A new blog called The Monkey Cage points us to an interesting graphic that argues the ideological differences between different regions in America is actually relatively small:
The graphic, taken from a new book that argues America isn't as ideologically polarized as our national political narratives would suggest, shows the average ideological bent of a resident in New York City to lean liberal by 0.2 out of a possible 1.0, and for the average ideological bent of someone living in rural Texas to be 0.4 conservative out of a possible 1.0. Both areas are closer to the overall national median, and to each other, than they are to the national ideological extremes.
Now, this is something that should be obvious to anyone who follows national politics from a big picture perspective. After all, despite the red state vs. blue state narrative that fixated in the national imagination during the 2000 and 2004 elections, very, very few states were decided by more than 25% in either election, and only five smallish states (D.C., Idaho, Nebraska, Utah and Wyoming) were decided by more than 30%. Most states were decided by less than 20%, an amount that, in the US House, would be considered worthy of heavy targeting in the next election cycle. Over 40% of the country is either a blue resident of a red state, or a red resident of a blue state (and that doesn't even include the 35% of the adult population that never votes at all). New York and Texas, often seen as the best examples of states that are diametrically opposed politically, actually only represent a swing of 19% in either direction. On average, the ideological and partisan differences between red states and blue states are not massive, even if they are significant enough to produce very different types of public policy.
Certainly, America is becoming more diverse ethnically and spiritually, not to mention that the income gap continues to widen at an alarming pace. However, the goal isn't to eradicate these differences, and reside in some state of bi-partisan, cross-cultural, perfectly wealth distributed harmonic consensus. Further, I feel that all generalized talk about how terribly divided our country is misses the point, at least politically. We don't need to end division altogether in order to make real change in American politics, we just need a coalition that represents around 60% of the country. If you have 60% support combined with 60% representation in political office, you can pass pretty much whatever legislation you like. With 67%, you can even change the Constitution to your liking. A huge number of people in America, over 100,000,000, would still disagree with you, and one could still argue that the country would remain divided 30% dissented on any particular subject. However, the political stalemate would be broken, and real legislative change would occur.
No matter the extent to which it is actually divided, unless we are bordering on a replay of the Civil War I just don't understand fretting over America being divided. No matter how flawed the execution of its ideals has often been, America is, at its core, both a republic and a nation that embraces ethnic and religious pluralism. Unlike virtually ever other nation on Earth, we don't have, and have never had, a national language, ideology, religion, ethnicity, or hereditary ruling class. We have never been united on those fronts, and we never should be united on those fronts. In fact, our lack of unity in these areas has long been one of the best features of America, and one of our greatest sources of strength. As one of the world's first post-colonial nations, first post-state religion nations, and first stable republics, our division and diversity long pointed the way to the future. Even the European Union, a post-nationalistic governmental organization which I believe is the way forward for humanity, was, in many ways, an idea that found its origin in the United States of America. It was an idea that came about after many people realized the dangers caused by ideological one-party states, ethnically homogeneous states, monarchies or theocracies. If history has shown anything, it is that those unified states are the problem, not diverse, "divided" states such as our own.
As long as no one is being denied equal economic, political or cultural opportunity, I don't care if America is divided ideologically, culturally or politically. I don't want unity--I just want 60% so we can make real political change at the legislative level. There is a real path for Democrats to achieve that goal by 2010, and for progressives to achieve that goal five to ten years later. If and when that happens, America will still be divided, but the overwhelming balance of political power will at least temporarily reside with progressives. And that is perfectly fine by me, on both counts.
We can either have unity, or we can have America. We can't have both. Personally, I'll choose America every single time.
At the time, I was less than convinced of the merits of Chris's suggestion of designating a group of Dem senators, equivalent to the Bush Dogs, as the ONOs.
There, I was arguing on general principles: in particular, that it was wrong (if not delusional) to think that Dem senators who regularly help the GOP in their hours of need did so out of fear or venality: it's clear beyond a peradventure that they are really most sincerely non-progressive.
However, having spreadsheeted Senate voting information for the 110th for the period up to the August recess, I thought I'd run some numbers to see quite how the ONOs match up to the non-ONOs when it comes to hurting the Dem party in the Senate.
The handiest way of doing this is to look at votes on Democratic rolls.
I pointed to the MTR as a useful tool for an aggressive minority, but without expecting that the 110th GOP would make such strikingly successful use of it.
The bare numbers are impressive enough: in the seven months running up to the August recess, there were roll call votes on 45 MTRs, of which 31 failed and 14 passed.
By contrast, over the whole 109th, there were 51 RCVs on MTRs, none of which passed.
As a prelude to the upcoming hostilities, I've spreadsheeted what I hope are all the Iraq-related RCVs in the House so far this year, including votes on related special rules, previous and motions to recommit.
If I've missed any, please let me know. (It took me a little while to work out that I'd missed the infamous RC 425 - because the accompanying (crappy) description of the measures (taken from the Clerk's site by the automatic gizmo that scrapes the data therefrom) does not mention the magic word Iraq!)
Following up my piece yesterday, with a look at a suggestion in the comments to the piece that Dem freshmen were more than averagely prominent in the ranks of the Bush Dogs and other mod dissenters.
Let's take two measures, comparing the performance of freshmen with non-freshmen.
(I make 44 of the former - but manual data entry is fraught, especially when the result of comparing two spreadsheet lists (rep names for the 109th and 110th)!)
Firstly, the proportions who voted for the infamous Iraq amendment on RC425 (HR 2206). Freshmen broke 19-25, nonfreshmen 67-115.
Testing these numbers on Compare 2 part of the quick-and-dirty stats package PEPI (download from here), none of the many tests offered show anything remotely like the magic p = <0.05 that we statistical autodidacts cling to like a blankie.
Secondly, I compared the rankings of each group for rebel votes (see earlier piece) using the same software, to see if the freshmen had a lower ranking (more rebel votes) than nonfreshmen.
This time, p-numbers were infinitesimal all round, which I take as a good sign that the difference might be judged statistically significant by someone expert in the craft!
More research needed, evidently!
[The PEPI gizmo, by the way, is just the job for this sort of exercise: R, unbeatable on regression models, is too cumbersome for back-of-the-envelope working.]
UPDATE: the PEPI output for the two tests is here and here.
(Great info on voting patterns in the 110th Congress, and provided more depth to the Bush Dog discussion. Check out the charts in the extended entry, especially the longer, second one - promoted by Chris Bowers)
Finally, I've managed to sit down and spreadsheet the 110th House rollcall votes up to the August recess. (I started before the Bush Dogs thing got going, but I can't help thinking that was a spur to getting me to finish it!)
So - let's look at the issue of rep loyalty.
There have been 846 RCVs in which Dem reps have cast 187,746 votes. Of these, 8,336 (4.44%) were cast against a Dem majority on the vote in question. Let's, for want of a better expression, call those rebel votes.
Do some Democrats vote for legislation like FISA because they believe that warrant-less, widespread spying on American citizens is actually a good thing for Alberto Gonzales to be doing, or do those same Democrats vote for legislation like FISA because they are afraid they will be portrayed as terrorist hugging, anti-American?
That's a question we ought to hang onto - perhaps refine a bit, but certainly hang onto.
(The danger in a place like this is that worthwhile research questions (or questions that are worthwhile investigating whether they're worthwhile research questions) get thrown out, chewed over and discarded without reaching any conclusions on which knowledge can be built.)
So - how do you go about formulating the question in a way that it could be answered with something more persuasive that gut feeling or anecdote?
(Perhaps gut feeling and anecdote are the best we can do. But I think we ought to explore before conceding that.)
Chris's piece from earlier today set me off thinking about the working out of the 1932 party realignment, in particular, the slow extinction of the Solid South over the generation following passage of the Civil Rights Act of 64.
The 90th House (list culled from the list of reps in 1-109 on the Voteview site) shows an ideological and geographical spread vastly different from that in the 110th. [I take the 90th as the first House elected after the Great Society reforms had been enacted - with a considerable bounce-back for the GOP after the pretty disastrous post-assassination elections of 64.]
For instance, there is no ideological separation between the parties. As measured by DW-NOMINATE 1st dimension scores, the most conservative Dem (Rarick LA-6) is the 432nd most liberal (out of 437 reps ranked) at 0.528. (The scale runs from -1 (most liberal) to +1.)
No fewer than 183 GOP reps were more liberal than Rarick!
I'm laboring under the handicap of not have read the research underlying this piece.
(Is there a PDF somewhere? Or, if it's published, a journal reference?)
But, as it stands, the proposition support of which is attributed to the academics in question looks pretty flaky: it's suggested that because rural areas showed a higher level of support for hetero-only marriage, but no similar differential existed on other policy issues, homo-bashing won Bush his reelection.
Where's the evidence of causation of the various issues on voting? What kind of analysis of the 04 National Election Survey have the writers done?
STATE BLOGS
