Today a front page article details the switch of our weekly poll to registered voters from adults (this weeks cross tabs, last weeks cross tabs). This has led to some fairly significant swings in the various demographic groups. Of significance, I note that there are a significant number of enthusiastic unregistered 30-something people who claim they are going to go to the polls and vote for Republicans. Effort needs to be made to understand who these people are, where they live, what their political ideology consists of, whether they will actually register and participate and whether it would be a better strategy to ignore them and build strength elsewhere or to try to improve our standing among them.
First let's look at Obama's support by age as well as the support for Republicans, and the Generic Congressional Ballot. This week is the first week in which registered voters are used and last week (adults) is shown in parenthesis.
Note the sample sizes in this week poll imply a margin of error on the order of 5-7%, so some of the swing but not all of the swing can be account for by the margin of error.
Obama Fav/Unfav (last week)
18-29 64/31 (86/12) -41
30-44 56/40 (40/55) +31
45-59 49/47 (64/30) -32
60+ 46/51 (38/56) +3
Republican Party fav/unfav (last week)
18-29 19/78 (7/87) +21
30-44 30/68 (49/42) -45
45-59 33/62 (22/69) +18
60+ 35/60 (26/63) +12
Gen Cong Ballot Dem/Rep-Not Sure (lastweek)
18-29 58/17-25 (60/4-36) -15
30-44 46/43-11 (22/54-24) +35
45-59 42/48-10 (43/29-28) -20
60+ 38/50-12 (32/45-23) +1
We see that among those 18-29 and those 45-59 Obama has less support among registered voters than among adults. This is expected due to the significant number of unregistered minorities and young people. Among those over 60 little change is observed.
The most interested results are those 30-44. In this range we see that there are a very substantial number of people with favorable opinions of the Republicans and intent to vote for Republicans in the congressional election, but who are not registered.
Next we have some numbers related to self reported voter enthusiasm.
This week (registered voters)
Def vote/vote/not likely/def not - unsure
Last week (adults)
Def vote/vote/not likely/def not - unsure
As this is a bit cluttered, I simply this by looking at the net change in voter likelihood comparing registered voters to adults to find:
Net Change (def vote+will vote - (not likely+will not))
A positive number indicates that the voters who reported themselves as registered see themselves as more likely to vote. This is what one would expect! However, the voters age 30-44 are reporting the reverse. More adults than registered voters say they will vote! Thus, there are a significant number of very enthusiastic voters between the ages of 30 and 44 who say they will vote and will vote for Republicans but who are not registered!
This raises some interesting questions. Firstly, who are these people? Are they neocons largely opposed to Obama on questions of national security but willing to be reasonable on other issues? Perhaps they are libertarians who prefer Obama's stance on national security but are upset with his level of economic intervention. Lastly they could be paleocons who disagree with Obama on most everything and for this reason are now becoming interested.
It seems important to find out who they are, and if there are any actions the Democrats could take to win their support or at least curtail their enthusiasm. It also makes the midterms appear as though they could hinge upon the Republicans ability to register these voters vs. our ability to register our young and minority voters. Lastly, there is the question of the impact these voters will have on the Republican party as it searches for an identity and is pulled in differing directions.
Edited to add some regional statistics:
NORTHEAST 74/22 (87/7) -28
SOUTH 35/62 (26/69) +16
MIDWEST 55/41 (60/36) -10
WEST 57/29 (57/37) +8
NORTHEAST 18/78 (6/84) +24
SOUTH 39/53 (61/32) -43
MIDWEST 33/62 (19/71) +23
WEST 28/70 (20/70) +8
Congressional Generic Ballot
Dem/Rep-Notsure (last week)
NORTHEAST 62/17-21 (61/5-34) -11
SOUTH 33/59-8 (19/59-22) +14
MIDWEST 45/43-12 (37/35-28) 0
WEST 46/41-13 (38/36-26) +3
Change in voter likelihood by region
(Def vote+vote-(not likely+def not))
From examining these it looks as though there are a significant number of unregistered Republican supporters in the south. However, it isn't showing nearly the enthusiasm increase that the 30'somethings are showing. The could be due to minorities in the southern states perhaps showing up as non-enthusiastic and obscuring the enthusiasm of the southern Republicans. Alternately, it could be showing that we have a partially disjoint set in which there are southern Republicans who are unregistered and unenthusiastic and 30-something Republicans located across the country who are unregistered and enthusiastic.