The Supreme Court Database (SCDB) is a publicly available database coded by Dr. Harold Spaeth and others. It includes detailed information about every supreme court case in United States history. While working on my project on modeling judicial bias I wrote an extension to the SCDB that included demographic information for the petitioner and respondent (race/gender if they were an individual, otherwise the type of entity, among a few other details) for a sample of nearly two thousand cases between the Jan 20, 1953 and Jan 20, 1969 (This timeslice represents the Eisenhower through Johnson presidencies), with even more partial examples. Eight hundred of these cases were selected randomly, the rest were added based on what could be automatically coded from search terms in the case name (if the petitioner was “United States” for example).
This extension has a few important limitations to note. Demographics were primarily coded by cross-referencing census data with information from the case and oral arguments. No data, or only partial data, could be found for about a fifth of randomly selected cases. These were replaced by new randomly selected rows to make up a total of two hundred manually coded cases per presidential term (800 overall). If there were some systematic reason why some of these data points were impossible to find, the existing data could be skewed on that axis.
I plan on continuing my work improving this extension, and to ultimately release it publicly. If this data seems useful to your research, or if you are inspired by it, please contact me at email@example.com and we can discuss collaborating.