Like almost everything else in Congress, Senate votes on Supreme Court nominations have become much more polarized along party lines in recent years. That was certainly true of the recent vote on President Obama’s nomination of Sonia Sotomayor. While all 59 Democrats who were present voted to confirm Sotomayor, 31 of 40 Republicans voted against confirmation.
An analysis of the Senate vote indicates that a combination of three factors accurately predicted support for confirmation among Republican senators: the 2008 presidential vote in a senator’s state, a senator’s personal ideology, and whether a senator was retiring from elected office. The 9 Republicans from states carried by Barack Obama voted 6-3 in favor of confirmation; the 31 from states carried by John McCain voted 28-3 against confirmation. The 6 Republicans who had announced that they were retiring from elected office voted 4-2 in favor of confirmation; the other 34 voted 29-5 against confirmation. Finally, 8 of 21 Republicans who were classified as moderates or moderate conservatives based on their voting records in the 110th Congress voted for confirmation; only 1 of 16 who were classified as strong conservatives voted for confirmation.
The results of a logistic regression analysis of the confirmation vote are shown below. This analysis is based on 38 Republican senators for whom an ideology score in the 110th Congress was available. Logistic regression is a statistical technique used to estimate the effects of several independent variables on a dependent variable when the dependent variable is a dichotomy as was the case with the confirmation vote. Two of the three predictors, Obama vote in the state and personal ideology, had effects that were statistically significant at the .05 level. The third predictor, retirement from elected office, just missed being significant at the .05 level. Taken together, these three variables correctly predicted the votes of 35 of 38 Republican senators including 28 of 30 votes against confirmation and 7 of 8 votes in favor of confirmation.
Results of Logistic Regression Analysis of Sotomayor Confirmation Vote Among Senate Republicans
These results tell us something about the politics of Supreme Court confirmation votes in the 21st century. They indicate that these votes are influenced by the same factors that influence votes on legislation–ideology and constituency opinion. The more conservative a Republican senator’s ideology and the more conservative a Republican senator’s state, the more likely he or she was to vote against confirmation.
The tendency of Republican senators who were retiring to vote in favor of confirmation may be explained by the fact that these senators did not have to weigh constituency opinion as heavily as their colleagues because they did not have to worry about facing the voters again. They could afford to vote based on their judgment of the nominee’s record and qualifications rather than partisanship. For this small group, the Sotomayor vote was more like Senate votes on Supreme Court nominees in an earlier, less partisan era.