Is it possible to predict lethal and non-lethal family violence? Te Puna Haumaru Seminar Series

The idea that cases of lethal family violence can be accurately predicted ahead of time is a widely held view among academics, policy makers and the news media. This premise has informed public policy internationally and has led to the development of risk assessment frameworks that purport to detect the risk of lethality based on risk factors such as strangulation, controlling behaviour, and sexual assault. However, few empirical studies have sought to examine whether this premise is supported empirically using appropriate longitudinal designs.

This presentation will report on a prospective study of nearly 40,000 family violence reports in Victoria Australia. The study examines the extent to which these risk factors can be used alone or in combination to predict lethal and non-lethal family violence. The presentation will discuss the limitations of extant research in this area and the implications of using prediction of lethality as the basis of criminal justice policy.

SPEAKER

Associate Professor Benjamin Spivak received his Ph.D. in 2016 and is employed at the Centre for Forensic Behavioural Science, Swinburne University of Technology. He has published over 45 peer reviewed papers on the topics of risk assessment, family violence, criminal justice contact among young people, and child maltreatment. For the past decade, Dr Spivak has worked on a programme of research examining risk assessment of family violence which has included development of prediction models and examination of disparities in prediction among demographic groups.