Latest poll:
SHN POLL: PATRICK THE FRONT RUNNER, VOTERS AGREE WITH GOV ON NEEDLE SALES
By Craig Sandler
STATE HOUSE NEWS SERVICE
STATE HOUSE, BOSTON, JULY 10, 2006â¦..Deval Patrick has become the clear front runner for the Democratic nomination for governor, and Kerry Healey would lose to any of the three Democrats if the election were held today, according to the latest State House News Poll.
… Those who said they were likely to vote in the Democratic primary preferred Patrick by 35 percent to Gabrieliâs 22 percent and Reillyâs 19 percent. This was the first major statewide poll to show Reilly in third place.
(Edited by Charley for copyright reasons. Read the rest here.)
Please share widely!
But. 35 + 22 + 19 = 76. That’s a lot of undecided voters to go around. Still, very, very, very good news for Deval.
35/22/19 may reflect the ratios among decided voters, but don’t most undecideds just hang up the phone on an automated poll? Isn’t support for one candidate a huge incentive to stay on and answer it?
Was this automated? It doesn’t seem to say either way on the website.
State House News is not automated.
I guess I read “telephonic, random-digit dialing survey” as implying push-button. But even if it’s a live person, doesn’t this undecided # seem too low? Have 4 out of 5 primary voters already made up their minds?
Where in that do you get push-button? Almost all polls use random-digit dialing, and these days unless they’re conducted over the Internet, they’re by telephone.
<
p>
Anyway, what you’re missing is the “if the election were being held tomorrow” factor. Poll questions are designed to get people leaning one way or another to commit. I bet there are a lot of people who picked a candidate in this poll that are still up for grabs if another candidate wows them (or their candidate stumbles).
What I meant in the last comment was that I read the phrase incorrectly. And the “if the election was held tomorrow” point is enlightening too. Each of those numbers must contain a lot of weak support. That says to me that while Deval must be acknowledged as the front-runner, the real gap between the candidates isn’t as wide as the poll suggests. Same with the big lead over Healey #s, which surely can’t hold up.
the 21% number is high for likely primary voters only 2 months before the election, at least compared to other polls recently, but they might be getting more undecided voters as it’s a live call. i doubt any less than 80% could possibly be undecided at this point if they’re already planning to probably vote in the primary.
I thought Gabrieli had the momentum?!? Can’t you feel it taking over? That’s the word on the street.
<
p>Yeah, I know, I’m a petty, sarcastic bastard. 😉
Wait a minute, you heard it too? It must be his burgeoning grassroot effort trying to trick us on the polls.
<
p>
/snark off
<
p>
As a Patrick supporter, I’m not going to let myself get too excited by this news. He needs to keep at it and convince more and more voters of Massachusetts that his vision is the best one available for the Bay State.
have not yet found their way to our Attorney General in his eighth year in office. 57% (35 + 22) are with someone else. The primary is in a little over two months.
<
p>
“I know Eliot Spitzer. Tom, you’re no Eliot Spitzer.”
<
p>
This poll is good news for Patrick, but it’s really, really bad news for Reilly.
That the sample of Likely Democratic Primary Voters is only 193? I thought that you needed about 400 to produce something that is considered reliable.
I guess that accounts for the potential 14 point swing created by the +/- 7.
Makes for good overall numbers, pitting the dems against Kerry Healey, but 200 still isn’t a terrible source of numbers for the Dem primary. If you crosstab that with other segments, like African Americans or whatnot, then the margin of error grows a bit high, but at 7% it’s still a fairly reliable poll in terms of showing that Deval has a lead. Plus, it’s the second or third poll that’s showed Deval with about 35% of the primary voters, so I’d think 35% sounds about right.
<
p>
Still a very close race and anything can happen, but I can’t say I’m not at least a little happy (but realistic) with the results. It just illustrates that voters are continuing to listen to Deval.
But even given a 7% margin of error, Deval is still in the lead.
If it’s a 7 percent margin of error, and assuming Reilly doesn’t gain or lose in any error, Deval and Gabrieli are in a statistical dead-heat. Deval could be as low as 28% and Chris as high as 29%. Of course that also means that the error could benefit Deval the other way and be detrimental for the others as well.
<
p>
The point is this poll sounds good for Deval, fairly neutral for Gabrieli, and bad for Reilly, though it’s far from accurate (most pollsters and political analysts consider a poll to be accurate when the margin is within 4%).
<
p>
One thing’s for sure: Reilly is going to start punching harder, he’s got nothing to lose.
a strongly-worded letter from Phil Johnston and Mike Dukakis! Don’t forget about that! 😉
I bet Reilly’s really scared about a letter from a former governor and an ineffectual party hack.
To disagree with the above, this poll cannot be interpreted as neutral for Gab, bad for Reilly. A 3-point difference in a 7-point-plus-or-minus poll is not reliable. Now, this poll probably will hurt Reilly by making him look like the loser; but that’s a statistical injustice, in other words, Reilly got the short end of it.
<
p>
I don’t recall seeing a 7-point poll before, 3 and 5 are more common. Seems to me someone should have gotten a larger budget for their polling.
the margin of error puts Gabrieli within range of first place, whereas Reilly can only hope for second according to the +/- 7%. According to the poll, as with the last one done by SHNS, Gabrieli is “still in the race” so it’s a neutral indicator for his campaign even though he slipped some.
<
p>
Just the numbers talking…
The 7-percent thing isn’t a yes/no cut-off, where if two candidates are within 14%, then they’re tied, and if they’re outside it, then one’s clearly ahead. What it means is something like, after comparing this sample size to all of Mass, the odds that Gabrielli is ahead of Patrick in Mass is 4%, while the odds that Reilly is ahead of Patrick are 2%, and the “margin of error” cut-off was 3%. I’m just making those numbers up, but they’re pretty close (I think the “margin of error” refers to the 3rd standard deviation, or the 97th percentile–any statisticians out there?)
So all a Gabs supporter can really say is, “This is good for Gabs because his odds of really being in the lead are about one in twenty, whereas Reilly is in trouble because his odds are one in forty.” It’s not actually much of a distinction.
i think the margin of error is 2 standard deviations, cause 1 is 67% confidence, 2 is 95%, and 3 is 99%, when it comes to normal distributions. i’m more of a physicist than statician, but that’s what i remember it being.
<
p>
<a href="http://en.wikipedia.org/wiki/Margin_of_error"the wiki is good if a poll says 95% confidence, then it’s ~1.96 standard deviations and if it’s 99% it’s 2.58 st.dev. also explains how to calculate probabilities of leads
The usual poll samples I see in the news are from about 1000 people. And they usually quote an MoE of 3%. I’ve always thought the MoE was based on one sigma, i.e., sqrt(1000)/1000 = 0.032. It all made sense. Oh well.
State House News seems to tend to have smaller sample groups than other polls. But my guess is that the margin of error estimates are done in a 1/sqrt(N) way, and 400 is when you get to 1/20 deviation. but 193 is about half, and so only 1.414 higher.
<
p>
However, when the margin is this big, the one thing you can get out of this poll is that Patrick has an almost 95% chance of being ahead of the other two candidates.
I’m curious to see the SHN article–what’s this about voters agreeing with the governor on syringe sales?? If people have bought his demagoguery, that’s a bad thing for public health and HIV prevention all around. May also encourage Kerry Healey to continue her sleazy campaign tactics on such things.
<
p>
Some other interesting results: asked to rate the job being done by the state legislature, 1.5% said “excellent,” 22.2% said “above average,” 7.7% said “average,” 47.5% said “below average,” and 19.9% said “poor.” Only 1.1% didn’t know or refused to answer. So a total of 67.4% think the legislature is doing a lousy job. Something to bear in mind.
<
p>
All the details are here.
Thanks, David. While it’s clear that the question was phrased in a way that didn’t mention any of the reasons why the state would make needle sales illegal (no reference to public health at all, in fact), it’s sad that so many people don’t understand that point othis crucial measure. And it gives Romney and Healey space for demagoguery.
My God, here we go again! Democrats celebrating the fact that an on coming train is going to run them down in November!
<
p>
PEOPLE, yes…I’m ESPECIALLY shouting at the Deval folks. There is more bad new here than good.
<
p>
TAXES are GENERAL election voters top concern. Two thirds of the GENERAL election voters think it’s important or some what important that we have a Gov from a different party than the one controlling the Legislature.
<
p>
Why is that important, some may ask? BECAUSE IT IS HOW THE REPUBLICANS HAVE BEEN BEATING US FOR 16 YEARS IN THIS STATE!
<
p>
What part of this concept do people not get? I’m not saying retreat on these issues. I’m saying we need to confront reality and develop EMOTIONAL strategies to counter them (PLEASE, no diatribes or charts about how it’s the property tax or MA is 48th in per capita tax burden and other eye glazing points.)
<
p>
The way many people on this site dismiss taxes as an issue or the runaway legislature, makes me wonder how “reality based” this blog is.
Frank, you’re missing the big points here.
<
p>
First of all, taxes was picked by 19% (and 27% of Republicans) as the most important issue. The next on the list – education, healthcare, the economy, housing, jobs — are all good Democratic issues and together add up to WAY more than 19%.
<
p>
Even just looking at the unerolled responses on taxes vs. Education: if the unenrolled voters broke such that most taxes voters end up voting R and most education voters end up voting D, we win. By a lot actually.
<
p>
Also, If respondents were given second choices of issues, or allowed to rate the importance of EACH of a list of issues, we would see very different results. And polling of the general electorate on whether we should cut the tax rate to 5.0% show that they are pretty evenly split. The other issues, in total, matter much more.
<
p>
Second, relax about two-thirds saying that it is very or somewhat important that the Governor is from a different party from the legislative majority. The first goverment specific issue on the big list only had 3.4% naming it. It’s just not what people are thinking about when they vote. When other issues are tested in the same way — increasing funds for education, for example — their are FAR more people saying that they are VERY important. The party issue fails in comparison.
<
p>
Only 27% of unenrolled voters said the issue is “very” important. I am sure that there are about 20 issues that would have double that number if they were tested, but they weren’t. People who said it was somewhat important, I might add, were saying that is is NOT very important.
<
p>
Finally, even if these issue responses were as bad as you suggest, the election question is right there for you to see. The Dems are up from 5 to 10 percent against Healy depending on which Dem. Unenrolled voters are split evenly between Reilly (the worst in the general) and Healy — which again, would mean a big win for us. When, and only when, those numbers change, you can talk about your theories of why we keep losing.
<
p>
You are right that it doesn’t do us any good to try to educate the voters about tax rates and so forth. It’s too hard, and it means trying to fight on the Republicans’ turf.
<
p>
But your concern about where the voters are just doesn’t match what this poll says is the case.
FactCheck, I disagree with your premise that if people say they care about education, health care, jobs, etc…is generally good news for D’s. I’d say: things are no different than 2002.
<
p>
Anyone who lists taxes as a top issue wants lower taxes…is highly likely to vote R.
<
p>
Anyone who lists education, jobs, economy as a top issue, however, is less certain to be a D vote. I agree that people concerned about these issues tilt towards D’s, but perhaps 60-40.
<
p>
For example, someone who lists education as a top priority probably leans towards the old-school Dem “let’s spend more” approach – smaller class sizes, higher teacher pay, more enrichment programs, new buildings.
<
p>
However, most MA voters also support centrist Clinton-type reform – MCAS as graduation requirement to stop illiterates from getting diplomas; charter public schools particularly in inner-cities; merit pay for best teachers, etc.
<
p>
I don’t think this poll has much good or bad for D’s. Big Picture remains similar to 2002. Bad for Dems: expensive primary battle, voters want check/balance against Dem Legislature. Good for Dems: Mihos pulls 5% to 10% from Healey, and Healey is no Mitt.
Frank’s right. Healey’s poll numbers are crap right now because she hasn’t been campaigning much, though the sleeping GOP giant is starting to wake. Once she hits it full-steam with her “law & order” running mate, she’ll be taking the Republican message of doom and gloom to the people through clever and pricey campaign ads and high-profile public engagements. Then watch her poll numbers rise.
<
p>
This is a likely attack against the Dems as a party:
<
p>
1. The Dems spend out of control.
<
p>
2. The Dems won’t cut your taxes like you demanded and are going to raise them.
<
p>
3. The Dems are going to open our state to a flood of illegal immigrants.
<
p>
4. The Dems want to give needles to junkies and little kids.
<
p>
5. The Dems are going to destroy the sanctity of marriage by letting gays wed.
<
p>
Etc, etc, etc… You can’t trust them to have the Corner Office as well. Keep a Republican as your governor.—
<
p>
Again, this is how they’re going to hit us. As Al Franken put it, “smears, fears, and queers.” Massachusetts isn’t as liberal as people think it is, especially Central and Western MA. So we need to fight back. We know we have the facts right and we know that we’ll do better. But we need to listen to Frank’s advice and find ways of countering with our own “emotional” tactics.
<
p>
Off the top of my head, Republicans have closed your schools, they’ve cut your police and fire departments down, made you less safe, put your kids’ futures in danger, created a “giant sucking sound” (a la Perot) of jobs and residents fleeing from our state, created the pothole that popped your tire…
<
p>
It’s late and I’ve tried to inject a little humor to this discussion, but we can fight back standing firm on our core values but presenting them in a way that instills a sense of urgency to our cause and disgust towards the GOP.
Good rundown of what the repubs will hit us with… and how we will hit them back (hopefully) with language that can not only compete with Republicans, but beat them.
<
p>
I especially liked the way you phrase it. Rhetoric is so important. I really love it!
I sure wish the SH poll was clear about which type of tax the respondents were most concerned about. Personally, I think property taxes are out of control, while income taxes are very stable. A nice steady 5% or 5.35% or even 6% rate on my income I can deal with (after all, if I am paying more income tax, it means I am making more income!) However, not knowing what next year’s 2-1/2 override will do to my budget – ouch, I don’t like that uncertainty, so of course I am concerned!
If the margin of error is +/- 7, that does not mean you can add or subtract to any numbers you choose. A while back, Kevin Drum went over this.
A poll like this is a prediction of which candidate will have more votes. Patrick’s real numbers, if not 35%, are more likely to be 34% or 36% than 28% or 42%. Likewise, Gabrieli’s real support is more likely to be 21% or 23% than 15% or 29%. Think of Deval’s and Gab’s numbers as two mound-shaped (normal) distributions centered at 35 and 22, with a tiny amount of overlap. The real Deval numbers are somewhere in that mound around 35, and same for Gab’s. Most likely, Deval’s real numbers are higher than Gab’s.
If you want the chance that the leading candidate will win, look in the boxes on this chart.
If the margin of error were 5%, and Candidate A led Candidate B by 2%, there’s a 69% chance A’s support is actually greater.
<
p> If you go to the Drum link above, it has an Excel spreadsheet that you can plug the numbers into. I plugged in the numbers, and the probability that Deval will beat Gabrieli is 99.24%. There’s no “statistical dead heat” here. Deval has an indubitable lead!
<
p>
Thanks to yellowdogdem for sharing the good news.
A 6 for html wizardry!
Rasmussen release another poll on the MA governors race today as well. Unfortuantely the cross tabs are only available if you pay for them, but it seems to fall in line with what we have been seeing recently. The best news is that all three of our candidates are leading Heally by about 40%-25% (with about 15% to Mihos and the rest undecided). For more info go to the Rasmussen site
Thanks for that link, cephme. Interestingly, the Rasmussen poll has Patrick with the largest lead over Healey, and his favorability rating is also above every other candidate.