Must Reads: Defending the KSM Trial | Dems Fundraise on Palin 'Lies' | Big Coal's PR Blitz | Chamber: Blocking Swine Flu Prevention? | Qaeda: Not a 'Direct' Threat?

Email Kevin Drum at

Sarah Palin, Tito the Builder, and Me

I have a cameo appearance in an important scene in the Palin book. Guess what? She didn't get it exactly right.

—David Corn
Afghanistan's Firebrand

Former Afghani parliament member Malalai Joya on warlords, women's rights, and failed US policy in her homeland.

—Sonja Sharp
GM's Rainforest Racket

In Brazil, people with some of the world's smallest carbon footprints are being displaced—so their forests can become offsets for SUVs.

—Mark Schapiro

Discuss the latest from Mother Jones with other readers in Comments Central.

Get FREE Mother Jones newsletters in your inbox three times a week.

Email Kevin

Drum's Links

2009
2008

In The Blogs

Chart of the Day

— By Kevin Drum | Fri November 6, 2009 11:33 AM PST

I barely even understand this chart, but it looks pretty cool, doesn't it? It's an analysis of the Senate vote on Tom Coburn's screwball amendment to defund political science research, which failed 36-62. The dark blue and dark red are nay votes, while bright blue and bright red are yea votes. Brendan Nyhan:

Each senator is placed at their estimated ideal point in the ideological space. The diagonal cutting line, which represents the best-fitting line dividing yes from no votes in the space, indicates that the vote reflected both the primary ideological division between the parties (in this case, cutting "wasteful" government spending) and the second "social issues" dimension (feelings toward pointy-headed academics?).

Sure. I guess I'll buy that. More charts for other votes here.

Actually, though, I think I'm more interested in the placement of senators themselves. Democrats are almost all bunched into a single grouping, with only four outliers. Republicans, by contrast, are spread through considerably more space on both the economic and social dimensions. That doesn't seem intuitively right to me, but it strikes me as more complimentary toward Republicans than Democrats. So tell me again why they want to defund pointy-headed political scientists?

36 Comments | Post Comment

Get Mother Jones by Email - Free. Like what you're reading? Get the best of MoJo three times a week.

Better yet, tell me why Evan

Submitted by BGinCHI (not verified) on November 6, 2009 - 12:10pm.

Better yet, tell me why Evan Bayh voted for this ridiculous nonsense.

I'm not sure how they

Submitted by Dan Miller (not verified) on November 6, 2009 - 12:37pm.

I'm not sure how they defined where to place the dots, but if it's based on recent votes, this makes sense. Democrats control the floor, so they bring up lots of bills that unite their caucus and divide Republicans; thus, Republicans look more spread-out because a few are always tempted to defect. The chart might have looked the opposite way if presented in 2005.

Two shocking votes

Submitted by The Grand Panjandrum (not verified) on November 6, 2009 - 12:44pm.

What possible reason could Jim Webb and Claire McCaskill have for voting FOR this amendment?

MO screwy!!

Submitted by Heather (not verified) on November 6, 2009 - 12:46pm.

And Kit Bond voting against????

outlier poli sci prof

Submitted by elisabeth on November 6, 2009 - 12:54pm.

I had lunch the other day with a friend who is tenured in Political Science. I asked him if there'd been a lot of e-chat etc about Coburn's amendment and he said, sure, lots of folks were dismayed, shocked, etc etc. BUT he said, he himself wasn't so sure that funding political scientists is the best use of federal funding -- "the money could be better used by for health and sustainability or it could be given to artists" he said to me, and added that the bulk of the funding goes to a big national study that has been going on for years that studies why people vote the way they do. And why do they, I asked. He said, so far the research says "party affiliation is the strongest predictor."
Surely that's worth every penny that has been spent!

elisabeth

Wonder how that correlates

Submitted by Anonymous (not verified) on November 6, 2009 - 2:52pm.

Wonder how that correlates to Prego/Ragu.

Reality has a liberal bias.

Submitted by Omega Centauri (not verified) on November 6, 2009 - 12:57pm.

As many have been pointing out. They can't afford reality to get through. And the social sciences crowd is pretty strongly liberal, so they are all deserving of a trip to the unemployment line.

It looks like the social sciences are making progress towards good epistemology (careful thinking processes), at least thats the impression I get when I lurk on crookedtimber. And that is the opoosite to kneeejerk emotional ideological thinking, which has become the hallmark of the republicans lately. Their approaches are diametrically opposed.

Republicans have no real idea what they're thinking about

Submitted by cld (not verified) on November 6, 2009 - 1:00pm.

I would suggest the Republicans are more spread out like that because they actually have no real idea what they're thinking about.

All they really know is what they can blather about to get them through the next election.

Clustering and Distance

Submitted by Nick Edwards (not verified) on November 6, 2009 - 2:04pm.

Kevin,

While there's less clustering in the Republican party, I think you need to consider the positions of the clusters as well. The Democratic party is tightly clustered, but in the middle of their half of the circle. The Republican party has a much larger spread on the arc of the circle, but their distance from the middle is a) fairly uniform, and b) much greater than the average Democrat's distance (with the obvious exception of the two Senators from Maine).

It's also fairly interesting that it looks like the median point between all the Senators is not that close to the center of the circle, rather it''s edged over to the right due to the weight of the majority Republican Senator being jammed up against the edge of the spectrum. (We may not have a center-right country, but we sure as heck have a center-right Senate).

To me the graph looks like a Democratic party that is almost uniformly clustered around an mildy progressive position with no one that could be considered particularly Liberal or Left, contrasted with a Republican party that's exploring the reaches of the lunatic fringe. That sounds a lot like reality to me. Almost every Democrat in major election cycles has run on a fairly mild version of liberalism, some government services, some civil/minority/womens rights, a little environmentalism, whereas Republicans have run on all kinds of flavors of crazy, Neocons, FYIGM Capitalists, Theocrats, etc.

Once upon a time you could probably have picked out the socialists, from the tree huggers, from the minority rights folks in the American left, but most Dems now seem to fall into that general vaguely left-leaning category.

As far as I can tell, the

Submitted by Muscat (not verified) on November 6, 2009 - 2:06pm.

As far as I can tell, the distribution says the following:

Senate Democrats vary from moderate to moderately liberal (the vast majority being moderately liberal) with general (and apparently moderate) agreement on social issues. They're mostly bunched up in the middle of their "side" of the graph with 4 outliers that are more moderate and are also on the Democratic extremes in terms of "social issues."

Senate Republicans vary from moderate to extremely conservative (the majority being extremely conservative). So Senate Republicans are more diverse but mostly because they trail off from being bunched up on one side of the graph, which makes them look more extreme. Which isn't necessarily complimentary, though it might be something they'd like to show off to their tea bagger constituents. There's also less agreement on social issues though about the same breadth of range as Democrats.

...though I'm also not sure what "Social Issues" really represents other than a catch-all "other" dimension given that I can't come up with something that would make Claire McCaskill and Kit Bond opposite outliers. Perhaps someone from MO could fill us in. Or perhaps that's not the right way to read it.

What are the axes again?

Submitted by Anonymous (not verified) on November 6, 2009 - 2:21pm.

This chart seems more complicated than it needs to be--I have no idea what that circle means, for example, unless it's just chart junk. But what calls the whole exercise into question for me is what's supposed to be represented on the y axis, "Social Issues". With an ordinary scatter plot you can project all the data points onto one axis and get some notion of how the points line up with respect to that variable. If we do that for social issues we see that tight cluster with a few outlier for Democrats, as Kevin notes, but also that there are Republicans not just on one side but on *both* sides of that Democratic cluster. What kind of measure for social issues would produce that pattern? Pick any current hot-button social issue, and Republicans line up on the opposite side from Democrats. Even on many of the less hot-button issues. Is there some sort of unintuitive averaging going on? I don't get it.

P = Brendan Nyhan Q = "Each

Submitted by Anonymous (not verified) on November 6, 2009 - 3:00pm.

P = Brendan Nyhan
Q = "Each senator is placed at their estimated ideal point in the ideological space. "

P and Q => FAIL

yeas, nays and agenda-setting

Submitted by Anonymous (not verified) on November 6, 2009 - 3:33pm.

Under Democratic control, I would expect Dems to vote pretty consistently with one another on most floor votes. Their party controls rules, procedures and timing, and would have (or should have) done its internal horse-trading before the bill got to the floor. Members of the Republican minority, OTOH, exercise what initiative they have by offering amendments, filibustering, etc., and less unanimity would be expected on a Coburn-sponsored no-hoper provocation among moderates, less-than-hard-right conservatives and members with parochial or pet interests. (Like, say, Kit Bond or George Voinovich angling for a post-Senate academic sinecure.)

Someone's Not Making Their Case...

Submitted by Howard (not verified) on November 6, 2009 - 4:26pm.

If this is the best chart that Political Scientists can come up with, I'd de-fund them too.

how ironic

Submitted by RobertWaldmann on November 6, 2009 - 6:33pm.

Your irony is appreciated. I think the pattern can be explained based on the title of Nyhan's post (I admit that is all I read). It seems the data are all votes from this congress. So classification is based on those votes then the lines are estimated with that classification.

The Democrats are all voting together, because they have to in order to win. If they aren't unanimous or nearly unanimous, they will lose the cloture vote which has become a necessary step for passage of any bill. Bills on which Democrats are divided are not brought to the floor. If a hypothetical Republican amendment might get significant Democratic support, said Democrats will have already demanded that it be in the bill brought to the floor (in committee, as a manager's amendment or however Reid is making the sausage).

In contrast, since Republicans don't control the process, they disagree in votes on the floor. This isn't a statement about ideological similarity. The process is keeping the many bitter disagreements between Democrats hidden from Nyhan, because the Democrats can't waste the whole Senate's time considering issues on which they haven't reached unanimity or near unanimity.

Secular Humanism.

Submitted by Omega Centauri (not verified) on November 6, 2009 - 7:17pm.

This was the big boggieman for the religious right. The social scientists are pretty much secular-humanism-central (if anyplace can be given that label). So of course they'd love to shut um down. Or at the very least lay down a marker, that tells their hardcore followers whose side they are on.

Now I have a bit of a problem with the graph too. I would expect two axes, to be for hopefully independent metrics (measurements on some ideological spectrum). But without at least a little bit of explanation as to what they mean they are not informative. Now I can imagine they might have used some sort of math analysis program, that came up with a two degree of freedom best fit, for two degrees of freedom that the computer algorithm decides work best. But, other than (perhaps) providing some mechanism for the social scientist to evaluate the probably of the prediction being good, I don't think it has any meaning for the general public.

Axis of The Right

Submitted by Anonymous (not verified) on November 6, 2009 - 9:21pm.

How the hell are social issues a separate axis?

That is the worst graphic

Submitted by Paul Camp (not verified) on November 6, 2009 - 9:39pm.

in the history of data presentation. Tufte would puke.

Um, its been this way since

Submitted by Doc Merlin (not verified) on November 7, 2009 - 6:38am.

Um, its been this way since I can remember. Republicans vary A LOT in terms of ideology. Democrats vary only by degree, but are very serious about that degree.

The republican party is made up of Libertarians, Social Conservatives, pro-corporate folks, and hawks. Each of these has major policy beliefs that clash with the others.

On the second dimension: The

Submitted by Anonymous (not verified) on November 7, 2009 - 7:01am.

On the second dimension: The voting behavior on many social issues obviously overlaps with that on economic issues -- at least when it comes to the behavior of members of Congress. Such behavior is therefore already explained by a "liberal versus conservative" dimension. A second orthogonal dimension could only capture information not contained in this first dimension, so it would include only those disparities in voting not reconciled with general liberal/conservative distinctions. Many of these, historically, happen to be driven by social issues that can occasionally divide the parties, when they are allowed to be voted on. While these secondary differences are clearly not very important in explaining Senators' vote choices overall, they can help systematically explain what might otherwise appear to be erratic or idiosyncratic behavior. Ultimately, all we can learn is how well a given vote was predicted by broader behavioral tendencies and who deviated from those expectations. Nothing more.

I agree.... Political

Submitted by Anonymous (not verified) on November 7, 2009 - 3:55pm.

I agree.... Political Science is more or less a hobby than an actual field of study.

Why should macroeconomics without mathematical rigor, or philosophy without the deep thought be funded in the first place?

Political science not a field of study?

Submitted by Anonymous (not verified) on November 7, 2009 - 5:13pm.

The economists just announced the Nobel Memorial Economics Prize is going to Elinor Ostrom. A political scientist. Whose work can be given a libertarian interpretation. Wake up Tom Coburn!

Burberry watches

Submitted by Burberry watches (not verified) on November 8, 2009 - 1:41pm.

Rolex Day-Date II watches
Rolex Daytona watches
Rolex Explorer watches

The value of diversity

Submitted by Randy (not verified) on November 9, 2009 - 7:53pm.

Which is better, to have your party speak, or to have it express diversity? It seems that you value diversity, and you think that the more liberal party should be more diverse. But in practice it is always the conservative party that is more diverse.

Do you value diversity for its own sake? Or do you value it only when it encompasses your own opinions?

My own experience is that the republicans that I talk to are much more open to differing opinions. I explained it this way to a friend, just the other day: "my democrat friends seem to look for a reason to argue, while my republican friends seem to look for a reason to agree."

I have a theory as to why this is true. The conservative group generally represents the historical status quo. The liberal group generally represents the historical urge to change the system--reformers, if you will. (These are the definitions of "conservative" and "liberal"...)

There is only one status quo, or rather, all of the status quo can be gathered up into one place (probably near the center of that red spot.) And everyone starts there, at some level. The definition of "status quo" is that it is what children are taught to revere and protect.

That status quo is never perfect. As each person ages they see the faults in the system and gain some desire to reform it. But there are an infinite number of posible reformers, each who want to move the system in a different direction. Generally speaking, each reformer will have one or two hot button issues that they care about.

So you start out with a political scatter-plot of reformers in all directions away from the center. And they are all pulling in separate directions. And they are all alone (or form small groups) and many of them die without ever accomplishing anything substantive except teaching the center to be more accepting of deviants. So you end up with a status quo-protecting conservative party with a high tolerance of diversity.

If the reformers want to get anywhere they have to start working together. Those who can get over their disagreements can form a significant force outside of the status quo--and have a real chance at effecting change. And since many of them only really care about one or two issues it is not too hard for them to vote-swap* and pretend to care about each others' issues when there is not direct conflict.

What happens then is that the liberal reformer movement establishes well outside of where any one reformer would like it to actually be--they each individually want to be one or two steps outside of the center; and they end up 3 or 4 steps out because that is the lowest common denominator between them. And they have to be quite dogmatic about the party line because that is the only way to enforce the vote-swapping agreements that allowed them to form a party in the first place. So, you end up with a liberal party that is very intolerant of any reforms (liberalism) outside of their own.

And to take the theory to the next logical step.... If that reformer party gains too much control and threatens the status quo too much then the reformers who cannot fit within that party must coalesce into a third party; probably quite diametrically opposed to the liberal party in power.

This is just the yin and yang of functioning democratic politics. Don't take it personally. But I know that if you are a democrat you will take it personally, because I'm outside of your orthodoxy. :-) You won't want to admit that you don't care passionately about each and every plank in the party platform, because that would endanger the bargains that you have made.

* I realized that "vote-swap" may sound like it has negative connotations. None are intended. This is the very heart of democracy--trading away my votes on issues that I don't care about in order to gain votes on issues that are more important to me. It is one of the core pillars of democracy. Democracy would fail to accomplish anything (and fail as a system of government) without vote-swapping.