Some analysts may wish to select a unit of analysis different from the four formats that we distribute. Any selection can be accomplished by parsing the identification variables (SCDB Case ID, SCDB Docket ID, SCDB Issues ID, and SCDB Vote ID). This code selects the first issue for the first docket associated with a particular citation, keeping two records if there is a split vote on that issue. This corresponds to the ANALU=0, or 4 in the legacy database.

Some analysts may wish to select a unit of analysis different from the four formats that we distribute. Any selection can be accomplished by parsing the identification variables (SCDB Case ID, SCDB Docket ID, SCDB Issues ID, and SCDB Vote ID). This code selects the first issue for associated with each docket, keeping two records if there is a split vote on that issue. This corresponds to the ANALU=0, 1, or 4 in the legacy database.

The Justice Centered Data contain one record for each justice participating in the dispute for the chosen unit of analysis. For example, each docket will have a record for each justice in the Cases Organized by Docket dataset. For some types of analysis it is useful to have one record for each unit of analysis with a column of data for each justice. This is the way in which the vote data was included in the legacy Supreme Court Database. This Stata program takes data from our long format and "flips" it to a wide format.

This code takes the justice-centered data and allows us to compute term-by-term percentages of cases that have one or more concurring votes. This code could be modified to do a similar aggregation for any justice-specific variable.