Category Archives: polls

Did I Get It Wrong on the Blair Poll?

Normally, when people feel I got something wrong in a post, I hear from them very quickly. So I was a little surprised when Aaron Kraut, the David Blair’s Communications Director, contacted me to say that I got it wrong in yesterday’s post on the poll recently released by Blair.

According to Aaron Kraut, “The topline poll question cited in our press release was asked before any further questions.” This would mean that the polling results were not skewed by the sorts of priming and message testing that occurred during the poll.

I asked to see the polling results because that, after all, would quickly settle the matter. To me, as I said in the original post, it doesn’t make much sense to still be message testing at this point. Blair’s poll doesn’t jibe with a recent independent poll by Data for Progress.

The Blair campaign won’t share their polling data, feeling that they shouldn’t have to prove that something is false. They have a point and that’s why I am writing this post. The poll was done by a highly reputable polling firm. The Blair campaign is quite emphatic that the numbers they presented were the topline and not the post-message testing numbers.

As Hans Riemer’s campaign pointed out, the numbers presented by candidates often skew in their direction if only because candidates tend not to release unfavorable polls. Campaigns release information selectively, as the Blair campaign did, but everyone knows that.

More generally, doing good polling is getting more difficult. As is often said, the only poll that matters is the one at the ballot box.

Share

Blair’s Bad Poll

David Blair recently trumpeted poll results produced by his campaign that claim he trails incumbent County Executive Marc Elrich by only a single point. They oddly left Hans Riemer out of the graphic in the blast email and press release, which present Elrich at 29%, Blair at 28% with 23% undecided. The poll was conducted by a highly reputable pollster.

Councilmember Hans Riemer’s campaign has repurposed this poll that has him in third place with 20%. They claim the poll shows Elrich falling but discount the better numbers for Blair because it came from his campaign. Their graphic excludes Blair just like Blair’s leaves out Riemer.

Except that the poll really show Blair’s weakness.

The poll was taken only after voters were primed with a bunch of messaging questions. Voters were asked questions related to Blair’s endorsement by the Washington Post and the Sierra Club combined with standard messaging. Blair’s campaign also asked negative questions about Elrich and Riemer’s longevity in office combined with a positive spin for Blair.

Priming can have large effects on poll outcomes. Beyond heavily skewing the information presented to voters, people like to please and are more likely to give an answer if they think it will make the interviewer happy. Yet even after all this priming designed to drive Blair’s numbers up and Elrich and Riemer’s down, Blair still trailed Elrich.

This message-testing poll suggests a few conclusions quite opposite from those presented by Blair as well as Riemer to a lesser extent.

First, Elrich almost certainly has a lead and quite possibly a strong one. If the Blair campaign had polling results that were at all good for him without priming questions, they would show them to us and even share the details.

These results instead suggest that Blair’s campaign is stalling despite his millions in spending. Blair’s omnipresence on television may not matter much when fewer people see the advertisements because they stream or scroll past commercials on their DVR. I have literally seen one Blair ad while streaming a YouTube video.

Second, Riemer is running uncomfortably well from the Blair campaign’s perspective. Just as the poll depresses Elrich’s numbers, it does the same to Riemer. Dropping Riemer from their graphic was hardly accidental. Blair is trying to convince people that it is a two-person race with Riemer faring poorly.

This is the logical purpose of the poll as no campaign is message testing at this late date. Campaigns have already settled on their plan and focused on execution. Other recent polls suggest that Blair and Riemer are statistically tied. My view is that Riemer has been running the best campaign of the three candidates, which would help explain why he hasn’t fallen behind Blair despite expectations and Blair’s very large wallet.

It doesn’t hurt that there are hundreds of thousands of dollars in expenditures funded by California donors on Riemer’s behalf separate from the campaign. (UPDATE: This is an anti-Elrich group that helps both Riemer and Blair.) Unlike four years ago, Riemer is the only councilmember challenging Elrich. Of course, Riemer’s campaign can’t have it both ways—the numbers understate Elrich’s support as well as his own.

Rather than convincing me that Blair is coming on strong and positioned well in the final weeks, this poll confirms his weakness.

Share

Marylanders Strongly Support Legalizing Marijuana

By Adam Pagnucco.

Part Two of Goucher College’s latest poll of Marylanders is out and one of the questions it asks is this: “Do you [support or oppose] making marijuana legal for recreational use in Maryland?” Here are the results on this question from the poll’s crosstabs. The margin of error is 3.8% for all voters and is higher for subsamples.

All adults

Oppose: 28%
Support: 67%
Sample size: 725

Women

Oppose: 31%
Support: 63%
Sample size: 383

Men

Oppose: 26%
Support: 71%
Sample size: 340

Registered voters

Oppose: 29%
Support: 66%
Sample size: 654

Democrats

Oppose: 18%
Support: 77%
Sample size: 352

Republicans

Oppose: 47%
Support: 50%
Sample size: 171

Unaffiliated

Oppose: 34%
Support: 60%
Sample size: 115

Age 18-34

Oppose: 16%
Support: 79%
Sample size: 219

Age 35-54

Oppose: 25%
Support: 74%
Sample size: 249

Age 55+

Oppose: 42%
Support: 50%
Sample size: 257

Black

Oppose: 21%
Support: 74%
Sample size: 217

White

Oppose: 33%
Support: 63%
Sample size: 442

Other races

Oppose: 22%
Support: 68%
Sample size: 65

No college degree

Oppose: 25%
Support: 70%
Sample size: 443

College degree

Oppose: 33%
Support: 62%
Sample size: 283

Montgomery/Prince George’s

Oppose: 24%
Support: 73%
Sample size: 232

Central/Baltimore Metro

Oppose: 26%
Support: 67%
Sample size: 333

Outside urban corridor

Oppose: 40%
Support: 57%
Sample size: 160

Conservative

Oppose: 42%
Support: 54%
Sample size: 203

Moderate

Oppose: 30%
Support: 65%
Sample size: 315

Progressive

Oppose: 11%
Support: 84%
Sample size: 180

Children in household

Oppose: 23%
Support: 73%
Sample size: 259

No children in household

Oppose: 31%
Support: 63%
Sample size: 362

Share

Crosstabs: Goucher College Poll on COVID

By Adam Pagnucco.

Part One of Goucher College’s latest poll of Marylanders is out and it asks several questions related to the COVID pandemic. The accompanying crosstabs spreadsheet is also available. Here are the results for all voters along with two crosstabs of interest to Seventh State readers: statewide Democrats and residents of Montgomery and Prince George’s counties. (Because of the size and partisan nature of those two counties, there will be some overlap.) The margin of error is 3.8% for all voters and higher for subsamples.

Question: Do you [approve or disapprove] of how Governor Larry Hogan has handled the outbreak of the coronavirus in Maryland?

Answer from all voters

Disapprove: 21%
Approve: 77%
Sample size: 654

Answer from Democrats

Disapprove: 17%
Approve: 81%
Sample size: 351

Answer from MoCo/Prince George’s

Disapprove: 20%
Approve: 76%
Sample size: 232

Note: 32% of Republicans disapprove and 65% approve, so Hogan’s ratings on this question are better with Democrats than Republicans.

Question: Do you think Maryland is moving [too quickly or too slowly] to ease restrictions and reopen businesses, or has the pace been about right?

Answer from all voters

Too slowly: 25%
About right: 57%
Too quickly: 16%
Sample size: 654

Answer from Democrats

Too slowly: 13%
About right: 66%
Too quickly: 20%
Sample size: 351

Answer from MoCo/Prince George’s

Too slowly: 18%
About right: 64%
Too quickly: 16%
Sample size: 232

Question: Are public schools in your area moving [too quickly or too slowly] to reopen for in-person classroom instruction, or has the pace been about right?

Answer from all voters

Too slowly: 29%
About right: 36%
Too quickly: 31%
Sample size: 654

Answer from Democrats

Too slowly: 11%
About right: 41%
Too quickly: 43%
Sample size: 351

Answer from MoCo/Prince George’s

Too slowly: 19%
About right: 39%
Too quickly: 36%
Sample size: 232

Note: 65% of Republicans say the schools are reopening too slowly so there is a huge partisan difference on this question.

Question: Please tell me if you’ve experienced these feelings [more often, less often or about the same] as you did before the outbreak of the coronavirus last year… stressed.

Answer from all voters

Less often: 13%
About the same: 38%
More often: 49%
Sample size: 654

Answer from Democrats

Less often: 12%
About the same: 35%
More often: 53%
Sample size: 351

Answer from MoCo/Prince George’s

Less often: 15%
About the same: 30%
More often: 55%
Sample size: 232

Question: Please tell me if you’ve experienced these feelings [more often, less often or about the same] as you did before the outbreak of the coronavirus last year… frustrated.

Answer from all voters

Less often: 10%
About the same: 34%
More often: 55%
Sample size: 654

Answer from Democrats

Less often: 11%
About the same: 34%
More often: 55%
Sample size: 352

Answer from MoCo/Prince George’s

Less often: 10%
About the same: 34%
More often: 56%
Sample size: 233

Question: Please tell me if you’ve experienced these feelings [more often, less often or about the same] as you did before the outbreak of the coronavirus last year… angry.

Answer from all voters

Less often: 16%
About the same: 49%
More often: 34%
Sample size: 654

Answer from Democrats

Less often: 16%
About the same: 47%
More often: 36%
Sample size: 352

Answer from MoCo/Prince George’s

Less often: 15%
About the same: 49%
More often: 37%
Sample size: 232

Question: Please tell me if you’ve experienced these feelings [more often, less often or about the same] as you did before the outbreak of the coronavirus last year… sad.

Answer from all voters

Less often: 13%
About the same: 41%
More often: 45%
Sample size: 654

Answer from Democrats

Less often: 13%
About the same: 34%
More often: 53%
Sample size: 352

Answer from MoCo/Prince George’s

Less often: 13%
About the same: 36%
More often: 51%
Sample size: 233

Question: How concerned are you—[very, somewhat, a little, or not at all]—about yourself personally or a close family member getting the coronavirus?

Answer from all voters

Little/not at all: 28%
Very/somewhat: 71%
Sample size: 654

Answer from Democrats

Little/not at all: 17%
Very/somewhat: 82%
Sample size: 351

Answer from MoCo/Prince George’s

Little/not at all: 24%
Very/somewhat: 76%
Sample size: 232

Note: Among Republicans, 49% are a little or not at all concerned and 50% are very or somewhat concerned. Democrats are much more concerned about family members getting COVID than Republicans.

Question: When the Covid-19 vaccine is available to you, do you think you’ll get it as soon as you can, wait and see how it’s working before you get it, only get it if required, or are you definitely not going to get the Covid-19 vaccine?

Answer from all voters

Already received at least one dose / will get it as soon as they can: 67%
Wait and see how it’s working: 14%
Only get it if required / Will not get the vaccine: 18%
Sample size: 654

Answer from Democrats

Already received at least one dose / will get it as soon as they can: 71%
Wait and see how it’s working: 15%
Only get it if required / Will not get the vaccine: 13%
Sample size: 352

Answer from MoCo/Prince George’s

Already received at least one dose / will get it as soon as they can: 69%
Wait and see how it’s working: 18%
Only get it if required / Will not get the vaccine: 11%
Sample size: 232

Question: Overall, do you think the Maryland state government is doing a(n) [excellent, good, fair, or poor] job distributing the COVID-19 vaccine?

Answer from all voters

Poor / fair: 64%
Excellent / good: 33%
Sample size: 654

Answer from Democrats

Poor / fair: 62%
Excellent / good: 36%
Sample size: 351

Answer from MoCo/Prince George’s

Poor / fair: 72%
Excellent / good: 27%
Sample size: 232

Question: When do you think the country will get the outbreak under control and be able to return to normal: in the next month or two, by the summer, before the end of the year, later than this year, or never?

Answer from all voters

Next month or two / by the summer: 17%
Before the end of the year: 39%
Later than this year / never: 40%
Sample size: 654

Answer from Democrats

Next month or two / by the summer: 11%
Before the end of the year: 43%
Later than this year / never: 45%
Sample size: 351

Answer from MoCo/Prince George’s

Next month or two / by the summer: 11%
Before the end of the year: 44%
Later than this year / never: 43%
Sample size: 232

Share

Leventhal Poll Shows Wide Open Executive Race

By Adam Pagnucco.

Council Member George Leventhal, who is running for County Executive, has released an internal poll from Celinda Lake’s firm showing a wide open race.  The poll surveyed 400 likely voters with a margin of error of 4.9 points.  We reprint the three-page polling memo below and have a few comments following.

 

So what can be gleaned from this?

First, voters aren’t paying attention yet.  No candidate draws more than 11% support in the first run of questions and undecided gets 58%.

Second, the margin of error (4.9 points) matters.  In the first run, that means there is no statistically significant difference between the six candidates.  For example, Bill Frick polls at 2% but, with the margin of error, could be at 6.9%.  Marc Elrich polls at 11% but, with the margin of error, could be at 6.1%.  In effect, no one is leading.

Third, even with the “engaged communications” part of the poll, Leventhal is still within the margin of error against Elrich.  As for the “final ballot” that shows Leventhal blowing out the rest of the field, it appears to depend on giving respondents additional information on Leventhal and not on the other candidates.  It “simulates Leventhal having a resources and communications advantage” that does not exist.  At the moment, he will be fifth in cash on hand once the latest public matching funds are disbursed.

Could Leventhal win?  Sure, it’s possible.  He is a four-term incumbent, a former chair of the county Democratic Party, has a geographically diverse base of small contributions and is working as hard as any candidate in the county.

But the simulations showing Leventhal leading the field right now reek of spin.  Voters aren’t fully engaged yet and no one other than David Blair has started mailing regularly and going on television.  We believe the poll’s real message, which is that no one has wrapped up this election quite yet.

Disclosure: the author is a publicly-listed supporter of Roger Berliner.

Share

Candidates Ask Supporters to Stuff Online Poll

By Adam Pagnucco.

Back in the days of Maryland Politics Watch, we would occasionally set up online polls for various races.  The polls had no validity, of course.  They could not screen for registered voters, much less those in a particular party or county, and even those that limited each IP address to one vote could be easily gamed through multiple devices and masking software.

But we did it anyway for two reasons.  First, we were greatly entertained by the thought of frenzied candidates snapping the whip over armies of interns and yelling, “Vote faster!  You’re not pushing the vote button fast enough!!”  And second, they got lots of eyeballs, or at least apparent eyeballs.  Your author once scrutinized the traffic coming into one of our better performing online polls and discovered that most of it was coming from a handful of IP addresses.

Eventually, we stopped.  The “polls” added no value to the readers’ understanding of the elections.  And they also turned into a huge waste of time for candidates.  After we posted yet another online poll in 2010, one candidate emailed and said, “I really hate these things.  I have to drop everything and start voting!”  We took that comment to heart.  From that point on, your author determined that it was only worth doing a poll if we could invest it with some kind of methodological validity.  That’s easier said than done with an online poll!

That has not stopped others.  On Wednesday, Bethesda Magazine did an online poll on the Council At-Large race, which it admitted was “not scientific,” with SurveyMaker.  The poll began making the rounds on Facebook and one political insider sent it to your author, breathlessly panting, “Unscientific, but very surprising!”  Then a complete unknown, Steve Solomon, took the lead spot and folks started to understand just how unscientific this poll was.

Solomon’s “win” was not an accident.  He is a sports radio host and he encouraged his listeners to vote for him on both radio and Twitter.

Solomon was not alone.  Neil Greenberger sent out a blast email asking his supporters to stuff the poll.  He told his list that while the poll was unscientific, “It is better to be vaulting in this poll than to be lingering.”  He even said, “You don’t have to be a registered voter or live in Montgomery County vote in this poll. Just let them know who you would like to see come out ahead in the June 26 Democratic primary.”

Now look.  We do not absolutely deplore all online polls.  They can be fun and buzzy, and if folks want to push buttons for kicks, that’s fine.  But it’s absolute cross-eyed tomfoolery to see them as containing any merit.  Candidates, listen up.  If you spend your time pumping worthless ca-ca like this instead of phone banking, door-knocking and raising money, your chances of winning will be about as high as the coyote’s chances of catching the road runner.  Now get back to work!

Share

Poll Launched in CD6

By Adam Pagnucco.

A poll has been commissioned in Congressional District 6.  Congressman John Delaney currently represents the district, but he is considering a run for Governor and many potential candidates are mulling a run for what would be his open seat.  The pollster called Merry Eisner Heidorn, a former General Assembly staffer and school board candidate, and she kindly provided details of the call.

The call was made by a live caller and lasted twenty minutes.  The first three minutes collected demographic information, including age, gender, zip code, county and party.  This was followed by questions on voting tendency, including whether the respondent understood what primaries were, voted for candidates or only on party label, had voted in gubernatorial as well as presidential elections, intended to vote in the 2018 primary and had voted in past primaries.  Then the caller asked about the respondent’s opinions on Donald Trump, Larry Hogan, the economy and other issues.

Next, the caller asked, “So if John Delaney runs for Governor, would you support his run for Governor?”  This was followed by five to seven minutes of favorability questions on three potential candidates to succeed him – Total Wine co-owner David Trone, State Senator Roger Manno (D-19) and Delegate Bill Frick (D-16).  The caller then zeroed in on Trone, asking about a series of issues pertaining to him and then asking how each impacted the respondent’s favorability towards Trone and the likelihood to vote for him.  The specific issues raised about Trone included the fact that he had never held office, had contributed money on behalf of his business to politicians of both parties, had run for office before and was a “successful businessman from Potomac.”  At the conclusion of the call, the pollster asked, “Now that we have talked about David Trone, has your desire to vote for him changed?”

This is a fairly standard bio- and message-testing poll.  The pollster is attempting to gauge support for a possible run in CD6 both across the entire sample and among a number of key sub-groups.  Trone is known to be considering a run in CD6 and has polled previously on the Montgomery County Executive race.  This poll along with Trone’s establishment of campaign office space will fuel further speculation on what race, if any, he will enter.  The entire Montgomery County political class is watching.

Share

Washington Post Poll Shows Hogan Vulnerability

By Adam Pagnucco.

Governor Larry Hogan loves to discuss his high approval ratings in polls, which have usually been in the range of 60-70%.  But a new Washington Post poll that examines his reelection prospects shows that they are well below his approval numbers and provides hope to Maryland Democrats.

The Post poll of March 16-19 has sample sizes of 914 adults and 841 registered voters.  The margin of error for those two groups is 4 points, growing to 5.5 points for a half-sample and 6.5 points for the 317 respondents who live in Maryland’s D.C. suburbs.  These margins of error must be kept in mind when reading the poll –  effectively, only large gaps are meaningful for small sub-groups.

With that significant caveat in mind, let’s examine data on Hogan’s reelection prospects.  The Post asked respondents the following question: “Thinking about Maryland’s Governor’s race in 2018… if Larry Hogan ran for re-election as governor, do you think you would vote for him OR for the candidate nominated by the Democratic Party?”  Among adults, 39% said they would vote for Hogan and 36% said they would vote for the Democratic nominee, an advantage of 3 points for the Governor.  Among registered voters, 41% said they would vote for Hogan and 37% said they would vote for the Democrat, a margin of plus 4.  So far, this looks very much like Hogan’s 4-point victory in 2014.

But the sub-group results are more interesting.  We compiled the Post’s sub-group data on this question in the presentation below.

Let’s recall the margin of error estimates above.  Margins of 10-15 points or less for small sub-groups are probably not very meaningful.  That said, many of the Governor’s strengths are predictable.  He does well with Republicans, Conservatives, Whites and rural residents.  He is weak among Democrats, liberals, African Americans and Prince George’s residents.  One item that stands out is his strength with seniors, with whom he has a 17-point advantage.  Seniors are among the most reliable voters in any election.

Now let’s compare the geographic results of this poll with how the Governor actually performed in 2014.

The Governor appears stronger in the poll in Baltimore and the Washington suburbs, but weaker elsewhere than in 2014.  This could be statistical noise due to large margins of error.  But it could also be the result of tax fatigue in some Democratic strongholds, like Montgomery (where voters recently passed term limits by 40 points) and Prince George’s (where the County Executive proposed a 15% increase in property taxes two years ago).  It’s hard to believe that the Governor is actually weaker in Anne Arundel and Howard, both of which have Republican Executives who are strongly favored for reelection.  (And a random question: what pollster combines Baltimore City and County in one estimate?  C’Mon, Man!)

The big takeaway from the poll is this: Larry Hogan will not be coasting to reelection.  Maryland is simply not wired that way.  It has too many Democrats, African Americans, liberals, immigrants and people who are either employed by or do business with government at some level to give any GOP statewide incumbent a blowout win.  From a purely political perspective, the Governor deserves credit for his focused message of tax cuts, job growth and reform (like redistricting) while trying his best to avoid distractions from the right, the left and Washington D.C.  His approach gives him a path to victory in a rather blue state.  But if the Democrats begin preparing now, play smart and field a good candidate for Governor, Larry Hogan can be defeated.

Share

Clinton Campaign Hires Latino Decisions for Polling

This is a smart decision by the Clinton campaign. Latino Decisions is the gold standard for polling Latinos, whose support will be critical if she hopes to secure the Democratic nomination and the Presidency.

The following is by Stephen A. Nuño:

The Hillary Clinton presidential campaign has hired Latino-owned firm Latino Decisions to join the campaign’s polling team, sources told NBC News Latino Wednesday.

The hiring of Latino Decisions brings aboard significant knowledge to the campaign about Latino voters that will continue to build up the outreach efforts of the Clinton campaign into the Latino community.

The Clinton campaign did not immediately respond to emails and a call about the hiring.

The Clinton campaign has gone on an unprecedented hiring spree of Latino outreach specialists, such as Amanda Renteria, their National Political Director, and Lorella Praeli, the campaign’s Latino Outreach Director.

The firm’s co-Principles, Matt Barreto and Gary Segura, bring with them a significant amount of academic heft to their polling operation. Barreto is a professor of political science and Chicano Studies at UCLA and Segura is a professor of political science at Stanford University.

Michael Jones-Correa, a professor of political science at Cornell, welcomed the news about Latino Decisions as a step forward for the Clinton team.

“LD provides essential policy relevant public opinion on Latinos in the US today, reflecting the very highest standards and best practices in the public opinion field,” said Jones-Correa.

The addition of a Latino-owned firm to the campaign comes after findings in a study by PowerPAC+ reported by NBC News which highlighted the significantly small number of minority firms hired by Democrats to work on their campaigns as paid consultants. Even though the vast majority of Latinos and African Americans vote for Democrats, about 98 percent of of the $514 million spent by the three national Democratic Party committees in 2012 went to non-minority consultants.

Much of the advocacy work by Latino Decisions has been with groups with deep ties within the Latino community, such as America’s Voice and National Council of La Raza.

Frank Sharry, founder and executive director of America’s Voice, has praised the work of Barreto and Segura.

“Latino Decisions has developed a methodologically-sound, data-driven approach to polling Latino voters. Given that the road to the White House cuts through the Latino community, the Clinton campaign is fortunate – and smart – to have them on board,” said Sharry.

Share