I was just looking at some of the breakdowns in the latest Quinnipiac poll on Lieberman-Lamont. Some of it is predictable. For example, of the likely Democratic primary voters who say they’ll vote for Lamont, 54 percent say their vote is more against Lieberman than for Lamont. Of those who say they’ll vote for Lieberman, 84 percent say their votes is more for Lieberman than against Lamont.
But one result is baffling. While the overall result of the poll shows Lamont ahead 51-45, among likely Democratic primary voters, there is a significant gender difference in the numbers. Among men, Lamont leads Lieberman 55-42. Among women, the two candidates are tied, 47-47. Intuitively, that doesn’t seem to make sense. Is it more evidence of the unreliability of polling in this August primary?