This is an adaptation of The Supreme Court Database maintained by Washington University in St. Louis ("WUSTL"). Two datasets are included: (1) a "case centered" dataset ("Case") that describes in each row of data a case decided by the court; and (2) a much larger "justice centered" dataset ("Justice") that describes in each row of data how a particular justice of that court voted on an issue within a case.

The WUSTL dataset provides extensive information on every decision of the United States Supreme Court since 1946 (excepting particularly insignificant determinations). The WUSTL website devoted to this data contains a lengthy explanation of the organization of the data and the meaning of each variable used therein. Thus, the column "issueArea" in the original WUSTL datasets has the same meaning and same encoding scheme as the column "issueArea" used in the two datasets presented here. Moreover, somewhat subjective determinations by WUSTL within the original datasets, such as whether a particular decision was "liberal" or "conservative," are not re-evaluated here; WUSTL's determinations are simply replicated. Research done using the original WUSTL datasets can thus be readily done using the representations of the data contained here.

This adaptation makes the WUSTL data more useful in several ways, however. First, the datasets here add columns with names ending in "S" and "F" that respectively provide a "short" and "full" explanation of the encoding system used by WUSTL for many of its columns. This permits users of the data to understand the meaning of (integer) values in those columns without extensive cross referencing with an online codebook provided by WUSTL that extends over dozens of web pages. Thus, in the original WUSTL dataset (and in the datasets here) the integer value 103 in the "petitioner" column would mean that the entity petitioning the Supreme Court for relief would be an "airplane manufacturer, or manufacturer of parts of airplanes." But one would not know this from the data value alone; one would have to consult a lengthy codebook. This dataset adds a column "petitionerS" that concisely identifies the petitioner as an "airplane manufacturer." It also adds a column "petitionerF" that more fully identifies the petitioner as an "airplane manufacturer, or manufacturer of parts of airplanes." The translation from WUSTL integer encodings to textual descriptions are taken from various spreadsheets not generally available but graciously provided by WUSTL to assist with this project. Thus, the textual descriptions of the data used here are fully consistent with the WUSTL understanding of the meaning of the integer codes.

Second, a few of the columns in the datasets refer to dates, persons and states for which there is more complete representation in the Wolfram Language through its framework for dealing with dates and its store of information on "Entities." Thus, dates of argument, reargument and decision are transformed here into DateObjects on which it is convenient to perform date arithmetic. And columns containing encodings of states and persons are augmented with columns with names ending in "E" that contain Entities. By way of example, the original data contains a column "respondentState" that contains an integer encoding of the state (or other geographic entity) in which the respondent resides (or is otherwise associated). For a particular case, it might thus contain a value 51. The datasets here add not only a column "respondentStateS" that contains "TX" (a postal abbreviation for Texas) and a column "respondentStateF" that contains "Texas" but also a column "respondentStateE" which contains Entity["AdministrativeDivision",{"Texas","UnitedStates"}] and whose properties can then be further queried using the Wolfram Language.

Users are cautioned that the justice centered dataset is over 1 gigabyte in size and it thus may take some time both to download and use the data on a local machine.