Perils of Polling in Connecticut

As the political world awaits the last Quinnipiac Poll before Connecticut's Democratic primary on Tuesday, University of Connecticut pollster Chris Barnes sees familiar and dark portents for Senator Joseph Lieberman. Recent polls have shown a close race between Lieberman and surging antiwar challenger Ned Lamont. Barnes, who is not polling the race, doesn't believe it. He's thought since May, when Lamont captured an impressive one-third of the votes at the Democratic state convention, that Lamont would win a comfortable August victory.

Science and experience inform Barnes. He maintains that Democratic primary polls exaggerate the number of moderate Democrats who will vote and understates the liberal turnout. Primaries are for the wings of each party, not the bodies. In order to find what political pollsters view as likely primary voters, they take what they can get.

Political polls, Barnes states, don't use enough "screens" to find likely voters because it's too expensive. Screens are those endless questions that test in many ways if the respondent really is going to vote or is just saying so to sound like a good citizen to the pollster.

The last serious statewide showdown among Connecticut Democrats between moderates and liberal took place in 1994. The candidate of the center was state Senator John Larson (now a left-wing disciple of Nancy Pelosi). Representing the liberal wing was state Comptroller Bill Curry. Curry enjoyed union support and a cadre of devoted activists. Like Lieberman this year, Larson won the state convention. He also preened with money and endorsements from other Democratic officials.

Barnes, who worked on the Larson campaign, recalls that Larson's pollster told the candidate was so far ahead he should save the money earmarked for a primary day field organization to spend in the fall contest against Republican John Rowland. Barnes was supposed to see the $100,000 effort to get Democrats to the polls. He despaired of the decision.

The Hartford Courant poll the Sunday before the September primary showed a statistical dead heat. Two days later, Curry thrashed Larson by 10 remarkable points. Barnes left politics to become a pollster.

Now the Director of Project Development at UConn's Department of Public Policy, Barnes thinks Lieberman should have taken a powder from the Democratic race in the spring and concentrated on winning as an independent in November. It's a mug's game for a moderate Connecticut Democrat to try to beat a jazzed liberal in a primary. Lieberman has tried to gin up his organization in the past two weeks but it is probably too late to deliver the difference he needs, says Barnes.

The attention paid to the race (and the money spent by both candidates) may drive up turnout to 30% of the state's 700,000 Democrats, but the impact of the August primary date is an imponderable. Though the Lieberman campaign has been dreading a low turnout in the typical 20% range, Barnes says a higher turnout bodes ill for the three-term incumbent. "It's the people who are excited about the war who will vote." And in a closed Democratic primary, that bodes ill for Lieberman.