The comment today is about the Sotomayor vote. A couple of weeks ago I had tweeted that I thought it possible that she would get 80 votes today, based on Lindsay Graham's likely (at that time) vote for her. If Graham voted for her, you would think that others less conservative than he would do so, as well.
Wrong. The 68-31 vote in her favor made a liar out of me. But I was wrong, as they say, for the right reason.
Turns out Graham was the biggest outlier in the final vote, by far. I make this claim relying on Keith Poole's recently-published ideological rank-ordering of the Senate and House, which are based solely on roll call votes before August. Check it out here: http://voteview.com/sen111.htm.
The accompanying graph shows the current Senate, ranked according to Keith's optimal classification analysis. The reds are Republicans and blues are Democrats. I've stacked up the nay voters and the yeah voters.
Notice there's a pretty seamless ideological explanation for the vote. There's a little raggedness near the cutting point, which you would expect. Martinez (R-Florida) makes a reasonably clean break in the vote. With Martinez the break point, the two Republicans we would have predicted to have voted for Sotomayor, but didn't, are Murkowski and Cochran. So far, so good, for a simple spatial model explanation of the vote.
Graham, of course, voted for Sotomayor. Ideologically, this leaves him hanging out there, as twelve more moderate Republicans chose to oppose her. Given Graham's rock-solid conservative credentials, it's hard to imagine this vote will hurt him with his base, and given his reasonable (and reasoned) approach to the hearings, his vote will lead some to believe --- those who don't follow Congress closely --- that he is more moderate than his overall voting record would indicate.
On to Vermont. Because of a discussion I had with a reporter today, my next post will likely be about figuring out what difference the small-state bias made in the cap-and-trade vote.
4 comments:
Thank you, Charles. That's a wonderful graph, and a good way to launch a conversation.
Interesting. But why do you think ideology now matters for confirmation votes, when as recently as 88 (or 94) it didn't?
In response to ewi, the interesting thing about past votes is that sometimes you had roll call votes on SC confirmations and sometimes you didn't. When you did, they were ideologically structured. Because the SC is closely balanced ideologically, more people are paying attention to nominations, and thus senators are more likely to be voting (and therefore voting ideologically).
That said, we're certainly in a more ideological age. Part of that fact is seen in another interesting pattern that has been pointed out to me --- the Republican senators who have announced their retirement are on the whole much more moderate than the rest of the Senate Republican caucus. The dynamics pushing the parties apart appear to be pretty relentless these days.
This is an almost one-dimensional ideological vote, but I have found that with two-dimensional 111th estimates a cutting line can additionally predict Cochran correctly as Nay, leaving only two "errors."
Post a Comment