As I said in my last post, looking at a single ideological score for each legislator doesn't really convey the complexity of the 83rd House of Representatives.
For example, Jacob Javits, of NY-21, had a similar "conservativeness score" to a bunch of Southern Democrats he probably voted with only rarely.
Here's the result, colored by party, and with larger letters indicating a district where Stevenson did better at the Presidential level:
The labels of the axes--"overall/economically conservative" and "international/socially" conservative--should be regarded as rough approximations at best. Hence the ''.
But Jacob Javits, while his "overall liberal" score is nothing special, has the single "most liberal" score in the entire House on the second dimension, placing him far away from the Southern Democrats.
We can see that most Democrats continue to be to the left of most Republicans on the first (overall) dimension, but that other coalitions are possible.
I also think it's interesting that the two dimensions are fairly correlated for Democrats, which suggests that the Democratic caucus mostly split in similar ways, no matter what the issue was. And, with most Southern Democrats more conservative than most Northern Democrats on both measures, the Democratic caucus probably usually split along regional lines. (Even though there weren't any real civil rights votes in the 83rd Congress that I noticed.)
On the other hand, the Republican caucus is sort of a shapeless blob, which means that it can be divided in more ways. That suggests that the Republican caucus split into a wider variety of factions than the Democratic caucus.
Also note the small cluster of New Jersey Democrats who were apparently more "socially liberal" than "economically liberal".
For example, here's roll call 18, which the A.D.A. called a motion to restore public housing funds:Comparing the two, we can see that some "socially conservative" (and often Southern) Democrats opposed the bill, while some "socially liberal" (and often urban) Republicans supported the bill, although it was fairly party-line.
Here's roll call 88, on the other hand--also scored by the A.D.A., and requiring court approval for wiretaps:
All Democrats voted for this, while the Republicans who voted for it are a somewhat different group than the public housing Republicans.
And here's roll call 16, which the A.D.A. characterized as a vote to extend a program where Mexican agricultural workers were "imported" without "minimum sanitation...working conditions" and so on:
We can also look at combinations of votes this way. Two consecutive votes on Hawaii statehood are particularly intriguing.
The first was a vote to recommit a statehood bill, and the second was a vote to pass it. These usually go the same way (recommit is usually read as "kill") but not here:
As you can see, many northern/liberal Democrats voted inconsistently: First to recommit the bill, then to pass it. An article I read suggested some may have been angling for AK statehood along with HI statehood, since HI was still a Republican state.
But many Southern Democrats, and a few "socially conservative" Republicans and "more economically than socially liberal" Democrats voted consistently, in the two votes, against HI statehood.
Here's a vote that more or less gets the "liberal wing" on its own, and even splits it: Roll call 137, which the A.D.A. described as a bill that would let Congress "force testimony by grant of immunity":
Finally, here's roll call 97, which was apparently to "establish a national committee on education":
Not really sure why that was such a party-line vote, but a somewhat interesting split beyond that.