Apprenticeships in homelessness: a quantitative study.
Cebulla, Andreas ; Goodwin-Smith, Ian
Introduction
The slowing of the Australian economy over the last half decade has
momentarily refocused public attention on the risk of unemployment and,
if arguably less so, the economic and social plight of young people. In
early 2014, unemployment among 15 to 24 year olds sits around twice the
national average, whilst that of 15 to 19 years olds remains close to
three times the national rate (ABS 2014; ACCI 2010).
While the causes of youth unemployment are complex and varied,
those first to lose their jobs, or those encountering barriers to labour
market entry, are often those least well educated or trained. Youth
training and training policy in Australia have remained fraught with
problems and are critiqued for failing to meet societal and economic
need and skill demands (McDowell et al. 2011). Apprenticeship training
has repeatedly been in the spotlight, in particular because of the
continued high dropout rate among apprentices and trainees (McDowell et
al. 2011; Karmel & Mlotkowski 2010a; 2010b), and the harsh economic
reality of having to meet living and training expenses on an
apprenticeship or traineeship award (Bittman et al. 2007; Schutz et al.
2013).
This paper explores a very specific social and personal crisis
faced by a small but far from negligible fraction of trainees and
apprentices: the risk of homelessness. Using data from a survey of
trainees and apprentices in South Australia, we estimate the scale of
the risk of homelessness for this group of young people. Homelessness
among apprentices and trainees remains a largely unreported and
unrecognised social phenomenon, despite its potentially adverse effects
on a person's ability to continue their training. Community housing
projects such as 'The house that builds people' project in
Canberra, (1) recognise the social need for improving the housing
situation of apprentices in a hands-on manner. In this paper, we seek to
make the empirical case for paying greater attention to the housing
situation of those undertaking apprenticeships and traineeships.
In the following sections, we first summarise some of the
literature on youth training in Australia, its successes and challenges,
and we discuss the crossover with the literature on housing instability
and homelessness among young people. We then move on to introduce the
specific aims and objectives of our study, before presenting our
approach and the survey data generated, and discussing its strengths and
weaknesses. The next section presents our survey results. The final
section of this paper draws together the evidence and presents
recommendations for improved housing and youth training policy and
practice.
Background and literature
Who are the apprentices?
In recent years, Australia's apprentices have become an
increasingly diverse group of learners, most notably with respect to
their age characteristics. As recently as 1995, three in four
apprenticeships and traineeships were taken up by young adults below the
age of 20; by 2009, this had decreased to fewer than two in five (NCVER
2011a: Table 3, own calculation). In the September quarter 2011, 14 per
cent of current apprentices and trainees were aged 45 and over, as were
13 per cent of new apprentices and trainees (NCVER 2011b, own
calculations). Just over half of all current apprentices and trainees
were younger than 25 years of age (NCVER 2011b).
Not everyone who starts a traineeship or apprenticeship also
completes his or her training. In fact, high course drop-out (around 48
per cent; see McDowell et al. 2011) has been a major issue for youth
training in Australia for some time, posing cost burdens on training
systems, as well as reducing prospects for sustained employment for
those dropping out. Low apprenticeship awards have been found to be a
factor contributing to course drop-out from training (Karmel &
Mlotkowski 2010a; Bittman et al. 2007), alongside workplace-related
causes (poor relationship with the employer/trainer), and lack of
support and loss of interest in the work (McDowell et al. 2011; Snell
& Hart 2008). The effect of awards has, however, not been uniform,
raising drop-out rates especially in non-trade occupations, such as
sales, and community and personal services, where wage premia upon
completion can be minimal or indeed negative. In general, expected
post-apprenticeship earnings tend to be more important in affecting
completion rates than (current) training wages per se (Karmel &
Mlotkowski 2010b; 2011).
Notwithstanding the primacy of post-apprenticeship earnings in
regards to completion rates, low earnings during traineeships or
apprenticeships can pose a significant risk to apprentices and their
training activities. Low earnings during training acutely affect current
living conditions, as trainees and apprentices manage on limited
financial resources. Apprentices' and trainees' expenses
include many direct costs associated with training. These often compete
with expenses associated with leisure activities retained from
pre-apprenticeship times as young people seek to maintain their
friendship networks. Reports on apprenticeships and the risk of drop-out
do not typically capture the broader picture of living as an apprentice
and, most notably, lack consideration of housing stability (NCVER 2010;
Karmel & Mlotkowski 2011). Likewise, reports on homelessness
typically fail to capture information adequately about apprenticeship
and other educational pathways (Chamberlain & Mackenzie 2008;
Homelessness Taskforce 2008), and their risk of homelessness.
The living arrangements of young learners in Australia
The living arrangements of young Australians have undergone a
transformation in recent decades. Young adults' extended education
and the resultant delay in financial independence and self-sufficiency
have increased the age until which young adults remain living with their
parents (ABS 2009; Mission Australia 2011). Many young people commencing
work-related training nonetheless choose to leave the parental home, or
are forced to do so as a result of travel-to-work distances that make
daily commutes between parental home and work difficult; others leave as
a result of parental conflict, including physical or sexual abuse (Crane
et al. 1996; Rosenthal et al. 2006; Martijn & Sharpe 2006). Those
who find themselves living away from the parental home typically face
the challenge of meeting high costs of renting, with or without the
financial backing of their parents (Burke et al. 2002a; Schutz et al.
2013). Individuals and households on low and moderate incomes have
increasingly been 'squeezed out of home ownership and both private
and social rental' (Vitis et al. 2010: 1) as wages have failed to
keep up with rising house prices and rentals, and eligibility criteria
for social housing have been tightened, while affordable housing stock
has decreased (Yates 2011). In a study using 1999 data, about one in
five job-seeking young adults cited housing difficulties as a
'major factor' in previously abandoning tertiary studies
(Burke et al. 2002b). Almost 40 per cent of students in receipt of Rent
Assistance from the Commonwealth Government stated that receipt of this
support had been a major factor in their decision in favour of studying.
Low income and a tight, often expensive housing market can combine
to cause an increased risk of housing instability and, ultimately,
homelessness. A large proportion of Australia's homeless people are
young adults or adolescents. On census night 2006, almost one
third--about 32,000--of all homeless people were aged 12-24 years (Muir
et al. 2009). A report by the Australian Institute of Health and Welfare
(AIHW 2011), using Supported Accommodation Assistance Program (SAAP)
data, concluded that about four per cent of SAAP clients aged 18 or over
had been in post-secondary education or training prior to receiving SAAP
support in 2010-11 (AIHW 2011: 29). Almost half of those people --48 per
cent--were living transiently with friends and relatives--typically
understood as examples of 'secondary homelessness'. (2) In an
earlier study, MacKenzie and Chamberlain (2008) had estimated that, in
2006, about 1,800 of Australia's almost 22,000 homeless youth aged
between 12 and 18 had been TAFE students.
More recent statistics about the receipt of the Youth Allowance
(YA) provided by the Australian Government Department of Human Services
(DHS), equally illustrate the risk of housing instability among young
people. YA is a means-tested benefit for 16-24 year olds studying
full-time, undertaking a full-time Australian Apprenticeship, or looking
for full-time work. A variant of YA is the YA-UTLAH: the Allowance paid
to those for whom it is deemed 'Unreasonable to Live at Home'.
This unreasonableness is defined as resulting from 'extreme family
breakdown (other than normal parent/adolescent conflict)',
'serious risk of your physical or mental health if you continue to
live at home', or if parents or caregivers 'cannot provide a
suitable home as they do not have stable accommodation' (CYA 2013).
In August 2012, 17 per cent of all YA recipient apprentices--761 in
total--received the YA-UTLAH, down from around 25 per cent in previous
years. (3) Just over a third of all YA-UTLAH recipients were living in
New South Wales (36 per cent), while around a quarter were based in
Victoria (27 per cent) or Queensland (22 per cent). Six per cent of
YA-UTLAH recipients lived in South Australia. These statistics represent
significant youth cohorts who cannot rely on parental support to avoid
or overcome the housing-related stressors that contribute to housing
instability and risks of homelessness.
For those affected, episodes of homelessness can entail not only
adverse social, physical and emotional effects, but also long-term
problems in work and labour market participation as a result of
interrupted or incomplete training and education. Conversely, housing
security and stability have practical merits in that they increase
opportunities for, and probabilities of, obtaining and retaining
employment (Dupuis & Thorns 1998; Mavromaras et al. 2012).
Study objectives and approach
The present research explores the prevalence of homelessness among
apprentices and trainees, focusing on those in South Australia and using
new survey data collected for this purpose. Specifically, the research
sought to record (i) current housing arrangements of apprentices and
trainees and (ii) past experiences of homelessness during traineeships
or apprenticeships. To address these issues, the study adopted a mixed
method approach that included stakeholder interviews (largely private
and third-sector youth training providers and business organisations),
focus groups and one-to-one interviews with apprentices, an online
survey of members of the Australian Council of Social Service (ACOSS)
network, a telephone survey of a sample of providers of the Australian
Apprenticeship Mentoring Program (AAMP) and, finally, a survey of
apprentices and trainees. The current paper focuses on the analysis and
findings of the latter survey, adding new statistical evidence.
Survey of apprentices and trainees in South Australia
In March 2013, 3,000 apprentices and trainees who were registered
at the time or during the previous 12 months in vocation and educational
training (VET) courses in South Australia (SA) were invited to take part
in a brief survey regarding their living arrangements. The survey asked
about the type of accommodation in which these apprentices and trainees
were living, whether they were living independently or with others, and
how much they spent on their accommodation. The survey also specifically
asked apprentices and trainees about accommodation arrangements when
they were away from their homes for block training, that is, intensive
class-based sessions held on campus.
The survey was facilitated by the South Australian Department of
Further Education, Employment, Science and Technology (DFEEST). DFEEST
maintained the state's register of apprentices and trainees. In
line with regulations protecting the identity of apprentices and
trainees, and the confidentiality of the research, DFEEST prepared a
survey sample on behalf of the researchers, drawing a random sample of
1,000 apprentices and 1,000 trainees who had been registered on SA VET
courses in February 2013. An additional sample of 500 former apprentices
and 500 former trainees who had cancelled or withdrawn from such courses
in the previous 12 months was also randomly selected. As at 8 February
2013, 11,738 apprentices and 28,761 trainees were registered in VET
courses in SA, while 5,387 apprentices and trainees had cancelled or
withdrawn from their courses in the preceding 12 months.
Past and present apprentices were invited to complete the survey
online. In addition, half of the group of current and past apprentices
were also sent a paper copy of the survey questionnaire, alongside a
reply paid envelope. Reminder letters to those who had not yet returned
their questionnaires or participated online were issued in late April
2013, with a deadline for participation of 10 May. Again, this work was
undertaken by DFEEST. Conscious of the challenges associated with
surveying apprentices and trainees, survey participants were offered the
incentive to be included in a draw for a tablet computer.
Two hundred and twenty-one (221) apprentices and trainees
participated in the survey, giving a response rate of 7.4 per cent.
Amongst the respondents were 106 current apprentices (response rate:
10.6 per cent), 81 current trainees (8.1 per cent), 21 past (i.e.,
cancelled or withdrawn) apprentices (4.2 per cent), and 13 past trainees
(2.6 per cent). Unfortunately, it was not possible to establish how many
of the non-returns resulted from out-dated address information that was
particularly likely to have adversely affected the survey of past
apprentices and past trainees. More generally, this survey, which sought
to measure the prevalence of homelessness among learners, suffered from
having to rely on surveying people with known addresses. It was thus
very likely to miss an unknown number of apprentices and trainees who
were no longer living at their address, including some who might have
been homeless at the time. Moreover, youth workers participating in the
qualitative part of this study told us that young homeless people,
especially young men, were typically reluctant to talk about their
experiences and, for this reason, would have been less likely to
participate in this survey. Our own efforts to engage young men with
previous homelessness episodes in this study confirmed this. Both
caveats imply that this survey was most likely to under-estimate the
full extent of the homelessness risk among apprentices and trainees in
South Australia.
Survey responses were weighted to reflect the sex, age and
vocational characteristics of the original sample frame (see Appendix A
for details on the weighting). Whilst low, the response rate was not
atypical for a survey of this type, as colleagues at DFEEST also
confirmed. Post-program monitoring surveys conducted by the Department
of Employment (formerly DEEWR) of its Job Services Australia and
Disability Employment Services initiatives typically achieve a response
rate of no more than 25 per cent (AIHW 2013). These surveys are
typically better resourced than our survey was and include, for
instance, telephone follow-up. Compared with our inclusion of past
trainees and apprentices, the DoE surveys are conducted much sooner
after program participation--about 3 months after participating in
employment services.
This said, low response rates risk bias especially where
non-response is not random, that is, when those not responding differ
markedly in their characteristics or experiences from those who do
participate in the survey. Where a low response rate also results in
small case numbers, the robustness of statistical analyses may be
reduced. While measures of statistical significance can help to
ascertain the latter, weighting of the response sample can seek to
correct for response bias. To do so, various weights were constructed
and tested for their ability to match the response sample to the initial
sample of 2000 current apprentices and trainees.
DFEEST was able to provide the researchers with simple
cross-tabulations of the sex, age (under 25, or older) and type of
apprenticeship or traineeship of those in the sample, which were used to
develop sample weights (see Appendix A). In the absence of more detailed
information about the sample population--notably cell frequencies across
the three variables--the match remained imperfect, in particular with
respect to the age distribution among apprentices and the sex
distribution among trainees. Importantly, however, weighting improved
the match of courses attended, in particular by trainees. As will be
seen later in this paper, regression analysis found that it was course
type rather than age or sex that was most strongly associated with the
risk of homelessness. For this reason, weighted data was preferred to
unweighted data for the descriptive analysis. Moreover, weights that
simultaneously, if imperfectly, corrected for sex, age and course bias,
were preferred over weights that only corrected for one or two of these
variables, as they generated the best match across the three indicators.
Findings
In this section, we report the main results from our research,
starting with descriptive statistics profiling apprentices and trainees
in South Australia captured in this survey, before turning to a more
detailed statistical analysis of apprentices and trainees with past
homelessness episodes. The small number of responses inevitably
restricts the scope of some of the data analyses, recording many
findings as statistically non-significant even when large in scale.
Regardless of this caveat, we report these results where they are
critical for describing and understanding the housing situation of young
people in training. In some instances, we report the absolute number of
responses to survey questions alongside percentage response rates to
acknowledge that these statistics were based on small case numbers.
Because of their distinct characteristics, we analysed current
apprentices and current trainees separately. Past apprentices and past
trainees, however, were combined into one group and briefly examined for
key information about housing and homelessness. Because of their small
numbers, no formal analysis of the survey returns of past students was
undertaken, although they were included in the regression analysis
reported below to increase its statistical robustness.
Descriptive analysis
Current apprentices and trainees in South Australia captured in the
survey differed on some key demographic characteristics and vocational
course choices. Both apprentices and trainees were, in the majority,
male, but this was much more the case with respect to the former than
the latter: while 93 per cent of current apprentices were male, almost
two-thirds of the trainees were (63 per cent; Table 1). Trainees also
tended to be older than apprentices: 69 per cent of current trainees
were 25 years of age or older, compared with 29 per cent of current
apprentices. The mean age of apprentices was 26 years (median: 20),
while the mean age of trainees was 35 years (median: 34). The youngest
apprentice in the survey was 16 years of age and the oldest was 66. The
youngest trainee was 15 and the oldest 60 years of age. Reflecting these
age differences, the majority of current apprentices were single and
never married (68 per cent; compared with 27 per cent of trainees),
while half of current trainees (50 per cent) were married (compared with
23 per cent of apprentices).
Current apprentices surveyed were predominantly studying technical
and manufacturing or building trades (86 per cent); while current
trainees included in the survey were, in particular, attending courses
in business and management, retail, personal services, or transport (65
per cent).
Living arrangements
Housing differed markedly between apprentices and trainees, again
quite likely reflecting differences in age and also partnering. Thus,
whereas 61 per cent of apprentices were living with their parents, this
was true for only 27 per cent of trainees (Table 2). Trainees were more
likely than apprentices to be living in accommodation that they owned
and typically shared with a partner: 46 per cent of trainees did so,
compared with 22 per cent of apprentices. Apprentices (13 per cent) and
trainees (17 per cent) were similarly likely to be renting their
accommodation, either by themselves, with partners, or with others. A
fraction of apprentices reported squatting without paying rent (1 per
cent).
Renting and home ownership
Even when living with parents or other family relations,
apprentices and trainees typically paid for or contributed towards the
cost of their accommodation: 39 per cent of apprentices were paying rent
to parents or other family, as were 11 per cent of trainees. Whereas
only 2 per cent of apprentices owned their property outright and 21 per
cent were paying off a mortgage, the respective statistics for trainees
who reported higher levels of home ownership were 11 per cent and 38 per
cent. A further 11 per cent of apprentices and 18 per cent of trainees
paid rent to a landlord. Twenty-three per cent of apprentices and 17 per
cent of trainees lived rent-free, while a further two per cent of
apprentices and one per cent of trainees paid rent to a friend.
Rooms per person
Measured in terms of the number of rooms, trainees lived in smaller
accommodation on average, reporting a median of 4.9 rooms available to
them for exclusive or shared use, compared with 5.4 rooms available to
apprentices. The mean number of rooms available to a trainee was 0.69,
compared with 0.78 in the case of apprentices. Neither statistics were
significantly different between the two groups. However, both were
markedly below the estimated Australian average of about 2 rooms per
person (OECD 2011). Trainees were also less likely than apprentices to
have exclusive use of a room: whereas 34 per cent of trainees said they
had no room for their exclusive use, the same was true for 15 per cent
of apprentices.
Incomes, costs of housing, costs of living
Mean mortgage or rent payments per week amounted to $174 (in 2013
AUS$) for apprentices and $243 for trainees. Median weekly payments of
$100 for apprentices and $246 for trainees indicate that accommodation
costs were skewed towards the lower end of the distribution in the case
of apprentices, but towards the higher end in the case of trainees.
Information about income was available for 73 apprentices and 77
trainees. With reported mean incomes of $627 per week for apprentices
(median: $600) and $654 for trainees (median: $747), apprentices spent
about 27 per cent of their income on rent or mortgage payments (median:
17 per cent), whereas trainees spent about 38 per cent (median: 31). The
above statistics for trainees were somewhat inflated by the inclusion of
a few very high values of rent or mortgage as a percentage of income of
trainees. Excluding those exceeding 100 per cent, the mean decreased to
33 per cent, whereas the median remained at 31 per cent. All these
estimates excluded amounts for Youth Allowance or AUSTUDY payment
received by nine (five per cent) respondents, of whom only five had
reported the amount (ranging from $45 to $350) received.
By comparison, average full-time adult weekly earnings in South
Australia reported in the 2012 Year Book Australia amounted to
$1,184.20, dropping across Australia to $559 for the youngest group of
full-time employees aged 15 to 19 years (ABS 2012a: 302-309). No
breakdown for South Australia was reported by the ABS (2012a). Other
sources indicate that, across Australia, owners with a mortgage spend an
average of 18 per cent of the average gross income on housing costs
(including repaying the principal outstanding on the loan); while
renters from private landlords spend about 20 per cent; renters from
state and territory housing authorities expend about 19 per cent of the
gross income on housing costs (ABS 2013). Thirty per cent expenditure of
gross income on housing is commonly used as an indicator of
'housing stress' among low-income households (AHURI 2014).
Financial stress
We asked apprentices and trainees whether they had sought help in
the last 12 months, or whether they had been unable to pay bills or had
gone without food or heat in their home 'because they had been
"short of money'". Close to a quarter (23 per cent) of
apprentices and a fifth (21 per cent) of trainees reported having had to
seek help from friends or family in the last 12 months because of
financial problems. About a fifth of apprentices (19 per cent) and
one-sixth of trainees (15 per cent) had been unable to pay car
registration or insurance, while a sixth of both apprentices and
trainees (14 per cent; 15 per cent respectively) could not pay their
telephone, gas, or electricity bills. Between five and six per cent of
apprentices and trainees had been unable to pay their mortgage or rent
on time; or had gone without meals (seven per cent of apprentices and
trainees) as a result of their financial problems.
Home moves
The majority of apprentices and trainees had either always lived at
their current address or had lived there since before they commenced
their apprenticeships or traineeships. However, 23 per cent of
apprentices (unweighted N = 26) and 11 per cent of trainees (unweighted
N = 14) had moved to their current address during the course of or just
before their studies began. In some small proportion of instances, these
moves had been motivated by the need or desire to be closer to the place
of work: 21 per cent of apprentices and 20 per cent of trainees who
changed address had done so for that reason. Four per cent of home moves
by apprentices were driven by the wish to be closer to the place of
training.
As was noted earlier, trainees spent on average 38 per cent of
their earnings on paying for accommodation. Perhaps unsurprisingly,
therefore, 20 per cent of trainees who had moved home had done so
because they wanted to reduce their mortgage or rent payments. This
compared with eight per cent of apprentices moving home for the same
reason; in addition, one in eight apprentices (12.5 per cent) returned
to his or her parental home.
Living without permanent address
Survey respondents were asked whether, since beginning their
apprenticeship or traineeship, and other than during block training,
they had ever been without a permanent place to live. Ten per cent of
apprentices (unweighted N = 11) and six per cent of trainees (unweighted
N = 8) reported such episodes since starting their VET course.
Typically, these apprentices and trainees ended up staying with
relatives. Seven of the 11 apprentices (64 per cent, unweighted) who had
experienced episodes without a permanent address reported having done
so, while a further five (45 per cent) had stayed with friends.
Likewise, two of the eight trainees (25 per cent, unweighted) who had
reported episodes of living without a permanent address had spent these
periods with relatives, while another two (25 per cent, unweighted) had
spent them with friends. While these apprentices appeared to have
arranged comparatively stable and secure temporary accommodation, three
of the 11 (27 per cent) apprentices without a permanent address in the
past reported to have lived in caravans or to have slept rough on the
streets. One apprentice reported staying in a boarding house or hostel.
In total, and using weighted data, we estimate that about eight per
cent of apprentices and trainees had experienced homelessness whilst
undergoing their training. If those who were able to return to live with
their parents or who found shelter with other relatives are excluded,
this statistic decreases to about three per cent and identifies the
proportion of apprentices and trainees who stayed with friends
('couch surfing': see McLaughlin 2012), in a caravan, a
boarding house or hostel, or slept rough. These percentages are subject
to some uncertainty as a result of the complex survey design,
heterogeneity within the sampled population, and low response rates. The
statistics should therefore be read as merely indicative of a
homelessness risk among apprentices and trainees. They are also likely
to underestimate the full extent of homelessness among apprentices and
trainees since, as discussed earlier, our sampling would have excluded
apprentices and trainees with no permanent address at the time of the
study. It is also reasonable to assume that previously homeless learners
were less likely and less willing to participate in this survey.
Statistics from the 2011 Census of Housing and Population suggest
that in South Australia 37.5 persons per 10,000 residents (all
Australia: 48.9) were homeless on Census night, defined as either living
in improvised dwellings, tents or sleeping out; living in supported
accommodation for the homeless; staying temporarily with other
households; staying in boarding houses; staying in other temporary
lodging; or living in 'severely' crowded dwellings (ABS
2012b). These homelessness rates varied with age groups, peaking at 65.6
per 10,000 among those aged 19-24 years. But others of similar age to
those of the apprentices and trainees surveyed for this study were
similarly affected by homelessness. Homelessness rates were 44.0 per
10,000 for those aged 12-18, 53.2 per 10,000 for those 25-34; before
dropping to 41 per 10,000 for those aged 35-44.
Different collection and estimation methods make a direct
comparison of the ABS statistics and this survey's statistics
difficult. On the face of it, the differences are considerable: an
average of 37.5 homeless persons per 10,000 Australian residents versus
between 3 and 8 per 100, that is, 300-800 per 10,000 apprentices and
trainees. But whereas the Census data refers to the Census night, the
survey data refers to any time during a person's apprenticeship or
traineeship. Not enough is known about the duration of homelessness
during training to allow these statistics to be adjusted for their
different accounting methods. This said, seen in the context of official
statistics and related research referred to earlier in this
contribution, these statistics suggest a notable risk of homelessness
among trainees, even when compared with others in the same age group.
Regressions analysis: who are the homeless apprentices now?
So far, the descriptions have focused on the general
characteristics of South Australia's apprentices and trainees
identified in our survey. The survey also sought to explore factors that
may be associated with the risk of homelessness among this group. Being
a retrospective survey that asked participants to recall incidents of
homelessness, and lacking the scope and scale for collecting detailed
life history data, it would have been inappropriately ambitious to
assume the capacity to identify actual causes of homelessness among
apprentices. However, the data allowed us to explore the present-day
training choices and living arrangements of those with past experiences
of homelessness.
For this purpose, we conducted a logistic regression analysis to
determine the current characteristics of young people who had
experienced homelessness compared with those who had not, using the
entire response sample of current and past apprentices and trainees.
This sample included 28 individuals who had reported previous incidents
of living without a permanent address. This analysis used unweighted
data, while controlling for differences between apprentices and
trainees, and current or past learners, as well as for sex and age. Only
whether a person was a trainee rather than an apprentice marginally
influenced the results, as trainees were slightly less likely than
apprentices to have reported episodes of homelessness. The difference,
however, was only statistically significant at the 10 per cent level
(Table 3).
Other more strongly differentiating characteristics were the
type--or industry--of the apprenticeship or traineeship, current living
arrangements, the receipt of Youth Allowance or AUSTUDY, and reported
incidence of financial stress.
In analysing the sector of the apprenticeship, building trades were
used as a comparison. Apprentices in this sector were least likely to
report past incidents of homelessness. Statistically significantly
higher risks of homelessness than among those in the building trades
were reported by apprentices/trainees in retail and personal services,
and transport; a diverse range of other certificated courses--including
in training and assessment; signwriting; occupational health and safety;
warehousing; and glass production; and, albeit to a lesser degree,
social and medical services.
Apprentices and trainees reporting episodes of homelessness were
statistically significantly more likely than others currently to receive
Youth Allowance or AUSTUDY payments. They were also more likely than
others currently to be paying rent to a friend rather than paying a
parent or family member--which, alongside living rent-free, was the most
frequent living arrangement among apprentices/trainees. Whilst outside
the five per cent level of statistical significance, albeit only
marginally so, people in training who had experienced homelessness were
more likely to have experienced other forms of financial stress because
they had been 'short of money' in the 12 months before the
survey. This included: not being able to pay electricity, gas or
telephone bills, mortgage or rent payments, car registration or
insurance; not making the required minimum payment on a credit card;
pawning or selling something because of the need for cash; going without
meals; being unable to heat their home; seeking financial assistance
from friends or family, or from welfare or community organisations.
These episodes of financial distress appeared not to be directly
associated with the amount of the training awards that
apprentices/trainees received. Separate analysis of the smaller number
of cases for whom income data were available failed to find a
statistically significant association between, on the one hand, current
income or expenditure on accommodation (gross or as a proportion of
income) reported at the time of the survey and, on the other hand, past
experiences of homelessness. However, we cannot be certain that income
or housing expenditure at the time of the survey had been the same or at
least similar to income or housing expenditure at the time when the
apprentices or trainees had been homeless.
Conclusion
The survey of apprentices and trainees in South Australia provided
some useful insight into the living arrangements of people attending VET
courses in the state. While the survey suggested that the majority of
apprentices and trainees had stable housing arrangements, typically
living with parents (most apprentices and many trainees) or living in a
partnered/married household and often in an owned property (particularly
trainees), it also found evidence of temporary homelessness. Thus, while
the majority of students appeared to be managing their housing
effectively, an estimated eight per cent of apprentices/trainees in
South Australia had experienced homelessness, staying with relatives,
with friends, or in a boarding house or hostel, living in a caravan, or
sleeping rough while being without a permanent address of their own.
Having only cross-sectional data that collected information
retrospectively and at a single point in time, it was not possible to
draw causal connections between training and homelessness. But our
statistical analysis is able to depict the current circumstances of
people with past experiences of homelessness. Although this shows no
direct statistical evidence of current training awards being associated
with past experiences of homelessness, apprentices/trainees who had
previously been homeless did also disproportionately report having
recently experienced financial distress. This was perhaps not a
surprising finding given the comparative large share of earnings that
apprentices and trainees in our surveys spent on housing alone, which
would have limited the resources available to cover the cost of other
daily necessities. Reflecting these stresses, apprentices and trainees
were disproportionately likely currently to --continue to--receive state
financial support, either in the form of the Youth Allowance or AUSTUDY.
Whilst not further reported here, the importance of YA and AUSTUDY had
been confirmed in conversations with youth workers and training
providers, who participated in the qualitative part of this study.
According to their accounts, many learners were unaware of the financial
support that was available, so helping those experiencing or threatened
by homelessness to obtain these additional funds was often a first step
to addressing a housing crisis. In their opinion, apprentices/trainees
often had insufficient support--including financial resources and, in
some instances, financial management skills--to pre-empt or resolve
their crisis.
The regression analysis also revealed that previously homeless
learners were significantly more likely than others to pay friends for
their current accommodation. This highlighted the importance of peer
group support alongside access to public welfare payments for sustained,
more stable living conditions among apprentices and trainees.
The roles of life experiences, evolving relationships and living
arrangements in buffering against housing and financial risk was
illustrated by the experience of trainees. The greater prevalence of
partnering and more settled social networks when compared with
apprentices may have protected trainees from the most extreme forms of
housing deprivation. While experiencing lower income than apprentices
and proportionately higher housing costs, they were less likely than
apprentices to have become homeless at some point during their
traineeship.
Although homelessness is not a widespread problem among learners,
just as it is not a widespread problem in Australian society at large,
it does have severe effects on an individual's wellbeing and, in
the case of learners, their capacity to continue and, indeed, complete
their training. The personal costs of homelessness during apprenticeship
or traineeship can be considerable, in particular if one considers the
long-term scarring effects of homelessness (Noble-Carr 2007; ABS 2012c).
In addition to this are the social costs associated with reduced
training outcomes and, not infrequently, dropping out of training due to
housing stress.
Although our survey was conducted in South Australia, Commonwealth
statistics and community activities such as the abovementioned 'The
house that builds people' initiative demonstrate that homelessness
and housing crises among apprentices and trainees are social issues that
are not confined to one state.
The educational setting that these young and sometimes older people
inhabit when they experience homelessness makes the prospect of tackling
the issue uniquely realistic. Whereas homelessness frequently emerges
unnoticed and is recorded only after it has become manifest, apprentices
and trainees at risk of homelessness are visible to their institutional
environment. This should make it possible to detect housing risk early,
perhaps even before it takes on the form of a personal crisis. In light
of these considerations, our research developed a number of
recommendations intended to improve the capacity of all those involved
with training and education, and youth work--including the apprentices
and trainees themselves--to be better prepared for detecting, preventing
and responding to the risk of homelessness. Most notably, we recommend:
Ensuring that training providers, including employers, maintain
a record of each apprentice's housing situation and that this is
verified regularly. Surprisingly, our qualitative research found
several instances in which even basic address details were neither
collected nor checked and updated.
Increasing awareness of Youth Allowance and the Living Away From
Home Allowance' among apprentices and promoting their uptake.
Improving the recording of support services requested by
apprentices/trainees under mentoring programs operated across the states
and territories. Monitoring under the AAMP, federal funding for which
ceased at the end of 2014, was geared towards recording retention
outcomes. More information should be collected systematically concerning
the types of supports requested and provided, and their circumstances.
Linking data across relevant programs. To understand fully the
housing situation and risk to apprentices, data regarding the
apprenticeship should contain--or be linked to sources
containing--information detailing the receipt of Allowances, which would
also permit better upkeep of address information.
Monitoring apprenticeship awards and income of apprentices in
relation to poverty indices and information on the costs of being an
apprentice. With few notable exceptions (e.g., Bittman et al. 2009;
Schutz et al. 2013), the living standards of apprentices are rarely
assessed, yet that knowledge remains essential for understanding housing
risks.
Table A.1. Current apprentices and trainees in SA--variables used for
weighting, weight effects
Current apprentices
Population Response Sample
Sample
Unweighted Weighted
% % %
Male 89.9 91.5 93.3
Female 10.1 8.5 6.7
Age: >=24 75.1 70.8 71.2
Age: 24+ 24.9 29.2 28.8
Agri-/ 0 0 0
Horticulture
Building Trades 21.8 21.7 23.1
Business & 0 0 0
Management
Retail & Personal 8.5 13.2 9.6
Services,
Transport
Social & Medical 0 0 0
Services
Technical & 63.2 63.2 62.5
Manufacturing
Trades
Other 6.5 0.9 4.8
Not available 0 0.9 0
N (unweighted) 1000 106 106
Current trainees
Population Response Sample
Sample
Unweighted Weighted
% % %
Male 55.9 58.0 63.0
Female 44.1 42.0 37.0
Age: >=24 33.0 46.9 30.9
Age: 24+ 67.0 53.1 69.1
Agri-/ 6.7 6.2 6.0
Horticulture
Building Trades 2.7 1.2 3.6
Business & 29.6 19.8 33.7
Management
Retail & Personal 34.7 32.1 31.3
Services,
Transport
Social & Medical 8.4 13.6 8.4
Services
Technical & 3.3 8.6 3.3
Manufacturing
Trades
Other 14.6 6.2 13.3
Not available 0 12.4 0
N (unweighted) 1000 81 81
Acknowledgements
We would like to thank the then Department of Families, Housing,
Community Services and Indigenous Affairs (FaHCSIA), now the Department
of Social Services, for funding this research under their National
Homelessness Research Agenda program (Project No 2013/5). We are
particularly grateful to the South Australian Department of Further
Education, Employment, Science and Technology (DFEEST), now the SA
Department of State Development, for its support in conducting the
survey. Finally, this research would not have been possible without the
help of the many youth, housing and homelessness organisations, youth
workers, and apprentices and trainees who volunteered their time,
knowledge and experience during the course of this study.
Endnotes
(1) 'The house that builds people', see
http://thehousethatbuildspeople.com.au
(2) For definitions, see
http://homelessnessclearinghouse.govspace.gov.au
(3) Figures provided upon request by then Department of Families,
Housing, Community Services and Indigenous Affairs, FaHCSIA.
References
ABS (Australian Bureau of Statistics) (2009) Home and Away: the
living arrangements of young adults, Australian Social Trends, Cat. No.
4102.0.
--(2012a) 2012 Year Book Australia, Cat. No. 1301.0.
--(2012b) Census of Population and Housing: Estimating
homelessness, 2011, Cat. No. 2049.0.
--(2012c) Australian Social Trends 2012. Life after homelessness,
Cat. No. 4102.0.
--(2013) Housing occupancy and costs 2011-2012, Cat No. 4130.0.
--(2014) Labour Force, Australia, Feb 2014, Cat. No. 6202.0.
ACCI (2010) Youth Employment, Issue Paper, Canberra/Melbourne,
Australian Chamber of Commerce and Industry, http://www.acci.asn.au
(accessed 5 March 2014).
AHURI (Australian Housing and Urban Research Institute) (2014) Do
current measures of housing affordability reflect wellbeing? AHURI
Research & Policy Bulletin. Issue 166, March 2014, Australian
Housing and Urban Research Institute, Melbourne.
AIHW (Australian Institute of Health and Welfare) (2011)
Government-funded specialist homelessness services: SAAP National Data
Collection annual report 2010-11: Australia, Cat. No. HOU 250.
--(Australian Institute of Health and Welfare) (2013) National
Indigenous Reform Agreement: PI 25-3 month employment outcomes
(post-program monitoring), 2012 QS. http://meteor.aihw.gov.au (accessed
8 March 2014).
Bittman, M., Reavell, R., Smith, G. & Battin, T. (2007) Living
standards of apprentices, www.une.edu.au/bcss Accessed 11/04/2012.
Burke, X, Pinkney, S. & Ewing, S. (2002a) Young adults and
housing, Research & Policy Bulletin, Issue 11. July, Australian
Housing and Urban Research Institute, Swinburne-Monash Research Centre.
--(2002b) Rent assistance and young adults' decision-making,
AHURI Final Report No. 6, Australian Housing and Urban Research
Institute, Swinburne-Monash Research Centre.
Chamberlain, C. Sc MacKenzie, D. (2009) Counting the Homeless 2006,
Canberra, Australian Institute of Health and Welfare.
Crane, R, Brannock, J., Ray, L, Campbell, J., Smeal, G. &
Atweh, B. (1996) Homelessness among young adults in Australia. Early
intervention & prevention, Hobart, Tasmania, National Clearinghouse
for Youth Studies.
CYA (Centrelink Youth Allowance) (2013) Statement by Young Person
form, http:// www.humanservices.gov.au (accessed 2 May 2013).
Dupuis, A. & Thorns, D. (1998) 'Home, home ownership and
the search for ontological security', Sociological Review, 46,
25-47.
Homelessness Taskforce (2008) The Road Home. A National Approach to
Reducing Homelessness, Canberra, Department of Families, Housing,
Community Services and Indigenous Affairs.
Karmel, T. & Mlotkowski, P. (2010a) How reason for not
completing apprenticeships and traineeships change with duration,
Adelaide, National Council for Vocational Education Research.
--(2010b) The impact of wages on the probability of completing an
apprenticeship or traineeship, Adelaide, National Council for Vocational
Education Research.
--(2011) The impact of wages and the likelihood of employment on
the probability of completing an apprenticeship or traineeship,
Adelaide, National Council for Vocational Education Research.
McDowell, J., Oliver, D., Persson, M., Fairbrother, R., Wetzlar,
S., Buchanan, J. & Shipstone, T. (2011) A Shared Responsibility.
Apprenticeships for the 21st Century. Apprenticeships for the 21st
Century Expert Panel Paper. Commonwealth of Australia,
http://www.australianapprenticeships.gov.au (accessed 15 March 2014).
MacKenzie, D. & Chamberlain, C. (2008) 'Youth Homelessness
2006', Youth Studies Australia, 27 (1), 17-25.
McLaughlin, P. (2012) 'Couch surfing on the margins: the
reliance on temporary living arrangements as a form of homelessness
amongst school-aged home leavers', Journal of Youth Studies, 2012
(4), 1-25.
Martijn, C. & Sharpe, L. (2006) 'Pathways to youth
homelessness', Social Science & Medicine, 62,1-12.
Mavromaras, K., King, D. Macaitis, K., Mallett, S. & Batterham,
D. (2012) Finding Work: Homelessness and Employment, Canberra,
Department of Families, Housing, Community Services and Indigenous
Affairs.
Mission Australia (2011) National Survey of Young Australians, Key
and emerging issues, http://www.missionaustralia.com.au (accessed 2
April 2012).
Muir, K., Mullan, K. Powell, A., Flaxman, S., Thompson, D. &
Griffiths M. (2009) State of Australia's Young adults: A Report on
the social, economic, health and family lives of young adults, Sydney,
Social Policy Research Centre/ University of New South Wales.
NCVER (National Centre for Vocational Education Research) (2010)
Australian vocational education and training statistics: apprentice and
trainee destinations. Apprentices and trainees destinations, Adelaide,
National Council for Vocational Education Research.
--(2011a) Overview of the Australian apprenticeship and trainee
system,
Adelaide, National Council for Vocational Education Research,
http://www. australianapprenticeships.gov.au (accessed 19 March 2011).
--(2011b) Apprentices and Trainees, September Quarter 2011,
Australian vocational education & training statistics, Adelaide,
National Council for Vocational Education Research.
Noble-Carr, D. (2007) The experiences and effects of family
homelessness for children. A Literature Review, Canberra, Australian
Catholic University, Institute of Child Protection Studies.
OECD (2011) Compendium of OECD Wellbeing Indicators, Paris,
Organisation of Economic Cooperation and Development.
Rosenthal, D., Mallett, S., & Myers, P. (2006) 'Why do
homeless young adults leave home?', Australian and New Zealand
Journal of Public Health, 30, 281-285.
Schutz, H., Bittman, M. Chan, S. Jakubauskas, M. & Buchanan, J.
(2013) The Changing Situation of Electrical Apprentices. Submission to
the Modern Award Review, Fair Work Australia. Prepared for the
Electrical Industry Stakeholder Consortium. Sydney, Workplace Research
Centre/Business School, University of Sydney.
Snell, D. & Hart, A. (2008) 'Reasons for non-completion
and dissatisfaction among Apprentices and Trainees: A regional case
study', International Journal of Training Research, 6(1), 44-73.
Vitis, L., Ware, V-A & Gronda, H. (2010) Intermediate housing
markets and effective policy responses, Shelter Brief 43. AHURI Research
Synthesis Service, Melbourne, Australian Housing and Urban Research
Institute.
Yates, J. (2011) 'Housing in Australia in the 2000s: On the
Agenda Too Late?', in H. Gerard and J. Kearns (eds) The Australian
Economy in the 2000s, Sydney, Reserve Bank of Australia.
Appendix A
Weighting of the survey of apprentices and trainees
Weights were constructed to match current apprentice and trainee
respondents to the original sample. The matching criteria were sex
(male/female), age (aged 24 or less/25 or over) and vocation. The
vocations were grouped as follows:
Agriculture & Horticulture
Building Trades
Business & Management
Retail & Personal Services, Transport
Social & Medical Services
Technical & Manufacturing Trades
Other
The weighting had only small effects on the gender composition of
the participating apprentices and trainees, and on the age distribution
and VET course distribution of apprentices. However, effects on the age
distribution and the vocational course profiles of trainees were more
marked (see Table 1).
Table A.1. Current apprentices and trainees in SA--variables used for
weighting, weight effects
Current apprentices
Population Response Sample
Sample
Unweighted Weighted
% % %
Male 89.9 91.5 93.3
Female 10.1 8.5 6.7
Age: >=24 75.1 70.8 71.2
Age: 24+ 24.9 29.2 28.8
Agri-/ 0 0 0
Horticulture
Building Trades 21.8 21.7 23.1
Business & 0 0 0
Management
Retail & Personal 8.5 13.2 9.6
Services,
Transport
Social & Medical 0 0 0
Services
Technical & 63.2 63.2 62.5
Manufacturing
Trades
Other 6.5 0.9 4.8
Not available 0 0.9 0
N (unweighted) 1000 106 106
Current trainees
Population Response Sample
Sample
Unweighted Weighted
% % %
Male 55.9 58.0 63.0
Female 44.1 42.0 37.0
Age: >=24 33.0 46.9 30.9
Age: 24+ 67.0 53.1 69.1
Agri-/ 6.7 6.2 6.0
Horticulture
Building Trades 2.7 1.2 3.6
Business & 29.6 19.8 33.7
Management
Retail & Personal 34.7 32.1 31.3
Services,
Transport
Social & Medical 8.4 13.6 8.4
Services
Technical & 3.3 8.6 3.3
Manufacturing
Trades
Other 14.6 6.2 13.3
Not available 0 12.4 0
N (unweighted) 1000 81 81
Table 1. Current apprentices and trainees in SA--sex, age, vocation
Apprentices Trainees
% %
Male 93.3 63.0
Female 6.7 37.0
>=24 years 71.2 30,9
25+ years 28.8 69.1
Agri-/Horticulture 0 6.0
Building Trades 23.1 3.6
Business & Management 0 33.7
Retail & Personal Services, Transport 9.6 31.3
Social & Medical Services 0 8.4
Technical & Manufacturing Trades 62.5 3.3
Other 4.8 13.3
N (unweighted) 106 81
Table 2. Type of current accommodation, apprentices and trainees, SA
Apprentices Trainees
% %
With parent(s) at home 61.2 26.8
Rented accommodation--with partner 5.8 7.3
Rented accommodation--shared (other than with
partner) 3.9 6.1
Rented accommodation--by yourself 2.9 3.7
Owned accommodation--with partner 22.3 46.3
Owned accommodation--shared (other than with
partner) 0 1.2
Owned accommodation--by yourself 1.0 4.9
Squatting (not paying rent) 1.0 0
Other 1.9 3.7
N (unweighted) 106 81
Table 3. Odds ratios from logistic regression: experience of
homelessness during apprenticeship or traineeship
Homeless Odd ratio Std. Err.
Trainee -1.41 0.80
Cancelled apprenticeship or traineeship 1.25 0.69
Female -1.16 0.83
Age -0.004 0.02
Industry/Sector(Building trades)
Retail & Personal Services, Transport 2.84 1.30
Social & Medical Services 3.31 1.81
Technical & Manufacturing Trades 1.70 1.19
Other 3.54 1.64
Receipt of AUSTUDY or YA 2.16 0.86
Paying rent or mortgage (Paying parent/s or other family
member)
Paying friend 2.64 1.02
Paying landlord/estate agent 0.50 0.65
Making mortgage payments to bank -0.67 0.94
Living rent-free -0.61 0.72
Financial stress in last 12 months 1.13 0.59
_cons -2.84 1.88
Statistical
Homeless Significance
Trainee *
Cancelled apprenticeship or traineeship *
Female
Age
Industry/Sector(Building trades)
Retail & Personal Services, Transport **
Social & Medical Services *
Technical & Manufacturing Trades
Other **
Receipt of AUSTUDY or YA
Paying rent or mortgage (Paying parent/s or other family
member)
Paying friend **
Paying landlord/estate agent
Making mortgage payments to bank
Living rent-free
Financial stress in last 12 months
_cons
Note: [chi square](16) = 42.30. Prob > [chi square] = 0.0004. Log
likelihood = -58.452911. Pseudo [R.sup.2] = 0.2657. No. obs = 191
Statistical significance: ** 5% level; * 10% level
