My friend Chris Bowers posted a thoughtful response to my column from this week, which argued that only a small fraction of liberal Democrats have grown disillusioned enough with President Obama to express disapproval. Chris is smart and reality-based about poll data, so his arguments are worth considering.
Chris points out that the liberal Democrat subgroup that I focused on (17% of adults in February 2009 according to Gallup's data, 15% in December) overlooks the small portion of liberals that do not identify or lean Democratic . He estimates that liberal-non-Democrats (or LNDs as he calls them) are now roughly 6% of adults, since Gallup reports that 21% of adults self-identified as liberal during all of 2009. I'll spare the details, but he extrapolates from the available Gallup data to conclude that Obama's approval rating among liberal-non-Democrats is now hovering at just over 50% after falling roughly 15 percentage points since February.
I won't quarrel with Chris' central argument: I have no doubt that a group of liberal-non-Democrats exists that rates Obama less positively than the larger group of liberal Democrats and whose rating of Obama has fallen over the course of the year at a rate similar to independents and moderate Republicans.
I also have no doubt that some liberals are feeling disillusioned. Again, as I reported in the column, the size of the liberal Democrat subgroup fell from 17% in February to 15% in December (I did not ask Gallup for numbers for other months, but the massive sample sizes involved -- over 11,000 in December and over 14,000 in February -- make even a 2 point drop meaningful). So a small sliver of liberals may shifted their allegiance to independent over the course of the year.
My point was mostly that the number of disillusioned liberal Democrats is small relative to disillusioned conservative Democrats and pure independents. Chris argues that liberal-non-Democrats are a small but nonetheless critical swing constituency. That's an interesting but different argument, and I think we can agree that it's not an argument about an erosion of the Democratic "base."
Chris also compares Obama's current approval ratings by ideology to his vote by ideology as reported on the 2008 exit polls to argue that "the only people who have become disillusioned with President Obama are liberals." I know these sorts of comparisons are popular, but I'm not a fan: They compare apples to oranges. Some Republicans will express approval of a Democratic president they did not vote for (and may never support), while some Democrats will express disapproval of a President of their own party, even if they would never consider voting for a future Republican opponent. I would be more convinced by this sort of comparison if it involved a vote or "reelect" question about 2012, rather than presidential approval.
All of that aside, this is a good conversation to continue.