Why had such a variance? Is one incorrect? Both?
There might be a number of reasons for the difference--including the prospects that the polls are really not dissimilar at all.
This was the first question discussed when Couvillon and Jim Brown discussed the polls and the Louisiana governor’s race, with me, Wednesday afternoon.
The major difference was the measurement of the expected African American turnout.
Kennedy measured a turnout of 20 and then 25 percent of the electorate.
Couvillon used 29 percent assuming a roughly 95% support for the democrat.
The pollster said the runoff produced a 28 percent turnout of the African-American community.
He also said the rain depressed an already depressed electorate, it could have impacted the democrats more than the republicans and that the psychology of elections often drives turnout. For example, if a candidate does not appear to have much of a chance to win, he will not get as large of a turnout, as desired.
Jim Brown, with 28 years of statewide office, discussed the lack of other competitive elections on the ballot that could drive greater turnout. He said there are now only two statewide elections--the Governor and Lt. Governor race—a limited number of sheriff races and other elections. He also compared this election to the John Breaux-Henson Moore US Senate race.