Session: Tackling selection bias in sentencing data analysis
Time: Thursday 5th July, 09:15 - 10:45
Convenors:
Dr Jose Pina-Sanchez (University of Leeds)
Dr Sara Geneletti (London School of Economics)
Abstract Details
Analyses of judicial decisions tend to focus on the duration of custodial sentences. These types of sentences are however quite rare (8\% of the total in England and Wales), which generates a problem of selection bias. We suggest a new approach based on finite mixture modelling, Bayesian statistics and aggregated views from judges, capable of modelling simultaneously all types of sentences. Distributions of the relative severity of the four major sentence outcomes (fines, community orders, suspended sentences, and custody) are specified into the same mixture model, making the most of the information available in sentence data.
Presentation downloads
Presenter: Jose Pina-Sanchez
Tackling Selection Bias in Sentence Data Analysis
The level of the session is: Accessible
Presentation details
Presentation 1
Start time: 09:15
Presentation title: Tackling selection bias in sentencing data analysis
Presenters:
Dr Jose Pina-Sánchez (University of Leeds)
Dr Sara Geneletti (London School of Economics)