Timothy M. Hagle

Goodness-of-Fit Measures for Probit and Logit

Some regression analysts find the R2 statistic to be of little utility. Others use it extensively while evaluating model performance. One of the shortcomings of probit and logit analyses is the lack of an analog to the Ordinary Least Squares (OLS) R2 statistic. To aid in the evaluation of model performance, several pseudo-R2s have been proposed. Dichotomizing a continuous interval-level variable results in distortions due to a loss of information. We do not know, however, the degree to which these distortions affect the pseudo-R2s vis-a-vis the OLS R2 that is based on the underlying continuous dependent variable. In this study we use simulation techniques to compare four common pseudo-R2s for probit and logit with the R2 that would be obtained under OLS regression. After making a correction to one of the measures, two of them perform quite well, comparing favorably with the OLS R2. The choice between them may be simply a matter of availability and ease of use.

Strategic retirements: A political model of turnover on the United States Supreme Court

The presidents ability to nominate justices to the United States Supreme Court is one of his most powerful tools for advancing his policy goals. To maximize the use of this tool it is important that he be able to determine how reasonable it is to expect a retirement from the Court during his presidency. Prior research is inconclusive as to whether Court retirements are politically motivated. In this study I develop and test a political model of Court retirements. The results from estimating the model using exponential Poisson regression show there to be an identifiable political element relating to the timing of retirements from the Court.

Increasing the Size of Minimum Winning Original Coalitions on the Warren Court

Clearly there may be resistance to unanimous decisions, e.g., Brown v. Board of Education, 347 U.S. 484 (1954), as well as those which are minimum winning. Because of this possible resistance, it may be advantageous for decisions and opinions of the Court to be decided by a large majority.2 Moreover, a minimum winning majority opinion is likely to result in an unstable precedent for the Court, which might be overturned if membership changes or if one of the justices in the majority changes his mind. Given the possible undesirability of minimum winning decisions and opinions, the majority opinion assigner might attempt to use his opinion assigning power to generate a final decision and an opinion that are larger than minimum winning. To understand how the size of the final decision or the opinion coalition can be enlarged, we describe the Courts opinion assignment proc-

Constitutional Process: A Social Choice Analysis of Supreme Court Decision Making. By Maxwell L. Stearns. Ann Arbor: The University of Michigan Press, 2000. 420p.

A not uncommon complaint concerning social choice models is that they tend to assume away everything of interest, leaving a model with little relation to the real world. The usual response is that one must start with the basic elements of a problem before moving to more complex and inclusive models. Unfortunately, few social choice models applied to legal studies have achieved sufficient complexity to be of significant value. Maxwell Stearns recognizes this problem and, in Constitutional Process, systematically presents a comprehensive social choice model and applies it to a complex legal doctrine.