The Recidivism of Homicide Offenders in Western Australia

Abstract

Popular perceptions concerning the recidivism of extremely violent offenders - in particular, homicide offenders - are contradictory. Perceptions vary from one extreme - that such offenders rarely commit further violent offences - to the opposite end of the spectrum, where it is thought that they remain high risks on release from custody. The present study, the first of its kind in Australia, draws on the records of persons arrested in Western Australia over the period 1984-2005 to provide accurate measures of the recidivism of this small but potentially dangerous group, and to make rigorous comparisons between specific types of homicide offenders.

Using survival analysis techniques, we accurately estimate and analyse the long-term rate of recidivism of 1,088 individuals arrested and charged for homicide in Western Australia between 1st January 1984 and 31st December 2005. The cases involve individuals arrested for domestic murders and other types of homicides (robbery and sexual murder), including attempted murder, conspiracy to murder, manslaughter (unintentional homicide) and driving causing death offences. Most arrests for homicide result in conviction and lengthy prison sentences; therefore, very long follow-up time is required to study the recidivism of such cases. Our database provides up to 22 years follow up time (for those arrested in 1984) and accounts critically for the first and any subsequent arrests, if they occur.

Recidivism is defined herein as re-arrest following release from prison for the signal homicide offence within the period of follow-up, and is considered in three separate ways: as re-arrest for another homicide, for another violent offence, or for any offence. Using both nonparametric and parametric analyses we estimate probabilities of recidivism and the speed with which it occurs. Risks of recidivism are compared across types of homicide, and we explore the role of covariate information such as gender, age, Aboriginality, and prior record.

Of the 1088 individuals, only 3 (all male non-Aboriginals) were subsequently arrested and charged with a distinct homicide offence relating to a single event in the follow-up period. But 40.3% were re-arrested for another offence (any offence) and 21.9% were re-arrested for another grave offence (robbery, blackmail and extortion, assault, sexual assault, abduction and other offences against the person, burglary, arson, child pornography and homicide) by the cutoff date. Accounting for censoring, among those committing a murder and subsequently released, we estimate a probability of 0.66 of being rearrested for another offence of any type. The corresponding probabilities for those convicted of manslaughter or for driving causing death were both 0.43. Having a prior record increased the risk of re-arrest substantially; for example, male non-Aboriginals arrested for murder with at least one prior arrest have an estimated probability of 0.72 of being rearrested for another offence of any type. Their estimated probability of being rearrested for another grave offence was 0.33.

Keywords: homicide, recidivism, survival analysis, dangerous offenders, parole