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: Marginally housed

KEY ISSUES: MARGINALLY HOUSED


SUMMARY OF KEY POINTS

  • There is no agreed definition of 'marginal housing'
  • Counting the Homeless 2006 (ABS cat. no. 2050.0) identifies and analyses a population described as 'marginal residents of caravan parks' and reports details of this population separately to and alongside the homeless population.
  • The ABS will undertake further analysis on this group and other characteristics which could be used to identify and understand the marginally housed
  • The ABS will, along with the Homelessness Statistics Reference Group, explore how to define and measure the marginally housed population in the Census, and present the findings alongside estimates of homelessness.


INTRODUCTION

Marginal housing is housing which is not adequate for reasons such as no security of tenure, or the dwelling may be overcrowded, or it does not meet 'minimum expectations' - it does not have basic facilities or adequate facilities. Not only should marginal housing be understood in its own right, some of those who do not have security of tenure, or have poor quality housing, or live in overcrowded dwellings may be at risk of homelessness in the future. Some of the submissions received in the Review urged the ABS to consider changes to the application of the cultural definition of homelessness to include those marginally housed in caravan parks as part of the homeless population, in line with those living in boarding houses.

An example of people living in marginal housing is those who live in caravans in caravan parks. The marginally housed may live in caravans in caravan parks because they have either exhausted all other forms of accommodation, or are waiting for supported accommodation or for other forms of assistance, or cannot access other housing (for example they cannot afford to rent conventional housing). The difficulty is in identifying the marginally housed separately to others in caravan parks such as holiday makers and those making a lifestyle decision to live in a caravan park.


USING THE CENSUS TO IDENTIFY THE MARGINALLY HOUSED

The Census of Population and Housing does not collect information on the security of tenure or the quality of the dwelling, even though for some it can be inferred based on tenure type. Overcrowded dwellings can be determined from the Census and have been discussed separately in the Overcrowding Chapter. The Census provides information on residents of caravan parks but it can not at this time, differentiate between those in caravans, cabins or houseboats. As Chamberlain and MacKenzie (2008) report, it is now more common to find cabins as the main type of accommodation in caravan parks, and often have better facilities than a caravan, such as a separate kitchen and bathroom and more than one bedroom. Chamberlain and MacKenzie (2008) report that somewhere between one-quarter and one-half of marginal residents in caravan parks may be living in cabins and this undermines the argument that they are part of the tertiary homeless population.

When using the Census to measure marginal housing in caravan parks, the analysis starts with people who are using caravan parks as a usual place to live, and then using demographic and dwelling characteristics to infer whether the occupants have no alternative place to live. The Counting the Homeless (CTH) (ABS cat. no. 2050.0) approach was to undertake separate analysis of a population called 'marginal residents of caravan parks' and comparing the characteristics of this group with those of selected homeless sub-populations also derived in CTH. In CTH persons who are marginal residents of caravan parks are defined as:
  • persons who were 'at home' on Census night AND
  • persons living in a caravan, cabin or houseboat and the dwelling is located in a caravan park/residential park or camping ground AND
  • persons with a tenure of 'renting', those who did not state their tenure, those with a tenure of 'rent free' and 'other tenure type' AND
  • dwellings where no person in the dwelling was working full-time (defined to be 35 or more hours per week).

CTH analysed this group in more detail, comparing key characteristics to CTH estimates for selected sub-populations of homeless people including those found in CTH to be in the boarding house and 'visiting friends and relatives' groups. CTH provides arguments for and against including this group in the homeless population and concluded that there are convincing arguments on both sides. The CTH analysis does not include those that have been identified as 'marginal residents of caravan parks' in the homeless population, stating that although this group do not have a strong financial position they could be living in cabin accommodation and CTH "treat 'marginal residents of caravan parks', outside of the tertiary [homeless] population" (CTH 2008).

The ABS has briefly analysed the records for the 17,497 persons included as 'marginal residents of caravan parks' group. The people in this group have a range of tenures which may indicate a diverse range of circumstances. For example the CTH estimate for 'marginal residents of caravan parks' includes 6,818 people who had 'not stated' for their tenure type. However of this group 2,099 people (or 31%) were fully imputed records (i.e. no person was identified at the dwelling on Census night). A further 1,571 (or 23%) were 'head counted' by the collector (forms were not completed however the collector ascertained the number of persons and the sex of the occupants and then the rest of the demographic characteristics were imputed). Some 1,062 additional records were partially imputed. The significant amount of imputation for this group means that it is impossible to draw conclusions about the 27% of the CTH marginally housed population who were imputed, nor for the 18% who did not state the tenure type, and all of these records, would need to be removed for future analysis to be undertaken to understand the characteristics of this group.

The following table shows selected demographics for all persons who were in a caravan, cabin, or houseboat in a caravan park / residential park or camping ground and who were 'at home' on Census night. This is broader than the definition used in CTH of marginally housed in a caravan park but provides a picture of usual residents of caravan parks, to enable further consideration of those who may be marginally housed (Fully imputed records and those where only a head count was obtained by the collector have been removed from the analysis).

As shown in the Table below, for the population who were renting, 35% were not in the labour force and 40% were employed, working either part or full time or away from work. Over half of renters were living at the same address one year prior to the Census, and a quarter had been living at the same address five years prior to the Census.

TABLE 1 - PERSONS IN A CARAVAN, CABIN OR HOUSEHOLD IN A CARAVAN PARK/RESIDENTIAL PARK OR CAMPING GROUND WHO WERE 'AT HOME' ON CENSUS NIGHT, Tenure by Selected demographics - 2006
Owned outright Owned with a mortgage Being purchased under a rent/buy scheme Rented Being occupied rent-free Being occupied under a life tenure scheme Other tenure type Not stated Total
no.
no.
no.
no.
no.
no.
no.
no.
no.
Labour force status
Employed – worked full-time
4,660
654
85
4,264
355
18
156
613
10,805
Employed – worked part-time
2,592
222
32
1,592
88
3
48
220
4,797
Employed – away from work
659
108
10
386
30
27
173
1,393
Unemployed – looking for full-time work
826
25
9
1,329
3
7
14
85
2,298
Unemployed – looking for part-time work
314
13
9
332
8
10
29
715
Not in the labour force
19,403
308
99
5,560
155
58
165
1,316
27,064
Not stated
1,079
20
9
469
17
4
22
1,248
2,868
Not applicable (b)
886
252
30
1,797
76
52
207
3,300
Usual address one year ago
Same as in 2006
23,837
1,132
170
8,255
431
71
314
2,033
36,243
Elsewhere in Australia
5,537
429
102
6,293
268
14
145
845
13,633
Overseas in 2005
176
6
4
411
7
10
69
683
Not stated
810
22
6
631
17
3
22
923
2,437
Not applicable
56
14
139
8
3
21
241
Usual address five years ago
Same as in 2006
15,685
601
77
3,948
186
45
183
1,371
22,096
Elsewhere in Australia
10,950
733
137
7,795
376
38
201
974
21,204
Overseas in 2001
443
36
6
714
45
17
111
1,372
Same as in 2005
2,015
128
42
1,503
72
3
39
276
4,078
Not stated
1,052
32
12
1,065
27
3
36
1,094
3,321
Not applicable
275
74
8
704
25
17
66
1,169
Total
30,420
1,604
282
15,729
731
89
493
3,892
53,240
– nil or rounded to zero (including null cells)
(a) Cells in this table have been randomly adjusted to avoid the release of confidential data.
(b) Includes children 0–14 years.


Further analysis is being undertaken by the ABS to gain a statistical picture of residents of caravan parks and to identify characteristics for a group living in caravan parks who could be identified as marginally housed. Focus group work will be considered after the 2011 Census to improve understanding of the circumstances of people enumerated in these settings.

In addition to those who are living in caravan parks, there are other forms of housing that would be considered to be marginal, such as those who live in substandard housing, do not have security of tenure and/or those who are living in overcrowded housing. Overcrowded housing is discussed separately in the Overcrowding Chapter.

The Census can not be used to give an indication of the adequacy or security of tenure of a dwelling beyond the proxy variables discussed above. However the ABS collects relevant data in the housing module included six yearly in the Survey of Income and Housing (SIH). These data were last collected in 2007–08. The SIH collects data from residents of private dwellings (excluding households from very remote areas). SIH data show that:
  • 16% of private dwellings were reported to have major structural problems
  • 57% reported that repairs had been undertaken in the 12 months prior to the survey
  • 2.6% of dwellings needed 1 or more bedrooms
  • Of renter households, 14% had no formal lease, and a further 14% were on month by month leases
  • 4.5% of reference persons reported being dissatisfied or very dissatisfied with their dwelling (ABS 2009c; ABS 2009b).


FUTURE CENSUS IDENTIFICATION

As a result of the review, including the feedback received through consultation and the submission process, and a further brief exploration undertaken into marginal housing, the ABS, with advice from the Homelessness Statistics Reference Group, will explore how to define the marginally housed population. The ABS will then determine if there are adequate characteristics, collected in the Census of Population and Housing which could be used to separately identify the population. Results of investigations will be presented alongside estimates of the homeless population.



Previous PageNext Page