PDA

View Full Version : New Pew Political Typology Released


Exapno Mapcase
05-08-2011, 12:28 PM
Every few years the Pew Research Center does a massive survey and tries to identify the coherent belief blocs in America. This year's Political Typology (http://people-press.org/2011/05/04/beyond-red-vs-blue-the-political-typology/) has just been released.

It's near book length (OK, not really, but 13 long pages with lots of graphs and charts) so I won't pretend to have absorbed it.

A couple of comments, though.

Pew divides the population into nine groups:
Staunch Conservatives take extremely conservative positions on nearly all issues – on the size and role of government, on economics, foreign policy, social issues and moral concerns. Most agree with the Tea Party and even more very strongly disapprove of Barack Obama’s job performance. A second core group of Republicans – Main Street Republicans – also is conservative, but less consistently so.

On the left, Solid Liberals express diametrically opposing views from the Staunch Conservatives on virtually every issue. While Solid Liberals are predominantly white, minorities make up greater shares of New Coalition Democrats – who include nearly equal numbers of whites, African Americans and Hispanics – and Hard-Pressed Democrats, who are about a third African American. Unlike Solid Liberals, both of these last two groups are highly religious and socially conservative. New Coalition Democrats are distinguished by their upbeat attitudes in the face of economic struggles.

Independents have played a determinative role in the last three national elections. But the three groups in the center of the political typology have very little in common, aside from their avoidance of partisan labels. Libertarians and Post-Moderns are largely white, well-educated and affluent. They also share a relatively secular outlook on some social issues, including homosexuality and abortion. But Republican-oriented Libertarians are far more critical of government, less supportive of environmental regulations, and more supportive of business than are Post-Moderns, most of whom lean Democratic.

Disaffecteds, the other main group of independents, are financially stressed and cynical about politics. Most lean to the Republican Party, though they differ from the core Republican groups in their support for increased government aid to the poor. Another group in the center, Bystanders, largely consign themselves to the political sidelines and for the most part are not included in this analysis.
What jumps out at me from this is that it is almost impossible to sort these groups into two parties. Except at the extremes, the middle groups have a jumble of positions that don't neatly align with either party. But, and here's the other big point, they don't align well with one another either. Anybody looking to form a viable third party out of the middle will find it an almost impossible challenge to identify a set of positions that would attract a permanent coalition.

The other major point that makes me an election thread is the chart on p. 3, How the Typology Groups Voted in 2008 and 2010.

You normally expect a lower turnout across-the-board for non-Presidential elections. But the two conservative groups actually showed a higher turnout in 2010 than 2008. (Libertarians and Disaffecteds also voted more Republican but the total percentage who voted went down as expected.)

The four groups that voted overwhelmingly for Obama all turned out in stupendously lower numbers for Democrats in 2010.

For me, the obvious takeaway is that this gives the Republicans a huge obstacle toward taking the White House, and probably also in the Congressional elections. While there is 18 months for a bombshell to fall that will Change Everything, right now there is no reason to expect that the Democratic voters who stayed home in 2010 won't show up in 2008 numbers for the 2012 election. That means even without knowing who the Republican candidate is, or what the issues will be, or how the economy will be doing (assuming no bombshells), you can call the election as a Democratic sweep similar to 2008 just from voting patterns.

All politics is local, I know. More Democratic Senators are up for re-election and in vulnerable areas. Highly motivated minorities turn out in numbers proportionally far higher than those who are nominally satisfied. The 10% who are Bystanders and weren't counted in any of the opinions may become motivated. A Mitch Daniels candidacy would take Wisconsin from Obama. Each side will pour a couple of billion dollars into the races. An independent candidate could siphon votes from the middle.

Still. Any Democratic candidate reading these results has to be pleased. The Republicans aren't starting from even. They are starting about 8 percentage points behind, or exactly where they were in 2008. They either have to keep the Democratic millions who stayed home in 2010 from voting for President or they have to find more than that many millions to switch or come out for the first time. Either is extremely unlikely according to this.

Lots of other info for political junkies in those 13 pages, so go to it.

appleciders
05-08-2011, 12:48 PM
...there is no reason to expect that the Democratic voters who stayed home in 2010 won't show up in 2008 numbers for the 2012 election. That means even without knowing who the Republican candidate is, or what the issues will be, or how the economy will be doing (assuming no bombshells), you can call the election as a Democratic sweep similar to 2008 just from voting patterns

I think that the major reason that Democrats who turned out in 2008 might not in 2012 is that there's no Messiah-like buzz around Obama this time around. He's no longer the tabula rasa that liberal groups can envision their own agendas on, nor can he run as credibly by promising to fix everything-- we've got four years of watching him struggle to do so. The economy is still bad, and we're not coming off of four years of George W. Bush. Obama may still win re-election, but he's no longer a fresh-faced outsider, and that changes things.

Simplicio
05-08-2011, 02:44 PM
So more Dem voters then Republicans voted in an election where the Dems won by a large margin, but in an election where the GOP won by a large margin the situation was reversed?

Thats not exactly ground-breaking.

Profound Gibberish
05-09-2011, 10:09 AM
Thanks for the link. Very interesting article.

My take on the Repubs is that their focus on "purity tests" and need for all to tow the same line no matter how absurd will ultimately hurt their ability to bring in the middle ground. They are hostage to the extreme fringe (and pretty much have been for decades), while I find that the Dems will take anyone under their tent.

furt
05-09-2011, 10:43 AM
Thanks for the link. Very interesting article.

My take on the Repubs is that their focus on "purity tests" and need for all to tow the same line no matter how absurd will ultimately hurt their ability to bring in the middle ground. They are hostage to the extreme fringe (and pretty much have been for decades), while I find that the Dems will take anyone under their tent.
Exactly. The dems are soooo much more willing to nominate candidates that, for example, differ from the party on abortion, or support school vouchers, or who oppose affirmative action.

Profound Gibberish
05-09-2011, 11:08 AM
Exactly. The dems are soooo much more willing to nominate candidates that, for example, differ from the party on abortion, or support school vouchers, or who oppose affirmative action.

That is not what I said. I said that I believe Dems are more accepting of differing views and allowing "non-believers" to participate meaningfully in their party.

Jas09
05-09-2011, 11:28 AM
Exactly. The dems are soooo much more willing to nominate candidates that, for example, differ from the party on abortion, or support school vouchers, or who oppose affirmative action.Well, let's look at one of those issues - abortion:

Pro-Life Dems in the House and Senate: 8
Pro-Choice Republicans in the House and Senate: 4 (all in the Senate, IIRC, and one is technically an IND now)

Pleonast
05-09-2011, 12:13 PM
The Pew typologies are interesting reads, but I'm never convinced they're objective portrayals of reality. I have the feeling that the groups are divided by rather arbitrary lines in the sand. I'd like to see numbers about variability within each type and a sensitivity analysis of the effects of small changes in how the types are defined.

Profound Gibberish
05-09-2011, 12:46 PM
The Pew typologies are interesting reads, but I'm never convinced they're objective portrayals of reality. I have the feeling that the groups are divided by rather arbitrary lines in the sand. I'd like to see numbers about variability within each type and a sensitivity analysis of the effects of small changes in how the types are defined.

If you really want to geek out on stats then check out http://fivethirtyeight.blogs.nytimes.com/

Pleonast
05-09-2011, 01:04 PM
I read Nate Silver regularly. :D

I'd love to see him do an analysis of the Pew typology.

Profound Gibberish
05-09-2011, 01:07 PM
I think it is highly likely he will. Stay tuned.

Exapno Mapcase
05-09-2011, 01:31 PM
The Pew typologies are interesting reads, but I'm never convinced they're objective portrayals of reality. I have the feeling that the groups are divided by rather arbitrary lines in the sand. I'd like to see numbers about variability within each type and a sensitivity analysis of the effects of small changes in how the types are defined.

You've got to tiptoe very carefully when you try to make the case for a typology that divides voters into fewer groups than zodiac signs. But what's the alternative? You're already down to groups of about 10%. If you try to halve that percentage how much information do you gain? Is there a point between 10% and individuals that is meaningful?

The value of Pew is in the ways that it undercuts the standard line of Democrats, Republicans, and Independents and shows that those groupings don't line up well with voting records or with opinions about issues. I found the big reveal to be that there doesn't seem to be a middle block of Independents, just an amorphous disgruntled mass. It's not clear from these answers how either party captures them or how to use them to break the impasse between the two extremes that can't agree on anything.

There's a good deal of evidence that the group that doesn't line up for parties early waits until the last minute - the last days of the campaign or even in the voting booth - to make up their minds. I'd say that Pew explains why, no matter how frustrating that may be to those who do commit early.

There are 18 months, 18 endless months, to accomplish the two things that swing elections: get the faithful to actually turn out and vote and give the unpersuaded a reason to vote for your side. Almost everything we'll hear from now until November 2012 will be one or the other. For the majority of people who are paying attention now, that's all noise. (I read Nate Silver, too, of course.) But for the people running campaigns it's life and death. That's the only thing that makes it interesting.

septimus
05-09-2011, 02:10 PM
The Pew typologies are interesting reads, but I'm never convinced they're objective portrayals of reality. I have the feeling that the groups are divided by rather arbitrary lines in the sand. I'd like to see numbers about variability within each type and a sensitivity analysis of the effects of small changes in how the types are defined.

From their own report:
Cluster analysis is not an exact process. Different cluster solutions are possible using the same data depending on model specifications and even the order in which respondents are assessed. Several different cluster solutions were evaluated for their effectiveness in producing cohesive groups that were sufficiently distinct from one another, large enough in size to be analytically
practical, and substantively meaningful. While each solution differed somewhat from the others, all of them shared certain key features. The final solution selected to produce the political typology was judged to be strongest from a statistical point of view, most persuasive from a substantive point of view, and was representative of the general patterns seen across the various cluster solutions.

I wonder if they'd provide the raw data so others can run their own clustering methods? In any event, cursory examination of the tables should make clear that there are indeed strong non-trivial correlations in play.

Pleonast
05-09-2011, 02:12 PM
Is there a point between 10% and individuals that is meaningful?
This is the sort of question I'd like to see answered. Even more, I'd like to see some justification that any grouping is meaningful. Maybe the reality is voters are on some multi-dimensional continuum that confounds any meaningful grouping. Or maybe some groups are meaningful and others are not.

Pew does make a good effort, certainly better than anything else. But I'd like to see some numbers indicating how good it really is.

~ ~ ~ ~ ~

Instead of grouping voters into types, where perhaps many voters only loosely fit to a group, I think it may be more useful to identify "poles" on the political map. Then "maps" could be developed to show voter populations near or far from each pole. We could answer questions about which types are closer to which other types. If a politician changes their stance in a particular "direction", who are they getting closer to, who are they distancing from?

L. G. Butts, Ph.D.
05-11-2011, 01:50 PM
I would like to point out to the viewers of this thread that there is a poll here on the dope to see where members lie with regards to their typology.

See here: SDMB Political Typology (http://boards.straightdope.com/sdmb/showthread.php?t=608009)

Bryan Ekers
05-12-2011, 01:30 PM
I would like to point out to the viewers of this thread that there is a poll here on the dope to see where members lie with regards to their typology.

Gee, wouldn't it be more useful if they told the truth?

BrainGlutton
05-12-2011, 02:12 PM
You normally expect a lower turnout across-the-board for non-Presidential elections. But the two conservative groups actually showed a higher turnout in 2010 than 2008. (Libertarians and Disaffecteds also voted more Republican but the total percentage who voted went down as expected.)

I think all of us here already knew that, and why.

BrainGlutton
05-17-2011, 08:54 AM
Points of particular interest in this typology:

1. "The most visible shift in the political landscape since Pew Research’s previous political typology in early 2005 is the emergence of a single bloc of across-the-board conservatives. The long-standing divide between economic, pro-business conservatives and social conservatives has blurred. Today, Staunch Conservatives take extremely conservative positions on nearly all issues – on the size and role of government, on economics, foreign policy, social issues and moral concerns. Most agree with the Tea Party and even more very strongly disapprove of Barack Obama’s job performance. A second core group of Republicans – Main Street Republicans – also is conservative, but less consistently so."

2. Staunch Conservatives are the oldest of the typology groups.

Who they are: More than nine-in-ten (92%) non-Hispanic white and 56% male. The oldest of the groups (61% ages 50 and older). Married (79%), Protestant (72%, including 43% white evangelical), and financially comfortable (70% say paying the bills is not a problem).

Lifestyle notes: Many are gun owners (57%) and regular churchgoers (57% attend weekly or more often), and fully 81% are homeowners. More watch Glenn Beck (23%) and listen to Rush Limbaugh (21%) than any other group.

3. The youngest typology group is . . . not the Solid Liberals . . . not the Libertarians . . . but the Post-Moderns.

Post-Moderns
13% OF ADULT POPULATION /14% OF REGISTERED VOTERS

Basic Description: Well-educated and financially comfortable. Post-Moderns are supportive of many aspects of government though they take conservative positions on questions about racial policy and the social safety net. Very liberal on social issues. Post-Moderns were strong supporters of Barack Obama in 2008, but turned out at far lower rates in 2010.

Defining values: Strongly supportive of regulation and environmental protection. Favor the use of diplomacy rather than military force to ensure peace. Generally positive about immigrants and their contributions to society.

Who they are: The youngest of the typology groups (32% under age 30); a majority are non-Hispanic white (70%) and have at least some college experience (71%). Nearly a third (31%) are unaffiliated with any religious tradition. Half live in either the Northeast (25%) or the West (25%). A majority (58%) lives in the suburbs.

Lifestyle notes: 63% use social networking. One-in-five (20%) regularly listen to NPR, 14% regularly watch The Daily Show, 10% read the New York Times. 31% trade stocks and 53% have a passport.

So, that's the wave of the future -- the political center-of-gravity, if it has one, of Generation Y. (http://en.wikipedia.org/wiki/Millennial_generation) (Of course the same generation also includes millions of Americans more likely to be found among the Dissaffected, the Bystanders, and of course the Hard-Pressed Democrats, among others.) And it's . . . an amazingly commonsensical world-view, these Post-Moderns seem to have.

Exapno Mapcase
05-17-2011, 04:41 PM
I think all of us here already knew that, and why.

AFAIK, the standard expectation for a midterm election is that voting levels drops among all groups. My expectation for 2010 would have been that both the Obama-leaning voters and the anti-Obama-leaning voters would have dropped, but the antis would just have dropped less. For me, the finding that the percentage increased, from a base of over 90% yet, is quite surprising.

In fact, it's staggering. You must not have a sense of what the numbers are. About 85 million voted in 2010. Over 180 million voted in the 2008 Congressional elections. So left than half as many people voted overall and one group representing a tenth of the population increased its percent? That may be unprecedented.

The problem is finding comparable numbers from earliest elections. None of the previous Pew typologies is from an equivalent year. 1999 is close but they don't have the same breakdown chart for voting. If you have any numbers that say that this has ever happened before I'd like to see them.

BrainGlutton
05-17-2011, 05:19 PM
In fact, it's staggering. You must not have a sense of what the numbers are. About 85 million voted in 2010. Over 180 million voted in the 2008 Congressional elections. So left than half as many people voted overall and one group representing a tenth of the population increased its percent? That may be unprecedented.

Yes, and we know why. There was no significant worsening of the economic situation from 2008 to 2010, nor any game-changing events but the health-care reform issue. The success of the Tea Party movement and all the rest was pure political/cultural backlash. I doubt they can sustain the momentum.

Peremensoe
05-17-2011, 05:39 PM
I found it was possible to get (at least) three different results on the test for myself, just based on the different random question selections and slight variations in my attitude about how to interpret certain wordings. Every answer I gave was true, within the limits of the test. (BTW, you can refuse to answer some questions and you'll still be sorted.)

Granted, I am vast and contain multitudes. Others may be easier to pin down. But it sure doesn't give me any confidence in the methodology.