Why do the justices on the U.S. Supreme Court vote they way they do? In an attempt to answer this question scholars have advanced two major models of decision making: the legal and the attitudinal models. In Chapter 7 we will test one element of the legal model – adherence to precedent. In this chapter we describe the attitudinal model and determine the extent to which it explains the justices' votes in the overruling cases.

The attitudinal model possesses two attractive features. First, it readily lends itself to testing. Second, the testing already conducted shows that this model can predict and explain the justices' votes.

Both the legal and the attitudinal models purport to explain the justices' votes and aspire to do so simply and parsimoniously. Neither model claims to represent reality. They focus instead on a set of crucial variables that presumably explain the behavior in question. Note the tension between the goals of parsimony and explanation. Extremely complex models, though perhaps more valid and reliable, are not useful because it is axiomatic that the more variables one uses the more one can decipher. Thus, for example, one arguably can explain every judicial decision on the basis of its facts and the relevant legal provision(s) at issue. But such an idiosyncratic focus would destroy a model's utility. Better to employ a handful of variables that explain a high percentage of the behavior.

THE ATTITUDINAL MODEL

The first comprehensive attitudinal model was the work of political scientist Glendon Schubert. On the basis of research done by psychologist Clyde Coombs, Schubert argued that the stimuli presented by the cases and the justices' policy preferences could be ordered along ideological dimensions.