Lessons learnt from implementing a robust evaluation of the Youth on Track early intervention scheme

Ms Mandy Loundar1, Mr Hamish Thorburn2

1Juvenile Justice NSW, Sydney, Australia, 2NSW Bureau of Crime Statistics and Research, Sydney, Australia

Randomised controlled trials are considered to be the ‘gold standard’ for research designs. However, implementation of a randomized controlled trial is often hindered due to the practical or ethical issues. Here we present our experience with setting up a randomised controlled trial for the Youth on Track early intervention scheme. Youth on Track is funded by NSW Government to coordinate services for 10 – 17 year olds who are at risk of long-term involvement in the criminal justice system. The scheme is delivered by non-government organisations and monitored by Juvenile Justice.

The NSW Bureau of Crime Statistics and Research is undertaking a randomised control trial (RCT) to measure Youth on Track’s impact on reoffending and other social outcomes. A number of challenges have been faced in the development of the RCT, of a statistical, ethical and practical nature. The presenters will share the lessons learnt from implementing this evaluation methodology, which is rarely applied in a juvenile justice setting.


Mandy Loundar has 17 years of experience in the youth and criminal justice sectors. She has worked in non-government, local council, and state government. Since commencing work with the NSW Department of Justice in 2009, Mandy has gained extensive experience developing and implementing criminal justice interventions for adults and young people. As the Manager of Strategic Projects for Juvenile Justice Mandy implemented and managed the Youth on Track early intervention scheme.

Hamish Thorburn is a Principal Program Evaluator for the NSW Bureau of Crime Statistics and Research. He joined the Bureau in 2016 after completing a Bachelor of Science/Arts (majoring in Mathematics and Statistics) at the University of Queensland in 2015. His previous research experience includes sentencing data, simulation models of court flows, efficiency of operating theatres in the Queensland Health system and applications of dynamic programming in sports.