CJB261

Author Adam Teperski and Stewart Boiteux
Published December 2023
Report Type Crime and Justice Bulletin No. CJB261
Subject Apprehended violence orders (AVOs); Domestic violence; Offenders; Recidivism / Re-offending; Victims; Policing; Evaluation reports
Keywords Domestic violence; Recidivism; Victims; Offenders; Policing; Apprehended violence orders (AVO)

Summary

Background

Apprehended Domestic Violence Orders (ADVOs) are civil orders which restrict a defendant’s behaviour towards victims of domestic violence (DV) to protect them from further violence. An ADVO breach is a criminal offence and can attract severe penalties, including custody.
ADVOs can be set at any length to meet a victim’s needs. However, orders are most commonly set at one of two standard lengths, 12- and 24-months.
We provide the first quantitative evidence examining the influence of ADVO length on offending. To do so, we match 2,897 defendants placed on 24-month ADVOs to 10,820 defendants placed on 12-month ADVOs between January 2016 and April 2018 in NSW.
We then compare DV offending (excluding breaches) and ADVO breach rates over 36 months comprising of: the first 12 months (when ADVOs were active for both groups); the next 12 months (when only those on longer orders were still subject to an ADVO); and the subsequent 12 months, (when both groups’ ADVOs had lapsed).

Key findings

Figure 1. Difference in quarterly DV and ADVO breach offending rates between defendants placed on 24- and 12-month ADVOs
 Figure 1. Difference in quarterly DV and ADVO breach offending rates between defendants placed on 24- and 12-month ADVOs
Figure 1 shows the differences in the quarterly rate of DV and ADVO breach offending between defendants placed on 24-month ADVOs and the matched group placed on 12-month ADVOs, after controlling for relevant demographic and offending characteristics.
We estimate that in the 5th, 6th, and 7th quarters after placement on an ADVO (when only longer orders remained active), defendants placed on 24-month orders were 1.8 percentage points (p.p.), 2.0 p.p., and 3.1 p.p. less likely to have a DV offence, respectively. This reflects large relative decreases in DV offending between 41% to 59%. In these periods we find corresponding increases of 3.4 p.p., 4.1 p.p., and 2.3 p.p. in the likelihood of an ADVO breach offence, respectively. There were no statistically significant differences in DV or ADVO breach offending between groups from the 9th quarter onwards (i.e., post ADVO).

Conclusion

Compared with 12-month ADVOs, 24-month ADVOs were associated with a lower probability of DV offending and a higher probability of ADVO breaching in the additional 12 months that the ADVOs were active.