Author Judy Trevena and Suzanne Poynton
Published May 2016
Report Type Crime and Justice Bulletin No. 190
Subject Diversion; Domestic violence; Policing; Prisons and prisoners; Recidivism / Re-offending; Sentencing; Statistical methods and modelling
Keywords domestic violence, prison sentence, suspended sentence, reoffending, deterrence, propensity score matching

Download this publication



To examine whether short prison sentences (up to 12 months) exert a deterrent effect for domestic-violence (DV) related offending.


Propensity score matching was used to compare time to reoffence among 1,612 matched pairs of offenders, in which one of each pair received a prison sentence of 12 months or less and the other received a suspended sentence of two years or less. Kaplan-Meier survival analysis was then used to examine time to the first proven offence committed after the index court appearance.


In the matched analysis, DV-related reoffending was not significantly different for people with suspended sentences and prison sentences. After 1 year, 20.3% of people given a suspended sentence and 20.3% of people given prison sentence had at least one new DV-related offence, and after 3 years the proportions were 34.2% and 32.3% respectively. These were not significantly different (HR 0.96, p=0.6).


Short prison sentences (up to 12 months) are no more effective in deterring DV-related reoffending than suspended sentences.

Download this publication