Report with Mara Liasson at
NPR this morning:
In the key battleground states that will decide control of the Senate this November, President Obama's approval numbers are lower than they are nationally – but not much lower.
That's the key finding in a new poll, conducted by Democrat Stan Greenberg of Democracy Corps and Republican Whit Ayres of Resurgent Republic, that sampled likely voters for NPR.
In the 12 states with competitive Senate races this fall, only 38 percent of likely voters said they approved of the way the president is handling his job. An index of all national polls shows the president's approval rating about four percentage points higher nationwide.
NPR's list of battleground states:
- Alaska
- Arkansas
- Colorado
- Georgia
- Iowa
- Kentucky
- Louisiana
- Michigan
- Montana
- New Hampshire
- North Carolina
- West Virginia
That omits four of the states in
Wikipedia's list of competitive seats—
- Minnesota
- Oregon
- South Dakota
- Virginia
—and you really don't need any numbers to see how this biases the sample in favor of Republicans, since three of those states have large populations with a very high proportion of Democrats among them (maybe Oregon should be moved out of the column altogether since the selection of the hilariously inappropriate
Monica Wehby as the Republican candidate at the end of May, but I don't yet see how we can be certain the seat is safer than Colorado or New Hampshire, and Wikipedia, which edits that page practically every day, doesn't either).
The larger point, though, is that this is a poll of something that
really doesn't exist. There is no "battleground" of disputed land between the respective territories of the Red and Blue armies. That's a
metaphor, and a fairly distorting one at that. What there is is a disparate collection of 12 or 16 "battlegrounds" each with its own parochial concerns, demographic character, and candidates, with no more in common politically with each other than they have with whichever undisputed states they resemble. Agglomerating their numbers, spreading our perception of Obama-hatred from Arkansas and Kentucky where it exists to Michigan and New Hampshire where it doesn't, can have no other purpose than to make things look more dramatic than they actually are. It literally
subtracts from our knowledge (since NPR doesn't show us the results for the individual states, which would add something). It's really dumb.
