| | Advanced Search


Trinity Rep to Debut Great Writers Series for 2014-2015 Season—Trinity Rep will present the Great Writers Series…

newportFILM To Present Last Two Outdoor Screenings of the Summer—newportFILM will host their last two outdoor screenings…

Narragansett Bay Ranked #5 as Best for Boaters in US—Providence has ranked as the #5 best region…

Chef Walter’s Flavors + Knowledge: Red Peppers and Potatoes Frittata—End of summer combination with red peppers and…

Rhode Island Ranked Third Worst in Nation for Online College Students—Rhode Island has ranked as the third worst…

Arrivals & Departures Highlight Emotional Day At Gillette Stadium—The New England Patriots and New England Revolution…

Block Calls Out Raimondo For Failing To Support 2010 Tax Reform—Republican candidate for Governor Ken Block has blasted…

Fung Blasts Block for Government Contract Extension and Refusal to Release Tax Returns—Republican candidate for Governor Allan Fung has called…

Saul Kaplan: 3 Simple Words to Revolutionize the World—How many people end business meetings with an…

Coming to Rhythm and Roots - Charles Bradley—Charles Bradley is experiencing a late life awakening.


Brown Poll Becomes Political Football Among Candidates

Saturday, August 07, 2010


Some candidates for state office and Congress seized upon a new Brown University poll yesterday as evidence that their campaigns are catching on with voters while other candidates skewered it for being a poor measure of public opinion.

The poll showed Providence Mayor David Cicilline leading the Democratic primary for the Congressional First District with 32 percent support. Gubernatorial candidates Frank Caprio and Lincoln Chafee were nearly dead even, with the General Treasurer holding a razor-thin lead with 27.9 percent and the former U.S. Senator close on his heels with 26.5 percent.

Large Numbers of Voters Undecided

One the most notable results of the poll was not where candidates came out—but the sheer number of undecided voters in several of the races. In the race for governor and the First District a little over 30 percent of voters said they had not made up their minds. In the Democratic primary for Attorney General, 60 percent were undecided.

  "Rhode Island politicians have always had to compete for attention with the beach and summer fun, and, as the poll indicates, it is a losing battle,” said Joshua Starr, a native Rhode Islander, former Clinton advisor, and Washington, DC-based pollster.

“That is why a plurality of voters are undecided in the two most competitive races—for governor and the Democratic nomination for the First Congressional District,” Starr said. “Until late August, the hardest choice should be, 'Should I go to the beach or go sailing?  And, once everyone gets the sand out of their toes, the campaigns will ratchet up their advertising and there will be movement in the polls.”

The gubernatorial campaigns for Moderate Party candidate Ken Block and Republicans John Robitaille and Victor Moffitt were more focused on the 30 percent of voters who were undecided than the single-digit support they garnered in the polls.   

“This poll should be an embarrassment for both the Caprio and Chafee campaigns. They are well known career politicians and cannot capture enough votes to win in November,” Robitaille said. He said the undecided voters would gravitate toward him as soon as they learned more about his positions on lower taxes, cutting spending, and standing up to unions.

The campaign for Chafee, on the other hand, took the poll as good news, even though he was slightly behind Caprio. Campaign manager J.R. Pagliarini pointed out that the poll was taken during President Clinton’s visit to Rhode Island on behalf of Caprio—and after Patrick Lynch dropped out of the Democratic primary.

“We’re very pleased with the results,” Pagliarini said. “We don’t see the progress on the Treasurer’s front, so we’re very encouraged by it.”

But Caprio spokesman Nick Hemond said the Treasurer would rise in the polls as voters learned more about the differences between the two candidates. “Many voters remain undecided, and as they learn more about Senator Chafee’s plan to raise taxes and Frank’s plan to cut wasteful spending and get government out of the way and out of the pocket of small businesses, we are confident in earning their support,” Hemond said.

60 Percent of Voters Undecided on AG Race

Democratic candidate for Attorney General Joe Fernandez garnered only 7 percent in the poll—behind Steve Archambault, with 11.4 percent, and apparent frontrunner Peter Kilmartin, with 21.3 percent.

But he sensed opportunity in the 60.3 percent of voters who hadn’t made up their minds.

“If this poll tells us anything with regard to the Attorney General’s race it’s that the race is wide open given that 60 percent of those responding are undecided,” said Fernandez campaign manager Dan Herkert. “The other candidates in this race are entrenched politicians, and neither has been able to move into a commanding position. All of this is good news for a candidate like Joe Fernandez who’s never run for office but who has the resources necessary to communicate his message to voters.”

A spokesman for Archambault also thought the campaign was in a good position. “We are within striking range of Peter Kilmartin,” said Rob Horowitz, noting that he was receiving about one percentage point of support for about every year he had been in public office. “With six out of ten primary voters undecided and advertising yet to begin we are right where we want to be.”

Kilmartin spokesman Brett Broesder said the poll was further proof that his message of cracking down on violent crime, mortgage fraud, and public corruption was resonating with voters. He declined to comment on criticism of the poll, saying Kilmartin could better use his time meeting with voters. “We’ll let our opponents spend time complaining about their lackluster showing in this poll,” Broesder said.

Candidates Say Poll was Poorly Done

Several candidates also ripped into Brown for failing to properly measure voter opinion. All three of the Congressional candidates behind Cicilline in the poll said its sample size of 174 voters was too small and its 7.4 percent margin of error was too big.

“I am disappointed in Brown for releasing the sample size,” said Bill Fischer, a spokesman for Bill Lynch, who had 15 percent of voters supporting him. “There is not a campaign in Rhode Island or the country that would rely on a sample size that small.”

"We knew something was suspect immediately because the numbers are significantly at odds with our own internal tracking polling,” added Rachel Miller, the campaign manager for David Segal. “But we were shocked when it took only an elementary glance at the numbers to see the methodological flaws.”

Anthony Gemma also blasted the poll yesterday, saying it was “statistically invalid and politically meaningless.”

Starr said it was ultimately up to readers to determine how valid the poll results were, since Brown had been up front in its press release about the margin of error in the two Congressional races. “The determination as to whether the results are credible depends on how the reader wants to use the information,” Starr said. “From my perspective, the results provide me a picture with broad strokes, which is that Cicilline has a lead but there are a lot of undecided voters with more than a month left in the campaign.”

Frontrunner David Cicilline was alone among the Congressional candidates in praising the poll results. “While we’re pleased that today’s Brown University poll shows that David's message is resonating across the First Congressional District, this campaign is about listening to voters and their concerns,” said campaign manager Eric Hyers.

Marion Orr, the director of Brown’s Taubman Center for Public Policy, which did the poll, told GoLocalProv yesterday that the university stood by the results of the poll. He declined to respond to the specific criticisms of the candidates.

Besides the margin of error and small sample size, the Block and Robitaille campaigns took issue with the fact that the poll included both Republican candidates when they asked voters who they would support for governor. They suggested that that painted an unrealistic picture of the election, since only one Republican candidate would be in it after the primary.

“It would have been preferable to have two ballots test in the gubernatorial race to test the statewide preference for each Republican candidate,” Starr said. “However, having both Republican candidates in the horserace question should not change the key information, which is that a plurality of voters are undecided and it is a competitive race.”


Related Articles


Enjoy this post? Share it with others.


Brown stands by the reults. What a joke. They never, ever describe their methodology, and without that information the validity of the data can never be verified by an independent observer. In this case they actually outdid themselves with goof-ups. First, they polled a governor's race in November that can never happen with two Republicans among the choices. Fortunately one-third of the sample said they didn't know, the only rational choice for an election that can never occur. As for the primary data, of course the number of undecided voters is large--in fact, most of them won't turn out on election day in September. If one were to accept the Brown sample as valid the turnout in the Democrat primary would need to be twice the normal, extremely improbable, but typical when you rely on people who say they are likely to vote in the primary, rather than using voter history in your selection, information readily availble but obviously not used by Brown. Further, since Brown refuses to fully disclose its data with hard numbers and cross-tabulation data, there is absolutely no way of knowing how, or even if, the samples (or was there only one sample with the primary interviewees only subsamples of the statewide sample and weighting). It makes a major difference. The Blackstone Valley, for example, will produce about one-third of the general election vote, but will make up just under 40% of the congressional primary vote.

Comment #1 by Victor Profughi on 2010 08 07

Well, it's always the province of those trailing to question the validity of the polls. That Brown used a relatively sample size doesn't mean the results are statistically insignificant.

In fact, using the margin of error with a 95% confidence rate, this poll clearly states that we can reasonably say Cicilline is leading the First Congressional race and Caprio/Chafee are within 7 points of each other.

The candidates questioning the validity of the poll are just mirroring the argument of conservatives in fall 2008 that the polls were all wrong and John McCain was about to pull of a decisive election victory. They're basically asking us - "Who do you believe, me or your lying eyes?"

Comment #2 by Jesse Tuggle on 2010 08 07

I'm not a candidate and I have no horses in any of these races. But I am a professional pollster with more than 40 years experience polling here in the Rhode Island market. When someone produces work as sloppy as Brown's, I have a professional obligation to point out probable problems with mysterious, unsubstantiated methodology.

Let me make it crystal clear that I don't know the numbers in any of the races polled by Brown, but my contention is that if one relies upon Brown's data for analysis, he or she knows nothing as well.

The 95% confidence rate and the 7% margin of error only has validity IF the sample was actually RADOMLY drawn from a valid population or universe THAT HAS BEEN PROPERLY PROPORTIONED TO REFLECT VOTER TURNOUT IN EACH RACE--GENERAL AND PRIMARY.

Since Brown consistely fails to provide any information that would permit an independent researcher to evaluate a sample's accuracy their results are expected to be taken completely by faith.

I wonder how many times the Brown group has presented a professional paper at a peer reviewed conference with a similar lack of any varifiable documentation, and just asked their colleagues to trust them because they are, after all, Brown and in the Ivy League. Never would wash, nor should blind acceptance of their statements of confidence levels and margin of error.

Get real,folks!

Comment #3 by Victor Profughi on 2010 08 07


Marion Orr, in defending his methodology by stating, in essence, it is valid because I say so, presents the classic logical fallacy of the argument from false authority:

Argumentum ad Verecundiam.

As Bertrand Russell wrote, "It is not what the man of science believes that distinguishes him, but how and why he believes it. His beliefs ... are based on evidence, not on authority or intuition."

The Brown/Taubman Poll is indeed meaningless -- except as a cautionary tale.

But don't fret, Cicilline. Orr and Brown have a LONG way to go before they can challenge you in the false authority (lying) department.

As they say on the streets, "one guy lies and the other guy swears to it."

Or, if you prefer: David Cicilline and Marion Orr are chatting. Cicilline is going on and on, and suddenly Orr interrupts to say, "Wait a minute, you're lying."

And Ciciline responds, "Granted, but hear me out."

Comment #4 by R. Parsons on 2010 08 08

Parsons, enough with the vendetta, you're tall, ominous tales are ridiculous. I don't know what you have against him but this is not the place to make up stories. It is appreciated when people (i.e. Victor P.) question the validity of the poll, this is good to know and valuable information. You continue to post utter junk on these blogs and it degrades the conversation/debate that could be taking place. I invite anyone to look at any of Parsons other posts to see the filth and hate that comes out of his mouth (hint, any time there is an article mentioning the mayor, you will see a grimy trail of laughable and juvenile stories hanging to the coat-tails).

Comment #5 by Adda Nielson on 2010 08 08

Thanks, Adda, for your comments. With a 40 year record of polling in Rhode Island and having had three local TV stations, the projo, and a hundred or so clients from both political parties, with both conservative and liberal prspectives, Parsons, name calling has no credibililty, as they put it in the court room. And for the record, my truthfullness has been good enough for judges in both federal and state courts to accept my testimony as an expert witness.

Comment #6 by Victor Profughi on 2010 08 08

Write your comment...

You must be logged in to post comments.