Process Addictions
Sejmjl308
IndependentCorrelatesofReportedGamblingProblemsAmongstIndigenousAustralians.pdf
Independent Correlates of Reported Gambling Problems Amongst Indigenous Australians
Matthew Stevens • Martin Young
Received: 25 June 2009 / Accepted: 21 September 2009 / Published online: 4 October 2009 � Springer Science+Business Media B.V. 2009
Abstract To identify independent correlates of reported gambling problems amongst the Indigenous population of Australia. A cross-sectional design was applied to a nationally
representative sample of the Indigenous population. Estimates of reported gambling
problems are presented by remoteness and jurisdiction. Multivariable logistic regression
was used to identify independent correlates of reported gambling problems amongst
individuals and their social and family networks. The analysis was stratified by remoteness.
Indigenous people living in remote locations reported significantly more gambling prob-
lems than those living in non-remote locations. In non-remote areas, being female, having
high personal income, being more socially connected (i.e. involved in an Indigenous
organisation or living in a household where all members were Indigenous) and reporting
community problems were associated with higher levels of reported gambling problems. In
remote areas, multifamily households, participation in sports and cultural events, and
reporting of community problems were associated with higher reported gambling prob-
lems, while having a relative removed from their natural family was associated with lower
reported problems. Problematic gambling is clearly related to the social and environmental
contexts in which it occurs. Harm minimisation policies that focus on a reduction in
crowding (especially in remote locations), increased public awareness of the negative
consequences of gambling, improving access to support services, and lifting the socio-
economic status of Indigenous people may reduce gambling related harm in the medium to
long term.
Keywords Aboriginal populations � Public health policy � Gambling
M. Stevens (&) � M. Young Institute of Advanced Studies, School for Social and Policy Research, Charles Darwin University, Darwin, NT 0909, Australia e-mail: [email protected]
123
Soc Indic Res (2010) 98:147–166 DOI 10.1007/s11205-009-9536-4
1 Introduction
1.1 ‘Indigenous Gambling’: Card Games
Gambling is a popular recreational activity in Australia with a national participation rate of
over 70% (Productivity Commission 1999). While the exact date of gambling introduction
to Indigenous Australians is unknown, card games were being played for small twists of
tobacco as early as the 1930’s, though most games were played predominantly for
entertainment within family and extended family groups (Dodd and Vaughn 1985).
Anthropological studies from the 1950’s suggest that card games had become a ubiquitous
activity across most parts of Australia, prompting Elders in some remote communities to
lament that gambling (and alcohol use) was not ‘Aboriginal’ and sometimes interfered with
Aboriginal Law and ceremony (Berndt and Berndt 1946–47; McKnight 2002; Tonkinson
1974). More recent research into gambling by Indigenous Australians in the 1980’s indi-
cated a mix of positive and negative consequences depending on the location of the
research and the perspective of the researcher. For example, research in a small community
(approximately 30 people) located in the Northern Territory found gambling to be largely
unproblematic, with the games acting to distribute money at a time when resource dis-
tribution was inequitable, as well as enabling large sums of money to be raised for major
purchases (Altman 1985). However, this and similar research noted that gambling did
cause some problems including a relaxation of traditional Indigenous kinship relations (e.g.
avoidance) (Altman 1985) and in some instances a neglect of children by parents who were
gambling (Goodale 1987). Other research from the same period highlighted more specific
negative affects of card games, including higher levels of anxiety amongst gamblers
compared with non-gamblers, increased tensions within families, poor nutrition of children
due to the reduced money available when gamblers lost, reduced school attendance by
children due to sleep deprivation when card games ran through the night, and large win-
nings being spent outside the community, often on alcohol (Hunter 1993; Hunter and
Spargo 1988; Martin 1993; McKnight 2002). In all of these studies females were identified
as the more avid gamblers, with men generally only participating in high stakes games
(Goodale 1987; Hunter 1993; Hunter and Spargo 1988; Martin 1993). A study in a remote
community in north Queensland found that money tended to be transferred from females to
males and that larger winnings were nearly always spent obtaining alcohol, which involved
chartering a plane or spending up to $200 on a return taxi trip to the nearest town where
alcohol could be purchased (Martin 1993). A recent review seeking an Indigenous voice on
gambling issues in New South Wales found that card games were still a significant leisure
activity amongst Indigenous people, and particularly for those living in rural communities,
but that popularity was on the decline due to access to regulated gambling and upon the
death of older people who instigated the games (Aboriginal Health & Medical Research
Council of NSW 2007).
1.2 Indigenous People and Regulated Gambling
The research during the past two decades into regulated gambling (e.g. electronic gambling
machines (EGMs) or ‘pokies’, casinos, racetrack betting etc.) by Indigenous people has
identified significant related harm in all jurisdictions where studies have been carried out
(Aboriginal Health & Medical Research Council of NSW 2007; Brady 2004; Cultural
Perspectives Pty Ltd 2005; Dickerson et al. 1996; Foote 1996; Holden et al. 1996;
McMillen and Donnelly 2008; McMillen and Marshall 2004; McMillen and Togni 2000;
148 M. Stevens, M. Young
123
Young et al. 2007). For example, a representative population survey carried out in
Queensland found that Indigenous people were not over- or under-represented in non-
gambler or recreational gambler groups, but were significantly over-represented in low,
moderate risk and problem gambler groups (Queensland Government 2006). This is sig-
nificant given that the limited data available from national and state level surveys indicate
that participation in regulated gambling by Indigenous people is no different to the general
Australian population (McMillen and Donnelly 2008; Productivity Commission 1999;
Young et al. 2007). EGM play, the activity most associated with problem gambling in
Australia, has consistently been identified as the most popular activity for Indigenous
people (Dickerson et al. 1995; McMillen and Doran 2006; Queensland Government 2007;
Young and Stevens 2009; Young et al. 2008b). Two of these studies found that regular (at
least weekly) gamblers spent on average between 20 and 30% of their income on gambling
(Dickerson et al. 1995; Holden et al. 1996). An observational study in the Darwin casino
found that 67% of Indigenous patrons were females and 76% of the Indigenous people
observed were in the EGM area of the casino (Foote 1996). A more recent observational
study in a large gambling venue in the Northern Territory identified similar proportions of
males and females in the venue, and again EGMs were the preferred activity (Fogarty
2008). The issue of ‘shame’ (i.e. embarrassment) also arose in a number of studies in
regard to help-seeking behaviour (Aboriginal Health & Medical Research Council of NSW
2007; Cultural Perspectives Pty Ltd 2005). That is people may have known they had a
problem, but did not seek out help due to ‘shame’ or as another study found, Indigenous
people were less aware of available services for problem gambling (Roy Morgan Research
2006). The issue of ‘shame’ is also a likely factor in under-reporting of gambling problems
in empirical studies, because individuals are too embarrassed to self-report that they may
have a gambling problem (Aboriginal Health & Medical Research Council of NSW 2007).
1.3 Problem Gambling—the Concept
The national definition of problem gambling used in Australia states that ‘‘problem
gambling is characterised by difficulties in limiting money and/or time spent on gambling
which leads to adverse consequences for the gambler, others, or for the community.’’(Neal
et al. 2005). However, there has been considerable debate on the best instrument or
‘screen’ to measure problem gambling (Battersby et al. 2002; McMillen and Wenzel 2006;
Productivity Commission 1999; Svetieva and Walker 2008). Currently, in Australia, the
preferred problem gambling screen is the Problem Gambling Severity Index (PGSI) (Ferris
and Wynne 2001). However, this screen, as with the earlier South Oaks Gambling Screen
(SOGS) (Lesieur and Blume 1987), has been criticised for its focus on the individual, while
ignoring harms to other people and the broader community (Svetieva and Walker 2008);
significant aspects inherent in the national definition. In addition, the valid use of problem
gambling screens in culturally and linguistically diverse populations has been questioned
(Battersby et al. 2002; Duvarci et al. 1997; Productivity Commission 1999; Svetieva and
Walker 2008).
Australia’s Indigenous population exhibits significant variation in spoken language,
cultural practices and in socioeconomic status generally, which follows an urban-remote
gradient from least to most disadvantaged (Australian Bureau of Statistics 2003, 2004b,
2008b). A deficiency of the previously cited studies was that of differing methodologies
were used, and they were unrepresentative of the total Indigenous population for there
respective geographic regions. Therefore, a national assessment of the level of gambling
related problems or their correlates is at present unavailable. However, the Australian
Independent Correlates of Reported Gambling Problems 149
123
Bureau of Statistics (ABS) collects national statistics on the Indigenous population and
recent social and health surveys have included a scale that captures people’s reports of
gambling problems for themselves, their family or friends (see below for exact wording).
Using data from a nationally representative survey of the Indigenous population, we report
estimates of reported gambling problems by remoteness and state/territory. Further, an
analysis of variables from regional, demographic, social, economic and health domains
were modelled to identify independent correlates of reported gambling problems for the
Indigenous population of Australia. A background to the measure of reported gambling
problems is now provided before the analysis is presented.
1.4 Measuring Gambling Problems Using the Negative Life Events Scale
The Negative Life Events Scale (NLES) is a regular survey module used by the ABS in
social and health surveys, one that that aims to measure the emotional and social wellbeing
of a person. It was developed in consultation with peak Aboriginal and Torres Strait
Islander bodies with the aim of being able to be used as a comparable measure of social
and emotional wellbeing between the Indigenous and non-Indigenous populations of
Australia (Australian Bureau of Statistics 2004b). Copies of the survey questionnaire can
be obtained from the ABS website (www.abs.gov.au). The NLES asks respondents ‘‘have
any of these things (list of stressors or negative life events) been a problem for you or your
family or friends during the last year?’’ Respondents then answer yes or no to a list of up to
16 ‘‘stressors’’ or negative life events listed in Table 1.
It is clear that this question is not a prevalence estimate for problem gambling. It asks
respondents if gambling has ‘‘…been a problem for you, your family or friends during the last year’’. Therefore, the NLES gambling problem item measures the reach or impact of
gambling problems throughout peoples’ social and family networks. This broader
Table 1 Negative Life Events items for each ABS survey
NLES item 2002 NATSISS
Remote Non-remote
Gambling problem 4 4
Divorce or separation 4 4
Death of family member or close friend 4 4
Serious illness or disability 4 4
Serious accident 4 4
Alcohol or drug related problems 4 4
Not able to get a job 4 4
Lost job, made redundant, sacked 4 4
Witness to violence 4 4
Abuse or violent crime 4 4
Trouble with the police 4 4
Mental illness 8 4
Member of family sent to jail/currently in jail 4 4
Overcrowding at home 4 4
Pressure to fulfil cultural responsibilities 8 4
Discrimination/racism 4 4
150 M. Stevens, M. Young
123
conception of gambling related harm is consistent with the Australian definition of problem
gambling provided in the previous section (Neal et al. 2005).
The reliability of the NLES when used in a non-ABS survey has been previously
reported (Kowall et al. 2007). Psychometric properties of the NLES were assessed in a
sample of Indigenous carers and householders from ten remote Aboriginal communities
across the Northern Territory (see Fig. 1). The study used a modified version of the NLES
that asked respondents ‘‘have any of these things [list of stressors] been a worry for you or
anyone else living in this house during the last year?’’ The item was presented in such a
way that the estimates of gambling problems (and other stressors) related to households as opposed to the ABS target of social and family networks. The NLES was found to perform
well psychometrically with respect to item endorsement, discrimination and internal and
external reliability (Kowall et al. 2007). The NLES item of ‘‘gambling problem’’ showed
good discriminatory characteristics in relation to the overall scale and other listed items, as
well as giving estimates that were similar to those collected in the 2002 National
Aboriginal and Torres Strait Islander Social Survey (NATSISS) for remote areas in the
Northern Territory (Australian Bureau of Statistics 2004c; Kowall et al. 2007). Therefore,
we are able to utilise the NLES with some confidence.
The Indigenous population of Australia displays marked heterogeneity with respect to a
range of demographic, social, cultural and economic characteristics across jurisdictions
and by degree of remoteness (Australian Bureau of Statistics 2002, 2003, 2004b, 2007b,
2008b). Consequently these characteristics need to be accounted for when interpreting the
Fig. 1 Accessibility/Remoteness areas for Australia and casino locations
Independent Correlates of Reported Gambling Problems 151
123
estimates derived from the NLES items. For example, Indigenous people living in remote
areas tend to have larger extended families because of the traditional Indigenous kinship
systems which still play a prominent part in the life of many remote Indigenous people
(Austin-Broos 2003). The implication of this kinship system on estimates for remote
regions, where a larger proportion of Indigenous people live in small communities (Aus-
tralian Bureau of Statistics 2002, 2007b), is that larger estimates may be expected because
the more extensive social networks. While this potential overestimation needs to be con-
sidered when drawing direct comparison of estimates of reported gambling related prob-
lems between remote and non-remote areas and between Indigenous and non-Indigenous
populations, it does not limit analyses identifying significant correlations between NLES
items and other demographic and socioeconomic variables collected in the surveys, as
these correlations are based on patterns of association rather than absolute estimates.
Another limitation of the NLES item on gambling problems is that no definition of
gambling problems is provided as part of the module. From this viewpoint, the NLES is a
subjective scale with respondents deciding what is and is not problematic with regards to gambling (and other NLES items such as alcohol and drug problems). This is of impor-
tance for the NLES where the meaning of gambling problems per se in the context of many remote Indigenous communities across Australia varies according to current social prac-
tices and significant historical influence (Long 1970; Roberts 1981). However, while there
are limitations of using the NLES item for reporting estimates of gambling related prob-
lems, it nonetheless provides the only available measure of reported gambling problems
that is collected consistently both spatially and temporally in national surveys for the
Indigenous population of Australia.
2 Methods
2.1 Data Collection and Access
The 2002 NATSISS collects a diverse range of geographic, demographic, social, eco-
nomic, cultural and health status related variables. The NATSISS was conducted using a
stratified multistage area sample, with a scope of all people aged 15 years and over living
in remote and non-remote areas of Australia and carried out every 6 years (Australian
Bureau of Statistics 2007a). Remoteness was classified according to the Accessibility/
Remoteness Index of Australia (see Fig. 1) (Australian Bureau of Statistics 2008a). A full
description of the sample design for the NATSISS can be found in the technical manual
available on the ABS website (Australian Bureau of Statistics 2004d, 2006a). For the
purpose of the current paper both the remote and non-remote samples of the 2002 NAT-
SISS are used. Furthermore, survey respondents aged less than 18 years were excluded,
due to some data items only being available for people aged 18 years and over. The survey
was weighted to reflect the Indigenous population of Australia and it is the weighted data
that is used in all analyses. Table 2 provides information on the sample size and weighted
population by remoteness for the NATSISS.
Due to confidentiality restrictions data was accessed over the internet via the ABS
Remote Data Access Laboratory or RADL (Australian Bureau of Statistics 2006b). Data
was not subject to ethics clearance because the analyses constitute secondary use of data.
However, confidentiality of survey participants is maintained according to strict protocols
by the ABS in adherence to the Census and Statistics Act, 1905.
152 M. Stevens, M. Young
123
2.2 Statistical Analysis
First, estimates of reported gambling problems are presented for by remoteness and
jurisdiction. This data was sourced directly from the ABS through the purchase of cus-
tomised tables to ensure consistency with previously published ABS data (except for
Australian Capital Territory (ACT) and Tasmania combined estimate, which was derived
from the Confidentialised Unit Record File (CURF) accessed via the ABS RADL). Second,
to identify multivariable adjusted correlates of reported gambling problems a logistic
regression model was used. In light of the significant social, economic and cultural vari-
ation in the Indigenous remote and non-remote populations (Australian Bureau of Statistics
2003, 2004b, 2008b), the analyses were stratified by remoteness (see Fig. 1 for remoteness
areas). The following analytical strategy was carried out separately for the remote and non-
remote sample. Unadjusted associations between explanatory variables and reported
gambling problems were assessed using logistic regression. Explanatory variables showing
a significant (p B 0.05) association with gambling problems were then assessed for col- linearity to ensure the assumptions implicit in the logistic regression model were main-
tained. Where two or more explanatory variables were significantly correlated then the
variable(s) showing the best explanatory power were retained for the next stage of the
analytic strategy. Next, significant explanatory variables were entered simultaneously into
a multivariable logistic regression and backward elimination carried out with removal of
variables set at p [ 0.05. Odds ratios with 95% confidence intervals are used to specify effect sizes in the final models, and estimates of reported gambling problems and the
distribution of the variables are also presented to assist in interpretation. All explanatory
variables contained in final models represent those having an independent association with
reported gambling problems and were significant at p B 0.05. Given only one time period is involved in the modelling process the design is cross-sectional. All analyses were carried
out using Stata v8.2 �
on the weighted data, which was accessed via the ABS RADL web
portal (Australian Bureau of Statistics 2006b). The suite of SVR (survey replicate methods)
commands were used to analyse data (Winter 2008), and confidence intervals were cal-
culated using the Jack Knife (jk1) method with 50 replicate weights.
3 Results
3.1 Estimates of Reported Gambling Problems by Jurisdiction
Estimates of reported gambling problems by remoteness and jurisdiction are presented in
Table 3. Reported gambling problems were lowest in ACT/Tasmania and Western Aus-
tralia and highest in the Northern Territory, Queensland and South Australia. Estimates of
reported gambling problems were higher in remote compared to non-remote areas for all
jurisdictions except New South Wales. Statistically significant differences between remote
Table 2 Sample (18 years and over) and scope for analyses, 2002 NATSISS
Sample size (n) Weighted sample (N)
Non-remote 3,796 69,337
Remote 4,727 182,061
Total 8,523 251,398
Independent Correlates of Reported Gambling Problems 153
123
and non-remote were observed for Western Australia, Queensland and the Northern
Territory.
3.2 Remote Multivariable Model
Approximately 0.5% of the observations were excluded from the remote multivariable
model due to missing data. The final multivariable model for Indigenous people living in
remote locations in 2002 included demographic, socioeconomic, social and cultural net-
work, interpersonal violence, and community problem variables (see Table 4). There were
significant differences between jurisdictions with Queensland (OR 2.31, 1.15 to 4.62) and
the Northern Territory (OR 2.46, 1.38 to 4.38) reporting significantly more gambling
problems than Western Australia. Additionally, both the Northern Territory and Queens-
land reported significantly more gambling problems than New South Wales and South
Australia. Victoria and the Australian Capital Territory/Tasmania had no respondents
sampled from remote areas. Household structure was associated with reported gambling
problems, with people living in three-family households more likely to report gambling
problems compared with one-family households (OR 2.04, 1.23 to 3.40). Having access to
a landline telephone was associated with a reduction in reported gambling problems (OR
0.69, 0.48 to 0.98). Four variables relating to social and cultural networks showed inde-
pendent associations with reported gambling problems. They included being involved in
special community activities (OR 1.70, 1.23 to 2.35), participating in sports or physical
activity (OR 1.47, 1.08 to 1.99), attending a sporting event or carnival (OR 1.30, 1.02 to
1.65), and attending a funeral (OR 1.82, 1.25 to 2.66). Respondents who didn’t know or
refused to answer if they had other clan/tribal members living in their communities were
more likely to report gambling problems (OR 2.52, 1.20 to 5.27) compared with those who
did not have other clan members in their community. Compared with respondents who did
not have a relative removed from their natural family, those who didn’t know or refused to
answer reported less gambling problems (OR 0.53, 0.36 to 0.77), as did those who had a
relative removed (OR 0.75, 0.54 to 1.06), though the latter was marginally non-significant.
Respondents who identified community problems of youth gang violence (OR 1.50, 1.17 to
1.93), alcohol (OR 2.15, 1.37 to 3.39), and physical assault (OR 2.35, 1.79 to 3.08) were
more likely to report gambling problems than those not reporting these community
problems. Lastly, respondents who were a victim of threatened or physical violence in the
Table 3 Estimates of reported gambling problems by remoteness and jurisdiction, 2002 NATSISS
Remote a
% (SE) Non-remote a
% (SE) Total a
% (SE)
Western Australia 13.2 (2.9) 3.6 (1.2) 8.1 (1.5)
New South Wales 8.7 (2.3) 10.3 (1.3) 10.2 (1.2)
Victoria – 13.3 (1.6) 13.3 (1.6)
Queensland 37.1 (10.7) 10.7 (1.6) 17.4 (2.9)
South Australia 19.3 (5.4) 16.5 (2.3) 17.2 (2.2)
Northern Territory 31.9 (4.1) 11.4 (2.9) 28.4 (1.4)
ACT/Tasmania a
– 7.9 (1.1) 7.9 (1.1)
Australia 26.4 (3.2) 10.2 (0.7) 14.6 (1.0)
a NATSISS estimates sourced from ABS (data cubes), except for ACT/Tasmania which was outputted from
the 2002 NATSISS Confidentialised Unit Recode File (CURF) accessed via the ABS RADL
154 M. Stevens, M. Young
123
T a b
le 4
L o
g is
ti c
re g re
ss io
n m
o d
e ls
fo r
re p
o rt
e d
g a m
b li
n g
p ro
b le
m s
a m
o n
g st
th e
In d
ig e n
o u
s p
o p
u la
ti o
n fo
r re
m o
te a n
d n
o n
-r e m
o te
lo c a ti
o n
s, 2
0 0
2 N
A T
S IS
S
2 0
0 2
R e m
o te
2 0
0 2
N o
n -r
e m
o te
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
A u
st ra
li a
– 2
6 .4
(3 .2
) 1
0 0
.0 –
1 0
.2 (0
.7 )
1 0
0 .0
R e g
io n
a l
S ta
te /t
e rr
it o
ry
W e st
e rn
A u
st ra
li a
1 .0
0 1
4 .8
(3 .2
) 2
4 .0
(0 .7
) 1
.0 0
4 .4
(1 .4
) 1
0 .2
(0 .3
)
N e w
S o
u th
W a le
s 1
.5 4
(0 .8
3 –
2 .8
7 )
1 2
.6 (2
.6 )
7 .0
(0 .3
) 2
.8 3
(1 .3
1 –
6 .1
3 )
1 1
.1 (1
.4 )
3 8
.4 (0
.6 )
V ic
to ri
a n
r n
r 0
.0 (0
.0 )
4 .0
2 (1
.8 3
– 8
.8 6 )
1 4
.4 (1
.8 )
8 .6
(0 .2
)
Q u
e e n
sl a n
d 2
.3 1
(1 .1
5 –
4 .6
2 )
3 5
.7 (1
0 .1
) 2
4 .8
(0 .6
) 2
.8 5
(1 .4
5 –
5 .6
2 )
1 1
.0 (1
.6 )
2 7
.5 (0
.6 )
S o
u th
A u
st ra
li a
0 .8
2 (0
.2 8
– 2
.4 0 )
1 9
.6 (4
.2 )
4 .8
(0 .1
) 5
.1 6
(2 .2
7 –
1 1
.7 )
1 7
.9 (2
.5 )
5 .9
(0 .1
)
N o
rt h e rn
T e rr
it o
ry 2
.4 6
(1 .3
8 –
4 .3
8 )
3 1
.4 (3
.9 )
3 9
.3 (0
.4 )
6 .0
6 (1
.6 7
– 2
1 .9
) 1
7 .3
(5 .9
) 3
.1 (0
.3 )
A C
T /T
a sm
a n ia
n r
n r
0 .0
(0 .0
) 2
.4 3
(1 .0
7 –
5 .5
5 )
7 .9
(1 .1
) 6
.3 (0
.2 )
D e m
o g
ra p
h ic
G e n
d e r
M a le
n s
2 6
.2 (2
.9 )
4 8
.2 (0
.3 )
1 .0
0 8
.4 (1
.0 )
4 7
.2 (0
.3 )
F e m
a le
n s
2 6
.8 (3
.2 )
5 1
.8 (0
.3 )
1 .5
9 (1
.1 6
– 2
.1 7 )
1 3
.4 (1
.2 )
5 2
.8 (0
.3 )
H o
u se
h o
ld c o m
p o
si ti
o n
M ix
e d
n s
1 1
.1 (2
.1 )
8 .2
(1 .2
) 1
.0 0
8 .5
(1 .1
) 4
5 .2
(1 .7
)
A ll
In d
ig e n
o u s
n s
2 7
.9 (3
.2 )
9 1
.8 (1
.2 )
1 .6
8 (1
.1 1
– 2
.5 3 )
1 3
.2 (1
.2 )
5 4
.8 (1
.7 )
H o
u se
h o
ld ty
p e
O n
e fa
m il
y 1
.0 0
2 2
.5 (2
.4 )
5 5
.0 (2
.5 )
n s
1 0
.6 (0
.9 )
7 6
.5 (1
.4 )
T w
o fa
m il
ie s
1 .1
9 (0
.8 6
– 1
.6 4 )
3 3
.6 (5
.8 )
1 9
.8 (1
.7 )
n s
1 5
.1 (3
.9 )
6 .9
(1 .2
)
T h
re e
fa m
il ie
s 2
.0 4
(1 .2
3 –
3 .4
0 )
3 8
.1 (6
.3 )
1 5
.4 (2
.2 )
n s
1 7
.7 (1
6 .5
) 0
.7 (0
.4 )
M ix
e d
/g ro
u p
0 .9
5 (0
.3 6
– 2
.5 2 )
2 4
.0 (7
.1 )
4 .4
(0 .9
) n
s 1
0 .1
(2 .5
) 5
.7 (0
.7 )
L o
n e
p e rs
o n
0 .5
6 (0
.2 8
– 1
.1 2 )
1 1
.8 (2
.8 )
5 .4
(0 .8
) n
s 1
2 .0
(2 .3
) 1
0 .2
(0 .9
)
Independent Correlates of Reported Gambling Problems 155
123
T a b
le 4
c o
n ti
n u e d
2 0
0 2
R e m
o te
2 0
0 2
N o
n -r
e m
o te
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
S E
S /c
o m
m u
n ic
a ti
o n
P e rs
o n a l
in c o
m e
$ 0
-1 8 5
[1 –
3 d
e c il
e ]
n s
2 3
.2 (3
.0 )
3 2
.4 (1
.9 )
1 .0
0 8
.8 (1
.3 )
2 9
.6 (1
.3 )
$ 1
8 6 -2
3 0
[4 –
5 d
e c il
e ]
n s
3 1
.6 (6
.5 )
2 5
.4 (2
.0 )
1 .0
2 (0
.5 9
– 1
.7 6 )
9 .7
(1 .9
) 1
5 .7
(0 .9
)
$ 2
3 1 -3
8 0
[6 –
7 d
e c il
e ]
n s
3 0
.2 (3
.1 )
1 8
.8 (1
.2 )
1 .1
3 (0
.7 1
– 1
.7 9 )
1 1
.6 (1
.6 )
2 0
.3 (1
.0 )
$ 3
8 1 -6
9 0
[8 –
9 d
e c il
e ]
n s
2 5
.2 (5
.0 )
1 6
.5 (0
.9 )
1 .0
5 (0
.6 8
– 1
.6 3 )
9 .8
(1 .4
) 2
1 .1
(1 .1
)
$ 6
9 1
? [1
0 th
d e c il
e ]
n s
1 6
.0 (2
.9 )
6 .9
(0 .9
) 2
.4 5
(1 .5
2 –
3 .9
5 )
1 7
.9 (3
.0 )
1 3
.3 (1
.1 )
T e le
p h
o n
e a c c e ss
(l a n
d li
n e )
N o
a c c e ss
1 .0
0 3 0 .7
(3 .7
) 5 6 .9
(2 .6
) n s
1 2 .3
(2 .1
) 1 8 .3
(1 .4
)
H a s
la n
d li
n e
a c c e ss
0 .6
9 (0
.4 8
– 0
.9 8 )
2 1
.1 (2
.5 )
4 3
.1 (2
.6 )
n s
1 0
.8 (0
.8 )
8 1
.7 (1
.4 )
S o
c ia
l &
c u
lt u
ra l
n et
w o
rk s
S o c ia
l a c ti
v it
y -3
m o n th
s
C o m
m u n it
y a c ti
v it
ie s
N o
t in
v o
lv e d
1 .0
0 2
0 .0
(2 .3
) 6
9 .8
(2 .2
) n
s 9
.4 (0
.7 )
8 3
.6 (1
.1 )
In v
o lv
e d
in a c ti
v it
ie s
1 .7
0 (1
.2 3
– 2
.3 4 )
4 1
.7 (4
.9 )
3 0
.2 (2
.2 )
n s
1 9
.6 (2
.4 )
1 6
.4 (1
.1 )
S p
o rt
s/ p
h y si
c a l
p a rt
ic .
D id
n o t
p a rt
ic ip
a te
1 .0
0 1 9 .6
(1 .8
) 5 9 .6
(3 .2
) n s
9 .1
(0 .9
) 6 7 .5
(1 .2
)
P a rt
ic ip
a te
d in
a c ti
v it
y 1
.4 7
(1 .0
8 –
1 .9
9 )
3 6
.9 (4
.8 )
4 0
.4 (3
.2 )
n s
1 8
.5 (1
.8 )
3 2
.5 (1
.2 )
S p
o rt
in g
c a rn
iv a l
a tt
e n
d a n
c e
D id
n o
t a tt
e n
d 1
.0 0
1 7
.4 (1
.8 )
4 8
.2 (2
.6 )
1 .0
0 9
.1 (0
.9 )
7 8
.9 (1
.3 )
A tt
e n
d e d
sp o
rt s
c a rn
iv a l
1 .3
0 (1
.0 2
– 1
.6 5 )
3 5
.1 (4
.3 )
5 1
.8 (2
.6 )
1 .6
1 (1
.1 2
– 2
.3 2 )
1 8
.5 (1
.8 )
2 1
.1 (1
.3 )
F u
n e ra
l a tt
e n
d a n
c e
D id
n o
t a tt
e n
d 1
.0 0
1 4
.1 (2
.0 )
2 5
.0 (1
.6 )
n s
8 .8
(0 .9
) 6
1 .9
(1 .7
)
A tt
e n
d e d
fu n
e ra
l 1
.8 2
(1 .2
5 –
2 .6
6 )
3 0
.7 (3
.6 )
7 5
.0 (1
.6 )
n s
1 4
.7 (1
.4 )
3 8
.1 (1
.7 )
156 M. Stevens, M. Young
123
T a b
le 4
c o
n ti
n u e d
2 0
0 2
R e m
o te
2 0
0 2
N o
n -r
e m
o te
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
In d
ig e n
o u
s o
rg a n
is a ti
o n
:
W a s
n o
t in
v o
lv e d
n s
2 3
.8 (3
.2 )
7 3
.8 (2
.0 )
1 .0
0 8
.3 (0
.8 )
7 2
.5 (1
.4 )
In v
o lv
e d
in n
s 3
4 .3
(3 .4
) 2
6 .2
(2 .0
) 1
.7 6
(1 .1
9 –
2 .6
2 )
1 8
.1 (1
.9 )
2 7
.5 (1
.4 )
C la
n m
e m
b e rs
in c o
m m
u n
it y
N o
1 .0
0 1
4 .1
(3 .6
) 2
1 .6
(2 .2
) n
s 8
.5 (1
.0 )
4 9
.3 (1
.6 )
Y e s-
c la
n m
e m
b e rs
1 .4
1 (0
.8 6
– 2
.3 0 )
3 0
.0 (2
.9 )
7 7
.1 (2
.4 )
n s
1 3
.9 (1
.2 )
4 8
.3 (1
.5 )
R e fu
se d /d
o n ’t
k n o w
2 .5
2 (1
.2 0 – 5 .2
7 )
2 8 .1
(9 .4
) 1 .3
(0 .4
) n s
6 .5
(3 .1
) 2 .4
(0 .4
)
R e la
ti v
e re
m o
v e d
fr o
m n
a tu
ra l
fa m
il y
c
W a s
n o
t re
m o
v e d
1 .0
0 3
1 .0
(4 .6
) 5
3 .6
(2 .6
) n
s 1
0 .1
(1 .3
) 4
1 .2
(1 .4
)
R e m
o v e d
fr o
m fa
m il
y 0
.7 5
(0 .5
4 –
1 .0
6 )
2 2
.6 (2
.2 )
2 8
.0 (1
.9 )
n s
1 2
.9 (1
.3 )
3 9
.3 (1
.3 )
R e fu
se d /d
o n ’t
k n o w
0 .5
3 (0
.3 6 – 0 .7
7 )
1 9 .8
(3 .1
) 1 8 .4
(2 .3
) n s
9 .2
(1 .6
) 1 9 .5
(1 .2
)
S o
c ia
l a
n d
c o
m m
u n
it y
sa fe
ty
T y
p e s
o f
c o
m m
u n
it y
p ro
b le
m s
N o
y o
u th
g a n
g 1
.0 0
1 4
.1 (1
.6 )
5 2
.0 (4
.1 )
n s
9 .3
(0 .9
) 7
3 .6
(1 .5
)
Y o
u th
g a n
g p
ro b
le m
s 1
.5 0
(1 .1
7 –
1 .9
3 )
4 0
.0 (3
.9 )
4 8
.0 (4
.1 )
n s
1 6
.0 (1
.6 )
2 6
.4 (1
.5 )
N o
a lc
o h o
l p
ro b
le m
s 1
.0 0
1 1
.8 (2
.2 )
4 5
.8 (3
.5 )
n s
9 .1
(0 .8
) 7
4 .5
(1 .5
)
A lc
o h o
l p
ro b
le m
s 2
.1 5
(1 .3
7 –
3 .3
9 )
3 9
.0 (3
.9 )
5 4
.2 (3
.5 )
n s
1 6
.6 (1
.7 )
2 5
.5 (1
.5 )
N o
p h
y si
c a l
a ss
a u
lt 1
.0 0
1 2
.7 (1
.4 )
5 8
.6 (4
.8 )
n s
1 0
.1 (0
.8 )
8 8
.5 (1
.1 )
P h
y si
c a l
a ss
a u
lt p
ro b
. 2
.3 5
(1 .7
9 –
3 .0
8 )
4 6
.1 (3
.8 )
4 1
.4 (4
.8 )
n s
1 8
.5 (2
.5 )
1 1
.5 (1
.1 )
N o
b re
a k
-i n
s/ th
e ft
n s
1 9
.0 (2
.2 )
5 8
.4 (3
.8 )
1 .0
0 7
.9 (0
.8 )
5 5
.8 (1
.5 )
B re
a k
-i n
s/ th
e ft
p ro
b .
n s
3 7
.1 (4
.4 )
4 1
.6 (3
.8 )
1 .7
1 (1
.2 1
– 2
.4 1 )
1 5
.0 (1
.4 )
4 4
.2 (1
.5 )
N o
fa m
il y
v io
le n
c e
p ro
b .
n s
1 4
.0 (1
.8 )
5 8
.3 (4
.4 )
1 .0
0 9
.7 (0
.8 )
8 6
.3 (1
.1 )
F a m
il y
v io
le n
c e
p ro
b .
n s
4 4
.1 (4
.3 )
4 1
.7 (4
.4 )
1 .6
8 (1
.1 7
– 2
.4 1 )
1 9
.7 (2
.4 )
1 3
.7 (1
.1 )
Independent Correlates of Reported Gambling Problems 157
123
T a b
le 4
c o
n ti
n u e d
2 0
0 2
R e m
o te
2 0
0 2
N o
n -r
e m
o te
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
O R
(9 5
% C
I) b
G a m
b li
n g
p ro
b le
m sa
% (S
E )
P o
p -n
d is
tr ib
u ti
o n
a
% (S
E )
P h
y si
c a l/
th re
a te
n e d
v io
le n
c e
N o
t a
v ic
ti m
1 .0
0 2
2 .5
(2 .7
) 7
7 .0
(1 .3
) n
s 9
.9 (0
.9 )
7 6
.6 (1
.1 )
H a s
b e e n
v ic
ti m
1 .8
8 (1
.4 2
– 2
.4 9 )
4 0
.4 (5
.0 )
2 3
.0 (1
.3 )
n s
1 4
.7 (1
.5 )
2 3
.4 (1
.1 )
H e a
lt h
a n
d h
e a
lt h
b eh
a v io
u rs
S e lf
-a ss
e ss
e d
h e a lt
h st
a tu
s
E x
c e ll
e n
t n
s 2
2 .7
(4 .0
) 1
5 .4
(1 .3
) 1
.0 0
8 .2
(2 .0
) 1
6 .3
(1 .1
)
V e ry
g o
o d
n s
3 2
.1 (3
.9 )
2 7
.1 (1
.6 )
0 .9
9 (0
.4 9
– 2
.0 3 )
9 .5
(1 .3
) 2
5 .5
(1 .2
)
G o
o d
n s
2 4
.0 (2
.7 )
3 5
.8 (1
.9 )
1 .7
5 (0
.8 7
– 3
.5 3 )
1 3
.3 (1
.7 )
3 1
.7 (1
.3 )
F a ir
n s
2 6
.9 (3
.4 )
1 5
.0 (0
.9 )
1 .7
9 (0
.9 1
– 3
.5 4 )
1 3
.2 (2
.0 )
1 8
.9 (1
.0 )
P o
o r
n s
2 6
.6 (6
.5 )
6 .6
(0 .8
) 0
.9 8
(0 .4
6 –
2 .0
6 )
8 .0
(1 .5
) 7
.7 (0
.6 )
W e ig
h te
d p
o p
u la
ti o
n (N
) –
– 6
9 ,0
7 7
– –
1 7
6 ,0
0 8
N o te
P o
p u
la ti
o n
d is
tr ib
u ti
o n
fo r
e a c h
v a ri
a b
le m
a y
n o
t su
m to
1 0
0 d
u e
to ro
u n
d in
g
n s
n o
t si
g n
ifi c a n
t (p
[ 0
.0 5 )
a n
d th
e re
fo re
n o
t in
c lu
d e d
in fi
n a l
m u
lt iv
a ri
a b
le m
o d
e l
n a
n o t
a p p li
c a b le
a s
d a ta
it e m
o r
c a te
g o ry
o f
v a ri
a b le
w a s
n o t
c o ll
e c te
d o r
c o d e d
re sp
e c ti
v e ly
n r
A C
T a n
d V
ic to
ri a
h a v
e n
o re
m o
te re
g io
n s
a n
d th
e T
a sm
a n
ia n
sa m
p le
d id
n o
t in
c lu
d e
re m
o te
re g io
n s
d u
e to
th e
sm a ll
In d
ig e n
o u
s p
o p
u la
ti o
n li
v in
g in
th e se
a re
a s
a E
st im
a te
s fr
o m
R A
D L
m a y
d if
fe r
sl ig
h tl
y fr
o m
p u
b li
sh e d
A B
S e st
im a te
s d
u e
to ra
n d
o m
p e rt
u rb
a ti
o n
o f
R A
D L
d a ta
b y
th e
A B
S to
e n
su re
c o
n fi
d e n
ti a li
ty b
O R
(9 5
% C
I) ,
o d
d s
ra ti
o a n
d 9
5 %
c o
n fi
d e n
c e
in te
rv a l
c In
d ig
e n
o u
s p
e o
p le
o f
m ix
e d
h e ri
ta g
e w
e re
re m
o v e d
fr o m
th e ir
fa m
il y
a s
a re
su lt
o f
g o
v e rn
m e n t
p o
li c y
to th
e m
id 1
9 7
0 ’s
d e p
e n
d in
g o
n st
a te
/t e rr
it o
ry la
w s
158 M. Stevens, M. Young
123
previous year to the survey reported more gambling problems (OR 1.88, 1.42 to 2.49) than
those not subject to threatened or physical violence.
3.3 Non-Remote Multivariable Model
Just over 3% of the observations were excluded from the non-remote multivariable model
due to missing data. The same domains were represented in the non-remote analysis as the
remote analysis with the addition of self-reported health (see Table 4). Compared with
Western Australia (the reference category), South Australia (OR 5.16, 2.27 to 11.7) and the
Northern Territory (OR 6.06, 1.67 to 21.9) had the highest estimates of reported gambling
problems, followed by Victoria (OR 4.02, 1.83 to 8.86), Queensland (OR 2.85, 1.45 to
5.62), New South Wales (OR 2.83, 1.31 to 6.13), and Australian Capital Territory/Tas-
mania (OR 2.43, 1.07 to 5.55). Female respondents (OR 1.59, 1.16 to 2.17) were more
likely to report gambling problems than male respondents, as were respondents living in
households where all residents were Indigenous (OR 1.68, 1.11 to 2.53) compared with
mixed households. Respondents in the highest income group of $691 per week or more
were more likely to report gambling problems (OR 2.45, 1.52 to 3.95) compared with the
lowest income earners. Respondents attending sporting events or carnivals (OR 1.61, 1.12
to 2.32) and those involved in an Indigenous organisation (OR 1.76, 1.19 to 2.62) reported
more gambling problems than respondents not attending or involved these activities.
Respondents who identified community problems of break-ins and theft (OR 1.71, 1.21 to
2.41), and family violence (OR 1.68, 1.17 to 2.41) reported more gambling problems than
those not reporting these community problems. Lastly, respondents who reported their
health as good (OR 1.75, 0.87 to 3.53) or fair (OR 1.79, 0.91 to 3.54) reported more
gambling problems than respondents who rated their health as excellent (and very poor if
reference category is changed).
4 Discussion
4.1 Estimates of Reported Gambling Problems
Estimates of reported gambling problems among the Australian Indigenous population
varied significantly by jurisdiction and by remoteness. The low levels of reported gambling
problems for non-remote locations in Western Australia was unsurprising as this is then
only jurisdiction that does not supply EGMs, the most harmful form of gambling, outside
of the casino in Perth (see Map in Fig. 1) (McMillen and Doran 2006; Queensland Gov-
ernment 2005; Young et al. 2007; Young et al. 2008a). The high estimates of reported
gambling problems in remote regions, particularly in the Northern Territory and
Queensland, may indicate that unregulated gambling (i.e. card games), which are a
ubiquitous activity in remote Indigenous communities (Martin 1993; Young et al. 2007),
are causing problems for people, their families and social networks. The evidence on the
role card games play in remote communities is fragmented and localised, with both
positive (Altman 1985; Goodale 1987; McMillen and Togni 2000; Paterson 2006, 2007;
Young et al. 2007) and negative (Hunter 1993; Hunter and Spargo 1988; Martin 1993;
McKnight 2002; McMillen and Togni 2000; Memmot et al. 2001; Young et al. 2007)
social outcomes identified. It is likely that the effects of unregulated gambling are highly
variable due to differing physical environs, histories of colonisation and religious influ-
ences (Long 1970). In addition, the gambling industry has expanded over the last 20–
Independent Correlates of Reported Gambling Problems 159
123
30 years with the opening of casinos in all capital cities and the introduction of EGMs into
bars and clubs throughout Australia (except Western Australia). The increase in gambling
opportunities has significant implications for Indigenous people living in remote localities
who make up to 30 trips a year per person to larger service centres (Memmot et al. 2006)
where EGMs and other gambling opportunities are readily available.
4.2 Urban—Remote Differences and Cultural Diversity
Variables from regional, demographic, socioeconomic, social connectedness, community
problems and health domains were independently associated with reported gambling
problems in the Indigenous population. Gender and household structure and composition
were associated with reported gambling problems although these varied between the
remote and non-remote analyses. Specifically, females reported more gambling problems
than males, but only in non-remote areas. The research literature suggests that Indigenous
females tend to gamble more frequently than males (Aboriginal Health & Medical
Research Council of NSW 2007; Holden et al. 1996; Martin 1993; McKnight 2002; Pat-
erson 2006, 2007; Young et al. 2007; Young et al. 2006), and this is likely to lead to more
problems. In addition, because females are the primary care givers and nurturers of chil-
dren, they are more likely to be impacted by others gambling (e.g. spouses and extended
family) and therefore more likely to report gambling problems. Unfortunately, because of
the way the reported gambling problem variable was collected, it is not possible to
ascertain whose gambling problem the respondent was referring to (i.e. whether it was the
respondent, a family member or a close friend).
Household composition and household structure were also associated with reported
gambling problems. Crowding is also a significant factor for remote communities with
large numbers of dwellings defined as multifamily households (see Table 3), which is
likely to contribute to the social problems associated with gambling (Australian Bureau of
Statistics 2002, 2003, 2004b, 2007b, 2008b; Bailie and Runcie 2001; Bailie et al. 2005).
This finding may reflect economic organisation of the household. For example, an
Indigenous person living by themselves in the Northern Territory does not necessarily
reflect social isolation. Sole residence may in fact be indicative of higher functionality and
economic independence. That is, Indigenous people living in lone person households do
not need to live with relatives and extended family as a means to access resources for day
to day living. Conversely, people living in multifamily households often do so as a means
of pooling resources and this allows the resident to share and borrow, which results in a
more constant supply of money available for gambling. Therefore, larger households can
support a mini-economy of gambling within a culture of reciprocal sharing (Austin-Broos
2003; Berndt and Tonkinson 1988; Moody 2008). Another explanation for this finding may
be the increased stress associated with crowded living conditions (Jones 1994, 1999;
Neutze et al. 2000), with the potential problems associated with gambling likely exacer-
bated (Gove 1979; Hunter 2007; Hunter and Spargo 1988).
In addition, all Indigenous households reported more gambling problems than those
with a mixture of Indigenous and non-Indigenous residents (although in remote areas less
than 10% of households contained a mixture of Indigenous and non-Indigenous residents,
compared with 45% of households containing a mixture in non-remote areas). This vari-
able may reflect the degree of integration into main stream Australian culture through
connections with non-Indigenous people. That is, it may be that Indigenous people who are
more socially connected to mainstream Australian culture are less likely to experience
gambling problems, either as a result of gambling less frequently or by being more aware
160 M. Stevens, M. Young
123
of the negative consequences of gambling. For example, in the 2002 General Social Survey
(GSS), which contains fewer than a 2% Indigenous sample, only 3.5% of the general
population in non-remote areas of Australia reported gambling problems (Australian
Bureau of Statistics 2004a), while 10.2% of respondents reported gambling problems in
non-remote locations for the equivalent Indigenous survey as reported herein.
There was also significant association between gambling problems and whether the
respondent had a relative removed from their natural family. Specifically, the association
was positive in non-remote locations and negative in remote locations. The culture of
Indigenous people living in remote areas is more intact compared with people living in less
remote locations of Australia. Additionally, the social and economic conditions in which
people live in remote areas is significantly worse than those of their urban and less remote
counterparts (Australian Bureau of Statistics 2002, 2003, 2007b, 2008b). Therefore, in
remote areas it may be that people who had a relative removed are less enmeshed within
Indigenous culture of the other people living in these regions, while in urban areas, it may
reflect an increased awareness of the negative consequences of gambling, though this
conclusion is speculative. The finding for remote areas is consistent with results from the
Western Australian Aboriginal Child Health Survey, with carer’s who were forcibly
removed from their natural family reporting gambling problems in their households at
twice the rate of other Indigenous carer’s (Zubrick et al. 2005). Another explanation comes
from a study of gambling and substance abuse in a remote North Queensland Indigenous
community identified trauma as a contributing factor in Indigenous participation in gam-
bling as a means of escape (Phillips 2003). If this is the case then it would indicate that
Indigenous people living in places with a higher percentage of the total population
Indigenous may actually be experiencing less trauma than their more traditionally orien-
tated (e.g. speaks language, visits tribal land etc.) Indigenous people, though this is again
speculative. Further analysis breaking down the inter-generational effects of having a
relative removed would go some way towards amplification of this finding.
Indigenous people in the highest personal income groups reported higher levels of
gambling problems for the non-remote analysis, but not for the remote analysis. For the
remote analysis, Indigenous people in these income categories reported the lowest levels of
gambling related problems (although not significant in the final model). This again may
reflect differing levels of acculturation into mainstream labour force markets (Australian
Bureau of Statistics 2008b). In non-remote regions it may be that Indigenous people in
higher income brackets are more likely to report gambling problems (due to awareness) or
it may be that they are being called upon to support other less financially well off relatives
in their pursuit of gambling. More detailed research is required to explain this finding.
4.3 Social Inclusion and Community Problems
Variables relating to Indigenous participation in social and cultural activities had a positive
association with reported gambling problems. There was some overlap between the remote
and non-remote analyses, though participation or involvement in these types of activities
was more important for Indigenous people living in remote locations. For remote locations
social and cultural activities included involvement in special community activities and
sports or physical exercise, and attendance at sporting carnivals and funerals. For non-
remote locations only attendance at funerals and involvement in an Indigenous organisa-
tion were independently associated with gambling problems. While other social and cul-
tural activities were not in the final models, they all showed the same directional
association with reported gambling problems (see Table 3). Historically, gambling has
Independent Correlates of Reported Gambling Problems 161
123
played a significant role in Indigenous Australians social life, particularly prior to self-
determination in 1967, when Indigenous people were excluded from virtually all enter-
tainment options available to other Australians. Card games have been ubiquitous since the
1930’s and were played when families came together (Berndt and Berndt 1946–47; Dodd
and Vaughn 1985; Tonkinson 1974), and they are still very common in remote commu-
nities across Australia (Martin 1993; McMillen and Donnelly 2008; McMillen and Togni
2000; Paterson 2006, 2007; Young et al. 2007). However, there is some evidence sug-
gesting Indigenous people are increasingly pursuing commercial forms of gambling, and
this is particularly so in larger regional centres (Aboriginal Health & Medical Research
Council of NSW 2007; Fogarty 2008; Foote 1996; Holden et al. 1996). Given that many
social events occur in larger regional centres, it may be that Indigenous people travelling to
these places are also partaking in available commercial gambling options. A recent study
supports this hypothesis, in that approximately 20% of the patrons in a large urban gam-
bling venue in the Northern Territory were Indigenous, yet Indigenous people make up
fewer than 10% of the urban population in the Northern Territory (Australian Bureau of
Statistics 2007c, 2008b; Fogarty 2008). However, it is worth pointing out that the variables
capturing involvement and participation in these activities are crude measures. Specifi-
cally, it is not known what types of community activities people were participating in, and
similarly it was not known where sporting carnivals and events are being held or whether
gambling was occurring at these venues or in social gatherings after the event.
The association between community problems was present for both the remote and non-
remote analyses, although the types of community problems that remained in the final
models differed. Types of problems in remote areas associated with gambling problems
were youth gangs, alcohol abuse, and physical assault problems. In 2008 the Federal
Government intervened in the management in 73 Indigenous communities across the
Northern Territory (Northern Territory Emergency Response Act, 2007) after a report on child abuse found rampant levels of sexual assault (Wild and Anderson 2007). The authors
of this report argued that child abuse and neglect was associated with alcohol and drug
abuse, overcrowding, pornography, gambling, and lack of employment opportunities.
Other reports have found virtually identical problems in remote Indigenous communities
across Australia (Department of Indigenous Affairs 2008; Mullighan 2008; Senate Com-
mittee 2008). These reports all highlight that community problems do not operate inde-
pendently and it is the sum of these problems that lead to ‘community dysfunction’.
The association observed between reported gambling problems and self-reported health
is not unexpected, though the non-linear association is difficult to further explain and was
not consistent between the remote and non-remote analyses. The use of self-reported health
in Australia’s Indigenous populations has received some attention, with one study, using an
earlier Indigenous social survey, recommending that it is not a reliable measure when
English is not the person’s first language (Sibthorpe et al. 2001). Given that 12% of
Indigenous Australians do not speak English at home (40% in remote locations) (Aus-
tralian Bureau of Statistics 2004b), along with varying meanings of health across culturally
diverse populations, further research is needed to understand the significance of this
observed association (Sibthorpe et al. 2001).
5 Conclusions: Policy Implication for Reducing Gambling Harm
The findings illustrate that gambling is closely tied to a range of social and environmental
domains that need to be considered when formulating harm-reduction strategies. In other
162 M. Stevens, M. Young
123
words, simply construing gambling as an isolated phenomenon that is causal of social
problems is somewhat limited and inaccurate. Gambling is produced by and produces
social issues; it is part of a complex of social relationships and processes. Harm reduction
strategies, to be effective, may need to include these broader contexts. Table 5 summarises
policy initiatives and the associated variable from the multivariable analysis that may
reduce gambling related harm amongst the Indigenous population of Australia.
As part of the Northern Territory emergency response, the Commonwealth government
will inject some AUD$680 million into housing in Indigenous communities in this juris-
diction over the coming years, and it is initiatives like this that are likely to reduce negative
impacts of gambling on families through reduced crowding and associated stresses when
living in these environments. Community initiatives aimed at improving community
cohesion and wellbeing and increased policing in communities, while reducing violence
and other community problems, would also be expected to reduce some of the negative
consequences of gambling. The association between reported gambling problems and
attendance at social and cultural events highlights the social nature of gambling and these
may be good places to promote awareness about the harms associated with excessive
gambling.
In conclusion, while the analyses conducted as part of this study constitutes the first
empirical analysis of reported gambling problems and their correlates across Australia for
the Indigenous population, we do not know what types of problems people are experi-
encing and more importantly, which gambling activities were associated with problems. In
addition, little is understood about the relationship between unregulated and regulated
gambling or how community level programs might ameliorate negative consequences of
gambling. Our lack of knowledge is significant in this regard. This paper has documented a
complex range of social factors affect gambling outcomes, and these outcomes may not be
divorced from the broader contexts of social and economic disadvantage in which Indig-
enous Australians are located.
Table 5 Independent correlates of gambling problems and policy implications
Significant independent correlates Policy implications
Multi-family households Crowded housing increases the chance of someone being affected by another persons gambling. High levels of overcrowding in remote communities may undermine other efforts to reduce gambling related harm
Income Improving employment and educational outcomes will increase disposable income, lessen time available for gambling, and improve individual ability to make an informed choice. Differences in the association between income and gambling problems in remote and non-remote areas may require different policy approaches for public health messages
Social connectedness (participation and attendance at social/cultural events)
Places where people gather provide good exposure for information and posters on gambling related harm and availability of counselling services, and also raise awareness about harm associated with gambling
Community problems and victim of physical or threatened violence
Community cohesion and wellbeing programs and improved policing of communities, while improving safety may also increase the community’s capacity to manage problems associated with gambling
Independent Correlates of Reported Gambling Problems 163
123
Acknowledgments The authors thank representatives of the Ministerial Council on Gambling for com- ments on an earlier draft of the manuscript. The authors thank Gambling Research Australia (GRA) for funding the research.
Competing interests None.
References
Aboriginal Health & Medical Research Council of NSW. (2007). Pressing problems: Gambling issues and responses for NSW Aboriginal communities. Sydney: NSW Government.
Altman, J. (1985). Gambling as a mode of redistributing and accumulating cash among Aborigines: A case study from Arnhem Land. In G. Caldwell, B. Haig, M. Dickerson, & L. Sylvan (Eds.), Gambling in Australia (pp. 50–67). Sydney: Croon Helm.
Austin-Broos, D. (2003). Places, practices, and things: The articulation of Arrente kinship with welfare and work. American Ethnologist, 30(1), 118–135.
Australian Bureau of Statistics. (2002). Housing and infrastructure in Aboriginal and Torres Strait com- munities, Australia, 2001. Canberra: ABS.
Australian Bureau of Statistics. (2003). Population characteristics, Aboriginal and Torres Strait Islander Australians, 2001. Canberra.
Australian Bureau of Statistics. (2004a). General Social Survey: Summary results, Australia, 2002. Can- berra: ABS.
Australian Bureau of Statistics. (2004b). National Aboriginal and Torres Strait Islander Social Survey, Australia, 2002. Canberra.
Australian Bureau of Statistics. (2004c). National Aboriginal and Torres Strait Islander Social Survey, Northern Territory, 2002. Canberra.
Australian Bureau of Statistics. (2004d). National Aboriginal and Torres Strait Islander Social Survey: Technical manual, 2002. Canberra.
Australian Bureau of Statistics. (2006a). National Aboriginal and Torres Strait Islander Health Survey: Expanded Confidentialised Unit Record File, Technical Manual, Australia, 2004–05. Canberra.
Australian Bureau of Statistics. (2006b). Remote access data laboratory (RADL) user guide. Version 4. Canberra.
Australian Bureau of Statistics. (2007a). ABS directions in Aboriginal and Torres Strait Islander Statistics, Jun 2007. Canberra.
Australian Bureau of Statistics. (2007b). Housing and infrastructure in Aboriginal and Torres Strait com- munities, Australia, 2006. Canberra: ABS.
Australian Bureau of Statistics. (2007c). Population distribution, Aboriginal and Torres Strait Islander Australians, 2006. Canberra: ABS.
Australian Bureau of Statistics. (2008a). Australian Standard Geographical Classification (ASGC). Can- berra: ABS.
Australian Bureau of Statistics. (2008b). Population characteristics, Aboriginal and Torres Strait Islander Australians, 2006. Canberra.
Bailie, R., & Runcie, M. (2001). Household infrastructure in aboriginal communities and the implications for health improvement. MJA, 175(7), 363–366.
Bailie, R., Stevens, M., McDonald, E., Halpin, S., Brewster, D., Robinson, G., et al. (2005). Skin infection, housing and social circumstances in children living in remote Indigenous communities: Testing con- ceptual and methodological approaches. BMC Public Health, 5(1), 128.
Battersby, M. W., Thomas, L. J., Tolchard, B., & Esterman, A. (2002). The South Oaks gambling screen: A review with reference to Australian use. Journal of Gambling Studies, 18(3), 257–271.
Berndt, R. M., & Berndt, C. H. (1946-47). Card games among the Aborigines of Northern Territory. Oceania, 17, 248–269.
Berndt, R. M., & Tonkinson, R. (1988). Social anthropology and Australian Aboriginal studies: A con- temporary overview. Canberra: Aboriginal Studies Press for the Australian Institute of Aboriginal Studies.
Brady, M. (2004). Regulating social problems: The pokies, the Productivity Commission and an Aboriginal community (No. 1036–1774 CNO: online only). Canberra: Australian National University, Centre for Aboriginal Economic Policy Research.
Cultural Perspectives Pty Ltd. (2005). Research into health promotion and best practice services for Indigenous diverse communities. Melbourne, VIC, Australia: Victorian Government Department of Justice, Office for Gaming and Racing.
164 M. Stevens, M. Young
123
Department of Indigenous Affairs. (2008). Western Australia State Government response to the Hope Report. Perth: Government of Western Australia.
Dickerson, M., Allcock, C., Blaszczynski, A., Nicholls, B., Williams, J., & Maddern, R. (1996). A pre- liminary exploration of the positive and negative impacts of gaming and wagering on Indigenous people in New South Wales. Sydney: Australian Institute of Gambling Studies.
Dickerson, M., Baxter, P., Boreham, P., Harley, W., & Williams, J. (1995). The impacts of the introduction of gaming machines on Queensland. Brisbane: Queensland Government, Department of Family Ser- vices and Aboriginal and Islander Affairs.
Dodd, R., & Vaughn, R. (1985). Aboriginal gambling and self-determination in Queensland. In G. Caldwell, B. Haig, M. Dickerson, & L. Sylvan (Eds.), Gambling in Australia. Sydney: Croon Helm.
Duvarci, I., Varan, A., Coskunol, H., & Ersoy, M. A. (1997). DSM-IV and the South Oaks gambling screen: Diagnosing and assessing pathological gambling in Turkey. Journal of Gambling Studies, 13(3), 193–206.
Ferris, J., & Wynne, H. (2001). The Canadian problem gambling index: Final report. Toronto: Canadian Centre on Substance Abuse.
Fogarty, M. (2008). Out of harms way: An Indigenous-centric approach to harm minimisation. Paper presented at the National Association for Gambling Studies 18th Annual Conference: Finding Com- mon Ground.
Foote, R. J. (1996). Aboriginal gambling: A pilot study of casino attendance and the introduction of poker machines into community venues in the Northern Territory. Darwin: Centre for Social Research, Northern Territory University.
Goodale, J. C. (1987). Gambling is hard work: Card playing in Tiwi society. Oceania, 58(1), 6–21. Gove, W. R. (1979). Overcrowding in the home: An empirical investigation of its possible pathological
consequences. American Sociological Review, 44(1), 59–80. Holden, A., Dickerson, M., Boreham, P., Harley, W., & Hogan, B. (1996). Long term study into the social
impacts of gaming machines in Queensland: An issues paper: The social and economic impact of gaming machines on Aboriginal and Torres Strait Islander communities in Queensland. Brisbane: Queensland Government.
Hunter, E. M. (1993). Aboriginal health and history, power and prejudice in remote Australia. Cambridge: Cambridge University Press.
Hunter, E. (2007). Disadvantage and discontent: A review of issues relevant to the mental health of rural and remote Indigenous Australians. Australian Journal of Rural Health, 15(2), 88–93.
Hunter, E. M., & Spargo, R. M. (1988). What’s the big deal? Aboriginal gambling in the Kimberley region. Medical Journal of Australia, 149(11–12), 668–672.
Jones, R. (1994). The housing need of Indigenous Australians, 1991. Canberra: Centre for Aboriginal and Economic Policy Research, ANU.
Jones, R. (1999). Indigenous housing 1996 census analysis. Canberra: Centre for Aboriginal and Economic Policy Research, ANU.
Kowall, E., Gunthorpe, W., & Bailie, R. (2007). Measuring emotional and social wellbeing in Aboriginal and Torres Strait Islander populations: An analysis of a Negative Life Events Scale. International Journal for Equity in Health, 6(18), 1–12.
Lesieur, H. R., & Blume, S. B. (1987). The South Oaks gambling screen (SOGS): A new instrument for the identification of problem gamblers. American Journal of Psychiatry, 144(9), 1184–1188.
Long, J. P. M. (1970). Aboriginal settlements: A survey of Institutional communities in Eastern Australia. Canberra: ANU.
Martin, D. F. (1993). Autonomy and relatedness: An ethnography of Wik people of Aurukun, western Cape York Peninsula. Unpublished PhD thesis. Canberra: Australia National University.
McKnight, D. (2002). From hunting to drinking: The devastating effects of alcohol on an Australian Aboriginal community. London: Routledge.
McMillen, J., & Donnelly, K. (2008). Gambling in Australian Indigenous communities: The state of play. Australian Journal of Social Issues, 43(3), 397–426.
McMillen, J., & Doran, B. (2006). Problem gambling and gaming machine density: Socio-spatial analysis of three Victorian localities. International Gambling Studies, 6(1), 5–29.
McMillen, J., & Marshall, D. (2004). 2003 Victorian Longitudinal Community Attitudes Survey. Melbourne: Victorian Government, Community Support Fund.
McMillen, J., & Togni, S. (2000). Study of gambling in the Northern Territory 1996–97. Sydney: Australian Institute of Gambling Research, University of Western Sydney.
McMillen, J., & Wenzel, M. (2006). Measuring problem gambling: Assessment of three prevalence screens. International Gambling Studies, 6(2), 147–174.
Memmot, P. C., Long, S. J., Bell, M. J., Taylor, J., & Brown, D. (2006). Between places: Indigenous mobility in remote and rural Australia. Brisbane: Australian Housing and Urban Research Institute.
Independent Correlates of Reported Gambling Problems 165
123
Memmot, P., Stacy, R., Chambers, C., & Keys, C. (2001). Violence in Indigenous communities. Canberra: Attorney Generals Department.
Moody, M. (2008). Serial reciprocity: A preliminary statement. Sociological Theory, 26(2), 130–151. Mullighan, E. P. (2008). Children on Anangu Pitjantjatjara Yankunytjatjara (Apy) lands, Commission Of
Inquiry, a report into sexual abuse. Adelaide: South Australian Government, Office of the Commissioner.
Neal, P., Delfabbro, P. H., & O’Neil, M. (2005). Problem gambling and harm: Towards a national defi- nition. Melbourne: Report prepared for the National Gambling Research Program Working Party.
Neutze, M., Sanders, W., & Jones, R. (2000). Estimating Indigenous housing need for public funding allocation: A multi-measure approach (No. 197). Canberra: Centre for Aboriginal and Economic Policy Research, ANU.
Paterson, M. (2006). A future now; not a past too late: An analysis of Aboriginal gambling in Australia. Paper presented at the ‘gambling: The future now’ the National Association for Gambling Studies 16th Annual Conference, Sydney.
Paterson, M. (2007). The regulation of ‘unregulated’ Indigenous gambling. Paper presented at the National Association for Gambling Studies 17th Annual Conference, Adelaide.
Phillips, G. (2003). Addictions and healing in Aboriginal country. Canberra: Aboriginal Studies Press. Productivity Commission. (1999). Australia’s gambling industries: Inquiry Report No.10. Melbourne:
Productivity Commission. Queensland Government. (2005). Games people play—Problem gambling among offenders in Queensland
corrections centres. Brisbane: Queensland Treasury. Queensland Government (Ed.). (2006). Queensland Household Gambling Survey, 2003–04. Brisbane:
Treasury and Department of Corrective Services, QLD Govt. Queensland Government (Ed.). (2007). Queensland Household Gambling Survey, 2006-07. Brisbane:
Treasury and Department of Corrective Services, QLD Govt. Roberts, J. (1981). Massacres to mining: The colonisation of Aboriginal Australia. Blackburn, VIC: Dove
Communications. Roy Morgan Research. (2006). The fourth study into the extent and impact of gambling in Tasmania with
particular reference to problem gambling. Hobart: Gambling Support Bureau. Senate Committee. (2008). Senate select committee on regional and remote Indigenous communities.
Canberra: Commonwealth Government of Australia. Sibthorpe, B., Anderson, I., & Cunningham, J. (2001). Self-assessed health among Indigenous Australians:
How valid is a global question? American Journal of Public Health, 91(10), 1660–1663. Svetieva, E., & Walker, M. (2008). Inconsistency between concept and measurement: The Canadian
Problem Gambling Index (CPGI). Journal of Gambling Issues, 22, 157–173. Tonkinson, R. (1974). The Jigalong mob: Aboriginal victors of the desert crusade. Menlo Park, CA:
Cummings Publishing Co. Inc. Wild, R., & Anderson, P. (2007). Ampe Akelyernemane Meke Mekarle ‘‘Little Children are Sacred’’: Report
of the Northern Territory Board of Inquiry into the protection of Aboriginal children from sexual abuse. Darwin: Northern Territory Government.
Winter, N. (2008). Programs for Stata. Retrieved December 19, 2008 from http://www.faculty.virginia.edu/ nwinter/progs/.
Young, M., Barnes, T., Stevens, M., Paterson, M., & Morris, M. (2007). The changing landscape of Indigenous gambling in Northern Australia: Current knowledge and future directions. International Gambling Studies, 7(3), 327–347.
Young, M., Morris, M., Barnes, T., Stevens, M., & Paterson, M. (2006). Indigenous gambling scoping study: A summary. Darwin: School for Social and Policy Research, Charles Darwin University.
Young, M., & Stevens, M. (2009). Player preferences and social harm: An analysis of the relationships between player characteristics, gambling modes, and problem gambling. International Journal of Mental Health & Addiction, 7(1), 262–279.
Young, M., Stevens, M., & Morris, M. (2008a). Problem gambling within the non-Indigenous population of the Northern Territory of Australia: A multivariate analysis of risk factors. International Gambling Studies, 8(1), 77–93.
Young, M., Stevens, M., & Morris, M. (2008b). Problem gambling within the non-Indigenous population of the Northern Territory of Australia: A multivariate analysis of risk factors. International Gambling Studies, 8(1), 77–93.
Zubrick, S. R., Silburn, S. R., Lawrence, D. M., Mitrou, F. G., Dalby, R. B., Blair, E. M., et al. (2005). The Western Australian Aboriginal Child Health Survey: The social and emotional wellbeing of Aboriginal children and young people (Vol. 2). Perth: Curtin University of Technology and Telethon Institute for Child Health Research.
166 M. Stevens, M. Young
123
Copyright of Social Indicators Research is the property of Springer Science & Business Media B.V. and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.
