Primary Outcomes (end points)
The primary outcomes are legal knowledge and conflict resolution.
We will use a difference-in-differences analysis, with time and individual fixed effects.
The research plan includes subgroup analyses by:
• Nature of the case
Indices will be built to address the issue of multiple hyopthesis testing. We will present robustness checks with the Fisher test, adjustment for multiple hypothesis testing The randomization was stratified on 1) regions (8), and 2) time to disposition. These stratification dummies will be added to the model.
We will also present multiple robustness checks. We will use the randomization inference tests (Fisher exact test) which have been shown to outperform standard models especially in small samples (Young, 2019). We will address the concern of multiple hypothesis testing by using the Sharpened False Discovery Rate (FDR) adjusted q-values (Anderson, 2008) as well as building indices. We will address the issue of attrition with the Lee bounds (Lee, 2009).