Dems Hold Wide Generic Congressional Ballot Lead in GOP-Leaning Poll
by Jonathan Singer, Tue Jun 26, 2007 at 06:24:47 PM EDT
It was a little over a year ago that my concerns over the Republican lean of Rasmussen Reports led me to generally refrain from posting their polls on this site. To get an idea of the impetus of this move, take a look at this graph put together by Professor Charles Franklin detailing how although Rasmussen's polling tended to that point to track with other national polling in trends, in actual numbers Rasmussen yielded a Bush approval rating fully 3-4 points higher than that of the composite of all other national polling.
Fast forward to this year. Professor Franklin's best estimate of President Bush's approval rating from the available national polling pegs the number at 29.1 percent, down about 5 points over the last two months. But where does Rasmussen put the President? According to the pollster's latest numbers, 37 percent of "likely voters" approve of the job the President is doing -- a number significantly higher than that of the consensus among almost every other national pollster. What's worse, Rasmussen does not pick up the trend viewed by other pollsters, with the President's approval rating down just 2 points from where it was across the month of April.
One might argue that the difference in numbers lies not in a Republican bias but rather in the fact that Rasmussen is polling likely voters rather than registered voters or adults. However, that does not account for the fact that Rasmussen is intentionally skewing its numbers to make Congress' approval rating appear to be much lower than that of the President by asking an unbalanced four-point question (excellent, good, fair, poor) for Congress and a balanced four-point question (strongly approve, somewhat approve, somewhat disapprove, strongly disapprove) of the President. (Professor Franklin explains elsewhere that the unbalanced question can yield lower numbers than the more balanced one.
Yet despite all of this (or perhaps in spite of it), I'd like to pass on a new poll from Rasmussen gauging likely voters' views towards the 2008 elections -- specifically the generic congressional ballot question. Even with the lean of Rasmussen Reports, Democrats still hold a 12-point lead on the question of whether likely voters would prefer to elect a Democratic or Republican Congress next fall. And all of this comes even while Rasmussen inflates the support for third party candidates, pegging their support at 6 points -- fully 70 percent higher than it has been over the last two decades or so and more than twice what it's been in either of the last two elections. If the Democrats hold this large of a lead even with all of that, it seems to me that they are in an outstanding position more than a quarter of the way to the 2008 elections.