FLOW MEASURES OF HOMELESSNESS
In the section: Complexities in estimating homelessness, both prevalence (point-in-time or stock) and incidence (or flow) measures of homelessness are discussed. An incidence or flow measure is an estimate of the number of people experiencing at least one period of homelessness over a given period of time, for example, over a 12 month period. Incidence measures have value in informing service provision by showing the potential demand for services over a given period.
For those who experience homelessness, the length of time in and nature of being homeless varies for different people. For some, being homeless is chronic and ongoing, whereas for others it is episodic. Research has indicated that homelessness among youth is more episodic rather than chronic (Robertson, 1991) and as a result, the characteristics of those identified in a prevalence measures may be different to those captured in an incidence measure.
In the Discussion Paper: Methodological Review of Counting the Homeless, 2006 (ABS cat. no. 2050.0.55.001) the ABS used the Review estimate of around 65,000 persons enumerated in the Census and likely to have been homeless on Census night, to estimate the possible number of people who may have experienced at least one period of homelessness over a year. ABS multiplied the published SAAP data for 2006-07, which show there were 118,800 clients and 69,100 accompanying children (AIHW 2008, tables 3.1 and 3.2), by the proportion of closed support periods for SAAP / CAP where accommodation was provided over the year in 2006-07 (39.9% for clients, and 57.5% for accompanying children) (AIHW 2008, tables 6.5 and 6.7) to derive an approximate total of 87,100 people accommodated in SAAP over the course of the year. This estimate is 6 times the estimate of 14,517 people accommodated in SAAP services on the 2006 Census night. Assuming the broad 6 to one relationship between SAAP accommodation throughout the year and Census night accommodation were to hold across other homelessness circumstances, the ABS estimated that the number of people who may have experienced periods of homelessness over the Census year may be between 340,000 and 440,000 people (a range of 5 to 7 times the reviewed estimate of 65,000 homeless people on Census night). No other data currently report on the scale of this aspect of homelessness in Australia.
The results from the 2010 ABS General Social Survey, available in late September 2011, will be analysed to provide further insight into the numbers of people who experience homelessness over an extended period of time (such as over the last 12 months, two years etc). Notwithstanding the limitations of this data source as outlined in section: Complexities in estimating homelessness, comparing the data for the different time spans will provide a perspective on the number of people experiencing homeless over time, as well as on the length of time of the most recent period of homelessness. Building up a profile, across successive ABS surveys, of the proportion of the population reporting past periods of homelessness will allow a much richer picture of the flow of people into and out of homelessness over time. Measuring repeat periods of homelessness allows the dimension of flows into and out of homelessness to be analysed.
In addition, the ABS will also look into any other data that may provide a more accurate estimate of the total numbers of people who may experience homelessness over a 12 month period. The ABS will review the data that will be released from the new Specialised Homeless Services collection, that replaced the SAAP collection on 1 July 2011, from which it is expected that flow measures for specialist services users will be available. Analysis of the Centrelink homelessness flags provides another potential opportunity to look at flows for this client population. The 5% Statistical Longitudinal Census Dataset also has the potential to provide some perspectives on flows into and out of homelessness over very long time periods.