The Supreme Court's Polarity: SCOTUSBlog has examined the voting patterns of the Supreme Court Justices in the last term and has some interesting statistics on the degree in which the
Justices agree with each other. I've made a graphical representation of the percentages of times justices agree in full, in part or in the judgment. There are a couple of threshold values, 90% and 85% -- the former tends to show close associations, while the latter weaker groupings. There are two main groups, right and left.
The right-leaning group has Rehnquist at its center, overlapping with Kennedy, O'Connor and Scalia, and with Thomas almost as strongly connected to Rehnquist (89%). Within that group, Kennedy agreed with O'Connor 87% of the time, and 85% of the time with both Thomas and Scalia. O'Connor and Thomas agreed only 78% of the time, though.
The left-leaning group consists of a very strong core group of Souter, Ginsburg and Stevens, who agree with each other 91% or 92% of the time, while Breyer is only slightly less closely associated, agreeing 91% of the time with Ginsburg, and 87% and 88%, respectively, with Souter and Stevens. The association between this group of four has somewhat less variation than that among the right-word five.
Now, the interesting thing is that the only association of more than 80% between the left-leaning four and the right-leaning five is between Breyer and O'Connor, at 82% (which happens to be closer than O'Connor and Thomas, at 78%).
I have made a pseudo-Venn diagram to illustrate some of the numbers (sorry I created the graphic counter-intuitively with the right-leaning group on the left and vice versa). A PDF from SCOTUSBlog is the source of the numbers.
