2050.0.55.002 - Position Paper - ABS Review of Counting the Homeless Methodology, Aug 2011
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 05/08/2011 First Issue
|Page tools: Print Page Print All|
KEY ISSUES: HOMELESSNESS AND YOUTH
As outlined in the Discussion Paper: Methodological Review of Counting the Homeless, 2006 (ABS cat. no. 2050.0.55.001), once the homeless operational groups decision rules had been applied to the Census, the numbers of 12-18 year olds who were identified as likely to be homeless were analysed. It is commonly assumed that the number of homeless youth enumerated in the Census will be underestimated as the households they are staying with may not know that they are unable to return to their home and will record their previous home address as their usual residence. Alternatively, the youth themselves may not identify themselves as homeless given the potential impact of being able to stay on in the household they are visiting. The ABS is undertaking a quality study after the Census on homeless school students which should inform on this assumption and help to further understand homeless young people.
In their CTH analysis, Chamberlain and MacKenzie classified 6,378 youth aged 12 to 18 years old who were enumerated in the Census as homeless. However, a separate CTH estimate of youth homelessness was derived by Chamberlain and MacKenzie (21,940 people) using sources other than the Census. The main source that was used was their National Census of Homeless School Students (NCHSS), adjusted using assumptions about the ratio of school to non-school homeless youth aged 12 to 18 years. In CTH the difference between the Chamberlain and MacKenzie Census measure and their separate estimate is assumed to relate to those youth who may be homeless who were staying with another household on Census night but had a usual address reported for them. CTH concludes "…we think the missing 15,562 are hidden within this category". CTH, pages 18 to 20 explains this methodology in more detail. The ABS has concluded that the Chamberlain and MacKenzie estimate of 21,940 homeless youth is not sufficiently robust to accurately inform on the numbers of 12-18 year old youth who were experiencing homelessness on Census night. The Chamberlain and MacKenzie estimate is based on school reports which capture youth homelessness over a week rather than on Census night; it uses an undercount adjustment methodology that is inconsistent between Censuses; and it applies an extrapolation to the non-school youth population that is both inconsistent with the stated methodology and which overstates the estimate. See pages 46 to 48 in the Discussion paper for more information. In addition, using an external source to provide an estimate of the possible number cannot inform on the characteristics or locations of homeless 12-18 year olds, limiting the usefulness of the data for policy and services planning purposes.
In analysing the characteristics of 12-18 year olds who had a usual address reported and were visitors on Census night, no characteristics have yet been identified that differentiate between those who were homeless and those who were visiting for other reasons. Additional analysis on the characteristics of the possible relationship status of the 39,966 12-18 year old usual residents visiting on Census night was provided in the Discussion Paper to try to understand any characteristics that can be used to distinguish potential homeless groups. This analysis used age, sex and Census District of usual residence to attempt to understand possible family units travelling together.
This chapter provides additional information on visitors' student status, labour force status and household type to help further inform on this group. The expertise of the HSRG will be sought to determine whether there are any characteristics to determine those who are more likely to be homeless, by utilising the data that have already been analysed and any other data using Census variables, or external datasets that may be useful to inform the analysis.
For the 2011 Census, the ABS has a focus on improving the enumeration of all people who may be homeless on Census night, and this is expected to lead to better Census based estimates of homeless young people from the 2011 Census. In particular, encouraging people who are 'couch surfing' to both ensure that they are reported in the Census and that 'no usual address' is reported for them will improve the estimation.
USING THE CENSUS TO CLASSIFY HOMELESS YOUTH
Using the 2006 Census data, the Review methodology estimated that there were 5,424 youth aged 12 to 18 years who were enumerated and could be reasonably classified as homeless on Census night. This accounts for 9% of the total homeless population. The ABS acknowledged in its Discussion Paper that this is an underestimate of the numbers of 12-18 year olds who were homeless in 2006 on Census night because it cannot distinguish between homeless 'couch surfers' with a reported usual address, and other visitors who reported a usual address elsewhere but were not homeless.
The following two tables compare the Review estimates with the CTH estimates. The remainder of this chapter will focus on where possible underestimation for youth homelessness may occur, including the rules applied to arrive at the current estimate, as well as further analysis of this group.
While the ABS does not define what age group should be considered when estimating youth homelessness, it had presented data for 12-18 year olds in the Discussion Paper in order to compare directly with the work done in CTH.
Some submissions provided to the Review, indicated that within the homeless sector, youth generally refers to those aged 12-24 years or 12-25 years. Because the Review numbers are identified directly from the Census, any age groups can be calculated and analysed. Later in this Chapter, statistics for persons aged 12 to 24 years are discussed. These can be used to analyse whether the characteristics of 12-24 year olds are more in line with the experiences of those who work with homeless youth.
Comparing Review estimates with CTH estimates
Table 1 - AGE DISTRIBUTION, Review, Counting the Homeless and Estimated Resident Population - 2006
Table 2 - AGE OF YOUTH VISITING PRIVATE DWELLINGS WHO REPORT HAVING A USUAL ADDRESS ELSEWHERE(a), 12-18 years - 2006
Table 1 represents the proportions of the homeless populations in each of the age groups for the Review and CTH estimates. It then compares them to the estimated resident population. While 12-18 year olds are the smallest homeless proportion in the Review estimate, this group spans a smaller age range than the other groups (which are ranges of 10 years or more).
When the CTH homeless estimate and the Estimated Resident Population for 12–18 year olds are compared (Table 1), homeless youth are over represented. The other homeless age groups are generally in line with their proportions in the total population (apart from the groups for 65 years and over, and under 12 years, which are both under represented in the homelessness estimates in CTH). When looking at the combined age range of 12-24 years, the differences are less pronounced. And while for the 12-18 year old group the CTH estimate is 4 times the reviewed estimate, for persons aged 19-24 years old the CTH estimate is only about 30% higher than the reviewed estimate.
Table 2 shows the proportion of 12-18 year olds in each of the homeless operational groups in the Review, in CTH before the NCHSS adjustment was applied, and in CTH after the addition of 14,656 youth in the persons 'staying temporarily with other households' (titled 'staying with friends and relatives' in CTH). This shows that the category containing 'couch surfers' staying in private dwellings accounts for the majority of homeless youth in CTH (77%). This is after the adjustment is applied to the Census data using the NCHSS. Prior to the adjustment it was 23%, and in the Review was 19%. Studies done by Banyule Nillumbik Youth Services Network (2010) and the Housing Young People Action Team (2005) found high proportions of young people who were homeless or at risk of homelessness staying with other households. In the Banyule Nillumbik Youth Services Network study of 98 homeless secondary school students aged 12 to 18 years in 2009, 43 were staying in caravans. Some of the families living in caravans were likely to be doing so because of the Victorian bushfires, but not all. Another 34 homeless youth were staying temporarily with friends/relatives, 12 were in supported accommodation, and 9 had no fixed address. In the Housing Young People Action Team study of 170 young people aged 15 to 25 years who were homeless or at risk of homelessness, 59 were aged 15 to 18 years. Across the group aged 15 to 25 years who were homeless or at risk of homelessness, 97 (57%) were staying with friends and relatives.
The potential numbers of youth who may be homeless on a single night, such as Census night, and those who have experienced it over a longer period, such as a year, may vary substantially. For those who experience homelessness, the length and nature of being homeless varies. 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 would result in a difference in the number of homeless youth counted on a single night as opposed to counting the number of youth experiencing homelessness over a longer period.
Since the Review estimates were published in the Discussion Paper in March 2011, some further analysis was conducted of estimates of homeless youth aged 12-18 years and enumerated in the Census, in order to understand the relatively high numbers enumerated in boarding houses on Census night compared to those who were staying with other households. Of the reviewed estimate of 900 youth in boarding houses, 74% were enumerated in a dwelling that was classified by the Census collector as a 'boarding house, private hotel', while the other 26% were enumerated as being in other dwellings. The Discussion Paper noted that the review had focussed on the CTH reclassification, as boarding houses, of dwellings that were enumerated by Census collectors as either private dwellings or other non-private dwellings and had not challenged the Census collector classification of boarding houses. In more recent analysis, about a dozen dwellings enumerated by collectors as boarding houses, with about 300 youth aged 12 to 18 years, enumerated in them who were mostly studying full-time, and with the other people in these dwellings generally being teachers, special care workers, and some clerical or administrative staff. The majority of these dwellings were in town centres across regional Australia with the few others being matched to mis-classified residential colleges or language schools. For those dwellings in regional town centres, secondary schools were found to be nearby to these dwellings and in a number of cases these dwellings aligned with known rural student hostels which support geographically isolated families without daily access to schools. Overall, about 400 youth aged 12 to 18 years were found to be included in this group in dwellings which appear to be legitimate student lodgings. Taking this information into account, the percentage of homeless youth aged 12 to 18 years staying in boarding houses would decrease to 10% and the percentage in the other categories would increase, including those staying temporarily with other households (21%).
A number of submissions noted that the Census is held on a Tuesday night in August (in winter, or the dry in Northern Australia) during the school term so it would be logical to assume a relatively lower proportion of young people away from home on a school night compared to non-school nights. In the 2006 Census there were 39,966 youth visiting other dwellings, or about 2% of the 1.9 million youth aged 12 to 18 years in 2006. However, the distribution across the ages is not uniform - a quarter were 18 years old and another 32% were either 16 or 17 years old, and only 10% to 12% in the younger ages.
It would be expected that some of those visitors on Census night would be staying with a parent living elsewhere or a grandparent or another relative who does not live at the youth's usual address. The ABS knows, from the 2006-07 Family Characteristics and Transitions Survey (ABS cat. no. 4442.0) that 36,400 youth aged 12-17 years stayed overnight with a natural parent living elsewhere at least once a week. When looking over a year, there were 21,300 youth aged 12-17 years who stayed at least 30% or more nights with their natural parent living elsewhere per year, of which 10,600 youth aged 12-17 years spent 50% or more nights per year with their natural parent living elsewhere. Many more youth (about 200,000) stayed overnight less frequently than once a week on average, but some would be expected to have been staying on Census night. The numbers of 18 year olds regularly staying with their parent living elsewhere, or the number of 12 to 18 year olds staying regularly with grandparents or other relatives are not known.
Further analysis of 40,000 visiting 12-18 year olds who reported a usual address elsewhere
The Discussion paper presented some preliminary analysis of the characteristics of the 40,000 12-18 year olds who were visitors on Census night and a usual address elsewhere in Australia was reported for them. The analysis was aimed at identifying which groups of youth, if any, were more likely to include significant numbers of youth who may have been homeless on Census night. Comments received during consultations indicated that in any of these groups there may be some homeless students, and no group was identified by stakeholders as more likely than any other group to include homeless youth.
As noted in the Discussion paper, while relationships between visitors, and between visitors and hosts, are collected in the Census, Census processing does not capture these relationships and the Census forms are destroyed after processing. Therefore, visitor relationships cannot be directly analysed to determine whether they are related to any of the persons in the household, or whether the visitors are related to each other. To better understand the characteristics of the 12-18 year olds visiting on Census night, age, sex, and the Collection District (CD) variables were used to infer relationship status which was then used to categorise visiting youth in the table below. As CDs are small geographic areas, groups of visitors that are all visiting the same dwelling, and all report the same CD of usual address elsewhere, and are present in age/sex combinations that are akin to usual family structures, it has been assumed that these groups are families visiting together.
TABLE 3 - VISITING YOUTH WITH A REPORTED USUAL ADDRESS(a)(b)
For youth visiting on their own without accompanying parents, if relationship status was available, this would be limited to identifying who was visiting another relative, such as their grandmother but could not identify whether it was because they were homeless, or because they were being cared for by their grandmother for the night while parents were away, or simply to spend time with the grandparent. The 2009-10 Family Characteristics Survey (ABS cat. no. 4442.0) reported on youth who had daily contact with their maternal grandparent. In lone mother families about 40,000 youth had daily contact, with even higher numbers of youth from couple families having daily contact.
Table 4 below shows the age groups of those youth who were visiting private dwellings and a usual address elsewhere was reported on Census night.
The data show that the numbers of youth visiting on Census night in 2006 increase with age, with over half of the youth visiting on Census night with a usual address reported elsewhere were aged between 16-18 years
TABLE 4 - AGE OF YOUTH VISITING PRIVATE DWELLINGS WHO REPORTED HAVING A USUAL ADDRESS ELSEWHERE(a)(b) - 2006
The following tables provide comparisons of the student status and labour force status for all 12-18 year olds reported in the Census, for those identified as visiting on Census night, and for those who were identified as homeless in the reviewed estimates.
Table 5 - ALL YOUTH(a), Student Status and Labour Force Status - 2006
TABLE 6 - THE 40,000 VISITING YOUTH(a), Student Status and Labour Force Status - 2006
Table 7 - ALL HOMELESS YOUTH(a), Student Status and Labour Force Status - 2006
For the visiting youth aged 15-18 years, 57% were students, 20% were not studying but were employed, and a further 7% have 'not stated' for both of these variables. Of those not studying, 41% were not in the labour force or were unemployed. For the whole 15-18 year old population, 77% were studying, 11% were not studying but were employed and 5% not stated on both variables. Of those not studying, 32% were not in the labour force or were unemployed.
Overall, visiting youth aged 15-18 years with a usual residence elsewhere were less likely to be students and more likely to be employed.
For homeless youth identified in the review, 28% were aged 12-14 years old with a high proportion of student status 'not stated' (23%).
For homeless youth identified in the review aged 15-18 years, less than half (47%) were students, 10% were not studying but were employed, while 17% have 'not stated' for both of these variables. Interestingly, 70% of those older homeless youth who are not studying are either unemployed or not in the labour force, compared with 41% for visitors and 32% for total youth.
The next two tables show the student and labour force status for two of the identified possible sub-populations of the 40,000 visiting youth, 1) those who are visiting on their own, or who may be visiting with one accompanying adult female, and 2) those youth in a visitor only household without any persons of parenting age.
Both of these visiting sub-groups had a lower proportion of 12-14 year olds, than did either the homeless population or the total 12-18 year old population. These sub-groups of visitors also had a higher proportion of 15-18 year olds who were studying compared to the homeless population but lower than for all 15-18 year olds. Of those who were not studying, in terms of being employed ( 59%) they were closer in line with the whole 15-18 year old population (67%) than with those who were identified as homeless (29%).
TABLE 8 - THE 20,000 VISITING YOUTH ON THEIR OWN OR WITH ONE ADULT FEMALE(a), Student Status and Labour Force Status - 2006
TABLE 9 - THE 1,000 VISITING YOUTH WITHOUT 'PARENTS' IN A VISITOR ONLY HOUSEHOLD(a)(b), Student Status and Labour force status - 2006
The next two tables show, for people in private dwellings, the household type for both the 12-18 year old visitors and those identified as homeless by the Review. The key differences between these two populations are the greater proportion of homeless youth in Group households (8%, compared with 4% for the households visited by youth) and the greater proportion of visiting youth in visitor only households (15% compared to 4% for homeless youth).
In looking at the 12-18 year olds who were visiting on Census night and had reported a usual residence elsewhere, and using the assumed family type (based on age, sex and CD of usual residence of the visitors) and the composition of the households they are visiting, around 60% of those youth visiting with two adults of opposite sex and parenting age (assumed to be couple family), were in visitor only dwellings.
For those youth who were visiting with a single male adult visitor who could be their parent, over half (53%) were staying with a one family household, and around a fifth were in visitor only households. In contrast those youth who were with a single female adult who could be their parent, 37% were visiting a one family household, 29% were visiting a lone person household and 23% were in visitor only households.
For a single 12-18 year old youth who were not visiting with anyone else, 71% were visiting a one family household (this will include couples with or without children or a single parent with children) and a further 15% were visiting lone person households. In contrast for multiple 12-18 year olds visiting another household together, 56% were visiting a one family household and 23% were visiting a lone person household, and for the 12-18 years olds visiting with a child/ren under 12 years, 59% were visiting a single family household, and 27% were visiting a lone person household.
TABLE 10 - VISITING 12-18 YEARS OLDS ON CENSUS NIGHT WHO HAD A USUAL RESIDENCE REPORTED ELSEWHERE(a) - 2006
TABLE 11 - HOMELESS YOUTH ENUMERATED IN PRIVATE DWELLINGS(a)(b) - 2006
Further analysis of these visitor groups will be undertaken and the results reviewed by the Homelessness Statistics Reference Group for guidance.
Homeless youth aged 12 to 24 years
Expanding the analysis of homeless youth to a 12-24 years age group aligns more closely with the age group targeted for youth homeless services. This wider age group has a higher proportion of visitors staying temporarily with other households (29% compared with 19% for the smaller, younger group).
The percentage of homeless youth 12-24 years is estimated to be 21% of the reviewed estimates of homeless, which is 4 percentage points higher than their proportion of ERP and one of the largest populations in the homeless population.
The table below shows that for homeless persons aged 12 to 24 years, the largest group are still those in supported accommodation for the homeless (36%) but now 29% are staying temporarily with other households, the second highest group.
TABLE 12 - HOMELESS OPERATIONAL GROUPS (a), 12-18 years, 12 to 24 years and All homeless persons - 2006
FUTURE CENSUS IDENTIFICATION
For the 2011 Census, the ABS has been working with homeless service providers in each state and territory to encourage accurate reporting of no usual address by all homeless people including those who are couch surfing. While this may improve the under-estimation of homelessness in 2011, including for youth, for the reasons outlined in this chapter, some homeless youth who are 'couch surfing' may still not be reported as having no usual address and therefore are unlikely to classifiable as homeless in analysis of Census variables.
To help understand the possible under-reporting of homeless youth in the 2011 Census, the ABS will undertake a quality study of homeless school students after the Census. This study will also scope a possible methodology for a more frequent survey of homeless school students.
The ABS will continue to undertake additional analysis to develop the final methodology for how youth are included homelessness estimates, drawing on the expertise of the Homelessness Statistics Reference Group.