Thursday, August 03, 2006

Q-Poll: If Your Enemy is Drowning, Throw Him Some Joementum

The new Q-Poll is out:

Momentum for Ned Lamont, the anti-war Connecticut U.S. Senate candidate, increases as he rolls to a 54 - 41 percent lead over incumbent Sen. Joseph Lieberman among likely Democratic primary voters, according to a Quinnipiac University poll released today.
...
In this latest survey, 5 percent of likely Democratic primary voters remain undecided, but 85 percent of voters say their mind is made up.
...
"Three months ago, Lamont was virtually unheard of, except perhaps on the blogs. As Democrats get to know Lamont better, they like what they see. Lamont has established himself as a credible alternative to Lieberman," Dr. Schwartz added.

Connecticut likely Democratic primary voters give Lamont a 46 - 14 percent favorability rating, with 20 percent mixed and 19 percent who say they don't know enough to form an opinion.


Keep in mind that likely Democratic voters are hard to identify. Also, none of these polls are taking the U to Ds into account at all. At last count there more than 10,000 U to Ds (and that number will rise until August 7) who are not being polled at all and, based on the fact that they changed their registration to vote in the primary, are highly likely to actually vote on August 8.

15 comments:

Anonymous said...

And Dan Malloy is closing the gap! It's pretty clear which race is going to keep folks up late next Tuesday.

Anonymous said...

Are you sure about U->D? I think Quinnipiac's methodology is random-dial followed by questions to determine likeliness of voting. Someone who recently registered specifically to vote in this primary would probably be considered a very likely voter.

85% committed in a primary is amazing, though, and Lamont's support is even harder.

Plus Lamont's bump comes almost entirely from a dramatic shift in his support from women (Joe had the lead among women 2 weeks ago, now down by 10). I'd be interested in what people think caused the shift.

Lamont's favorables went up 10 points, unfavorables only +1. Lieberman's attacks aren't having any traction. Though Lieb's favorables are up 2, so all those mailers aren't for naught.

I'm also interested in seeing what happens to Lieberman's support in the under $30K demographic with Jackson and Sharpton coming to town - that's Joe's only leading demo right now.

Anonymous said...

The Lieberman campaign is rivalling the Big Dig for the ineffective utilization of limitless resources

It is also clear that depsite not being on Colin McEnroe holiday card list John Droney knows a lot more about CT politics than his critics

Anonymous said...

So I'm walking into work this morning at 90 State House Square in Hartford. I take the escalator up to get some coffee at the food court and the office at the top of the escalator that has been empty since Travelers moved out of it 2 years ago has some evidence of activity. If you look between the panes of smoked glass, you can see that the room is full of tables and on those tables are phones...like 50 of them! Now, knowing that Lieberman's Campaign address is a PO Box at the State House Square post office, logic would dictate that this is a phone bank for Joe...bring on the Kiss diarama!

Gabe said...

Are you sure about U->D? I think Quinnipiac's methodology is random-dial followed by questions to determine likeliness of voting. Someone who recently registered specifically to vote in this primary would probably be considered a very likely voter.

Am I 100% sure? No, but I believe that Quinnipiac is random-dialing registered Democrats and I also think that their list of Ds is not updated daily and is thus not capturing the U-Ds..

Does that sound plausible?

Anonymous said...

It's plausible, but ordinarily drilling through a registered-voter list is considered a sin in polling methodology. Those who give their phone numbers on the registration forms are self-selecting to a degree that could influence the numbers -- for example, older voters, or those oriented towards privacy concerns or leaning libertarian, will leave the numbers off, and skew the poll. Matching names to white-page lookups (easier when it's for a candidate and not affected by the do-not-call list) skews the poll against low income, since younger and poorer voters change numbers / addresses more frequently.

The one thing that gives me pause is that today's q-poll has no general election questions whatsoever. I don't know if they hung up on indies/repubs, or just left that data off (since it's a 2-week and not a 4-week poll).

Either way, selecting a valid and truly random body to poll from is the biggest challenge in conducting a poll.

David said...

Trying to ID likely primary voters is more of an art than a science. Having said that, the Lamont/Lieberman race seems over with only 5% undecided and 85% saying their mind is made up on their choice - it will be difficult for Lieberman to buck these numbers and the trend lines. On the Governor's race, with 13% undecided and 36% who may change their vote, the outcome is much more up in the air. I think Malloy's campaign has built up a lot of momentum in the remaining days of the campaign - the Hartford Courant, NY Times and the endorsement here (the last two will help with Lamont voters who are undecided on Guv), he has outmailed DeStefano significantly among voters and the TV war now seems a draw in terms of air time. There seems to be energy in his campaign. However, Malloy may be getting too negative and that could backfire on him with undecided voters if he doesn't balance it with more positive images. Aside from union endorsements, DeStefano's campaign seems to lack energy and enthusiam. He may be able to coast it out and win or there is still enough time to reenergize before Tuesday.

Anonymous said...

Here is the interesting point in the Q Poll regarding the Governor's race: the numbers haven't changed since the June poll.

On June 8 the Q poll had the race 46% JDS to 35% DM with 18% undecided.

The August 3 poll has the race 48% JDS to 38% DM with 13% undecided.

Here is the DM spin: "Malloy cuts into big JDS lead, momentum is on his side."

Here is the JDS spin: "Months into the race and his lead has changed."

Here is the reality: this race is up for grabs. Given the attention given to the Senate primary, the numbers for the Governor's race are understandably "soft" for both candidates.

Primaries usually are won by the ground game and GOTV strategies, not mass media buys. But is that really the case in this race? Or will the outcome be determined by the Lamont/ Lieberman battle? That is the real question in the Governor's race. And, in hindsight, if either candidate had had the forsight to endorse Lamont, this race would be over too.

Anonymous said...

david-

(the last two will help with Lamont voters who are undecided on Guv)

You know what would help with Lamont voters? An endorsement. Otherwise expect a lot of blank votes on the Governor's column. (In the 2000 general, over 10% of those who voted for president left the Senate race blank.)

David said...

No argument that an endorsement would be much better, but flipping his support now to Lamont would be transparently opportunistic and would offend both Lieberman and Lamont voters at this point. All he can do now is emphasize his opposition to the war and highlight his supporters who are also with Lamont - like me. I was always concerned about which line would be better to be on for Malloy because turnout in this election is coming from the U.S. Senate race and Lamont people may just stay on the B line even if they don't know much about DeStefano. Having Lamont people leave the polling place without voting for DeStefano is still a partial victory for Malloy.

Anonymous said...

The Q poll didn't call only reg. Democrats. I'm an unafiliated voter and they polled me. Not so scientific.

Anonymous said...

David said "Trying to ID likely primary voters is more of an art than a science."

Not really.. its almost like you think you know how to do it and no one else does. There have been plenty of democrat primaries in this state in the last 10 years. I'm sure both campaigns have all the check offs for every one of those primaries and have purged deceased or out of staters and have identified their voters. And a good voter file and vote history will get the job done. Volunteers, phones and a good gotv effort (which is most important) is all you need for a primary.

Anonymous said...

Dave - I'd vote Lamont / DeStefano, though I'm at school and have been voting in California, and after thinking it over, thought it'd be unethical to vote in the CT primary too. That said, I don't think it'd upset Lamont supporters, but it might be more than offset by LieberVoter disgust.

ctrevolution - I think that quinnipiac polls run 4-5 days, and end on Sunday or Monday. So it probably captures part of the "endorsement bump."

anon 1019 - they call many more, usually, and determine who's likely to vote based on a set of questions on voting history and interest in the current race. See here to read about likely-voter models.

I'm curious though, 1019, if they asked you about general election preferences - who you'd vote for in a Lieberman / Lamont / Scheslinger race, or a [Dem TBD] / Rell race? I want to know if there are more numbers in the waiting back there :)

Anonymous said...

Thirteen survey organizations ( ... Quinnipiac, etc ... ) ask vote questions of all registered voters and later apply screen questions to select likely voters. These organizations often report results for both registered and likely voters.

Most of the others (Battleground, Democracy Corps, ICR, Insider Advantage, Rasmussen, SurveyUSA, and TIPP) typically define likely voters with screen questions and ask the vote question only of those who pass the screen. The Fox/Opinion Dynamics has alternated between screens for registered and likely voters.

Gabe said...

Matt, Anon, David, et al. -

I called teh Q-poll and this is what they told me:

They are calling a random selection of all citizens and having people self-identify their party affiliation. The $6MM question is how the u-ds self-identified. Your guess is as good as mine...