Empirical scholars of the United States Supreme Court, Jeffrey Segal and Harold Spaeth have long contended that Supreme Court decisions are based primarily upon the ideological beliefs of the justices, and that ideology alone accounts for over 60% of the total voting variance of the Court. However, recent scholarship demonstrates that this conclusion is only made possible through ecological inference. In the bivariate regression that purported to establish this hallmark conclusion, Segal and Spaeth aggregated their voting data into percentages. This resulted in exaggerated findings. Recent scholarship has corrected Segal and Spaeth’s mistake by modeling justice votes using a logistic regression that does not manipulate the dependent variable before analysis is performed. The new findings demonstrate that Segal and Spaeth’s model loses about two-thirds of its explanatory value. Ideology, therefore, is not as dominant of a force upon the Court as attitudinal modelers previously thought. Still, however, it remains an important variable in the judging equation. This paper explores various methodological issues that judicial politics scholars will confront when modeling justice ideology using logistic regression estimated with maximum likelihood. Topics covered include substantive interpretations of the model, the “correct” independent variable to be used, applications of the technique, and an empirical assessment of how well ideology models explain judicial choices made by the Court.