Extrapolating general election scenarios out of primary result preferences is like comparing apples to oranges in this case. In fact, data from a series of recent SurveyUSA state polls directly contradicts the belief that because Obama has performed relatively worse than Hillary among certain demographic groups in several recent primaries, he will therefore perform badly among those groups later. The data also strongly suggests that the reason Hillary is ahead of Obama in several states’ general election matchups against McCain is not because Obama polls poorly among these more traditional moderate/conservative white voters, but instead it is because he currently polls relatively poorly among more liberal, Democratic-partisan, female voters – a specific demographic which is very likely to return to the Democratic fold in November.
On the other hand, the demographic groups which are currently not in Hillary’s corner are very unlikely to move to that corner in the Fall. Let me explain here, as some of this is really counter-intuitive.
A few weeks ago, in the direct aftermath of "Bittergate", SurveyUSA released its now monthly series of general election polls for 14 states where Obama and Hillary were matched against McCain:
http://www.surveyusa.com/...
The polls included ones for a number of important swing states. Looking simply at just the bottom line of these polls, both Obama and Hillary had geographic strengths and weaknesses. Both Democrats led McCain in California, Massachusetts, New York, Minnesota and Oregon (although in the case of both Minnesota and Oregon, Hillary’s lead was just 47-46). Both Democrats were behind in Alabama, Kentucky, Kansas, New Mexico and Virginia (although previous SUSA polls have shown New Mexico and Virginia to be competitive). As far as the other states, Obama had a lead over McCain in Iowa and Wisconsin, while Hillary was behind in Iowa and tied with McCain in Wisconsin. Missouri and Ohio results were kind of a mirror image of Iowa and Wisconsin, with Hillary ahead of McCain, but Obama behind (though in the case of Ohio by only 2 points behind).
What I did here is I looked at the internals/cross-tabs for these polls to see if any of the data correlated to the belief that in certain states Obama currently does worse in general election matchups because of his supposed weakness with more traditional moderate/conservative voters. I looked at all the demographic categories and, using a spreadsheet, compiled averages for the different demographic groups -- kind of a primitive sample data set, if you will. I only used data for the 8 true swing states out of these 14 – Iowa, Minnesota, Wisconsin, New Mexico, Missouri, Ohio, Oregon and Virginia. (I strongly believe that California, Massachusetts and New York will go Democratic in November regardless of who the Democratic nominee is while whoever the Democrat is, he or she will lose Alabama, Kentucky and Kansas.)
Overall Results:
When averaging the data for the 8 "sample" swing states, you can see that, on average, Obama and Hillary perform very similarly:
McCain 47.3 – Hillary 45.9 (advantage 1.4 to McCain)
McCain 46.3 – Obama 46.6 (advantage 0.3 to Obama)
The difference here is only 1.7 points between how Hillary and Obama perform. This is actually very similar to the current RCP nationwide average:
http://www.realclearpolitics.com/...
Hillary -- 0.6 advantage over McCain
Obama -- 1.5 advantage over McCain
Difference between Hillary and Obama = 0.9 points.
Now, to the internals of the SUSA poll sample:
Gender:
Among Females:
McCain 41.6 – Hillary 51.6 (advantage 10.0 to Hillary)
McCain 43.4 – Obama 48.8 (advantage 5.4 to Obama)
Among Males:
McCain 53.4 – Hillary 40.0 (advantage 13.4 to McCain)
McCain 49.6 – Obama 44.3 (advantage 5.3 to McCain)
As you can see, Hillary has a 4.6 point advantage over Obama with women, while Obama has an 8.1 advantage over Hillary with men.
It gets even more interesting when you carefully examine the internals of several particular swing states. It appears that in Missouri Obama actually performs significantly better among male voters than he does among female voters. In the state (which Obama loses to McCain by 8 points overall, while Hillary wins by 1 point overall), Obama is down to McCain by only 2 points, while Hillary is down by 4 points. Among women, on the other hand, Hillary leads by 6 while Obama is down by a whopping 13 points.
In Ohio (which Obama loses to McCain by 2 points overall, while Hillary wins by 11 points overall), Obama performs slightly worse than Hillary among men (McCain 50 – Hillary 47; McCain 50 – Obama 43), but again the deciding factor appears to be female voters. Among Ohio women, it’s a 2 point advantage of Obama over McCain, but a whopping 23 point advantage of Hillary over McCain. In other words, the overriding reason why Hillary is ahead of McCain in Missouri and Ohio, while Obama is behind, is because of the way female voters are currently predisposed. Both the Missouri and Ohio gender breakdowns are, simply said, politically unprecedented. When you look at 2004 exit polls from these states you can see that there was no such great gender disparity in the results. Missouri was actually a bit of an anomaly in that women voted 54 Bush to 45 Kerry, while men voted 52 Bush to 47 Kerry. Ofcourse, this is nowhere close to the huge disparity in the current polling. Ohio was much more representative of general voting trends in 2004, with men favoring Bush 52 to Kerry’s 47, while women were split 50-50.
http://www.cnn.com/...
http://www.cnn.com/...
Another state of note is Virginia. In that state, McCain currently beats Hillary by 24 points among men, while he is ahead of Obama by only 4 points among men. Among women, on the other hand, Hillary is behind McCain by 8 points, while Obama is behind by 12. Like Missouri, this is a state where Obama performs better among men than among women. Again, this is rather unprecedented. Based on 2004 exit polls, Bush beat Kerry 59-40 among men, while women were split 50-50. The fact that Obama currently performs so well among men in Virginia (down to McCain by only 4 points, while Kerry lost by 19 points among men) indicates that the state may indeed be competitive for Obama in the general. If you extrapolate the numbers, assume for a second, that Obama’s numbers among men hold, while the women numbers return to their 2004 levels. That scenario would put Obama on the brink of winning; a little added push from female voters could put him over the top.
http://www.cnn.com/...
Hillary’s strength among women was one of the main factors enabling her 9-10 point victory in Pennsylvania. It is interesting to note that Hillary performed very close to predicted percentages in the different regions of the state – except for one area, the Philadelphia suburbs. Here, many of the polls predicted a slight Obama victory. When the numbers came in though, Hillary had won the suburbs by a slight margin. The reason was undoubtedly her support among women.
Age:
Among 18-34 year olds:
McCain 44.0 – Hillary 49.3 (advantage 5.3 to Hillary)
McCain 39.5 – Obama 54.8 (advantage 15.3 to Obama)
Advantage 10.0 to Obama over Hillary among 18-34 year olds.
Among 35-49 year olds:
McCain 51.8 – Hillary 42.5 (advantage 9.3 to McCain)
McCain 50.6 – Obama 42.4 (advantage 8.2 to McCain)
Advantage 1.1 to Obama over Hillary among 35-49 year olds.
Among 50-64 year olds:
McCain 47.0 – Hillary 46.3 (advantage 0.7 to McCain)
McCain 47.4 – Obama 46.4 (advantage 1.0 to McCain)
Advantage 0.3 to Hillary over Obama among 50-64 year olds.
Among 65+ year olds:
McCain 44.6 – Hillary 47.6 (advantage 3.0 to Hillary)
McCain 47.8 – Obama 42.3 (advantage 5.5 to McCain)
Advantage 8.5 to Hillary over Obama among 65+ year olds.
No revelations here ... Bottom line is this: Obama has 10 point advantage over Hillary among young voters, while she has an 8.5 advantege over Obama among older voters. Voters in the middle (35-64 yo) are almost evenly split. If Obama becomes our nominee he obviously needs to work on getting the trust and acceptance of older voters. His ideas on social security may help him to achieve this goal.
Race:
Among Whites:
McCain 50.6 – Hillary 43.8 (advantage 6.8 to McCain)
McCain 50.5 – Obama 42.4 (advantage 8.1 to McCain)
Advantage 1.3 to Hillary over Obama among white voters.
Among Blacks:
McCain 21.7 – Hillary 63.0 (advantage 41.3 to Hillary)
McCain 14.3 – Obama 84.3 (advantage 70.0 to Obama)
Advantage 28.7 to Obama over Hillary among black voters.
Among Hispanics:
McCain 30.5 – Hillary 61.5 (advantage 31.0 to Hillary)
McCain 33.0 – Obama 58.0 (advantage 25.0 to Obama)
Advantage 6.0 to Hillary over Obama among Hispanic voters.
As you can see, Hillary’s advantage over Obama when it comes to white voters in the general election is a miniscule 1.3 points. Even though the margin here slightly "favors" Hillary, it should be noted that McCain actually "tops out" at a higher number (50.6) when facing Hillary than when he faces Obama (50.5). I’m not a statistician, so it would be hard for me to obtain a margin of error for this. Needless to say, I don’t believe that, statistically, there’s any difference in how Obama and Hillary perform against McCain among white voters in swing states. Any significant differences that do exist in some of these particular states are instead driven largely (if not entirely) by gender. Thus, Hillary would likely have an advantage among white women.
It should be noted that for black and Hispanic voters, the sample sizes are relatively small. I included data only for states where the black or Hispanic population was at least 9% (or 50 or more participants in the SUSA poll). In effect, the black sample here is derived only from data from Virginia, Missouri and Ohio. Since New Mexico was the only state in the sample of 8 swing states with a statistically significant Hispanic population, I decided to also use the Hispanic data from California and average out those numbers with the New Mexico data (I figured that the New Mexico Hispanic population is quite different from the overall Hispanic population due to historical factors, but averaging with the more "representative" California Hispanic subgroup should give us a more accurate picture. It should be noted, however, that the California and New Mexico Hispanic data samples were very similar to each other in terms of preference of Hillary vs. McCain and Obama vs. McCain).
Political Party:
Among Democrats:
McCain 16.1 – Hillary 75.9 (advantage 59.8 to Hillary)
McCain 20.0 – Obama 72.0 (advantage 52.0 to Obama)
Advantage 7.8 to Hillary over Obama among Democrats.
Among Republicans:
McCain 88.3 – Hillary 9.4 (advantage 78.9 to McCain)
McCain 84.3 – Obama 13.3 (advantage 71.0 to McCain)
Advantage 7.9 to Obama over Hillary among Republicans.
Among Independents:
McCain 54.6 – Hillary 36.9 (advantage 17.7 to McCain)
McCain 48.0 – Obama 43.4 (advantage 4.6 to McCain)
Advantage 13.1 to Obama over Hillary among Independents.
It appears that Hillary has a 7.8 advantage over Obama among Democratic voters, while Obama has a 7.9 advantage over Hillary among Republican voters. Among Independents, Obama has a relatively large 13.1 advantage over Hillary in terms of performance against McCain in swing states.
Political Ideology:
Among Conservatives:
McCain 81.9 – Hillary 14.8 (advantage 67.1 to McCain)
McCain 79.8 – Obama 16.5 (advantage 63.3 to McCain)
Advantage 3.8 to Obama over Hillary among Conservatives.
Among Moderates:
McCain 40.5 – Hillary 53.3 (advantage 12.8 to Hillary)
McCain 38.4 – Obama 55.0 (advantage 16.6 to Obama)
Advantage 3.8 to Obama over Hillary among Moderates.
Among Liberals:
McCain 17.5 – Hillary 73.1 (advantage 55.6 to Hillary)
McCain 21.4 – Obama 71.9 (advantage 50.5 to Obama)
Advantage 5.1 to Hillary over Obama among Liberals.
This is where we go further into disputing the belief initially mentioned in this diary that Obama does not perform as well as Hillary among moderates and conservatives. The counter-intuitive truth confirms that it is indeed Obama who is stronger among conservative and moderate voters in general election matchups against McCain (these polls, we must remember, were conducted post-Wright and indeed immediately in the aftermath of "Bittergate"). Hillary’s major advantage in this category is among liberals.
Even in states where the conventional wisdom would say that Obama’s relative disadvantage against McCain is due to his poor showing among more traditional voters, we can see that it is largely a result of a poor showing among liberal voters instead. In Ohio, for instance, we can see that among conservatives, McCain gets 81 to Hillary’s 17, while McCain beats Obama by a narrower 76-20 (thus, advantage to Obama among Ohio conservatives). Among Ohio liberals, Hillary does better, beating McCain 80 to 16, while Obama only manages to beat McCain by 69 to 24. .
Abortion:
Among Pro-Life Voters:
McCain 66.3 – Hillary 28.5 (advantage 37.8 to McCain)
McCain 64.5 – Obama 29.3 (advantage 35.2 to McCain)
Advantage 2.6 to Obama over Hillary among Pro-Life Voters.
Among Pro-Choice Voters:
McCain 33.4 – Hillary 59.8 (advantage 26.4 to Hillary)
McCain 33.1 – Obama 60.1 (advantage 27.0 to Obama)
Advantage 0.6 to Obama over Hillary among Pro-Choice Voters.
Interstingly, Obama outperforms Hillary among both pro-life and pro-choice voters. He actually does somewhat better among pro-life voters than among pro-choice voters. This implies a possible correlation already discussed of Hillary doing better with liberal women, many of whom are pro-choice. Even in a more traditional state like Ohio, among pro-life voters, there does not appear to be a significant difference between Obama and Hillary (McCain 61 – Hillary 35; McCain 62 – Obama 31). There is, however, a significant difference among pro-choice voters in Ohio (McCain 27 – Hillary 69; McCain 34 – Obama 58). Nevertheless, for the most part the abortion issue is really a non-issue at least in terms of how Obama and Hillary would
perform against McCain in November.
Gun Issue:
Among Gun Owners:
McCain 55.4 – Hillary 38.8 (advantage 16.6 to McCain)
McCain 55.3 – Obama 38.1 (advantage 17.2 to McCain)
Advantage 0.6 to Hillary over Obama among Gun Owners.
Among Non-Gun Owners:
McCain 38.5 – Hillary 54.0 (advantage 15.5 to Hillary)
McCain 37.0 – Obama 55.5 (advantage 18.5 to Obama)
Advantage 3.0 to Obama over Hillary among Non-Gun Owners.
This is where the data gets fun. First thing to remember here, these polls were conducted in the days following the "Bittergate" explosion. Second, in virtually every state here (Iowa, Minnesota, Wisconsin, New Mexico, Missouri, Ohio, Oregon and Virginia) hunting is a popular pastime. Nevertheless, there is no evidence whatsoever of any sort of anti-Obama backlash among gun owners. Hillary’s advantage of 0.6 among gun owners is statistically insignificant. (Just like with white voters, Hillary’s margin is slightly larger, but McCain "tops out" at a higher number when facing Hillary (55.4) than when facing Obama (55.3).)
Religion:
Among Regular Church-goers:
McCain 56.5 – Hillary 36.6 (advantage 19.9 to McCain)
McCain 54.1 – Obama 39.0 (advantage 15.1 to McCain)
Advantage 4.8 to Obama over Hillary among Regular Church-goers.
Among Occasional Church-goers:
McCain 40.0 – Hillary 53.3 (advantage 13.3 to Hillary)
McCain 41.9 – Obama 51.4 (advantage 9.5 to Obama)
Advantage 3.8 to Hillary over Obama among Occasional Church-goers.
Among those who almost never attend Church:
McCain 38.0 – Hillary 55.5 (advantage 17.5 to Hillary)
McCain 37.4 – Obama 55.0 (advantage 17.6 to Obama)
Advantage 0.1 to Obama over Hillary among those who almost never attend Church.
Among Evangelicals:
McCain 62.4 – Hillary 31.8 (advantage 30.6 to McCain)
McCain 61.8 – Obama 31.1 (advantage 30.7 to McCain)
Advantage 0.1 to Hillary over Obama among Evangelicals.
Among Non-Evangelicals:
McCain 40.8 – Hillary 52.6 (advantage 11.8 to Hillary)
McCain 39.1 – Obama 54.8 (advantage 15.7 to Obama)
Advantage 3.9 to Obama over Hillary among Non-Evangelicals.
This data addresses the second part of "Bittergate's" guns and religion "controversy". Like with the gun issue, there is no evidence whatsoever of any sort of anti-Obama backlash among the more religious voters. In fact, the data seems to indicate that relative support for Obama is actually highest among the most religious and lowest among those who are occasional church-goers.
Summary:
The crux of my argument here is this (and I apologize for making this diary so long, but I really wanted to back up my points with data, not just throw out an unsupported hypothesis): Currently, contrary to conventional wisdom, when matched against McCain in crucial swing states Obama performs better than Hillary among men, moderates, conservatives, Independents, Republicans, and regular church-goers (and obviously with black and young voters as well) while Hillary performs better among women, liberals, Democratic partisans, and occasional church-goers (as well as older voters and Hispanics). In order to "solidify" their winning coalition for November, both candidates must reach out to the groups among whom they are doing relatively worse.
If Hillary becomes the nominee, she thus has to reach out to men, moderates, conservatives, Independents, Republicans, regular church-goers, blacks and young voters.
If Obama becomes the nominee, he has to reach out to women, liberals, Democratic partisans, and occasional church-goers, Hispanics and older voters.
The question Democrats must ask ourselves is this:
Who do you think will have an easier time reaching out and bringing these particular groups into the Democratic winning tent in November:
With the prospect of the pro-life McCain cementing a rabidly conservative Supreme Court majority for the next generation, one that would very likely overturn Roe v. Wade (btw, despite John Roberts’ and Samuel Alito’s claims to the contrary during their confirmations, they have been more than willing and able to overturn precedents while on the bench so far) as well as the prospect of the GOP continuing to institute countless other policies on social issues, health care, economics, as well as foreign policy which most liberal Democratic women find anathema, I just don’t see this demographic not coming back to the Democratic fold in November if Obama is the nominee. Older voters may be, relatively, harder to reach; but there is no reason why Obama should not be able to expand his reach to those over 65 years old by engaging this group. Hispanics, likewise in my opinion, will wholeheartedly support Obama if he becomes our nominee. Like liberal women, this group will ultimately not vote against their own interests by favoring a member of a party which is for the most part rabidly anti-immigrant.
On the other hand, I believe Hillary would have a much harder time (if not an impossible task altogether) in trying to get non-liberal Independent and Republican men over to her side. Many in this demographic simply find her anathema. At this point, it might also be hard for Hillary to get African-Americans and younger voters to back her with the same enthusiasm with which they will undoubtedly back Obama if he is our nominee.
To summarize, I will list the demographic groups below where Hillary has an advantage over Obama when it comes to general election matchups against McCain, and vice-versa (I have bolded the demographics where advantage is 3.8 points or higher):
Clinton advantage among:
8.5 65+ Voters
7.8 Democrats
6.0 Hispanics
5.1 Liberals
4.6 Women
3.8 Occasional church-goers
1.3 Whites
0.6 Gun owners
0.3 50-64 Voters
0.1 Evangelicals
Obama advantage among:
28.7 Blacks
13.1 Independents
10.0 18-34 Voters
8.1 Men
7.9 Republicans
4.8 Regular church-goers
3.9 Non-Evangelicals
3.8 Conservatives
3.8 Moderates
3.0 Non-Gun owners
2.6 Pro-Life
1.1 35-49 Voters
0.6 Pro-Choice
0.1 Almost never church-goers
Again, who do you think will have an easier time reaching out and bringing these particular groups into the Democratic winning tent in November:
Obama – needs to attract women, liberals, Democratic partisans, and occasional church-goers, Hispanics and older voters into the Democratic winning tent (hint: the only struggle may be with older voters).
or
Hillary – needs to attract men, moderates, conservatives, Independents, Republicans, regular church-goers, blacks and young voters into the Democratic winning tent (hint: good luck with most of the above).
I will end this diary with the above question; the answer should be obvious to anyone interested in ensuring Democratic victory in November.
