From 2018-2019, I created a novel dataset about the British Columbia Supreme Court. The dataset was designed to facilitate both descriptive (e.g., how have decisions’ delivery time evolved over the last forty years) and predictive analysis (e.g., what factors best explain that evolution from a statistically significant perspective). The goal of the descriptive analysis is straight-forward: describe the evolution of decisions’ length and delivery time over the last forty years as well as any other notable trends. The goal of the predictive analysis is also straight-forward: identify which dataset variables are most likely to predict longer decisions (i.e., word count) and slower delivery time (i.e., the number of days it takes to issue decisions). 

To facilitate that analysis, I reviewed all reported BCSC decisions in Quicklaw from 1980, 2000, and 2018 and hand-coded 17 variables according to a coding manual: (1) case name; (2) citation; (3) registry; (4) whether hearing information was provided; (5) the date the hearing concluded; (6) how many court days were required for the decision; (7) the date the decision was released; (8) how many days lapsed before the decision was released; (9) judge; (10) moving party-type; (11) responding party-type; (12) subject; (13) word count; (14) use of headings; (15) self-rep information; (16) appeal information; (17) whether the decision was oral or written. While error checking occurred, like any other hand-coded dataset, some errors are likely.

Some results of the descriptive analysis are below. The full descriptive and predictive analysis can be found here.