Last night the House of Representatives voted overwhelmingly to censure Representative Charlie Rangel (D-NY), 333-79. Only two Republicans (Peter King of New York and Don Young of Alaska) voted against the censure resolution, but Democrats were more evenly divided, with 170 supporting the resolution and 77 opposed. Interestingly, the divide among Democrats appears to be related to race.
To quantify this, the following graph tracks last night's vote to censure Rangel against the percentage of whites in each member's congressional district. It looks only at Democrats who voted on the resolution. Each member is represented on the graph by a +.
As you can see, most of the Democratic votes to censure Rangel come from districts in which whites form a majority, while most of the votes against censuring him come from districts where whites are not a majority (although they might form a plurality). Let's drill this down by looking only at districts where whites are not a majority, and see whether there is a relationship between the censure vote and the percentage of African-Americans in the congressional district.
Once again, we see a pretty strong relationship. Districts where there are fewer African-Americans (and thus the non-white population in the district would be composed more of Hispanic- or Asian-Americans) make up the bulk of the censure votes here, while districts with larger proportions of African-Americans make up the bulk of votes against censure.
We can bring both of these graphs together to form this pretty straightforward chart.
As we can see, district racial composition seems to be a fairly good predictor of how a Democratic member voted. Yet it's unclear as to why this is the case. It's very unlikely, for instance, that the censure of Rangel was such a salient issue that members were listening to their constituents, who were demanding that he be censured or not. It's worth noting here as well that several Southern white Democrats with fairly large numbers of African-Americans in their districts (e.g. Steven Cohen of TN-9 and John Barrow of GA-12) voted in favor of censure. So, it might be that members who are personally or professionally closer to Rangel were less likely to vote for censure, and those members happen to come from minority white districts.
In support of this hypothesis, it's worth noting the following chart. It tracks membership in the Congressional Black Caucus (CBC) against the censure vote. (Rangel is a veteran member of the CBC.)
Only one member of the CBC voted in favor of censure, Artur Davis (AL-7), who is not returning to the 112th Congress; meanwhile better than 75 percent of Democrats not in the CBC voted in favor of censure.
Minimally, we can say the following: among Democrats, race was a notable predictor of yesterday's roll call vote to censure Rangel. Whether it was a causal factor (e.g. members from districts with large proportions of African Americans felt compelled to vote against censure) or whether it was indirect (e.g. friends or associates of Rangel were compelled to vote against censure, and they tend to come from districts with large proportions of African Americans), it is impossible at this point to say. Nevertheless, the correlation between race and the vote to censure among Democrats is noteworthy.