Is statistical risk of sex offending related to the harm that is caused?

Prof. Mark Kebbell1

1Griffith University


There are reliable and valid risk assessment tools that can be used to predict the statistical risk of a person reoffending sexually. One such example is the Risk Matrix 2000. Nevertheless, statistical risk may not be related to the harm caused by an offence. In this paper, the effectiveness of the Risk Matrix 2000 is assessed in terms of ability to predict risk of reoffending for people on a sex offender register. Further, quality of risk prediction is also assessed with regards ‘harm’ caused to victims using an adapted Harm Index. The Matrix 2000 predicts sexual reoffending accurately but is not so effective in predicting Harm. Harm is less directly related to statistical risk and worthy of additional consideration.


I am a Professor of Forensic Psychology at Griffith University and Queensland President of the Australian New Zealand Association of Psychiatry, Psychology and Law. My interests concern identifying high-risk offenders.