Preface

Youth Homelessness in Australia 2006 is the first report from Counting The Homeless 2006. This project is a national data collection funded by the Australian Government in partnership with all state and territory governments. The project has been coordinated by the Homelessness Taskforce Branch in the Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA) who put in a great deal of effort to establish Counting The Homeless 2006 as a fully supported national project. The 2006 data collection follows the earlier work on the ABS Census data in 2001.

The Australian Bureau of Statistics (ABS) is a key partner in this project. A good deal of systematic attention went into improving the Census count of homeless people in 2006. The analysis of the main Census data file will be undertaken within the framework of the Australian Census Analytic Program (ACAP), where researchers work closely with ABS staff.

The analysis of school data is supplemented by information from the SAAP National Data Collection, administered by the Australian Institute of Health and Welfare (AIHW). This important data has been collected since 1996.

Counting The Homeless 2006 has been supported by peak bodies such as Homelessness Australia (HA), the Council to Homeless Persons (CHP) and the National Youth Coalition for Housing (NYCH), as well as by officers in the various government departments responsible for homelessness programs. Finally, the school census would not have been possible without the dedicated input of school welfare staff across Australia.

David Mackenzie,
Institute for Social Research, Swinburne University

Chris Chamberlain,
Centre for Applied Social Research, RMIT University

August 2008

[ Return to Top   Return to Section]

Executive Summary

• This report presents the main findings from the third national census of homeless school students and estimates the total number of homeless young people aged 12 to 18. The report uses two definitions of homelessness.

• First the cultural definition of homelessness is used to estimate the number of homeless young people. This definition distinguishes between primary, secondary and tertiary homelessness. The Australian Bureau of Statistics (ABS) uses this definition to enumerate the homeless population.

• Second, the report uses a service delivery definition of homelessness to estimate the number of homeless school students. The service delivery definition includes the categories in the cultural definition, but it takes into account that service providers often assist young people after they cease to be homeless. Schools were asked to include young people in their icensus return if they had been homeless within the last three months and were recieving continuing support.

HOMELESS SCHOOL STUDENTS

• There were 9,389 homeless school students in census week 2006 using the service delivery definition compared with 12,227 in 2001. Student homeless has decreased overall.
• The rate of homelessness decreased from 10 homeless school students per 1,000 of the school population in 2001 to seven homeless students per 1,000 in 2006.

HOMELESS YOUTH POPULATION

• The number of homeless youth decreased from 26,060 in census week 2001 using the cultural definition to 21,940 in 2006 (Table 1). The number of homeless youth decreased in all states and territories, except for Western Australia where there was an increase from 3,508 to 4,280.
• The rate of homelessness Australia-wide has decreased from 14 cases per 1,000 of the youth population to 11 cases per 1,000.

Table 1: Homeless youth and rate per 1,000; 2001 and 2006

		2001		2006
	N	Rate	N	Rate
NSW	6,242	10	4,987	8
Vic	4,663	10	3,896	8
Qld	6,381	18	4,469	11
WA	3,508	18	4,280	21
SA	2,394	17	2,129	15
Tas	1,008	21	770	16
ACT	400	13	307	10
NT	1,464	69	1,102	50
Australia	26,060	14	21,940	11

VULNERABLE GROUPS

• Young people from alternative family types are more at risk of becoming homeless than students from conventional nuclear families. Two-thirds (67 per cent) of the homeless students were from blended or single parent families, whereas only 15 per cent of the homeless students came from families where both biological parents were together.
• Indigenous youth are more at-risk of homelessness than non-Indigenous young people. The research found that 19 per cent of homeless school students were Indigenous, whereas 3.9 per cent of young Australians 12-18 years identified as Indigenous at the 2006 Census.
• Young people who have been in the state care and protection are over-represented in the homeless student population. Schools reported that 15 per cent of homeless students were known to have been in state care and protection, whereas, only 0.5 per cent of 10-17 years olds were in out of home care.

EXPLAINING THE HISTORICAL TREND

Table 2: Estimates of homeless youth aged 12 to 18; 1991 to 2006

	1991	1994	2001	2006
Age 12 to 18	8,500-10,800	21,000	26,060	21,940
Rate per 1,000	N/A	12	14	11

• Table 1 shows the estimated number of homeless youth aged 12–18 years from 1991 to 2006.
• The number of homeless youth aged 12 to 18 doubled from about 10,000 in 1991 to 21,000 in 1994. Over the same period, the number of young people aged 16 to 17 on Young Homeless Allowance also doubled from 4,500 to 9,900.
• The early 1990s was a period of deep economic recession with high youth unemployment (nearly 30%). Most homeless students dropped out of school quickly and many became long-term unemployed.
• A decrease in unemployment does not necessarily result in a decrease in homelessness. Youth unemployment declined after 1994 dropping to 18 per cent in 2001, but youth homelessness continued to increase to 26,060 in August 2001.
• The number of homeless youth decreased from 26,060 in census week 2001 to 21,940 in 2006. The two factors most likely to be associated with the change are early intervention and an improved labour market for young people.
• The youth unemployment rate was 18 per cent in June 2001 and this had decreased to per cent by June 2006. During the same period, youth homelessness decreased by per cent. It seems unlikely that the decline in youth unemployment was either large enough, or sudden enough, to explain much of the decline in youth homelessness.
• The major change that did occur after 2001 was the increase in early intervention services targeting homeless and at risk teenagers. The establishment of the Reconnect program between 1999-2003 was a major Australian Government early intervention initiative that has achieved good outcomes for homeless and at risk students.
• School welfare resources have been increased substantially in some states and territories since the late 1990s. Many schools facilitate family reconciliation so that at-risk students remain at home and homeless student return home. School welfare staff also support young people who need to live independently. These days it is more common to find schools working with community agencies on a range of issues, including support for homeless students.
• Early intervention is not always successful and many students still fall through the early intervention net. Nonetheless, the implementation of early intervention appears to account for most of the decrease in youth homelessness since 2001.

[ Return to Top   Return to Section]

1 Introduction

This report presents the main findings from the third national census of homeless school students. The research was carried out at the same time as the Australian Bureau of Statistics (ABS) was conducting the National Census of Population and Housing in August, 2006. The report also estimates the total number of homeless young people aged 12 to 18, using information from the school census combined with SAAP data. The school census findings will attract some interest in their own right, but the main purpose of the research was to produce the data necessary to support the ABS analysis of the homeless population.

The ABS used the cultural definition of homelessness to enumerate the homeless population on Census night. The cultural definition distinguishes between 'primary', 'secondary' and 'tertiary' categories of homelessness. Data was collected using a 'service delivery' definition of homelessness, which included the categories of the cultural definition. Service delivery definitions are broader and take into account that service providers often assist young people after they have ceased to be homeless. In the initial data collection we used the same service delivery definition as was used at previous censuses (MacKenzie and Chamberlain 1995; Chamberlain and MacKenzie 2002).

1.1 Definition of Homelessness

The cultural definition of homelessness is now widely used by policy makers and researchers in Australia (House of Representatives 1995; Northwood 1997; Department of Health and Family Services 1997; Charman, McClelland, Montague and Sully 1997; Driscoll and Wood 1998; LenMac Consulting 2005; Shelter WA 2005). It is a broad definition that takes into account the various circumstances of homeless people as they move around, and it is a definition that can be operationalised using the census.

The basic idea underpinning the cultural definition is that there are shared community standards about the minimum accommodation that people can expect to achieve in contemporary society (Chamberlain and MacKenzie 1992). The minimum accommodation for a single person (or couple) is a small rental flat with a bedroom, living room, kitchen and bathroom and an element of security of tenure provided by a lease. Below the cultural 'standard' people are homeless and above the minimum they are housed. This has led to the identification of 'primary', 'secondary' and 'tertiary' homelessness on Census night.

Primary homelessness includes all people without conventional accommodation, such as people living on the streets, sleeping in parks, squatting in derelict buildings, or using cars or railway carriages for temporary shelter.

Secondary homelessness includes people who move frequently from one form of temporary shelter to another. On census night, it includes all people staying in emergency or transitional accommodation provided under the Supported Accommodation Assistance Program (SAAP). Secondary homelessness also includes people residing temporarily with other households because they have no accommodation of their own and people who are staying temporarily in boarding houses.

Tertiary homelessness refers to people staying in boarding houses on a medium to long- term basis, defined as 13 weeks or longer. Residents of private boarding houses are homeless because they do not have a separate bedroom and living room; they do not have kitchen and bathroom facilities of their own; and their accommodation is not self-contained. They are homeless because their accommodation does not have the characteristics identified in the minimum community standard.

The cultural definition identifies people as homeless on the basis of their housing circumstances on census night, but this does not mean that there are three distinct 'segments' in the homeless population. Most homeless people move around between various forms of temporary accommodation (McCaughey 1992; Hanover Welfare Services 1995; Chamberlain and MacKenzie 1998, Ch.2; Bartholomew 1999, Ch.6). In a study of 4,200 homeless people Chamberlain, Johnson and Theobald (2007, Ch.3) found that 92 per cent were in the secondary category (moving around), but half (49 per cent) had slept rough and 85 per cent had been in a boarding house at some point. Most homeless people are transient.

Service providers often use definitions of homelessness that are broader than the cultural definition because they work with people who are 'at risk' as well as people who are attempting to return to secure accommodation. Our data collection used the same service delivery definition as was used at previous censuses. Schools were asked to identify young people as homeless using the core categories in the cultural definition (primary, secondary and tertiary homelessness). However, schools were also asked to include young people in their census return if they had been homeless within the last three months and were in need of continuing support. This allowed welfare staff to identify young people using their service delivery definition, but it also provided data in the core categories of the cultural definition. The results for the census of homeless school students are initially presented using this service delivery definition.

1.2 Structure of Report

Chapter 2 explains in broad terms the two-step methodology used to produce an estimate of the homeless youth population in the age group 12 to 18 years. A more detailed explanation of how the population estimate is constructed is given in Chapter 4 together with the relevant results.

Chapter 3 presents the main findings from the census of homeless school students, using the service delivery definition. The findings include: the number of homeless young people in schools; the number in state and Catholic schools; where they were staying in census week; the social characteristics of the sample; and those groups most at risk.

Chapter 4 provides estimates of the homeless population aged 12 to 18. These are the findings that will be used to supplement the ABS analysis of homelessness. This analysis uses the cultural definition following the precedents established at previous censuses. The ABS does not include people who are 'at risk', or formerly homeless people when conducting a census count. The estimate was 21,940, down from 26,060 in 2001.

Chapter 5 examines evidence from the field visits and the case studies to describe what is happening for young people on the ground. We describe what early intervention means in practical terms. Chapter 5 also identifies the strategies that help young people to move to independent living if family reconciliation is not possible.

Chapter 6 summarises the key findings from the research. We argue that improved early intervention strategies mainly account for the decrease in youth homelessness between 2001 and 2006.

[ Return to Top   Return to Section]

2 Methodology

The methodology used to estimate the size of the homeless youth population 12 - 18 years of age is a two-step process. The first step involves enumerating the number of homeless school students, who can be counted because they are still attending school. The second step involves estimating from the number of homeless school students to the broader population of this age group. This estimation is briefly described below but explained in more detail in Chapter 4.

The national census of homeless school students was carried out in the second week of August 2006, at the same time as the Australian Bureau of Statistics (ABS) was conducting the fifteenth National Census of Population and Housing on August 8.. On July 26 we sent a letter by fax and email to the Principals of all government and Catholic secondary school across Australia (N=2,025). It gave them details about the proposed research, explained who supported the project, and asked for their cooperation. The letter indicated that the ABS was implementing a special strategy to count homeless people at the 2006 census, and that the census of homeless school students was part of this strategy. Following the precedent established in the first census (1994), we did not include non-Catholic private schools in the research.

Principals were informed that their school would receive two forms. The first will 'ask for your best estimate of the number of homeless youth in your school, based on your local knowledge, and some brief details about these young people'. The second form will 'provide space for a case study which will inform a deeper understanding of what is happening to homeless students'. Schools were instructed not to record student names on either form.

Principals were asked to nominate one person to oversee the data collection in their school. It was suggested that the school counsellor or student welfare coordinator was 'probably the best person'. Alternatively, it 'could be the Vice-Principal responsible for student welfare'. Schools were told that the forms would be sent by fax and email and could be returned in either format.

The census used a method based on collating 'local knowledge'. This is a diverse body of everyday knowledge which emerges naturally in communities such as schools. Teachers and students rarely quantify this experiential knowledge, but in most schools at least a few people will know if a young person is homeless. The young person may tell a friend, confide in their favourite teacher, or approach a welfare coordinator for help. It is also common for other students to convey this information to a staff member at some point. The census asked one person in each school to bring together this disparate local knowledge into a quantified estimate of the number of homeless students.

On August 6, the census forms were faxed and emailed to all schools with a request that they be returned one week later. During the weeks following the census, we had telephone contact with about 300 schools to answer queries and check census returns. Most remote schools with Indigenous students were telephoned to clarify how the definition of homelessness should be operationalised. Table 2.1 shows that 99 per cent of schools completed a census return (2,017 schools out of 2,025). The response rate was comparable with the response rate at previous censuses.

Table 2.1: Response rate for census of homeless students, 2006 (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Response rate	99	99	99	99	100	100	100	100	99

Schools were asked to provide a case study of a homeless student where they had detailed knowledge of what had happened. They were asked to 'tell the story of why the young person left home', what has happened since and how they are managing at school. We also asked for details about age, gender, family structure, length of time homeless and effective strategies. Schools returned 560 case histories. This extensive qualitative data provided important insights into what is happening on the ground. In this report, we use a small amount of qualitative data to illustrate various points. In these cases, some details have been changed to observe our commitment to confidentiality and to protect the privacy of young people and their families. All names are fictitious.

Following the census, the research team carried out 173 field visits to schools. Schools were visited in every capital city and in most states there were some visits to schools in regional centres and country towns. One aim of the fieldwork was to ask staff how they had gathered information on homeless students and to check for undercounting. Another aim was to talk to welfare staff about 'best practice', and to gather information on how they were able to intervene effectively. We draw on the fieldwork to help interpret the quantitative results.

The second step in our analysis involves estimating the overall youth homeless population aged 12 to 18. The homeless youth population aged 12 to 18 includes school students, Tafe students, unemployed teenagers and a small number of young people who have full-time work. If we could establish the proportion of school students in the homeless population, then we could estimate the number of homeless young people overall. The only large data base that can be used to estimate the proportion of school students in the population is the SAAP National Data Collection. In order to 'estimate up' we make the assumption that the characteristics of the youth population in SAAP reflect the characteristics of the homeless population overall. In Chapter 4 we explain this methodology in detail.

[ Return to Top   Return to Section]

3 Census of homeless school students

The instructions for school staff explained the cultural definition of homelessness in plain language without reference to 'primary', 'secondary' and 'tertiary' homelessness. The instructions stated that a young person is considered homeless if they have left their family home and are living in any form of temporary accommodation including:

1. no conventional accommodation (e.g. streets, squat, car tent etc.)
2. temporary accommodation with friends or relatives or moving frequently between various forms of temporary accommodation
3. emergency accommodation in refuges or other crisis accommodation
4. other medium to longer term accommodation for people who have experienced homelessness (e.g. hostels and youth housing programs)
5. living in a single room in a boarding house.

We also used a service delivery definition which took into account that schools often provide assistance to students after they have ceased to be homeless. Welfare staff were told to include these young people in their return if they:

6. have been homeless within the last three months and were in need of continuing support.

School welfare staff were asked to record the number of young people in each of the catagories listed above (see appendix 1). This enabled us to ascertain the number of young people who were homeless according to the cultural definition of homelessness and the number of young people who were homeless according to the broader service delivery definition.

This chapter presents the main findings from the census using the service delivery definition. Chapter 4 presents the main findings using the cultural definition and estimates the overall homeless population aged 12 to 18.

Before analysing the main results, it is necessary to discuss the issue of undercounting. This is always a concern in any enumeration of homelessness. First, we summarise how we estimated the undercount at previous censuses. Then we explain how we estimated undercounting this time. After that we report: the number of homeless students; the number in state and Catholic schools; their accommodation at the time of the census; the social characteristics of the sample; and those groups most at risk.

3.1 Undercounting

Following the first national census of homeless school students in May 1994, we made 100 field visits to schools across the country. We found that some welfare staff were worried about undercounting, particularly in large schools where welfare staff thought there could be homeless students who had not come to their attention. We acknowledged the problem in our report, but we made only a small correction because we had not systematically discussed the issue with every school during fieldwork. We estimated an undercount of between 5 per cent and 10 per cent (MacKenzie and Chamberlain 1995).

Following the second census of homeless school students, we had telephone conversations with about 500 schools. These conversations confirmed that many welfare staff were worried about undercounting, and that it was more difficult to estimate the number of homeless students in large schools. In 2001, we made an adjustment for undercounting based on school size (Chamberlain and MacKenzie 2002). For large schools with an enrolment of 800 or more, we made an adjustment of 10 per cent. For medium-sized schools with an enrolment of between 400 and 799 we made an adjustment of five per cent, and for small schools (less than 400 students), we made no adjustment. The overall adjustment was 6.7 per cent.

In 2006, we made 173 field visits to schools following the census, more than we had initially planned. The sample was chosen purposively from schools that reported three or more homeless students. Our estimate for undercounting relates to this population. This time our fieldwork was systematic. We investigated how the census data had been gathered in each school that we visited, and we tried to make a judgment about the reliability of the information.

Two-thirds of the schools provided census figures that were confirmed during the field visits. In these schools we concluded that there had been no undercounting. Everything 'checked out' and there were no grounds for thinking that the school had missed a significant number of homeless students. The welfare team had a clear method for gathering the information. They were also confident that the school had procedures in place so that a student who was out of home would come to the attention of welfare staff. For example, the Principal of a high school in Brisbane told us that her census return:

... was not an estimate ... We know our homeless kids. I got together with the welfare team and we worked it out. We look after our kids if they are out of home.

In a regional city in Western Australia, the Coordinator of Indigenous Education explained:

John and I sat down with every teacher in the school and we identified the kids who were out of home. Then we checked them against your definition.

The Deputy Principal of a high school in Victoria said:

The figures are accurate. Annie and I went through every kid in the school. It took a whole afternoon to do it.

The Principal of a small high school in Tasmania said:

It wasn't a guess. I know every kid in my school. We had six homeless students.

In one-third of schools, there was evidence of undercounting. This occurred for different reasons and our estimate of the undercount was based on our discussions with welfare staff. In some cases, the undercount occurred because the welfare teacher had used a restricted definition of homelessness when supplying the information. One school in the Northern Territory reported one young person sleeping rough. However, when questioned about young people 'couch surfing', the welfare person said:

Yes, we get plenty of those ... we have at least 10 at the moment.

It was common for welfare staff in large schools to report that it was difficult to estimate the number of homeless students. For example, Doreen worked as a counsellor in a school with 1,600 students. She reported 12 homeless students in census week, but she said:

Of course there could have been others. I reported the students that I was working with. There's lots of couch surfing - there could have been another 10 couch surfers, easily.

In a large school in Western Sydney, the Head Teacher Welfare had recorded the five homeless students that she was working with. However, as the interview progressed:

It was clear that she knew about other cases ... apparently kids are often out of home ... Later I got her to talk about Pacific Islander kids who couch surf. This was a 'big issue' in the school and there are a lot of them ... Then she remembered some kids who had been squatting a few weeks back ... the more we talked it became clear that there could have been another 15 homeless students.

In a high school in North Queensland, the school guidance officer recorded eight homeless students because 'I wanted to give you the cases I knew about'. In our discussion she reported that 'there could easily have been another 12'. In another school in New South Wales, the Principal said, 'there could easily have been another 10 that we didn't know about'. In these cases, welfare staff did not want to guess when reporting their census figures, so they recorded the number of homeless students they were working with.

However, when we talked with welfare staff about the issues in their school, they were usually able to make a judgment about the reliability of their figures. Most welfare staff had a lot of 'local knowledge'. In some cases, they thought there could be 'two or three couch surfers' they did not know about. In other cases, they said homelessness was common and the figure 'could easily be double'. On the basis of the information gathered during field visits we estimated the undercount for each state and territory.

3.2 Number of homeless students

All tables in this chapter present findings using the service delivery definition. Table 3.1 presents the overall findings for 2006, compared with 2001. First, there is the raw data from the census assuming 100 per cent response rate. It shows 7,795 homeless students in 2006 compared with 11,461 homeless students in 2001. Second, there is the adjustment for undercounting. It was 20.5 per cent in 2006 compared with 6.7 per cent in 2001. Third, Table 3.1 shows the final figure from the census after the adjustment for undercounting. There were 9,389 homeless students in 2006 compared with 12,227 in 2001. The rate of homelessness decreased from 10 homeless students per 1,000 of the school population to seven homeless students per 1,000.

Table 3.1: Estimate of homeless students and rate per 1,000

	2001	2006
No. of homeless (raw data)	11,461	7,795
Adjustment for undercounting (%)	6.7	20.5
No. of homeless (final figures)	12,227	9,389
Rate per 1,000 of school population	10	7

Table 3.2 compares the number of homeless students by state and territory for 2001 and 2006. The number of homeless students decreased in all states and territories except Western Australia where there was an increase from 829 to 1,155. The rate of homelessness was down significantly in five states: from 10 to six per 1,000 in Victoria; from 15 to nine per 1,000 in Queensland; from 14 to 11 per 1,000 in Tasmania; from 14 to eight per 1,000 in the ACT; and from 37 to 25 per 1,000 in the Northern Territory. However, the rate in the Northern Territory was between two and four times higher than in other states.

Table 3.2: Homeless students and rate per 1,000; 2001 and 2006

		2001		2006
	N	Rate	N	Rate
NSW	3,063	7	2,404	6
Vic	2,886	10	1,993	6
Qld	3,073	15	1,993	9
WA	829	7	1,155	9
SA	1,018	11	953	10
Tas	574	14	372	11
ACT	324	14	207	8
NT	460	37	312	25
Australia	12,227	10	9,389	7

How do we interpret these findings? One possibility is that the number of homeless students has declined because many young people drop out of school soon after they become homeless. They remain homeless but are either unemployed or not in the labourforce. However, many welfare staff made comments such as:

Most are still at school ... one has gone ... we have a case management approach (Guidance officer, North Queensland).

All eight are still here. School is the only good thing in their lives (Assistant Principal, New South Wales).

All six are with us. I see them every day (Counsellor, Tasmania).

These comments contrast with what we were told after the first census. At that time, welfare staff reported that students who became homeless often left school at about the same time. Homeless students still have a higher risk of dropping out than other students. However, early intervention is well established in many localities and determined efforts are being made to help homeless students. Evidently many schools now hold on to their homeless students for longer as they try to assist them to sort out their issues and to remain at school.

The second explanation for the decline in homelessness is that schools are better at facilitating family reconciliation and assisting students to remain at school. Nowadays, schools and community agencies work more cooperatively than in the past. There was certainly fieldwork evidence that early intervention was succeeding for some students. Overall, there has been a significant decrease in the number of homeless school students, although the extent of the decrease across the country is uneven.

3.3 State and catholic schools

About four-fifths (79 per cent) of secondary schools are state schools and one-fifth (21 per cent) are Catholic schools. Table 3.3 shows 94 per cent of homeless teenagers were in the state system. There were 8,826 homeless students in state schools and 563 homeless teenagers in the Catholic system.

Table 3.3: Homeless students in state and Catholic schools (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
State	91	91	95	98	95	100	98	92	94
Catholic	9	9	5	2	5	0	2	8	6
	100	100	100	100	100	100	100	100	100

Nonetheless, 32 per cent of Catholic schools reported homeless students in census week, and another 20 per cent reported cases in the preceding 12 months. However, the numbers were usually small in Catholic schools. In contrast, 53 per cent of State schools reported homeless students in census week and another 24 per cent knew of cases in the preceding 12 months. The numbers were low in most state schools as well - 13 per cent reported 10 or more cases - but the problem of homelessness is more widespread in the state system.

3.4 Accomodation

Schools were asked to distinguish between young people who were currently homeless and young people who had been homeless within the preceding three months, but were in need of continuing support. Table 3.4 shows that three-quarters of the students were currently homeless and one-quarter were recently homeless. There was some variation between the states. In New South Wales, Victoria, South Australia and the ACT about 70 to 75 per cent were currently homeless. In Queensland, Western Australia and Tasmania about 80 per cent were currently homeless. In the Northern Territory most were currently homeless.

Table 3.4: Currently homeless and recently homeless students (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Currently homeless	71	70	77	80	74	78	72	92	75
Recently homeless	29	30	23	20	26	22	28	8	25
	100	100	100	100	100	100	100	100	100

Table 3.5 shows that 84 per cent of the homeless young people were in various forms of temporary accommodation. When school students first leave home they typically stay with other households, and then they begin to move from one friend's place to another if they remain homeless for any significant time. A typical example is James, a 14 year old student, who lived with his mother but had a bad relationship with his step-father. James reported that his step-father was 'dealing' and there were 'visitors at all times of the day and night'. The conflict between James and his step-father increased and James began 'staying away for a few nights' without permission. After a major argument, James left home and was taken in by another family. However, this arrangement broke down after four weeks and since then James had been couch surfing.

Table 3.5: Accommodation of homeless students (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Secondary
Friends, relatives, moving around, other temporary	79	80	88	86	87	80	60	94	84
Refuge, hostel, transitional housing	20	19	11	13	10	17	39	5	15
Primary Streets, squat, car, tent	1	1	1	1	3	3	1	1	1
	100	100	100	100	100	100	100	100	100

Another 15 per cent of the homeless students were in SAAP accommodation, such as youth refuges, hostels, medium to long-term housing or community placements. In the ACT schools reported that 39 per cent of their homeless students were in some form of government supported accommodation. The equivalent figure for New South Wales and Victoria was about 20 per cent, and it was lower in the other states and territories. These figures may be a function of young people's access to SAAP. If a young person becomes homeless in the ACT, all the SAAP services are nearby and therefore homeless students are more likely to find a SAAP place if one is available. On the other hand, if someone becomes homeless in a small regional centre in Queensland or Western Australia, there may be no SAAP service in their community and they may not be able to access a SAAP place elsewhere.

Table 3.6 gives information on the 25 per cent of homeless students who were in the process of returning to secure accommodation. Just under two-thirds of these young people were either boarding with other households or living in private rental accommodation, often sharing with other teenagers. These sorts of living arrangements can be unstable because young people have low incomes and are inexperienced at living independently. Welfare staff often provide ongoing support to these students. A typical example would be a school that was supporting a 16 year old girl who was boarding with another family. The school was monitoring the situation and a member of the welfare team had weekly meetings with the student to 'discuss attendance, punctuality, work completion ... what seems to work best for this student is regular meetings and to continually affirm her goals'.

Table 3.6: Accommodation of recently homeless students (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Rent flat or house / board with friends	68	62	54	70	68	79	53	29	63
Foster parents	13	20	30	8	14	8	19	0	18
Back with parents	15	11	15	22	13	10	23	71	15
Other	4	7	1	0	5	3	5	0	4
	100	100	100	100	100	100	100	100	100

Table 3.6 shows that 15 per cent of the 'recently homeless' were back with their parents. When young people return home welfare staff are often anxious about whether the reconciliation will be permanent and will tend to 'keep an eye' on the student.

Another 18 per cent of the 'recently homeless' were with foster parents. Some may have returned to foster families after a breakdown in family relations, but most of them were probably homeless students who had been placed with foster families. If family reconciliation is impossible, placement with a foster family is often considered a better option than independent living for younger teenagers. There has been a lot of criticism of the state care and protection system when it does not work well for young people. There are clearly many issues about the adequacy of state care and protection and leaving care arrangements. However, in the qualitative data provided by the schools there were also cases where foster care had worked well for students who were homeless.

3.5 Social characteristics

The first census of homeless school students found that 56 per cent of the students were female and 44 per cent were male. The second census recorded identical results (55 per cent female and 45 per cent male). Table 3.7 shows that the gender composition of the population remained the same in 2006 (57 per cent female and 43 per cent male). There were some differences between the states but the Northern Territory was the only jurisdiction that recorded a majority of males.

Table 3.7: Homeless students by sex (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Male	38	40	46	45	45	41	48	54	43
Female	62	60	54	55	55	59	52	46	57
	100	100	100	100	100	100	100	100	100

Most students are young when they have their first experience of homelessness and some move in and out of home a number of times before making a 'permanent break' from their family. Table 3.8 shows that one-quarter (25 per cent) of the young people were aged 14 or younger, 43 per cent were aged 15 or 16, and just over one-quarter (28 per cent) were 17 to 18. The distribution across the age groups was similar in New South Wales, Victoria, Queensland and South Australia. In Western Australia, 35 per cent of the homeless students were aged 14 or younger and in the Northern Territory it was 57 per cent. In both states, we suspect that Indigenous students were over-represented in the younger age groups.

Table 3.8: Homeless students by age group (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
14 or younger	25	20	25	35	20	17	18	57	25
15-16	42	41	48	44	43	44	33	32	43
17-18	29	36	27	17	26	36	48	11	28
19 or older	4	3	*	4	11	3	1	0	4
	100	100	100	100	100	100	100	100	100

* less than 0.5 per cent

3.6 Risk factors

Three groups are over-represented in the homeless population: Indigenous students; young people from single parent and blended families; and teenagers who have been in state care and protection. Most people from these backgrounds will not become homeless, but these social characteristics are often referred to as 'risk factors'.

Indigenous students

The 2001 census found that Indigenous people were over-represented in all sections of the homeless population (Chamberlain and MacKenzie 2003, p.5). In 2006, for the first time we asked schools to identify how many of their homeless students were Indigenous. Table 3.9 shows that 3.9 per cent of Australians aged 12 to 18 identified as Indigenous at the 2006 census, but 19 per cent of homeless school students were Indigenous. Indigenous young people were over-represented in the homeless population in all states and territories. In New South Wales 3.7 per cent of the youth population identify as Indigenous but 13 per cent of homeless students were Indigenous. In Western Australia half (49 per cent) of the homeless students were Indigenous as were nearly three-quarters (71 per cent) of the homeless students in the Northern Territory.

Table 3.9: Indigenous homeless students and Indigenous youth by state (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
% Indigenous students	13	5	17	49	14	11	7	71	19
% Indigenous youth, 12-18*	3.7	1.1	5.4	4.8	3.0	6.4	2.0	39.1	3.9

*Source: 2006 ABS Census

The Indigenous population is geographically diverse. Some Indigenous students live in remote communities, others live in country towns and a significant proportion live in capital cities. Nearly one-third (31 per cent) of Indigenous youth were in New South Wales, 28 per cent were in Queensland, 13 per cent were in Western Australia and 11 per cent were in the Northern Territory.

During the school data collection, we contacted most schools in remote communities and spoke with school staff. Our discussions focussed on the difficult issue of identifying 'homelessness' in an Indigenous context, where it is common for young people to stay with members of their extended family. In most remote communities, we were told that there were no young people who could reasonably be classified as 'homeless'. In these communities, children moved around between extended family, but they were said to be 'properly cared for' and attended school. A minority of schools reported homeless students. These were said to be 'dysfunctional' communities where there were high levels of violence, alcohol misuse, domestic violence and sexual abuse. These were some of the communities that have received media attention associated with the Federal Government's initiative to stop child abuse.

Indigenous people tend to be more mobile than non-Indigenous people and it is common for adults to travel to other communities to fulfil their kinship obligations. Sometimes children accompany parents, but other times children are left behind to be cared for by family members. In some communities this is effective but in others it is not, particularly if there are other problems in the community. In one household in the Northern Territory, three young people had been left in the care of various aunts, uncles and cousins. The school reported:

There are often no adults around ... the young people are without food or money or relatives to ask for food or money ... it appears the adults leave town for some reason.

In these circumstances, schooling is put at risk and students are effectively without adult supervision or support.

Single parents and blended families

Australian families have changed over the past 40 years. Divorce is now far more common than in the 1960s, as are single parent households, de facto relationships and blended families. Young people who come from alternative family types are more at risk of becoming homeless than students who grow up in conventional nuclear families. Table 3.10 shows that only 15 per cent of the homeless students came from families where they were living with both biological parents. In contrast, two-thirds (67 per cent) of the young people were living in blended families or with single parents. Schools provided many case histories that illustrated conflicts in these families.

In most blended families, the disputes were between the young people and their step- parents. The conflicts were over issues such as household rules, the young person's lifestyle, personality differences and a range of domestic issues. The conflicts had often been simmering for many months but a particular event brought the issue to a head. The conflict then escalated out of control and the young person left home. A typical case was Jane, 16, who resented her new step-father and was 'hurt that her mother usually sided with him'. During a bitter family argument her step-father became violent and her parents told her to leave:

The transition has not been easy ... she has ongoing problems with accommodation ... but she will not return home.

Mediation will work if both parties are prepared to compromise, but in this case Jane was not willing to return home.

Table 3.10: Family situation before leaving home (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Biological parents	17	19	14	11	10	18	14	11	15
Blended/single parent family	67	69	68	69	68	73	71	41	67
Foster parent(s)	6	6	6	8	2	2	11	1	5
Other	10	6	12	12	20	7	4	47	13
	100	100	100	100	100	100	100	100	100

Many single parents have financial problems because there is only one income and this can exacerbate family tensions. Others find it emotionally exhausting dealing with teenagers on their own. According to one Principal, 'some parents just seem to burn out'. A minority of single parents also have unresolved personal issues:

This lad's mother has a problem with drugs, alcohol, violence and money ... he moved out of home a number of times ... but finally left at 15 because of her violence ... he is couch surfing and moving frequently from friend to friend.

When conflict occurs in single parent families, there is no alternative adult who can take on the role of mediator and the likelihood of the young person leaving home is greater.

Table 3.10 shows that 13 per cent of young people had left 'other' family situations and that in the Northern Territory this figure was 47 percent. This category includes households where there were no biological parents present. In some cases the family unit had disintegrated because one or both parents had died, and in other cases one or both parents had abandoned the young person. In the Northern Territory, many Indigenous students were classified as 'other' because they were part of extended families.

State care and protection

Children and young people may be placed in the care of the state under a range of provisions according to state legislation. In 2005, there were 25,065 children and young people under care and protection orders in Australia (AIHW 2006, p.34). Roughly two-fifths were in the age range 10-17 years. The number of children and young people in out of home care has been increasing over the past decade from 13,979 in 1996 to 23,695 in 2005 (AIHW 2006, p.60). Only five per cent were in residential care, more than half were in foster care (54 per cent) and 40 per cent were living with relatives.

Young people in out of home care are a particularly vulnerable group although some placements work satisfactorily:

The school has supported her through meetings with the police and DOCS. She has continued at school and is apparently well supported by her foster family.

On the other hand, there are cases where young people are deeply disturbed by what has happened to them in care:

... placed in care by the Department. Since then the student has been in three crisis care placements ... currently the Department has placed him in the care of a family friend - not a permanent placement - and the Department is applying to the Children's court for him to become a ward, and it is likely that he will end up in a hostel.

These young people are particularly vulnerable and require intensive support.

Table 3.11: Homeless students in state care and protection (%)

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
% in state care and protection	13	17	16	19	12	13	20	8	15

Overall, 0.5 per cent of young people aged 10 to 17 were in out of home care in 2005. However, schools reported that 15 per cent of their homeless students had been in the state care and protection system (Table 3.11). Young people in state care and protection are particularly vulnerable.

3.7 SUMMARY

This chapter has reviewed the main findings from the third national census of homeless school students. There were 9,389 homeless students in census week compared with 12,227 in 2001. The rate of homelessness decreased from 10 homeless school students per 1,000 of the school population in 2001 to seven homeless students per 1,000 in 2006.

We found that 3.9 per cent of Australians aged 12 to 18 identified as Indigenous at the 2006 census, but 19 per cent of homeless school students were Indigenous. Indigenous young people were over-represented in the homeless population in all states and territories.

We also found that young people who come from alternative family types are more at risk of becoming homeless than students who grow up in conventional nuclear families. Only 15 per cent of the homeless students came from families where there both biological parents were together. In contrast, two-thirds (67 per cent) of the young people were living in blended or single parent families.

The third group who are over-represented in the homeless student population are young people who have been in the state care and protection system. Overall, 15 per cent of homeless students had been in state care and protection.

[ Return to Top   Return to Section]

4 How many homeless youth?

This chapter estimates the total number of homeless young people aged 12 to 18 by state and territory. Our main finding is that the homeless population aged 12 to 18 declined from 26,060 in census week 2001 to 21,940 in 2006.

The chapter follows the methodological precedents established in the analysis of the 2001 census (Chamberlain and MacKenzie 2002). For this analysis the cultural definition of homelessness is used, because the ABS only counts homeless people in the primary, secondary and tertiary categories of homelessness. The ABS does not count people who are 'at risk', or formerly homeless people who need continuing assistance. As explained previously, the census of homeless school students provides infomation on the number of homeless students using the cultural definition. The cultural definition is the basis for all calculations in this chapter.

4.1 Estimating up

Table 4.1 shows the number of homeless school students by state and territory using the cultural definition of homelessness. The homeless population aged 12 to 18 includes school students, TAFE students, unemployed teenagers and a small number of young people who have full-time work. If we could establish the proportion of school students in the population then we could estimate the number of homeless young people in each state and territory. The logic of 'estimating up' is similar to inferring from a random sample to a population. For example, if school students were 50 per cent of the homeless population aged 12 to 18 in New South Wales, then the homeless population in that state would be 3,400 (1,700x 100/50 = 3,400).

Table 4.1: Homeless school students, states and territories, 2006

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Number of homeless school students	1,700	1,401	1,527	922	761	289	148	287	7,035

The only large database that can be used to 'estimate up' at the present time is the National SAAP Data Collection at the Australian Institute of Health and Welfare. The National SAAP Data Collection is well established and has undergone many refinements and quality improvements over the past decade. It records information on all clients who use SAAP services throughout Australia. In 2005-06, 93 per cent of SAAP agencies participated in the data collection and 82 per cent of clients provided informed consent.

In practical terms, it is impossible to produce a random sample of the homeless population and even the ABS census count underestimates the number of young people couch surfing at the early stage of homelessness. In order to estimate up we make the assumption that the characteristics of the youth population in SAAP reflect the characteristics of the homeless youth population overall. This assumption underpinned previous analyses, but it cannot be independently verified. So long as it remains reasonable to assume that the proportion of school students in SAAP is reflective of the broader homeless youth population, then the SAAP data can be used for this purpose.

Our method of estimating up has been refined at each census, so we begin by reviewing how we estimated up on previous occasions. Then we explain how we will estimate up this time, drawing on information from 86,000 young people aged 12 to 18 who have received assistance under the SAAP program over the past five years.

4.2 Method 1

The first national census of homeless school students was carried out in May 1994. The census identified 7,700 school students using the cultural definition of homelessness. At that time there was no ongoing SAAP National Data Collection, but there were SAAP censuses carried out in May and November each year. We used data from the national SAAP census in May 1994 to estimate the proportion of school students in the homeless population (Chamberlain 1999, pp.23-26). There were usually between 12,000 and 14,000 people recorded in SAAP censuses and about 1,800 people would have been between 12 and 18. This was a relatively small dataset from which to estimate up to the overall population and it did not provide a reliable breakdown by state and territory.

Table 4.2 shows that the proportion of school students in the homeless population was 36.5 per cent and we calculated that there were 21,000 homeless young people aged 12 to 18 at that time (7,700 x 100/36.5 = 21,000). The national SAAP data also provided information on the proportion of TAFE students (5.5 per cent) and the proportion of young people who were unemployed or outside of the labour force (57.0 per cent). We used this information to estimate the numbers in these groups (Table 4.2).

The overall figure of 21,000 homeless young people aged 12 to 18 was used to make a statistical correction to the 1996 census because the census undercounted homeless youth aged 12 to 18. The reasons for this undercount have been explained elsewhere (Chamberlain 1999, Ch.3; Chamberlain and MacKenzie 2003, Ch.3).

Table 4.2: Education and labour force status, homeless youth; 1994

	Proportion in SAAP	Number of homeless youth
	%	N
School student	36.5	7,700
TAFE student	5.5	1,200
Unemployed (including not in labour force)	57.0	12,000
Other	1.0	200
	100.0	100.0

Source: Chamberlain and MacKenzie (1998, p.107)

4.3 Method 2

SAAP censuses were discontinued in 1996 when the SAAP National Data Collection began, so we needed an alternative data source in 2001. The annual SAAP National Data Collection records information on all persons using SAAP services from 1 July each year to 30 June the following year. It collects information on support periods but also uses an alphacode which allows some analysis of individual clients. An important advantage of the annual data collection is that it is a much larger database than a census count. In the year preceding the 2001 census young people aged 12 to 18 used SAAP services on 17,800 occasions (Chamberlain and MacKenzie 2002, Ch.4). In 82 per cent of cases (N=14,600) there was sufficient information to establish whether the young person was a school student, TAFE student, unemployed or not in the labour force. We used this as the best indicator of the proportion of school students in the population and the database was large enough to disaggregate the figures by state and territory.

Table 4.3 summarises three sets of figures. First, it gives the number of homeless school students in 2001 using the cultural definition. Second, it shows that school students utilised 32.6 per cent of support periods in SAAP in 2000-01, but there was marked variation in the proportion of school students by state and territory. Third, it estimates the number of homeless school students by state and territory (Table 4.3). There were just over 6,200 homeless young people in New South Wales and Queensland, 4,600 in Victoria, 3,500 in Western Australia, and smaller numbers in the other states and territories. These figures were used to make the statistical correction to the 2001 Census (Chamberlain and MacKenzie 2003, Ch.3).

Table 4.3: Estimated number of homeless young people aged 12 - 18; 2001

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Homeless school students	2,116	1,898	2,029	635	778	385	212	432	8,485
% of students in SAAP	33.9	40.7	31.8	18.1	32.5	38.2	53.0	29.5	32.6*
Number of homeless	6,242	4,663	6,381	3,508	2,394	1,008	400	1,464	26,060

Source: Chamberlain and MacKenzie (2002, Ch.4). * This is an average proportion of homeless students in SAAP Australia-wide reweighted according to the number of homeless youth. Previously, in Chamberlain and MacKenzie (2002, Ch.4) an unweighted proportion was presented.

4.4 Method 3

For 2006 we have used averaged figures to estimate up, derived from SAAP data covering the five years from 1 July 2001 to 30 June 2006. During this time young people aged 12 to 18 used SAAP on 87,000 occasions. These are known as 'support periods' and they can be for differing lengths of time. In 86,000 cases there was information on whether these young people were school students, TAFE students, unemployed (including not in the labour force), or in full- time work. The advantage of using averages calculated over five years is that they smooth out fluctuations in the data set and provides a better indicator of long-term trends. Table 4.4 shows the number of cases used to make the estimates by state and territory. It can be seen that there were over 20,000 cases in New South Wales and Victoria and even in the ACT and the Northern Territory the estimates were based on more than 2,000 cases.

Table 4.4: Sample size for estimating the proportion of students in SAAP

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
Number of cases	24,890	21,432	16,733	5,552	8,613	3,916	2,672	2,265	86,073

Table 4.5 shows that school students used 34.3 per cent of support periods in SAAP in the five years preceding the 2006 census. Table 4.5 also shows that there was significant variation in the proportion of school students by state and territory. The proportion of school students in the ACT was 48.2 per cent, followed by Tasmania (37.5 per cent), Victoria (36.0 per cent), and South Australia (35.7 per cent). In Queensland and New South Wales 34.1 per cent of those in SAAP were school students. The proportion of school students was 26 per cent in the Northern Territory and 21.5 per cent in Western Australia. These figures have been used to estimate the overall number of homeless young people aged 12 to 18 in census week.

Table 4.5: Students as a proportion of SAAP support periods for homeless youth; 2001-2006

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
% of students	34.09	35.96	34.17	21.54	35.74	37.51	48.20	26.05	34.31

Table 4.6 shows that the number of homeless young people aged 12 to 18 decreased from 26,060 in census week 2001 to 21,940 in 2006. The number of homeless young people was down in all states and territories except Western Australia where there was an increase from 3,508 to 4,280. Overall, the rate of homelessness across the country decreased from 14 cases per 1,000 of the youth population to 11 cases per 1,000.

Table 4.6: Homeless youth and rate per 1,000; 2001 and 2006

		2001		2006
	N	Rate	N	Rate
NSW	6,242	10	4,987	8
Vic	4,663	10	3,896	8
Qld	6,381	18	4,469	11
WA	3,508	18	4,280	21
SA	2,394	17	2,129	15
Tas	1,008	21	770	16
ACT	400	13	307	10
NT	1,464	69	1,102	50
Australia	26,060	14	21,940	11

There were some significant differences between the states (Table 4.6). In 2006, the rate of homelessness was lowest in New South Wales (8 per 1,000), Victoria (8 per 1,000) and the ACT (10 per 1,000) and these states also had the lowest rate in 2001. There was a sharp drop in homelessness in Queensland where the number of homeless youth fell from 6,380 to 4,470 and the rate dropped from 18 cases per 1,000 to 11 cases per 1,000.

The rate of homelessness has also decreased significantly in Tasmania, down from 21 cases per 1,000 to 16 cases per 1,000, and the rate decreased slightly in South Australia, down from 17 cases per 1,000 to 15 cases per 1,000. There was also a large decrease in the Northern Territory, down from 69 cases per 1,000 to 50 cases per 1,000, although the rate of homelessness was still much higher in the Northern Territory than elsewhere. Some caution should be exercised in the case of the Northern Territory result because many Indigenous youth do not attend secondary school.

Table 4.7: Homeless youth in education and labour force; 2006

	NSW	Vic	Qld	WA	SA	Tas	ACT	NT	Aus
School student	1,700	1,401	1,527	922	761	289	148	287	7,035
TAFE student	495	384	280	377	146	93	14	41	1,830
Unemployed*	2,714	2,034	2,582	2,887	1,196	381	136	747	12,677
Full-timework	78	77	80	94	26	7	9	27	398
Total	4,987	3,896	4,469	4,280	2,129	770	307	1,102	21,940

Table 4.7 shows the number of homeless young people in different segments of the homeless population by state and territory. Overall, it can be seen that there were just over 7,000 school students; about 1,800 TAFE students; about 400 who had full-time work; and 12,700 youth who were either unemployed or not actively looking for work, including a few hundred in part-time work.

Table 4.8 expresses the same data in percentage terms. In the ACT and Tasmania, about half of the homeless young people were in school or TAFE and the other half were unemployed. In New South Wales, Victoria, Queensland and South Australia, between 40 and 45 per cent of the homeless teenagers were in education and about 55 per cent were unemployed. In Western Australia and the Northern Territory, one-third of the young people were in education whereas two-thirds were unemployed. Evidently, the participation rate of homeless young people in education varies from state to state.

Table 4.8: Homeless young people in education, unemployment or full-time work (%)*

	ACT	Tas	NSW	Vic	Qld	SA	WA	NT
Education	52	50	44	46	40	43	31	30
Unemployed	45	49	54	52	58	56	67	68
Full-time work	3	1	2	2	2	1	2	2

* Education includes secondary school and TAFE

4.5 Summary

This chapter has reviewed three different methods to 'estimate up' from the census of homeless school students to the youth population overall. Following the first census of homeless school students we used data from the national SAAP census in May 1994 to estimate up. This was a relatively small data set from which to estimate up to the overall population.

Following the second census of homeless school students, we used information from the SAAP National Data Collection relating to the 12 months prior to the census. This was a larger data set containing 14,600 cases and it was possible to disaggregate the data by state and territory.

This time we have used SAAP data covering the five years from July 2001 to June 2006 to estimate up. We used information on 86,000 cases to estimate the proportion of school students in SAAP.

We found that the number of homeless youth decreased from 26,060 in census week 2001 to 21,940 in 2006. The numbers were down in all states and territories except Western Australia where there was an increase from 3,508 to 4,280. Overall, the rate of homelessness across the country decreased from 14 cases per 1,000 of the youth population to 11 cases per 1,000.

[ Return to Top   Return to Section]

5 Early Intervention in Practice

In 1989, the Human Rights and Equal Opportunity Commission published its landmark report, Our Homeless Children (the 'Burdekin' report). The report noted that some homeless young people continue their schooling, but 'probably the majority leave home and school almost simultaneously' (HREOC 1989, p.271). On the basis of the evidence presented to the Inquiry, the Human Rights Commissioners concluded that schools did little to assist homeless students. However, the Commissioners also argued that 'schools and teachers represent a critical resource which we must use effectively if we are to address the difficult issue of child and youth homelessness' (HREOC 1989, p.278). Their recommendations were 'lost' in media discussion of the report that focused on 'street kids' (Fopp 1989; National Clearinghouse for Youth Studies 1989). At that time, there were a few pioneering projects on homeless students but schools were not regarded as important sites for early intervention.

The issue of homeless school students came to public attention following the release of findings from the first national census of homeless school students which identified 11,000 homeless students in census week (MacKenzie and Chamberlain 1995). Following the census, we made field visits to 100 schools throughout Australia. We found that schools and community agencies were not working closely on youth homelessness and that most homeless students dropped out of school (Mackenzie and Chamberlain 1995, p.26).

Since that time, there has been an explicit turn in youth policy towards building an early intervention and prevention capacity in schools and local communities. One of the first initiatives undertaken by the Howard Government in 1996 was on youth homelessness. A Prime Ministerial Taskforce was set up in 1996 to oversee 26 'early intervention' pilot projects, including an extensive evaluation. The Taskforce reported encouraging results from the pilot projects and the Australian Government established the Reconnect program in 1999. This was a major initiative to provide support for homeless students and young people at risk. There are now 98 Reconnect services across Australia with about 200 early intervention workers. Most state and territory governments have strengthened the welfare infrastructure in schools since 2000, and several state and territory jurisdictions have focused specifically on youth homelessness as part of their overall homelessness strategies. SAAP services are also more 'early intervention' oriented than they were 10 years ago. The investment in early intervention over the past five to 10 years might be expected to have had an impact and there is evidence from this research to suggest that this is the case.

In this chapter we use data from the case studies and the field interviews to discuss what is happening on the ground for homeless students. In some cases welfare staff assist young people who are 'at risk' rather than actually homeless. In other cases welfare staff assist young people who have either run away from home or have been 'kicked out'. For a homeless student there are a limited number of pathways: they can return home following family reconciliation; they can remain at school but continue to be homeless, usually staying in temporary accommodation; they can be assisted to live independently while continuing at school; and they can leave school in which case they usually remain homeless. Figure 5.1 sets out these trajectories. We begin by looking at strategies that can be employed to avoid homelessness, then we focus on strategies that are used with students on the different pathways.

Figure 1: Pathways for homeless students

5.1 Avoiding Homelessness through mediation

Schools are often the 'first to know' agency when young people experience family problems. Before a young person runs away or is kicked out, there has usually been a period of worsening conflict within the family. In some cases, school welfare staff can identify 'at risk' students because their behaviour changes in unexpected ways. In other cases young people self refer to welfare staff for assistance and, occasionally, parents ask for help. Where incipient family issues are leading towards the likelihood of the young person leaving home, family mediation can often succeed in defusing conflict so that homelessness is averted. One school reported that:

This student has been involved in disputes with her parents over an extended period of time. We are having counselling sessions ... and they are coming to the point where family mediation will be possible.

An experienced counsellor told us she has a range of strategies if young people are thinking of leaving home:

If they come to me before they leave home I can often stop them. First I tell them how hard it will be ... then I might suggest that I have a talk with Mum or Dad ... or perhaps a family conference.

Mediation requires that both parties are prepared to accept the ground rules of the mediation process and the impartiality of the mediator. Various people can initiate mediation including the student, the parent(s), a year coordinator or the school welfare person. In one case, a 16 year old girl approached her school counsellor for assistance because her mother had drug and alcohol issues. The counsellor attempted to initiate family mediation but the mother did not want to be involved. In these cases, welfare staff may turn to professional agencies such as Reconnect for assistance, and it can take time for both teenagers and parents to agree to counselling. In some schools, Principals use their 'authority' to initiate mediation.

In another case, the school counsellor referred the young person to Reconnect:

Referral made to Reconnect to assist with the relationship between father and daughter ... This was resisted by female student. However, she agreed for me to mediate between her and her father, which I conducted through phone calls and individual meetings and, finally, through a joint meeting with her and her father.

Mediation improved the relationship and the young person remained at home. Reconnect deals with many cases of at risk student involved in family conflict with a high level of success (Department of Family and Community Services 2003).

There are cases of young people 'at risk' where it is inappropriate to attempt family reconciliation. A welfare coordinator was concerned about two brothers aged 14 and 15. The older brother had reported that 'their step-father was uncontrollably angry and was hitting them ... and that drug use was normal in the family'. Welfare staff made a referral to 'child protection' and after further investigation the children were placed in care with a relative. Family reconciliation did not occur, but 'throughout this turmoil the one constant in their lives has been school ... the boys know they are safe here and taken seriously'.

There is evidence that family reconciliation is happening with some success on the ground.

5.2 Returning Home

When young people run away from home or are kicked out, it is common for this event to come to the attention of someone in the welfare team, particularly if the school has a comprehensive pastoral care program. In cases where the conflict is between parents and their teenage children over behavioural issues, the chances of children returning home are good. In one example, a fifteen year-old girl had a serious fight with her father about her behaviour at home and she ran away. Over the next four weeks she stayed with various friends. The school knew about her circumstances and reported that 'the Principal, the school counsellor and her year coordinator have offered support and guidance'. No outside agencies were involved and eventually the student returned home following a family conference organised by the school counsellor.

It is common for such arguments to be about how much freedom a student should have or about adherence to household rules. In another case, the parents of a 14 year old boy contacted the school after their son had left home. There had been a dispute between the boy and his step-father about how late he could stay out and the boy had 'taken off'. The guidance officer found out where he was staying from other students and contacted him. The student had stayed at various friends' places but was too scared to go home. The guidance officer coordinated a family mediation session and the student returned home shortly afterwards.

Another fifteen year-old girl wanted more freedom to 'stay out late and meet boys'. The parents disagreed and the girl 'took off' to stay with friends. The school guidance officer attempted mediation by talking with both the parents and the girl. Both parties were prepared to engage. The girl returned home after three weeks and the guidance officer noted that this outcome was due to the parents' 'willingness to accommodate some demands' and to 'make some changes'. The girl also compromised.

The case of a 17 year-old girl was unresolved at the time the Student Welfare Coordinator wrote this case study. There had been ongoing family conflict for some years, particularly because Dad 'had never got over the fact that his daughter had had an abortion'. The student conceded that she 'consistently pushed the boundaries'. After the girl was kicked out of home following a dispute about curfew times, she went to live with another family. The placement broke down after five weeks. After this the girl moved from one friend's house to another. The welfare coordinator arranged mediation through a local community service. Discussions were underway between the student and her parents. The student wanted to return home and family reconciliation seemed possible.

If parents have mental health problems or drug and alcohol related issues, then it is more difficult to resolve family conflict. Mediation between parents and children is more likely to be successful where the conflict is about a young person's attitudes or behaviour. Mediation will work if the parent(s) and the young person are prepared to think about each other's point of view and to make compromises. It is more difficult to resolve conflict if there is family violence, possibly fuelled by entrenched drug and alcohol use, or where sexual abuse has occurred. Young people who have been in the state care and protection system typically come from families that have serious issues, and in these cases family reconciliation is less likely.

5.3 Homeless but still at school

Some homeless students are unwilling to return home because of what has happened in their family, but are committed to remaining at school because this is the one point of stability in their life. Others want to remain at school but they need support as they grapple with the other difficulties in their lives. In most states, 60 to 70 per cent of the students who were recently homeless had moved to some form of independent accommodation, not back home.

One young woman, aged 15, had been brought up by her father after her mother had died. Father had several failed marriages behind him and the girl had been 'the victim of physical, emotional and mental abuse for a number of years'. The home situation became so intolerable that she:

Left home to stay in a park ... then she moved between different friends ... these arrangements broke down quickly ... after that she spent approximately a month staying in a youth refuge ... last week she moved in with her new boyfriend's family.

For a while her attendance at school was erratic, but the school guidance officer kept in contact with the young person 'which made her feel cared about' and encouraged her to return to school. With the student's permission, the guidance officer informed teachers about her situation and she was given extensions on school work where needed. At the same time:

The school gave the young person clear expectations about what was required for her to stay connected to school, when and who she needed to check in with about her situation, information about what assessment needed to be completed from the time she missed, what agencies she should contact and so forth.

This young woman had not moved on to independent living, and the school's youth support coordinator continued to provide support both inside and outside of school. Her year coordinator was also providing assistance and arranged for her to drop a subject. The girl was catching up on assessment and 'attending school regularly'.

In some schools, a great deal of support is mobilised to assist students who become homeless. In one school, the school counsellor reported that a 16 year old boy was linked in with a number of adults who played a significant role in his life. They included the school counsellor, his home group teacher and a youth worker from a local agency. In another school, a 16 year old girl had a good relationship with her student welfare coordinator and her Year 12 coordinator. Another student had a friendship with the school counsellor and her art teacher. Homeless students often remain at school because they have supportive relationships with adults who become 'significant others' in their lives. If homeless students do not have these relationships, they are more likely to drop out of school given the other practical and emotional difficulties that they have to deal with.

Some schools allow homeless students to take on part-time enrolments to ease the pressure on them. Other schools offer students extra help and more time for assignments. Other schools offer individual tutoring to reduce the risk of marginal students dropping out.

In many schools, there is a concerted effort by school staff and local agencies to provide support for homeless students:

I believe the pastoral care system in this college is central to helping our students resolve issues. We put them in contact with organisations where they can get support in their times of need. The students are provided with information that inform about services on campus and in the community. Home group teachers also provide counselling.

Although there is evidence that many homeless students stay at school for extended periods of time, this state of transition cannot continue indefinitely. Some homeless students remain at school for a considerable time before they move on to independent living, but they do need a lot of support. Most homeless students cannot move on to independent living until they access youth allowance at the 'unable to live at home' rate.

5.4 Living Independently

The provision of support for independent students is an integral part of early intervention. Once young people acquire sufficient income, then they can look for more stable accommodation. In some cases, they live with other families on a longer-term basis paying 'rent' or 'board' for their accommodation. Others move into private rental accommodation with other young people, and some move into longer-term supported accommodation.

Louise, 16, left home after a long period of conflict with her mother. Her mother had a new partner and the student could not adjust to the changed dynamics in the household. Louise approached the college youth worker for help. The school 'made contact with her mother to let her know her daughter was OK' and the school contacted the local Reconnect service. It was not possible to facilitate reconciliation, and the school advocated for Louise when she approached Centrelink for financial assistance. It took two months for financial assistance to be approved, but this enabled the student to move in with another family where she had her own room and paid board. The school reported that 'Louise continues to access support and counselling at school', but she attends regularly and her studies are on track.

Another school was supporting Serge,17, who had left home after conflict with his step- father. The school supported him when he applied for youth allowance at the 'unable to live at home' rate. Serge was now living with another family where he had his own room and paid board. The school noted:

He attends all his classes and is expected to graduate with a year 12 certificate. He is still coming to counselling and has started to visit his mother at the weekends.

Other students move into shared accommodation after they receive income support. Some welfare staff reported that this was a more 'risky' option and their preference was for independent students to board with other families. Welfare staff observed that young people who wanted to share flats often had limited options either because real estate agents did not see them as 'good' tenants or because rents were prohibitively expensive in their community. One girl aged 16 left home in a country town because her mother had 'drug, alcohol and mental health issues'. Rents in this town were 'cheap' and she was living with 'two other young people in a flat'. The school was providing counselling, some financial assistance and 'regular follow-up to ensure that she is on track'.

In another case, two brothers aged 15 and 16 had been left with an uncle when their mother 'returned to Vietnam'. The relationship with the uncle broke down and the boys couch surfed until they were awarded youth allowance. They were now living in a small flat:

The school supports the boys by providing text books, uniforms and some expenses for excursions ... The boys get support from myself and their year coordinator. The older boy said that he would have turned into a drug dealer, if we hadn't helped.

Some young people are able to access various types of supported accommodation where they have their 'own place' but on-going support is provided. Jade's mother was involved in a car accident and had sustained brain damage. Mother became 'aggressive and at times physically violent'. Family relationships began to fall apart and Jade left home. She spent time with her grandparents, then several weeks with various family friends. The school principal and the counsellor had contact with the parents but they resisted reconciliation and 'longer term options needed to be considered'. The school knew of a student house run by a local church:

There is a paid worker who lives upstairs. The house has up to four young people who attend local schools. There is excellent support for students and strong links are maintained with local schools.

Jade went on the waiting list and 'after some interviews and meals at the house', she was offered a place.

Another school was supporting an 18 year old Sudanese refugee. His father had been killed in the civil war and he had been separated from his mother who was 'in a refugee camp in Northern Uganda'. The student had moved a number of times since arriving in Australia, but a local youth agency had found him transitional housing where he was living with two younger brothers:

He is committed to study ... We have paid for his fees, books and uniform. It is a challenge for him to manage home and school, learning how to cook, budget and manage a household ... he is managing well considering the difficult circumstances ... earlier this year he received news that his mother was still alive and he was able to speak to her on the telephone.

There are several factors that affect whether homeless students can make a successful transition to independent living. First, it depends on whether they have the maturity to manage the additional responsibilities of running a flat, paying bills and cooking and cleaning for themselves. Some students are able to do this but others need continuing support while they acquire living skills.

The second factor which affects the transition to independent living is whether Centrelink assesses that it is unreasonable for the young person to live at home so they qualify for youth allowance at the higher rate. Where there is evidence of violence or sexual abuse or where the parents refuse to let the young person return, then the teenager can normally access youth allowance at the 'living away from home' rate. However, if the parents present as 'reasonable' or if they insist that the young person can return home, then access to the allowance is often delayed until it is clear that reconciliation has failed.

In most states, school welfare staff reported that they were regularly consulted by Centrelink social workers when students applied for the unable to live at home allowance. This contact was seen as beneficial, particularly in more serious cases. The exception was New South Wales where school counsellors reported little contact with Centrelink social workers. Any changes that make it more difficult for 16 and 17 year olds to access the living away from home allowance would tend to increase homelessness.

A third important factor is the affordability of housing in the local community. If housing continues to become less affordable this would have a serious impact on those on the lowest incomes, including homeless young people.

5.5 Early Intervention in practice

When we carried out fieldwork following the first census of homeless school students, we found that most homeless students dropped out of school and few schools had early intervention strategies in place. Our experience during the fieldwork in 2006 was different. These days it is common to find schools undertaking early intervention and school welfare staff are much better at facilitating reconciliation between young people and their parents. Schools are also better at supporting homeless students to remain at school and schools are better at assisting students to move to independent accommodation if they cannot return home.

Effective early intervention involves schools working with welfare agencies in their local community. Schools that are effective usually have the following characteristics:

• a holistic philosophy that gives as much priority to well-being and inclusion as academic achievement;
• good leadership from a principal who places a strong emphasis on educational achievement, but who also believes that the school should provide a supportive and caring environment for all students;
• a well-organised and active welfare team which meets regularly and oversees the provision of support (including case management) for at risk and homeless students;
• a strong pastoral care program with sufficient time for teachers to build strong long-term relationships with a small group of students;
• special programs for at-risk students including anti-bullying programs and activities designed to improve self-esteem;
• different educational pathways for students including vocational and academically oriented options and the provision of part-time enrolments.

This is an idealised list of good practice characteristics. It is relatively rare to find schools that score highly on all points. While it is common to find schools espousing good philosophical propositions, it is less common to find such philosophy deeply implemented in practice. Many schools where at-risk students are a relatively small group have a limited capacity to hold onto their most problematic students. On the other hand, schools with a larger proportion of disadvantaged students have often adapted creatively to the challenge of meeting the needs of these students, but complain that they do not have sufficient resources for the size of the problems they have to deal with.

Pastoral care is about building effective relationships between a small group of students and a particular teacher who is expected to take a special interest in their progress and well- being. Pastoral care teachers are often the first to know when a young person is in trouble at home. However, in many schools pastoral care programs have been squeezed out by all the other demands for time in the curriculum. While our fieldwork suggests that in general, schools are clearly doing better than a decade ago, there are still many homeless students who eventually drop out of school. In most states, over 50 per cent of homeless youth were no longer in education.

Schools that have effective early intervention strategies in place are well linked in with the youth and welfare services in their local community. But there have to be the appropriate services in the community to link with. Such links were uncommon when we carried out the first census of homeless school students. At that time many secondary schools operated as self-sufficient worlds focused on classroom teaching and the curriculum. Community agencies had relatively few contacts with their local schools. We commented that 'for early intervention to succeed, this cultural "Berlin Wall" must be beached' (Chamberlain and MacKenzie 1995, p.44). By 2006, we can say that this 'Berlin Wall' is well and truly down. These days most schools have links with a range of community agencies, including Reconnect if there is a service available in their local area.

SAAP agencies are now more aware of 'early intervention' and more likely to explore reconciliation. In Victoria a major initiative has been the Family Mediation and Reconciliation Program which has funded early intervention strategies for working with homeless youth in SAAP.

There is evidence of positive change, but the provision of early intervention is clearly incomplete in terms of overall need.

[ Return to Top   Return to Section]

6 Discussion

There are two main findings from this research. First, the number of homeless school students decreased from 12,227 in 2001 to 9,389 in 2006, using the service delivery definition of homelessness. Second, the number of homeless youth aged 12 to 18 decreased from 26,060 in 2001 to 21,940 in 2006, using the cultural definition of homelessness. First, we review changes in the number of homeless youth from 1991 to the present. Then we comment on the possible reasons for the decline in youth homelessness.

The Senate Standing Committee on Social Welfare (1982) concluded that it was impossible to give precise figures on the number of homeless youth, and all estimates prior to 1994 were based on rudimentary methodologies. In 1989, the Burdekin Report also commented that it was impossible to provide exact figures on homelessness, but the report estimated that there were 20,000 to 25,000 homeless young people aged 12 to 24 at any point in time (HREOC 1989). Early research by MacKenzie and Chamberlain (1992) concluded that a more realistic estimate was 15,000 to 19,000 on a typical night in 1991, including 8,500 to 10,800 homeless youth aged 12 to 18 (Chamberlain and MacKenzie 1998, Ch.7).

Table 6.1: Estimates of homeless population aged 12 to 18; 1991 to 2006

	1991	1994	2001	2006
Age 12 to 18	8,500-10,800	21,000	26,060	21,940
Rate per 1,000	N/A	12	14	11

Sources: MacKenzie and Chamberlain (1992); Chamberlain and MacKenzie (1998); Chamberlain and MacKenzie (2002)

N.B. The time series point in time estimates are based on the cultural definition of homelessness

The first reliable estimate of youth homelessness was based on the 1994 national census of homeless school students (MacKenzie and Chamberlain 1995). Table 6.1 shows that the number of homeless youth aged 12 to 18 apparently doubled from about 10,000 in 1991 to 21,000 in 1994. There was corroborating evidence for this interpretation: over the same period the number of young people aged 16 to 17 on Young Homeless Allowance also doubled from 4,500 to 9,900 (Chamberlain and MacKenzie 1998, p.110). The early 1990s was a period of deep economic recession when unemployment was above 10 per cent much of the time and youth unemployment was closer to 30 per cent. The sharp increase in unemployment in the early 1990s undoubtedly exacerbated homelessness. Our fieldwork in schools indicated that most homeless students dropped out of school quickly (Mackenzie and Chamberlain 1995), and many of these homeless teenagers became long-term unemployed. However, it does not follow that a decrease in unemployment necessarily results in a decrease in homelessness. Youth unemployment declined after 1994 dropping to 18 per cent in 2001, but youth homelessness continued to increase to 26,060 in August 2001. It could be that there is a time lag between a decline in unemployment and a decline in homelessness.

However, it is likely that homeless teenagers are amongst those groups least able to take advantage of improved labour market opportunities. It may be that there has to be a large drop in unemployment before there is a significant reduction in homelessness. How then do we explain the decline in youth homelessness since 2001?

Homelessness is widely accepted to be the result of both 'structural' and 'individual' factors (HREOC 1989; House of Representatives 1995). Individual factors relate to specific characteristics of young people or to the events that happen in particular families. Young people whose family support networks are weak are more vulnerable to homelessness, and this includes young people who have been taken into state care and protection. Family breakdown is a major cause of homelessness and can be a result of: sexual, physical and emotional abuse; conflicting expectations about appropriate behaviour on the part of young people and their parents; conflicts between young people and step-parents; and the incompatible personality characteristics of family members that exacerbate interpersonal conflicts. These individual factors may lead to a breakdown in family relationships with the result that young people either leave home or are forced to leave home.

Structural factors relate to the contextual factors that impede young people making the transition to independent living. One important structural factor is the state of the local labour market which affects whether young people can access employment and gain sufficient income to pay rent and other living costs. A second structural factor is the affordability of housing which affects how much people have to pay to rent independent accommodation or to join shared households in the private market. Other structural factors include the regulations that govern access to welfare payments for homeless students who wish to remain at school.

The characteristics of the local labour market have often been assumed to be the prime determinant when it comes to explaining increases or decreases in the level of homelessness (Neil and Fopp 1992). However, youth unemployment declined during the second half of the 1990s while over the same period youth homelessness increased. The relationship between these variables changed after 2001. In June 2001 the youth unemployment rate was 18 per cent and this had declined to 15 per cent by June 2006. During the same period, youth homelessness came down by 16 per cent. The small drop in the unemployment rate probably enabled some homeless teenagers, including school students, to obtain either casual or part-time work. However, it seems unlikely that the decline in unemployment was either large enough, or sudden enough, to explain much of the decline in youth homelessness.

The major change that did occur after 2001 was the increase in early intervention services targeting homeless and at risk teenagers. Early intervention can achieve family reconciliation by supporting homeless teenagers and their families to resolve difficulties. Where family reconciliation cannot be achieved, early intervention can assist homeless students to remain at school and make a successful transition to independent living. In the last five to seven years, there has been a significant increase in the provision of early intervention services in schools and local communities.

The establishment of the Reconnect program in 1999 was a major Australian Government early intervention initiative to reduce youth homelessness. Reconnect targets young people aged 12 to 18 in order to achieve:

• family reconciliation wherever practicable, between homeless young people or those at risk of homeless, and their families; and
• engagement of young homeless people, or those at risk of homelessness, with employment, education, training and community (DFACS 2003, p.22).

The program was implemented in phases and was not fully operational until 2003. There were 29 services funded in December 1999. This was followed by two additional funding rounds with another 44 services becoming operational progressively from July 2000. A further 25 services were funded from July 2001, although not all became operational immediately (DFACS 2003, p.22). By 2003, there were 98 Reconnect services across the country, most having either two or three early intervention workers.

The evaluation of Reconnect (DFACS 2003, p.8) found that the program had achieved good outcomes for young people and their families, particularly in 'improving stability in young people's living situations' and 'achieving family reconciliation by increasing the capacity of families to manage conflict and to improve communication'. The main evaluation findings were that:

• the number of young people reporting good or very good skills in managing family conflict increased from 12 per cent to 44 percent after their engagement with Reconnect, and the number of young people reporting poor or very poor skills decreased from 66 per cent to 16 per cent;
• parents reported similar improvements in their capacity to manage conflict, up from 11 per cent to 47 per cent after engagement with Reconnect workers;
• three-quarters of the young people and parents reported overall improvement in the situation that led them to Reconnect;
• more than three-quarters of the young people and 80 per cent of parents expressed satisfaction with their case worker and their ease of access to Reconnect (DFACS 2003, pp.9-11).

During our field visits, we found that welfare staff often referred to Reconnect as an effective program, but not all schools have access to a local Reconnect service.The Reconnect program was approved for a further cycle of recurrent funding of $85m by Minister Nigel Scullion in 2007.

School welfare resources have been increased substantially in some states and territories since the late 1990s. As described in Chapter 5, many schools now use early intervention strategies to facilitate family reconciliation, but school welfare staff also provide support to young people who cannot return home and need to live independently. Some schools operate case management programs and most schools routinely work with community agencies on a range of issues, including support for homeless students. Working with community agencies was rare a decade ago but it is now commonplace. SAAP services are not early intervention services, but the issue of early intervention was raised as a strategic issue during SAAP IV. It is now more common to find SAAP services making referrals to Reconnect or contacting families to investigate the possibility of early intervention.

The two factors most likely to be associated with the decrease in youth homelessness are early intervention and the improved labour market for young people. The labour market has improved and it is easier for school students to find part-time or casual employment. However, the small decline in youth unemployment between 2001 and 2006 cannot explain most of the decrease in youth homelessness.

On the other hand, Australia's early intervention capacity has developed considerably over the past 10 years, and particularly over the past five years. The Reconnect program is a major success story as is the Youth Support Coordinators program in Queensland. These days, many schools work cooperatively with a range of community agencies and a lot effort goes into supporting homeless students to remain at school. Not all of these efforts are successful and many students still fall through the early intervention net. Nonetheless, early intervention appears to account for most of the decrease in youth homelessness since

[ Return to Top   Return to Section]

References

Australian Institute of Health and Welfare (AIHW) (2006) Child Protection in Australia 2004-05. AIHW cat. no. CWS 26. Canberra: AIHW (Child Welfare Series no. 38).

Bartholomew, T. (1999) A Long Way from Home: Family Homelessness in the Current Welfare Context, Geelong: Deakin University Press.

Chamberlain, C. and MacKenzie, D. (1992) 'Understanding Contemporary Homelessness: Issues of Definition and Meaning', Australian Journal of Social Issues, 27(4), 274-297.

Chamberlain, C. and MacKenzie, D. (1995) House of Representatives Report on Aspects of Youth Homelessness: A Summary and a Commentary, Melbourne: Centre for Youth Affairs, RMIT.

Chamberlain, C. and MacKenzie, D. (1998) Youth Homelessness: Early Intervention and Prevention, Sydney: Australian Centre for Equity through Education.

Chamberlain, C. (1999) Counting the Homeless: Implications for Policy Development, Canberra: Australian Bureau of Statistics, Catalogue No. 2041.0.

Chamberlain, C. and MacKenzie, D. (2002) Youth Homelessness 2001, Melbourne: RMIT University.

Chamberlain, C. and MacKenzie, D. (2003) Counting the Homeless 2001, Canberra: Australian Bureau of Statistics, Catalogue N0.2050.0.

Chamberlain, C., Johnson, G. and Theobald, J. (2007) Homelessness in Melbourne: Confronting the Challenge, Melbourne: RMIT Publishing, Informit E-library.

Charman, M., McClelland, A., Montague, M. and Sully, V. (1997) Strengthening Attachments: The Report of the Prevention of Youth Homelessness Project, Melbourne: Brotherhood of St Laurence.

Department of Family and Community Services (2003) I'm looking at the Future: Evaluation Report of Reconnect, Canberra: DFACS.

Department of Health and Family Services (1997) Tender Document for the Enumeration of Homelessness in Australia (mimeo).

Driscoll, K. and Wood, L. (1998) A Public Life: Disadvantage and Homelessness in the Capital City, Melbourne: City of Melbourne.

Fopp, R. (1989) 'Press Coverage of the Burdekin Report', Youth Studies Australia, 8(4), 10-12.

Hanover Welfare Services (1995) Hanover Stats and Facts: A Triannual Report on Homelessness, South Melbourne: Hanover Welfare Services.

House of Representatives (1995) A Report on Aspects of Youth Homelessness, Canberra: Australian Government Publishing Service (The Morris Report).

Human Rights and Equal Opportunity Commission (1989) Our Homeless Children, Canberra: Australian Government Publishing Service (The Burdekin Report).

LenMac Consulting Pty Ltd (2005) Sustainable Housing After Homelessness, Final Research Report to the National SAAP Coordination and Development Committee.

MacKenzie, D. and Chamberlain, C. (1992) 'How Many Homeless Youth?', Youth Studies Australia, 11(4), 14-22.

MacKenzie, D. and Chamberlain, C. (1995) 'The National Census of Homeless School Students', Youth Studies Australia, 14(1), 22-28.

McCaughey, J. (1992) Where Now? Homeless Families in the 1990s, Australian Institute of Family Studies, Policy Background Paper, No. 8.

National Clearinghouse for Youth Studies (1989) Responses to Burdekin: Government and Non-Government Responses to Our Homeless Youth, Hobart: NCYS.

Northwood, K. (1997) 1996 Census: Homeless Enumeration Strategy, Canberra: Australian Bureau of Statistics.

Neil, C. and Fopp, R. (1992) Homelessness in Australia: Causes and Consequences, Victoria: CSIRO Division of Building Construction and Engineering.

Rossi, P. (1989) Down and Out in America: the Origins of Homelessness, Chicago: University of Chicago Press.

Shelter WA (2005) The Boarding House System in Metropolitan Perth, Shelter WA Occasional Paper 2005-2.

Senate Standing Committee on Social Welfare (1982) Report on Homeless Youth, Canberra: Australian Government Publishing Service.

[ Return to Top   Return to Section]

Appendix 1: Census form

[ Return to Top   Return to Section]

Appendix 2: Diagram

 

[ Return to Top   Return to Section]