Table B.1 reports the results of the opinion assignment models. The data consist of majority opinion assignments to 34 justices during the 1946 through 2015 terms. Unless noted otherwise, the data were obtained from the Supreme Court Database. The dependent variable indicates whether each justice was assigned to write the majority opinion in each case (1 = assigned; 0 = not assigned). I conduct two separate analyses: one for the 5,844 cases in which the chief justice assigned the majority opinion (N = 41, 598) and one for the 1,093 cases in which the senior associate justice (SAJ) assigned the majority opinion (N = 6, 112). Modeling opinion assignments is especially complex because “the likelihood that a chief will assign the opinion to one justice is dependent on the probability that the chief justice will assign it to another member of the majority.” Because my models include an interaction term with an assignment specific variable (winning margin), I follow Maltzman and Wahlbeck by utilizing a random effects estimator for case. Therefore, I employ multilevel logistic regression models with random intercepts for case. However, my results are also robust to including random intercepts for justice. Both models include the justices’ SCIPEs for the Big Five (Extraversion, Conscientiousness, Agreeableness, Neuroticism and Openness), as well as the following control variables: