Pollster.com

Articles and Analysis

 

Riehle: Hierarchical Vote Analysis

Topics: 2008 , Barack Obama , Cook Political Report , John McCain , Thomas Riehle

Today's Guest Pollster article comes from Thomas Riehle, a Partner of RT Strategies.

If it were a different month on the calendar - say, October - the Obama campaign might be concerned to see that the groups most likely to be truly undecided, not leaning even a little toward Obama or McCain, comprise some voters he must be counting on:

  • Women (21% undecided), women ages 40-64 (24%), women ages 65 and older (25%), women with less than a 4-year college degree (24%), and
  • Registered voters in the Northeast (20%) and Great Lakes (20%).

Moreover, no one at the Obama campaign can be happy to see that the vote is currently tied among women with a college degree or more (43% Obama - 42% McCain) - highly educated women having become one of the most reliably Democratic groups in the electorate. Obama may start to win back support from among the relatively large group of McCain supporters currently to be found among women who voted for Clinton in Democratic primaries or caucuses (25% now support McCain), college-educated Clinton primary voters (28% McCain), moderate or conservative Clinton primary voters (23%) - but right now, those are some significant defections.

The calendar says June, not October, and undecided voters eventually will make up their minds. All in good time. For now, the McCain camp and Obama's camp are looking for indications of subtle trends moving in the early stages of the general election. For that kind of tracking, many campaigns use a tool called the "Hierarchical Vote." It divides support into 7 categories, from most pro-McCain to most pro-Obama, and tracks movement from one category to another across the 7 categories of support. Tracking changes month-to-month in the Hierarchical Vote overall among all voters (and within 50 or more subgroups) gives a campaign the insight needed to focus resources on the groups who are ready to move right now.

On the Cook Political Report website, you will find posted a Hierarchical Vote analysis from the past four Cook Political Report / RT Strategies polls (March through June). I hope you find it a unique and useful way to delve under the familiar topline vote totals and see what's really going on as we approach Independence Day. We'll keep updating this Hierarchical Vote table as Election Day approaches - just as the campaigns will do. Please let me know what you think and what you learn in reviewing these results.

 

Comments
OGLiberal:

I don't want to treat women voters as some monolithic bloc but I have to think that a majority of those college educated voters are pro-choice, anti-Iraq war, and for universal healthcare. If they vote for McCain simply because Hillary lost to Obama in the Dem primary - and McCain ends up winning - they'll only have themselves to blame for the band hand they'll be dealt with McCain as president. I have to think that when the emotion dies down - perhaps starting with Hillary's and Obama's unity event which just ended a little while ago - the women voters will start to focus on the 2-3 Supreme Court justices the next president will likely have the opportunity to appoint...and then you'll see the polls shift to more traditional numbers re: women voters.

____________________

carl29:

The only problem with this assumption is that according to all research, marital status is what define much more than anything else the way women vote, even among educated women. In 2000 GORE won MARRIED, COLLEGE EDUCATED WOMEN by 9%; however, he won UNMARRIED, COLLEGE EDUCATED WOMEN by 42%, imagine the same level of education but + or less the spouse made a more than 30% difference. In 2004 KERRY lost married, College educated women by 6% and won unmarried, college educated women by 34%.

Another group of women: UNION MEMBER, MARRIED WOMEN went in 2000 for Gore by 20%; however, UNION MEMBER, UNMARRIED WOMEN went for him by 51%, just the spouse make the difference. In 2004, Kerry won UNION MEMBER, MARRIED WOMEN by 13% and UNION MEMBER, UNMARRIED WOMEN by 43%.

As you can see marital status makes a huge difference with women. Could this have something to do with men, who tend to be more republican? Could it be that they influence their wives in such way? Sad, no?

____________________

Uri:

@carl29: Another possible explanation might be that within the same age group of younger women, those who are married are more likely to be religious or conservative. (Have no idea if that's necessarily true, though it fits with people that I know).

Another explanation might be that married women are more likely to already have or plan to soon have children, and that having children renders one more socially and economically conservative.

____________________

onelightonvoice:

As I have said before, these women will come home to the Dem ticket. The hurt feelings take time to overcome. This was the first day of both Clinton and Obama campaigning together. The bounce hasn't even finished yet - despite what the pundits say. Give it two months, and then analyze the polls.


____________________

Uri:

There's something I don't understand.
Before Obama won the nomination, about one quarter of Clinton supporters said that they will vote for McCain.

When Obama clinched and bumped, the narrative was that these are the Clinton people coming home.

Obama is still leading, and now suddenly we have one quarter of Clinton voters still saying "McCain".

Why have we not seen these shift? And where did the bump come from if not from Clinton supporters?

____________________

carl29:

Dear Uri: It is true that older people tend to be more religious but the connection seems to be marriage.

"Unmarried men have fewer ties to organized religion than married men. Only 20 percent go to church every week, compared to 36 percent of married men and 38 percent of unmarried women. About half of unmarried men never go to church, compared to only 31 percent of married men."

"38 percent of UNMARRIED women" go to church every week compared to 36% of MARRIED men; however, MARRIED MEN vote republican, conservative, in big margins. Therefore, if religion was the "link" to voting, shouldn't UNMARRIED WOMEN vote republican, conservative, at the same rate, if not more, of married men, since both groups are similarly religious? Just a question worth of analysis.

____________________

Uri:

@carl29:

I'm not sure if you understood my argument. I argued that there are demographic and political differences that may affect whether a young woman of a certain age is married, and that these same factors may affect her political choices.

For example, my impression is that young women from backgrounds that value marriage more (e.g., devout Catholics, jews, evangelical Christians, etc.) are more likely to get married earlier compared to women who are less religious within these backgrounds or from other backgrounds (e.g., reform Jews) who live the "single life" longer and also turn to live in cities rather than suburbs.

Also, as I mentioned, I think children play a factor. Having children seems to be a good motivator for some people to become more conservative, for example on crime.

Another factor which may account in a minor way towards the difference may be the gay community, which would unfortunately be less likely to be married because of our current laws, and which would be much more likely to lean democrat.

____________________

carl29:

But I think that you are assuming that single women must be young. When pollsters talk about single women, they mean divorce, which is in the raise, widow, whether old or young, or never married, which chances are these are the young gals we think of when talking about single women. The divorce nurse working the shift at the hospital is a single woman.

Single women tend to be poorer than married women(remember, we are not talking about the "Sex and the City" type of gals); therefore, it is probable that these unmarried women, whether mothers or not, vote more democrat than republican. You know, they need as many government help as they can get. Do you really believe that a single mother who needs all government programs in order to raise her children votes republican? I am myself the daughter of a single mother
( I am neither born nor raised in the US, so don't worry I was not living off your taxes) and can tell you how much help these women need. I am myself a married woman and can tell you: My life is extremely different that my mother's. When you married you get a lot of advantages: an extra paycheck which gives you a better life style than the regular divorce or single mother.

____________________

Uri:

@carl29: I am not assuming that single women must be young. I assumed that the vote by marriage statistic was adjusted for age.

And I didn't imply at all that single women would not vote democrat; I implied that married women would be more likely to vote republican because they are more likely to be suburban and have kids.

Also, as you know, I'm not from around here either (though like you, I do pay taxes, just don't get stimulus checks), and I was raised by a single mother so I know what it is like.

____________________

carl29:

With all due respect, dear Uri. You really confused me by talking about young women when I was talking about marital status. It's like if you are talking Chinesse and I'm talking Japanese. My point is that among women, marital status is very significant with regards to voting. Unmarried women, single mothers, widows, divorced, never married, tend to vote democrat; however, married women tend to vote republican. Why? I don't know.

All I can speculate is that women seem to be influenced by their husbands, men who happen to be republican by large margin. I can testify that based on what I see from my married sisters in law, the husbands are republicans, so they vote rep; however, my divorced sister in law votes democrat. The three women don't hold college degrees and attend church every week. Actually, the divorced one married very young, around 18 years of age, so we can conclude that this is the most conservative one, but this is the democrat. The only difference between her and her sisters is that she is an unmarried women with 2 teenagers to support. When I think of unmarried women, I think of these women, not the girls with mini-skirts running after guys.

I remember that Dick Morris, the guy that used to work for Bill and Hillary, and who is now a republican, once wrote an article about the GOP and unmarried women vote. His argument was that the fact that the GOP "owns" the married women vote for the most part but loses badly among unmarried women, young and old, doesn't bold well for the survival of the GOP in the long run. Why? Because with the divorce rate, more and more former married women are becoming part of unmarried women bloc that has not warm to the GOP. Let alone those profesional, progressive, feminist women who regardless of marital status will always be democrats.

____________________

eternaltriangle:

What is an interesting change is that education seems to benefit the GOP, rather than the Democrats (consistently across a number of polls). I suspect post-grads are still breaking for the Dems.

____________________

AJbaker:

Stop being so patronizing about the Democratic women now supporting McCain. We are not comming back. We are not acting from emotion and we are well aware of where McCain stands. Many of us have watched him for decades. We are a naturally skeptical group and because we supported Hillary we were able to give a cold survey of Obama and found him lacking in the qualifications which are necessary to even consider a candidate for the Presidency. Obama will do or say anything to be popular -- his pathetic performance before AIPAC was an example -- what sane grown up would voluntarily utter loose cannon statements about Jerusalem as the undivided capital of Israel? For a foreign policy major, he pulls some really mind-blowing clunkers. He is also the only candidate in this primary season who made false accusations of racism. He has a truly apparent disdain for feminism if not for women in general and white women in particular -- this may have to do with his resentment against his mother. Look at his books -- compare the index results for mothers and fathers. He can orate all he wants about his daughters -- his body language gives him away every time. He is in the process of moving the Democratic party to the right and most of us are not interested. The other group not interested are the conservative Hillary supporters who are closer to McCain on the issues in the first place. Think Reagan Democrats. Anyway, don't wait up -- we're not coming 'home' any time soon.

____________________

carl29:

OK AJbaker. Stay where you are!!!

P.S: We are fine without you guys.

____________________

desimba:

I would like to comment on why unmarried women tend to vote more Democratic than married women. I do not believe that it is their husbands influencing their votes directly as some have opined. Rather my hypotheses are:

1) As they are better able to purchase the goods and services that they need through their private purchasing power, they are more likely to favor a smaller government and see a reduced role of the welfare state. Not surprisingly, this is associated with a preference for lower taxes and Republican policies.

2) The left has always talked about revolution (not of the American revolution kind or the industrial revolution kind but like the French revolution kind). Married women are likely to be afraid and rightfully so, of what follows such a revolution, whether property rights would be respected in a dictatorship of the proliteriat and are hence more likely to go with the Republicans who abjure violence and are more likely to guarantee private property rights.

(Blog: alonelycapitalist.blogspot.com)

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR