Sunday, June 22, 2014

Are We That Polarized? Or Have We Just 'Sorted' Ourselves?

When discussing American politics nowadays, it's hard to get a word in edgewise before the topic of "polarization" rears its head and becomes the dominant buzzword of the conversation.

We look at the U.S. Congress and see an institution, in both the U.S. House and Senate, where the parties are further apart then they were right after the end of Reconstruction.  In fact, we’re nearing the end of any known moderates in the Congress who reach across the political aisle to the opposition; if they do, they are endangered of being ‘primaried’ by members further to the right or left of themselves. 

We see voters who self-identify with one party or the other vote for their party’s candidates over 90% of the time, with the era of the split-ticket voter now nearing its end.  In fact, strong partisans are the ones most likely interested in following political campaigns, while those voters who self-identify as independent aren’t really all that interested in campaigns. 




Commentators contend that the edges of the two parties, the “wings” of both the GOP and Democratic parties, have taken over the discussion and debate, with no real sense of compromise within this era of partisan conflict. 

Political scientists have gotten into the debate, with two camps firmly entrenched into whether America is polarized: some say it’s not polarization, but rather “political sorting” that has made it look like we’re polarized, and that there is indeed a political center to American politics.  Another camp argues sorting doesn’t explain it all, and that yes, we are indeed polarized against one another due to party loyalty to the political camps we belong to.

Enter into this continuing debate a recent report by the Pew Research Center that has sent both sides of the debate into critical analysis of either “ah-ha, see” or “that’s not what is truly happening.” 

Pew’s report on polarization showed that the median Democrat and Republican are more ideologically divided than in the past, and that the edges of the ideological spectrum have been filled with “consistently minded” individuals that has grown over time, while the “moderately mixed middle” has dropped. 

In 1994, the distribution of responses to the Pew’s scale of political values was pretty much a bell curve diagram; now, in 2014, that bell curve has flattened in the middle, while rising on the ends. 




What Pew determines is that Republicans have not only shifted to the right, while Democrats have shifted to the left, but that the ‘consistent’ sides of the political aisle look at each with disfavor, thereby increasing the polarization within our political landscape.

Those on both sides of the polarization debate have taken the Pew findings to task; for example, Norm Ornstein, a noted analyst of Congress, argues that it’s not both parties that have moved into their respective corners, but rather the polarization is asymmetrical, with Republicans moving much further than Democrats have.

Alan Abramowitz, a political scientist at Emory University, argues that the racial divide between the parties, along with the fact that the political center has shrunk and the poles at either end of the political spectrum have grown, leads one to believe that polarization is at the heart of the matter. 

On the other hand, political scientist Matthew Dickinson contends most observers of Pew’s report only scratched the surface, and that if they had gone deeper, they would have discovered that “majority [of respondents] do not have uniformly conservative or liberal views. Most do not see either party as a threat to the nation. And more believe their representatives in government should meet halfway to resolve contentious disputes rather than hold out for more of what they want.”

As Dickinson contends, Americans have ‘sorted’ themselves into their respective political camps, so that more liberals consider themselves Democrats and more conservatives consider themselves Republicans.  But that there’s still a viable middle that wants to see both parties work together. 

So, are we sorted or are we polarized? 

First, look at how the electorate views itself. Using the 2012 National American Election Study (ANES), we can break down the electorate into 7 general categories, ranging from “strong Democrat” to “independent” to “strong Republican.” 

Voters have distributed themselves along the partisan spectrum in what appears to be an upside-down bell curve, with the “strong partisans” at either end making up sizeable segments of the electorate, with 14% of pure independents in the middle.



And while the “independent-Republicans” and “independent-Democrats” could be combined with their pure independent brethren to make up a sizable middle, those independent-partisans vote at nearly the same levels as their strong partisans when it comes to casting ballots for their party’s candidates.

So, less than 15% of the 2012 electorate identify themselves as not only open to voting for either party, but are the most consistent over time in switching their votes back and forth (in 2012, they broke for Obama 54-46%). 

So, if voters have indeed ‘sorted’ themselves into partisan preferences, are they ideologically more cohesive in defining themselves as pure partisans or not?

In other words, if you ask someone who is “extremely conservative” or “extremely liberal” or even “moderate/middle of the road,” where would they classify themselves in terms of party affiliation, especially when it comes to the ‘strong partisans’ versus the pure independent?

We can find out that answer from the 2012 ANES data as well.

Note: Not Very Strong & Lean-Independent Partisans are not included in this analysis.

Both extremely liberal and extremely conservative respondents share something in common: the vast majority consider themselves strong partisans.  In fact, among those who simply consider themselves either liberal or conservative, significant numbers (47% for conservatives and nearly 60% for liberals) identify themselves as strong partisans in what we would expect (that is, liberals are strong Democrats while conservatives are strong Republicans). 

While there appears to be some voters who are extremely liberal and are strong Republicans and others who are extremely conservative but are strong Democrats, it would appear that the ideologically driven amongst us have indeed sorted themselves into the respective political camps as strong partisans.

In fact, one could argue that over the past forty years, an ideological “sorting” has come about, due to strong ideologues finding a home within the political parties.


The data comes from the yearly American National Election Studies survey of respondents in the years 1972-2012 (no data is available from 2006, 2008, or 2010). 
The above gif shows three columns of responses from 1972 to 2012:

  • blue, which indicates the percentage of those who identified as "very liberal and liberal" and who identified themselves as Strong Democrats in terms of party affiliation during the time period;
  • purple, which indicates the percentage of those who identified as "moderate/middle of the road" ideologically and who identified themselves as Pure Independents in terms of party affiliation during the time period; and,
  • red, which indicates the percentage of those who identified as "very conservative and conservative" ideologically and who identified themselves as Strong Republican in terms of party affiliation, during the time period.
In 1972, among “very liberal/liberal” identifiers, one third said they were strong Democrats; four decades later, nearly 60% of those very liberal/liberal identifiers said they were strong Democrats.

Among “very conservative & conservative” identifiers, 28% in 1972 said they were strong Republicans; forty years later, nearly 50% said they are strong Republicans. 

And it appears that it’s just not the liberal or conservative ideologies who have found their respective homes within the two major parties: moderates have become more ‘independent’ as the parties have become home to the strong ideologues.




If you then look within the party supporters and their ideological tendencies, you’ll find adherence to ideology that is associated with party identification.

Within each party’s strong adherents, we see intense affiliation to either end of the ideology spectrum: in 2012, 42% of strong Democrats said they are liberal or very liberal, while nearly 75% of strong Republicans said they are conservative or very conservative.

Compare those ideological identifications to when the same survey question was asked forty years ago: in 1972, only 31% of strong Democrats said they were either “very liberal or liberal,” while only 38% of strong Republicans said they were either “very conservative or conservative.”

Meanwhile, the pure political independent has become even more middle of the road moderate, going from 42% in 1972 to 64% in 2012. 

So if we see that sorting has occurred due to ideology and partisanship, can we see if there’s polarization going on as well? 

As a starting point, what is polarization?  According to one set of scholars, polarization can be viewed as both a ‘state and process:’ as a state, polarization “refers to the extend to which opinions on an issue are opposed” while a process, polarization means “the increase in such opposition over time.” 

In looking at different issues and the opinions that strong Democrats, strong Republicans, and pure independents have, we can see a variety of levels of polarization at work from the 2012 electorate (again, using ANES data):

On whether government or private medical insurance should be provided:



On whether the federal government should make it more difficult to buy a gun:



On gay marriage:



On whether newer lifestyles are breaking down society:



On whether the world is changing and we should adjust to those changes:



On abortion:



For these issues and values, there is a clear difference between what strong Democrats and strong Republicans hold, with independents typically in the middle.

On other issues, the differences may not be as stark, as on:

The death penalty:



Taxing millionaires:



Global warming:



On all the policies and perspectives, however, there appears to be real differences between the way that the strong partisans view and support, or oppose, issues.

In fact, there are real differences to how partisans look at the opposition party, both in terms of how they view their opposition in ideological terms and whether they simply like or dislike their opponents.

Based on ANES Data from the 2012 election, we find the ideologies within those who identify as a Democrat (combined strong Democrat, not very strong Democrat, and independent-Democrat), Republican (combined strong Republican, not very strong Republican, and independent-Republican) or pure Independent:



It is obvious, from the above data, that Republicans are much more "conservative" in their ranks than Democrats are "liberal" (you have to add extremely liberal, liberal, and slightly liberal within the Democrats to equal the same proportion of extremely conservative and conservative within the GOP).

In asking partisans (strong & not very strong) and independents (leaners and pure independents) on how they view each party ideologically, both strong partisans tend to see the opposition as more ‘ideological’ in their viewpoint, with Democrats seeing the GOP as more “extremely conservative/conservative” and Republicans seeing the Democratic Party as more “extremely liberal/liberal.”





When asked a feeling thermometer towards both parties (with 1 being absolute dislike and 100 being absolute like), a pattern emerges with how each party’s strong partisans tends to rate the other party.






So do we see sorting or polarization as the culprit of the dysfunction of the recent American political landscape?  I think the answer is yes to both: sorting has occurred, and polarization has occurred as well. The question becomes: how do we deal with this sorting and polarization in a system of government that inherently creates conflict but forces compromise in order to achieve anything? 

That may be an answer that requires all perspectives to recognize the fact of our current system, and then deliberately decide to ignore the environment and do what is best. The problem continues, however: what one would define as "the best" is open to interpretation by both parties, the elites/elected officials, and the voters themselves.