Last week, slinkerwink posted this diary highlighting the Reuters tracking poll. Reuters had shown Bernie Sanders pulling even with Clinton nationally following the Iowa caucuses, much like the Quinnipiac poll that showed Clinton only up 44-42.
What was the response to this from DKos commenters?
It was that Clinton still led handily among likely voters.
At that time, she still led 63-34 in Reuters’ "likely democratic primary voter” screen.
And to be sure, Reuters is not the best pollster by any measure. Reuters’ “likely democratic primary voter” screen excludes all self-identified independents, even though self identified independents made up about 24% of the electorate in the 2008 primaries. Even in a state with ultra-closed primaries like New York, exit polls say that self-identified Independents made up 12% of the electorate (with another 1% self-identified Republicans).
For that reason, the “likely democratic primary voter” screen tends to underestimate Sanders’ support — because he does well with Independents.
That was then
...
This is now
Over the past week, Sanders has gained steadily even among Reuters’ restrictive "likely Democratic Primary Voter screen.” It now stands at:
|
Feb 12 |
Feb 5 |
Sanders |
46% (+12%) |
32% |
Clinton |
54% (-11%) |
63% |
And just for the record, registered voters have remained basically tied over the past week:
But what about among Hillary's so-called "firewall” of African American support:
Since Feb 1, following the Iowa caucuses, Sanders has cut Hillary’s 69-24 lead among African American registered voters to a 58-37 lead. Note I am using registered voters rather than likely voters here because the Reuters website won't display African American likely voters:
|
Feb 12 |
Feb 1 |
Sanders |
37% (+13%) |
24% |
Clinton |
58% (-11%) |
69% |
Ruh-roh.