I recently posted the graph of my estimates of the two parties’ congressional candidates. In that post, I wanted to emphasize that moderation still exists, even in this polarized age. To highlight that point and make the plots prettier, I smoothed out the distributions.

However, that smoothing hid another very interesting take-home point from the 2012 candidate scores. There appears to be evidence of bimodality (two peaks) not only across the parties—that’s good old polarization—but also within the parties. Here are the unsmoothed plots that make that clear:



No, those aren’t Halloween ghosts. It looks like both parties have two distinct wings, a moderate one and an extreme one. This visual inspection is backed up by test statistics from the Hartigan dip test for unimodality.

Feel free to download the estimates for all the 2012 congressional candidates here. The explanation of how I generated them is here.

We haven’t seen this before in roll call-based ideal point estimates, and I don’t think I’ve seen it before in previous years’ survey estimates (this is something I need to go back and check). So this could be something new under the political sun.

What could be causing this? Perhaps new electoral forces like the Tea Party on the right and Occupy Wall Street on the left are forcing candidates to pay lip service to dogma in some new way. And what happens after the election? Will this internal schism go away? Or does this presage a new battle between liberal liberals and liberal moderates, and between conservative conservatives and conservative moderates?

Your guess is as good as mine, though. Any ideas?


Here are two graphs representing the distribution of 2012 US House and Senate congressional candidate ideological positions. Higher (more rightward) scores are more conservative, lower (more leftward) scores are more liberal. Click on the plots for higher resolution versions:



A couple of things can be seen clearly from these two pictures:

  1. There are two distinct distributions of scores, representing the two political parties. They are distinct; or, in other words, the parties are ideologically polarized. Democrats are liberal, and Republicans are conservative.
  2. There is a significant amount of overlap between the party bell curves. That is, there are plenty of conservative Democrats who are more conservative than a number of liberal Republicans (and vice versa). Even in an age of polarization, the candidate pool is not completely divided, unlike Congress in recent years. This replicates a finding about the Congress of the mid 90s by Stephen Ansolabehere, Jim Snyder, and Charles Stewart from over a decade ago.
  3. On average, Senate candidates are slightly more centrist than House candidates. This makes sense given the larger, more heterogeneous states that they seek to represent, relative to the smaller and more extremist House districts.
  4. It appears the candidate pool of the parties in 2012 is roughly symmetrically polarized.


  1. These scores are based on candidate positions expressed in survey responses, campaign statements, web sites, etc., as compiled by Project Vote Smart.
  2. They represent 722 House candidates from 419 districts and 64 Senate candidates from 33 states with elections this year. Not all candidates were scored because of a lack of data, but it’s a small number in that position.
  3. I have jointly classified all candidates into a common space, which simply means that House and Senate scores are comparable.
  4. More details about how I generated these scores can be found in a companion post that I wrote to keep this one more lean.
  5. The underlying scores are preliminary and subject to change, but I’m making them available to anyone interested in the name of transparency in another companion post here.
  6. You can find out more about my research on legislative ideology here.

Political scientists have been trying to summarize politicians’ ideological preferences for a long time. The most well accepted version of these are called ideal point estimates. These are measures of inherently unobservable preferences that are estimated from observed behavior. I see you voting in favor of a higher minimum age and regulating carbon dioxide as a pollutant, and I infer you’re probably a liberal. Or maybe you vote in favor of the Canadian oil pipeline as well as against “Obamacare” and I think you’re probably a conservative. As a sign they’ve hit the (nerdy) big time there’s now even a great XKCD comic about Keith Poole’s and Howard Rosenthal’s DW-NOMINATE scores.

The observed behavior that is most commonly employed are the Yea and Nay votes taken on roll calls in legislatures like Congress. These are very attractive to use as the raw data for ideal points for many reasons, one of which is that there is almost always an embarrassment of data. I’ve used them extensively in my research; here is a paper I recently published with Nolan McCarty on state legislative roll calls.

But they’re not perfect, for two reasons. First, a candidate at election time may present a different platform to voters than he actually uses as a guide to voting on roll calls once he achieves office. Second, by definition, they are only available after an election. This means we can’t get information on the losing candidate in state or district. This is a much more serious problem than the first.

An attractive alternative observable data is the candidate survey. In my opinion, the best candidate survey these days is administered by Project Vote Smart. It has been in the business of surveying tens of thousands of federal and state candidates for office since the mid 1990s. The questions it asks are numerous, well-phrased, and stretch across nearly all of the contentious political terrain you’d want them to. The results of their survey, which used to be called the National Political Awareness Test (NPAT) and is now the Political Courage Test (PCT), is published in a variety of formats for voters to use. The idea is that this makes it easier for voters to find out information on the policy preferences of candidates of whom they might otherwise know very little. The organization appears to be without a hint of partisan bias, as a nice bonus.

There’s another problem, one you might have guessed. Not every candidate answers the survey; in fact, fewer and fewer candidates do as time goes on. Many obviously feel that doing so could be an electoral liability now or in the future; better instead to refuse to be pinned down on many questions of policy specifics.

So Project Vote Smart figured out a solution in 2010 and now again in 2012. It would research answers to a subset of their candidate survey using good old fashioned research brawn. So nearly all of the congressional candidates in 2012 for nearly all of the congressional districts and all the states that are having elections to the House of Representatives and the Senate are represented in their 2012 Vote Easy tool. The tradeoff for this broad coverage is that only a small subset of policy stances could be researched for the many hundreds of candidates this year.

I’ve built on their work by merging their deeper but narrower NPAT with the smaller but broader Vote Easy. This gives us the best of both worlds. And the most important step is to estimate ideal points from this merged survey data. I’ve done this using a Bayesian two-parameter, one-dimensional item response model, implemented in the R statistical environment with Simon Jackman’s invaluable pscl package and visualized with Hadley Wickham’s powerful ggplot2 package.

How valid are these scores? One way to assess their external validity is to assess their convergence with measures taken from unrelated data. Luckily for me, just such an external data source exists in the form of Adam Bonica’s candidate scores for 2012. Bonica’s candidate scores correlate with my own at a level of r=0.88, which is quite high, especially as both of our measures are measured using no data in common and different estimators. The advantage of my method, though, is that it allows me to jointly classify candidates and voters, something I’ll be returning to in my blog in the coming days before the election.

For more technical details, you can consult a paper I cowrote on congressional voting with Jon Rogowski in part by using this data amalgam. You can find out more about my research on legislative ideology here.

Normally, I write something in 2012 for publication in 2013-2014 about what happened back in 2008 or 2010. Interesting, but not as much fun as it could (should) be. So, without further ado, here are the results of my exercise. Here are the plots of the two parties in 2012, and here are the underlying scores.

Big thanks to Chad Levinson, a political science PhD candidate at the University of Chicago, for helping me gather the survey data from Project Vote Smart.

Click here for my scores for the 2012 House and Senate congressional candidates.

Graphs of the distributions can be found in this post, and an explanation of how I came up with these scores is in this post.

The fields in the spreadsheets are as follows:

  • stdist: Congressional district for House candidates
  • st: State abbreviation
  • party: D, R, or X (independent)
  • pid: –1,0,1 (equivalent to party)
  • full.name: Self-explanatory; sorry for screwups with accent marks and the like.
  • incumbent: 1 if incumbent, 0 if challenger
  • crp.id: Center for Responsive Politics identification number
  • npat.id: Project Vote Smart candidate id
  • score: Candidate ideal point or ideological position estimated from survey response as described here
  • sd: Measure of uncertainty around the point estimate in score
  • perc: Percentile ranking within the pool of all 2012 candidates, House and Senate. So a percentile score of 84.5 for  Mia Love (R) in Utah’s 4th District indicates Love ranks as more conservative than 84.5% of all 2012 candidates.
  • perc.r: Percentile ranking within the pool of 2012 Republican candidates, House and Senate. So Love scores 70.3, which indicates she is more conservative than 70.3% of all 2012 Republican candidates: that is, she is certainly quite conservative, even within her own party.
  • perc.d: Percentile ranking within the pool of 2012 Democratic candidates, House and Senate. Love’s opponent, Jim Matheson (D) with a percentile score of 1.6, indicating that he is more conservative than all but 1.6% of 2012 Democratic candidates. In other words, Matheson is extremely conservative for a Democrat, which is not surprising given the conservative character of Utah’s 4th district.

Last week, I explained how a bunch of moderate and liberal Republicans getting elected is consistent with a larger story—my expectation that the new 112th Congress will be the most polarized yet, even more so than the record-setting 111th.

How will that happen? On this election day, I’ll detail one path that has received relatively little attention.

While everyone focuses on the Republicans’ pickup opportunities—of which there are many in the Senate and the House—fewer observers have taken a cumulative look at the consequences of open seats won by the incumbent party. That is, when a Republican replaces a retiring (or defeated) Republican, and similarly for Democrats. Since not all partisans are alike ideologically, it behooves us to examine more closely who is replacing whom.

I will address Senators today, and only Republicans. I will focus in this post on the latter for a very simple reason. Because of the coming Republican wave, only three Democratic open seats exist which other Democrats have a (virtual) lock on. These are Connecticut, Delaware, and West Virginia. The former are both liberal states, but I have no prior legislative voting behavior to predict the likely ideologies of Richard Blumenthal and Chris Coons, respectively, with respect to the incumbents Chris Dodd and Joseph Biden/Ted Kaufman. Nor can I tell if WV Governor Joe Manchin will be more or less liberal than Carte Goodwin, Robert Byrd’s replacement by Manchin himself.

So are the new Republican Senators going to be more conservative in the 112th Congress? Based on evidence culled from my research, and a little speculation, I will say yes with a large degree of confidence.

To proceed, I will divide Republican open seats going Republican (ROSGR) into five categories, based on the state of evidence that the seat is becoming more conservative, or not.

In the first category are those ROSGRs that I have strong evidence are becoming more conservative. This is because of the well-known finding in political science by Keith Poole and Howard Rosenthal that politicians “die in their ideological boots” (see here for an ungated version of Keith’s paper). That is, politicians rarely change their stripes. Thus, we can look at their prior voting records in other legislative chambers—federal and state—to predict their Senatorial ideology.

  1. Florida. Incumbent Mel Martinez retired and was replaced by George LeMieux (via a pick by Governor Crist). Both were moderately conservative; about three quarters of congressional Republicans were to their right. Marco Rubio, former speaker of the Florida House of Representatives and victor against Crist in the Republican primary, is a few clicks to their right (but less than I’d expected).
  2. Missouri. Retiring Kit Bond is a moderate; he’s about as conservative as LeMieux or former Senators Spencer Abraham or Pete Domenici. Roy Blunt is the current Representative and minority whip from the 7th Congressional District in Missouri. He’s also the likely victor on Tuesday. He’s compiled a voting record that’s 20 percentiles more conservative than Bond, about in the range of Mike Enzi of Wyoming.
  3. Ohio. Retiring George Voinovich was one of the most liberal Republicans in the Senate. Representative Rob Portman (OH-2) , his nearly-guaranteed successor, is 20 percentiles more conservative. His ideology is about the same as that of former Senator and majority leader Bill Frist.

In the second category, we have ROSGRs where I have informed speculation and a strong degree of confidence (but not nearly so much as the first category where I have evidence) that will be more conservative than the previous incumbents. Both are the open seats resulting from the defeat of incumbents in primaries, on grounds that the latter were too liberal.

  1. Utah. Senator Bob Bennett is in the moderate half of congressional Republicans; that’s still pretty conservative by national standards. Still, Utah is one of the most—if not the most—conservative states in the union. It can electorally support a much more conservative Senator. While I have no voting record for Mike Lee, every bit of evidence from the Republican primary tells me that he’s likely to be far more conservative.
  2. Alaska. Assuming Joe Miller wins, that means he’d replace Lisa Murkowski, a decidedly moderate Republican. The primary and general campaigns have clearly staked ideological territory where Miller is to the right of Alaska. As in Utah, Alaska is one of the most conservative states in the US; it can easily support a more conservative Senator.

In the third category, we have a ROSGR where I’m making an informed guess with less confidence than the previous category. There’s only one of these:

  1. Kentucky. Jim Bunning is a pretty conservative Republican; only about a fifth of congressional Republicans are more to the right. That causes me to move this race out of the last category. On the other hand, Republican nominee Rand Paul is running a distinctly ideologically conservative and libertarian campaign. This leads me to believe he’d be more conservative than Bunning, but I’m not as sure.

In the fourth category, we have a ROSGR where I have evidence that there won’t be much change in ideology after the incumbent leaves. There’s only one of these.

  1. Kansas. Despite his reputation, Sam Brownback is about in the middle of his party for conservatism, and is retiring to run for governor. Representative Jerry Moran (KS-1) is a lock to win Brownback’s seat, and he’s just about as conservative.

In the last category, we have the Republican counterpart to Delaware and Connecticut: a race where I have no evidence nor good speculation about how the winner will compare to the retiring incumbent.

  1. New Hampshire. Judd Gregg is a moderate Republican, but that’s probably about the carrying capacity of the state for conservatism. Kelly Ayotte is the former state attorney general, and not much can be gleaned from the campaign (apart from her defeat of more Tea Party-favored Ovide Lamontagne).

So, to sum up, of the eight open seats being vacated by Republicans and likely taken again by Republicans, 5 will become more conservative with a high degree of confidence, 1 will become more conservative with a medium degree of confidence, 1 will stay the same, and 1 is unclear.

Moreover, I didn’t include Pennsylvania’s Arlen Specter, who quite clearly would have been an example of a ROSGR had he remained a Republican. He was the most liberal Republican in the Senate at the beginning of the 111th Congress–more liberal than either Susan Collins or Olympia Snowe. Had he stayed a Republican, he would most likely have been defeated in a primary by Representative Pat Toomey (PA-15), which would have set up the same general election we are seeing today. Likely victor Toomey is very conservative; only 20 percent of Republicans are more to the right. In fact, I see him as more conservative even than Rick Santorum. So, depending on how you look at this case, the case for ROSGRs going more conservative is even stronger.

All of this is one reason why the 112th Republicans will be the most conservative, and the 112th Congress will be the most polarized yet. This is the “dog-bites-man” story; while the “man-bites-dog” story of the coming liberal House Republicans is merely an entertaining sideshow.

My coauthor and friend Andrew Gelman, in the course of plugging my previous post about incoming Republican moderates and liberals in the 112th Congress, asks me:

There’s only one thing I wonder about. Even if everything Boris writes is correct–and I have no reason to doubt him–he’s still only coming up with 10 moderate Republicans, out of a total of 200 or so. That’s not a lot.

Well, I found two more (Debicella in CT-4 and Bruun in OR-5), so it’s now 12. Of course, not all of them will get elected, so only a fraction of those will get a chance to be the fightin’ moderates of the Republican Class of 2010.

Congress has been getting more ideologically polarized since the 1950s, with a dramatic rise since the mid-1970s. This means that Democrats are becoming ever more liberal, and Republicans are becoming ever more conservative. In the 110th Congress, both the House and the Senate are more polarized than they have ever been (since the Civil War and Reconstruction). The 111th was more polarized still. Here’s Nolan McCarty’s plot of Congressional polarization:


I predict the 112th will be the most polarized yet. That’s not saying much; the trend has been very strong. Or in other words, the story of increasingly conservative Republicans is “dog bites man.” However, in a wave election, quite a few liberal districts and/or liberal candidates from the winning side will get through. In 2010, that’ll be liberal Republicans; in 2006 and 2008, it was conservative Democrats. That’s “man bites dog,” and therefore newsworthy in its own right (or so I think as an amateur journalist).

Nolan McCarty, Keith Poole, and Howard Rosenthal, in a 2009 AJPS paper identify the advance of polarization by two paths: 1) districts electing representatives more like them (eg, liberal districts electing liberals and conservative districts election conservatives), and 2) for a given district, Democrats becoming more liberal, and Republicans more conservative. The former effect is called sorting and the latter effect is called divergence. Here’s a freely accessible but older version of the paper; the final version is gated here.

Of course, the irony is that these moderate and liberal Republicans will be uniquely vulnerable in the post-wave elections of 2012 and beyond. Djou and Cao, for example, got in under extraordinary circumstance; so will some of the Fightin’ Liberal 12 I identified. That’s the sorting effect which has been cutting down on the number of liberal Republicans and conservative Democrats in the past few decades.

But once they’re defeated, they’ll be replaced by considerably more liberal Democrats. Until then, they are considerably more conservative than the Democrats they have or will have defeated. This is the divergence effect.

Thus, Djou is undoubtedly a liberal Republican in HI-1. But Colleen Hanabusa is much, much more liberal, as shown in her voting record as a Hawaii state senator. If she attains office, she’d be as liberal as Senator Patty Murray (WA) or Senator Patrick Leahy (VT). Thus, conservatives have to face the fact that the alternative to electing a liberal Republican is often electing an even more liberal Democrat.

A special Senate election is being held next Tuesday in Massachusetts to finish Edward Kennedy’s term. The candidates are Martha Coakley (D), and State Senator Scott P. Brown (R).

The election is particularly noteworthy for a number of reasons. First, recent polls show Brown matching or even exceeding Coakley’s electoral support, in one of the most liberal states in the entire country (and one that hasn’t elected a Republican in four decades). See Pollster.com’s aggregation of polls here, and Intrade’s political market for the election here. Second, the consequences of a Brown victory could be the derailment of the Democratic health care reform proposal, if all Senate Republicans maintain party unity.

To my mind, the election is fascinating for another reason. Brown is attracting very positive national and state Republican and conservative attention. On the other hand, State Assemblywoman Dede Scozzafava attracted very negative attention from conservatives in her special election campaign for the 23rd Congressional District of New York.

Brown is actually a liberal Republican who is to be found to the left of Dede Scozzafava! So why, then, the enthusiasm gap in support for the two? This post documents this assertion, and then answers this puzzle.

Citing my ongoing research on ideology in state legislatures in an earlier blog post, I made some waves by arguing that Scozzafava was actually a conservative Republican in a particular context. That context was the New York State legislature, where Republicans are exceedingly liberal relative to the rest of the country. In fact, she was actually located slightly to the right of the average Republican in the legislature. Despite this, there was a firestorm of opposition to her, leading to an insurgent challenge by Doug Hoffman under the Conservative Party label and her subsequent withdrawal from the campaign.

What about Scott Brown? How liberal or conservative is he? We have evidence from multiple sources. The Boston Globe, in its editorial endorsing Coakley, called Brown “in the mode of the national GOP.” Liberal bloggers have tried to tie him to the Tea Party movement, making him out to be very conservative. Chuck Shumer called him “far-right.”

In 2002, he filled out a Votesmart survey on his policy positions in the context of running for the State Senate. Looking through the answers doesn’t reveal too much beyond that he is a pro-choice, anti-tax, pro-gun Republican. His interest group ratings are all over the map. Business and gun rights groups typically rate him very highly, labor and and environmental groups have rated him both middling and high over time. The teacher’s union rated him low in 2001, and high in 2005.

All in all, a very confusing assessment, and quite imprecise. So how do we compare Brown to other state legislators, or more generally to other politicians across the country? My research, along with Princeton’s Nolan McCarty, allows us to make precisely these comparisons. Essentially, I use the entirety of state legislative voting records across the country, and I make them comparable by calibrating them through Project Votesmart’s candidate surveys.

By doing so, I can estimate Brown’s ideological score very precisely. It turns out that his score is –0.17, compared with her score of 0.02. Liberals have lower scores; conservatives higher ones.

Brown’s score puts him at the 34th percentile of his party in Massachusetts over the 1995-2006 time period. In other words, two thirds of other Massachusetts Republican state legislators were more conservative than he was. This is evidence for my claim that he’s a liberal even in his own party. What’s remarkable about this is the fact that Massachusetts Republicans are the most, or nearly the most, liberal Republicans in the entire country!

Plot of state legislative parties.

Of course, while the Republicans here are liberal, Democrats are incredibly liberal. In comparison to them, Brown is a conservative. He was also the most conservative of the tiny handful of Republican State Senators.

Perhaps the most important context in which Brown can be considered a conservative is the electoral one. We’re talking about Massachusetts here, one of the most liberal states in the country, delivering 62% of the vote for Barack Obama, in comparison to 36% of the vote for John McCain. And as liberal as Brown may be, he’d be far more conservative than Edward Kennedy (-.92), or Martha Coakley (no score as she has never been a legislator, nor has she filled out the Votesmart survey – but ACORN has given her its top rating). And the third party candidate here, Libertarian Joseph L. Kennedy (no relation to the famous ones), is not a viable candidate nor is he palatable to mainstream conservatives relative to Brown.

In other words, what began as a puzzle turns out not to be much of oneat all. It makes perfect sense that Scott Brown, a liberal Massachusetts Republican, has attracted Republican and conservative support. He’s perfectly suited for his liberal state electorate. Dede Scozzafava, in fact considerably more conservative than Scott Brown was not nearly so well matched to her intended constituency, the relatively conservative 23rd District that had returned moderate conservative John McHugh since the 1992 election.

What this shows, however, is that the conservative base in the United States, far from dragging their party moblike into an unelectable extreme, has made the decentralized decision to support the realistically best candidate they can relative to the context in which he’s being elected. The 23rd special district election can also be seen in this light; throwing Scozzafava overboard made far more sense in the context of that electorate.

« Previous PageNext Page »