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 Print Page Print all pages in this productPrint All  
Contents >> Contents >> Key issues: Homelessness and domestic and family violence

KEY ISSUES: HOMELESSNESS AND DOMESTIC AND FAMILY VIOLENCE

SUMMARY OF KEY POINTS

  • The complex nature of domestic and family violence, together with limitations of the Census in estimating homelessness in the circumstances in which people experiencing such violence may find themselves, may result in an underestimate of homelessness using Census variables.
  • For the 2011 Census the ABS will repeat the very successful 2006 practice of jurisdictional lists and the 'green sticker' approach for supported accommodation arrangements. This ensures that those who are seeking refuge from domestic violence and staying in supported accommodation will be included in the estimates. The ABS has also 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 may be staying with friends and relatives.
  • The next Personal Safety Survey, to be run in 2012, is currently testing the inclusion of questions on housing arrangements upon separation from their partner. While this survey will have the same limitations as the General Social Survey, i.e. it is unlikely to interview anyone who is currently homeless, it will provide insight into the housing arrangements and circumstances of those who leave home due to violence.
  • The ABS will draw on the expertise of the Homelessness Statistics Reference Group to further understand homelessness due to domestic violence.


INTRODUCTION

As outlined in the Discussion Paper: Methodological Review of Counting the Homeless, 2006 (ABS cat. no. 2050.0.55.001), under-identification in the homeless estimates is likely for those who were victims of domestic or family violence and were homeless on Census night. Feedback from the Review consultations indicated that women escaping from domestic violence come from all backgrounds and ages. It is also likely that some will report in the Census the usual address from which they have fled, and hence will not appear as homeless, or they may not be included on a Census form at all.

Service data show that domestic/family violence is a significant reason for women to be homeless, or at risk of homelessness. The SAAP support period data for 2006-07 showed that family/domestic violence was the most common reason for women seeking assistance (AIHW 2008). For both women with children, and lone women aged 25 years and over, it was the main reason (55% and 37% respectively). For lone women aged under 25 years, the reasons were more varied, with relationship/family breakdown the most common reason (21%), followed by domestic/family violence (16%).

The 2005 Personal Safety Survey (ABS cat. no. 4906.0) found that 5.8% or 443,800 women aged 18 years and over had experienced violence in the 12 months prior to the survey. This had declined from 7.1% in 1996. Of these women, 242,000 had experienced physical assault (3.1% of women, down from 5.0% in 1996). The survey also found that 31% of women who were physically assaulted in the previous 12 months, were assaulted by a current and/or previous partner (ABS 2006b).

However, not all populations experience violence at the same rate. For Aboriginal and Torres Strait Islander Australians aged 15 years and over, in 2008, nearly a quarter (23%) had been the victim of physical or threatened violence in the last 12 months. The proportions of Indigenous women and men who had been the victim of violence was the same (ABS 2008a).

The majority of submissions to the methodological review noted that the Census was an appropriate means to derive homeless estimates, despite challenges in enumerating and identifying sub-population groups, in particular women and children escaping domestic and family violence. Estimates are likely to be underestimated due to the complex nature of domestic and family violence, together with the limitations in both enumerating this homeless group and being able to accurately classify them as homeless in the Census data. Staying with other households and reporting a usual address mask homelessness in the Census dataset.


ASSESSING CENSUS DATA

SAAP data provide views of male/female representation in this homeless context that can be compared with the Census based estimates.

The Table below shows, that around 2 in 5 of the people identified as homeless in the reviewed 2006 Census estimates were women. However this proportion varied across the age groups. It was highest for the younger age groups with 52% of those aged 12-18 years being female. The proportions also varied across the different homeless groups. It was lowest in boarding houses (25%) and highest in supported accommodation for the homeless (50%). When looking at the proportions in supported accommodation, the Census results were lower than the proportion of female clients in SAAP administrative data. In using the number of clients with at least one period of accommodation in their support period spanning Census night in 2006, 58% of clients were female and 49% of accompanying children were female. When combining clients and accompanied children, 55% of people with support periods spanning Census night in the SAAP system, and during which support period some accommodation was provided, were female, slightly higher than the 50% in the reviewed estimates for Census night SAAP residents. It is highly likely that the support period measure overstates the female proportion on any night, as women tend to have longer support periods, on average, than men.

TABLE 1 - HOMELESS OPERATIONAL GROUPS(a), Sex and Age - 2006
AGE GROUP (YEARS)
Under 12
12–18
19–24
25–34
35–44
45–54
55–64
65+
All homeless persons
no.
no.
no.
no.
no.
no.
no.
no.
no.

PEOPLE WHO ARE IN IMPROVISED DWELLINGS, TENTS OR SLEEPING OUT
Male no.
422
343
415
825
1,065
941
588
394
4,993
Female no.
361
287
245
496
525
418
255
183
2,770
Total
no.
783
630
660
1,321
1,590
1,359
843
577
7,763
% female
%
46
46
37
38
33
31
30
32
36
PEOPLE IN SUPPORTED ACCOMMODATION FOR THE HOMELESS
Male no.
2,330
1,245
909
1,245
1,214
892
432
317
8,584
Female no.
2,157
1,477
1,144
1,562
1,198
636
288
284
8,746
Total
no.
4,487
2,722
2,053
2,807
2,412
1,528
720
601
17,330
% female
%
48
54
56
56
50
42
40
47
50
PERSONS STAYING TEMPORARILY WITH OTHER HOUSEHOLDS
Male no.
1,034
510
1,537
2,618
2,036
1,375
1,139
787
11,036
Female no.
970
546
1,274
1,771
1,112
1,082
1,169
619
8,543
Total
no.
2,004
1,056
2,811
4,389
3,148
2,457
2,308
1,406
19,579
% female
%
48
52
45
40
35
44
51
44
44
PEOPLE STAYING IN BOARDING HOUSES
Male no.
116
463
1,324
2,190
2,480
2,558
1,861
1,587
12,579
Female no.
78
439
803
769
598
576
414
572
4,249
Total
no.
194
902
2,127
2,959
3,078
3,134
2,275
2,159
16,828
% female
%
40
49
38
26
19
18
18
26
25
PEOPLE IN OTHER TEMPORARY LODGING
Male no.
41
67
200
262
245
186
109
83
1,193
Female no.
43
46
141
155
127
93
77
94
776
Total
no.
84
113
341
417
372
279
186
177
1,969
% female
%
51
41
41
37
34
33
41
53
39
ALL HOMELESS PERSONS
Male no.
3,943
2,628
4,385
7,140
7,040
5,952
4,129
3,168
38,385
Female no.
3,609
2,795
3,607
4,753
3,560
2,805
2,203
1,752
25,084
Total
no.
7,552
5,423
7,992
11,893
10,600
8,757
6,332
4,920
63,469
% female %
48
52
45
40
34
32
35
36
40
(a) Cells in this table have been randomly adjusted to avoid the release of confidential data.


When looking at Census based homeless estimates, for registered marital status of homeless persons aged 18 years and over, females had a lower proportion who had never been married (51% compared to 60% for males), with a higher proportion who were in a registered marriage (23% compared to 14% for males). Being married was highest for those homeless women who were staying in improvised dwellings, tents or sleeping out (33%) and those who were staying temporarily with other households (32%). While social marital status (including de facto relationships) is also available on the Census, it is of limited value for those who have been identified as homeless, because social marital status is determined using current living arrangements. This cannot be identified for those who were visitors on Census night nor for those in non-private dwellings, particularly for those who were separated from their partner on Census night.

When looking at the labour force status of homeless women and men aged 15 years and over in the Reviewed estimates, a large proportion did not have their labour force status reported (19% for women and 24% for men). A lower proportion of women in the homeless estimates were employed than men (24% compared 28%) and nearly half (49%) of women were not in the labour force, compared to 37% of men.

When looking at those in the reviewed homeless estimates who reported no usual address, a higher proportion of women reported no usual address than did men (41% compared to 36%). In contrast, women had a lower proportion who were classified as homeless and were reported as being 'at home' on Census night (56% compared to 62% for men). This is consistent with the higher proportions of men in improvised dwellings, tents and sleeping out, and in boarding houses.


ADDITIONAL DATA SOURCES TO INFORM ON HOMELESSNESS AND DOMESTIC AND FAMILY VIOLENCE
    The ABS welcomes any additional information that may help to distinguish which visitors on Census night who reported a usual address were homeless due to escaping violence, including other sources of data required to supplement the Census data.

    The new AIHW Specialist Homelessness Services administrative collection, that replaced the SAAP collection on 1 July 2011 will help to inform on those who seek services as a result of domestic or family violence.

    In addition, the ABS is testing questions for potential inclusion in the PSS 2012. These could cover information about a person's housing arrangements the last time they separated from a violent current partner and their housing arrangements at the end of their last violent previous partner relationship. The ABS is proposing to seek information from respondents who have experienced current partner violence, to establish whether they have ever separated from their violent current partner and had to leave their home, and if so, where they went the last time they separated. The ABS is also seeking to establish from respondents who experienced violence from a previous partner, when they left their last violent previous partner, whether they had to leave their home, and if so, where they went when the relationship finally ended: for example, whether they stayed with a friend or relative, slept rough, stayed in a refuge or shelter, stayed in temporary accommodation e.g. motel etc., or elsewhere. If they went to multiple places, they will be asked for the place in which they spent the most time.

    While not a complete picture of where people went every time they separated during all relationships, if the testing is successful, this will provide an indication of what accommodation was used by people the last time they separated from their violent partner/s.


    FUTURE CENSUS IDENTIFICATION
      For the 2011 Census, the ABS will repeat the successful 2006 practice of jurisdictional lists and the 'green sticker' approach for supported accommodation arrangements. This generated a count very close to Census week SAAP counts compiled by AIHW, but one with richer information about the homeless. This ensures that those who are seeking refuge from domestic violence and staying in supported accommodation will be included in the estimates based on Census data.

      The ABS has also 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 may be staying with friends and relatives. While this may improve the under-estimation of homelessness in 2011, some people who have escaped domestic or family violence may still not be reported as having no usual address, or not be recorded on a Census form at all, and therefore they cannot be identified in homeless estimation of the Census dataset.

      The Census based estimates of homelessness by those escaping domestic or family violence are acknowledged as being underestimates, but using administrative data on those seeking support from services, and other sources such as the data collected from the ABS General Social Survey, and in future surveys such as the next Personal Safety Survey, will help to provide a broader understanding of the homeless circumstances of these people.

      The ABS will draw on the expertise of the Homelessness Statistics Reference Group to further understand homelessness due to domestic violence.



      Previous PageNext Page