Data Understanding Phase
The data are provided by Victoria Legal Aid (VLA), a semigovernmental organization that
aims to provide more effective legal aid for underprivileged people in Australia. Over 380,000
applications for legal aid were collected from the 11 regional offices of VLA, spanning 1997–
1999, including information on more than 300 variables. In an effort to reduce the number of
variables, the researchers turned to domain experts for assistance. These experts selected seven
of the most important variables for inclusion in the data set: gender, age, occupation, reason for
refusal of aid, law type (e.g., civil law), decision (i.e., aid granted or not granted), and dealing
type (e.g., court appearance).