The researchers selected judges in Crown Point, LaPorte, South Bend, and Valparaiso, Indiana (three men and five women); Los Angeles and Santa Barbara, Calif. (three men and two women); Albany, N.Y. (two men and one woman); Laramie, Wyoming (one man); and Washington, D.C. (five men and six women). These locations offered a wide variety of political cultures, broad patterns of political gender dynamics, and different partisan political traditions. Each judge was observed on at least two occasions, with each session lasting between 2 and 4 hours. The matters observed included arraignments, pretrial motions, bail requests, plea agreements, and sentencing rulings. For each judge, the researchers obtained between 10 and 20 "instances" of judicial decision making. A coding sheet was used to assess whether the judges were using "masculine" or "feminine" voice traits in their decision making. The focus was on the theorized female voice traits of "community" and "context" and on the purported male voice traits of "individualism" and "rules." There were four central findings. First, male and female voice traits were exhibited by judges of both genders. Second, female judges were more likely to rely on the prosecutors in issuing their rulings, and male judges were more likely to side with the defense. Third, the females were more likely to use inclusive and procedural judicial styles. Fourth, male judges were more likely to use both the consensual and authoritarian styles. In addition to the gender of the judges, the study identifies several possible explanations for the findings, including age and length of judicial tenure, method of judicial selection, political party, ethnicity, type of caseload, previous legal training and experience, and other extrajudicial political factors; however, perhaps due to the small sample size, the regression equations did not show these factors to be significant. 5 tables and 44 references