on a false metric.  But using the legislative election results as Stokes would have done would produce a skewed metric.

Even if the proposed formula did not face these problems, trying to codify this fairness doctrine in Constitutional language is akin to making the ghost of Donald Stokes the commission’s public member in perpetuity.  This is simply not something that should be written into the Constitution.

In fact, recent changes to Ohio’s legislative redistricting process which were approved by voters there last year, includes a fairness provision that provides sufficient leeway for the members of their commission.  It says simply that: “the statewide proportion of districts whose voters, based on […] election results during the last ten years, favor each political party shall correspond closely to the statewide preferences of the voters.”

While, the full provision does use what I believe to be a false metric by including non-legislative elections, the language is broad enough that it allows for each decennial commission to negotiate its meaning while incorporating emerging standards, such as the principle of “communities of interest” which has been largely ignored in New Jersey’s process.

More importantly, the Ohio standard also states quite clearly, and I quote, “[n]o general assembly district plan shall be drawn primarily to favor or disfavor a political party.”

And it is on this standard that SCR 188 fails miserably.

Because this resolution was introduced less than four weeks ago, I have not had the same opportunity to run vote simulations on potential outcomes, as I am sure its supporters have been doing for the past few years.  However, I have been crunching numbers in New Jersey for long enough to know when something smells fishy.

The process in paragraph 2.d. claims to create competitive districts, but actually entrenches a permanent Democratic majority by using a tortured definition of the word “competitive.”

In reality, competitive districts drawn using this provision in the 2021 process would almost certainly range from a smaller but definite Democratic advantage to an absolutely solid Democratic advantage.

While this outcome might be in line with the fairness doctrine, it defies any common sense meaning of the word “competitive.”

For most voters, the word “competitive” means that either party has a decent shot of winning the seat.  It does not mean that one party simply won’t lose as badly in a certain district as it will elsewhere in the state.

Over the past two decades, I have had the privilege of hearing the opinions of hundreds of thousands of New Jerseyans.  And I can say with certainty that our state’s residents want a truly competitive legislative map.  Indeed, you need to look no further than election returns which consistently show that competitive elections produce higher turnout.

So I am left to wonder why the drafters of this resolution would use the word “competitive” to describe an outcome that is not competitive according to voters’ vernacular?

I am left with only one conclusion.  This is a bald-faced attempt to pull the wool over voters’ eyes; making them complicit in a process that will only serve to increase their cynicism about politics.

Anyone reading the ballot question and interpretive statement about creating competitive districts would come away with a far different interpretation of what that means than what the proposed Constitutional language will actually produce.

I fully endorse revisiting how our Legislative Redistricting Commission operates.  But if a fuller process for public input is a good idea for the commission, then it should also be a good idea for the legislative process by which these constitutional changes are proposed.

Therefore, I urge you to table this resolution. 


Thank you.