A key aim of the research is to support early identification of groups at risk of homelessness. The analysis identifies the most important factors for predicting homelessness, based on modelling using cross-agency service use and demographic information.
The main predictive model covers all presentations to homelessness services and is built on quarterly records in 2014/15 and 2015/16. The model shows:
- 32% of homelessness presentations can be attributed to 1% of the NSW population.
- The 1% group at highest risk are more than 30 times more likely to access homelessness services in the next year than the general population.
- In addition to broad service use such as welfare services and Medicare, the most important risk factors for the group at highest risk are repeated homelessness service use, police-recorded victim incidents, social housing history, and court appearances. Aboriginal people are overrepresented in the group at highest risk.
