Exploring the effects of community disadvantage and remoteness on Indigenous and non-Indigenous peoples’ risk of reincarceration

Abstract

Community disadvantage and a person's residential geographical location are believed to be risk factors for crime. This research aimed to go beyond examining individual-level risk factors for reincarceration and explored the impact of community disadvantage and residential geographical location on Australia's Indigenous and non-Indigenous peoples’ risk of reincarceration post-release. Descriptive analyses, logistic regression and Cox proportional hazards models were conducted using survey and linked administrative data for 1238 prisoners. We found no relationship between residential geographical location and reincarceration for either Indigenous or non-Indigenous people. Moreover, no relationship between community disadvantage and reincarceration was found for non-Indigenous people, however, results indicated community disadvantage to be a protective factor for Indigenous people. Potential explanation for this perplexing finding is discussed, as are potential implications for how we view and measure community disadvantage for Australian Indigenous people.

Keywords

Indigenous reincarceration social disadvantage remoteness kinship

Introduction

Australia's Indigenous prison population is at a record high with 12,092 Indigenous people currently serving time in prison (ABS, 2020). Accounting for 2% of Australia's population aged 18 and over, Australia's Indigenous people are over-represented in the prison system by an age-adjusted factor of 13, making up 29% of Australia's total prison population on any given day (ABS, 2020). Efforts to close the gap between Indigenous and non-Indigenous incarceration rates in Australia have been unsuccessful, and Indigenous incarceration numbers continue to grow. Prior studies that attempt to explain Indigenous over-representation at the most general level do so primarily by questioning whether Indigenous people are arrested and incarcerated more often than non-Indigenous people (Willis, 2008). Few studies, however, focus on Indigenous recidivism and prison re-entry to help explain the over-representation of Indigenous people in prison. Those that do, have largely focused on examining differences in recidivism between Indigenous and non-Indigenous people for specific types of offenders (i.e. non-aggravated assault or burglary (Weatherburn, 2010) and violent offenders (Willis, 2008)), and have only focused on Indigenous males (Wundersitz, 2010). The majority have only been able to explore a limited number of explanatory factors (e.g. Giles & Whale, 2016). As such, what we know about prison re-entry and the risk of reincarceration for Indigenous people is limited. Although prior research has indicated that Indigenous people have higher reincarceration rates than non-Indigenous people and has identified a small number of factors that provide an understanding of the causal mechanisms involved (Ryan et al., 2019; Ryan et al., 2020), there remains a clear need for further research to uncover a greater number of these causal mechanisms. For example, almost no research examines whether differences in the social ecology of Indigenous and non-Indigenous peoples’ communities explain Indigenous/non-Indigenous differentials in reincarceration risk.

Understanding more about the risk of prison re-entry is critically important to the design of treatment, reintegration, and support programs for ex-prisoners that intend to reduce their risk of reincarceration. A common recommendation is that programs should be individually tailored to address criminogenic needs. From arrest through to release from prison, the criminal justice system's response to offending behaviour, and society's, is focused almost exclusively on the individual, implicitly characterising crime as an individual-level problem, and thus failing to acknowledge the wider structural and socio-ecological drivers of offending behaviour. From this hyper-individualistic perspective, it is argued that the individual simply chooses to commit a crime, and as such should be held accountable and made to take responsibility for their behaviour. However, choices are not made in a vacuum, but rather influenced by wider social and structural mechanisms. Crime is generally the by-product of larger social problems that exist beyond the individual; therefore, not all blame can, or should, be placed on the offender (Bernie & Messerschmidt, 1995; Rose & Clear, 1998; Sampson & Laub, 1993). A major element of these factors is a person's environment in which they live and interact. A person's environment can provide or constrain opportunities for development and social progression due to community level social disadvantage (i.e. the comparative lack of social and economic resources; Wokstrom & Treiber, 2016) as well as a person's geographical location (i.e. city/urban vs. rural/remote; Jobes et al., 2004; McCausland & Vivian, 2010). For these reasons, we need to understand more about the ecological mechanisms that influence Indigenous and non-Indigenous prison re-entry if we desire effective treatment, reintegration, and support programs for ex-prisoners.

Australia is the sixth-largest country by landmass, but only has a national population of approximately 25 million people (ABS, 2018). Given Australia's population is widely dispersed across both disadvantaged and affluent communities in city/urban areas, and rural, remote, and very remote areas, Australia's communities are quite diverse in structure and disadvantage and vary greatly in how offending behaviour is influenced by these differences (Allard, 2010; Allard et al., 2012; Cunningham, 2007; Jobes et al., 2004). Studies that have examined the reoffending-ecological relationship have found (a) mixed results when examining the effect of community disadvantage (Huebner et al., 2007; Grunwald et al., 2010; Kubrin & Stewart, 2006; Kubrin et al., 2007; Wehrman, 2010) and (b) that geographical location is important for successful re-entry due to community differences in accessibility and availability of services, treatments, and social support networks, all of which are important to the re-entry process (Hipp et al., 2009, 2011; Kubrin & Stewart, 2006; Rose & Clear, 1998). In Australia, there is a lack of ecological research that focuses on the risk of reincarceration and even less on Indigenous people. Consequently, what we know about ecological risk factors is predominantly informed by research using a theoretical framework of social disorganisation that has largely been informed by a white urbanised American perspective.

Using this perspective to explain differences in Indigenous and non-Indigenous peoples’ risk of reincarceration is problematic due to important historical, political, cultural, social, and geographical differences between the United States and Australia. For example, it is argued that Indigenous people who reside in rural remote communities ‘cannot be said to have full civil rights’ due to their inadequate access to services (Cunneen et al., 2014), and that geographical location is not just a problem for very remote communities, but also for rural communities in Australia as services and funding for services are predominantly focused on capital cities (Law Council of Australia, 2018).¹ Furthermore, research examining the geographical distribution of chronic and costly offenders in Queensland Australia, found that the distribution of these offenders in the community was not random, and the characteristics of the communities where chronic and costly offenders were most likely to reside included high levels of disadvantage, remoteness, and high proportions of Indigenous youth (Allard et al., 2012). As such, identifying and understanding how one's environment affects a person's risk of reincarceration is critical for legislators, policy, and program makers, especially when it comes to reducing Indigenous over-representation in prison, and the overall prison population.

Community disadvantage

Resource deprivation, otherwise known as a concentrated disadvantage or community disadvantage, has been consistently found to have a strong association with criminality in ecological studies (Eitle et al., 2006; Huebner et al., 2007; Kubrin & Stewart, 2006; Land et al., 1990; Sampson et al., 2002; Shaw & McKay, 1942; Stahler et al., 2013). However, some authors have suggested that these studies often place too much emphasis on concentrated disadvantage and neglect the role of affluence as a protective factor (Sampson et al., 2002). Ecological studies should examine both ends of the disadvantage spectrum for communities to identify predictors of crime.

Prior ecological studies that have examined the disadvantage–crime relationship have reported mixed results. Kubrin and Stewart (2006) followed a group of 4630 ex-prisoners for 12 months after release from prison and found that those who returned to communities characterised by high levels of concentrated disadvantage were significantly more likely to be re-arrested within the follow-up period, compared with those who returned to communities characterised by lower levels of concentrated disadvantage (Kubrin & Stewart, 2006). However, Kubrin and Stewart (2006) also conducted a comparative study that examined the relationship between disadvantaged and affluent communities and recidivism and found that disadvantage increased the risk of recidivism in affluent communities by less than five per cent, and in other communities by more than 40%. These findings indicate that individuals from disadvantaged communities were more likely to recidivaite, compared to those who reside in more affluent communities, thus highlighting the importance of differentiating between affluent and disadvantaged communities when examining reoffending/reincarceration risks.

However, research by Huebner et al. (2007) found that concentrated disadvantage was not associated with an ex-prisoner's probability of reconviction. These findings are supported by those from Wehrman (2010) who examined the effects of disadvantage, and the interactive effect of disadvantage and race on an individual's probability to re-offend, with no significant effect being found for either. Nevertheless, other research has found a significant relationship between disadvantage and reoffending for a specific type of offender only. Grunwald et al. (2010) examined the effect of concentrated disadvantage and social capital on the probability of reoffending and found that when controlling for social capital, the concentrated disadvantage was a significant predictor of reoffending for repeat drug offenders only. These mixed results suggest that a community's level of disadvantage may have a direct or indirect effect on the risk of reincarceration, and therefore warrants further investigation.

Residential geographical location

The geographical location of an individual recently released from prison may influence their ability to successfully reintegrate post-release (Kubrin & Stewart, 2006; Rose & Clear, 1998), in part due to issues of accessibility to services, treatment, and support, both at an institutional and community level. Where people reside after release from prison (i.e. city, urban, rural, remote, and very remote) may directly affect their ability to access important services and treatment options and may strengthen or hinder social support networks important to facilitating a successful re-entry (Hipp et al., 2009, 2011; Kubrin & Stewart, 2006; Rose & Clear, 1998). Those who reside in remote communities are likely to have reduced access to services, treatment options and support, and as such may be at increased risk of reincarceration (Allard et al., 2012).

Few studies have examined the effect of an individual's residential geographical location on their probability of incarceration and even fewer have examined whether this has an impact on re-entry. Cunningham (2007) examined the effect of the geographical location of the residence on reoffending in the Northern Territory (NT) in Australia. Of the 3597 young offenders included in the evaluation of a pre-court diversion scheme, 54% lived in either a regional centre or a remote Indigenous community (Cunningham, 2007). Cunningham (2007) found that individuals who resided in areas with high accessibility (21%), compared to medium accessibility (26%), and low accessibility (28%) to services, treatment, and social support networks were significantly less likely to re-offend.

These findings are supported by Hipp et al. (2009, 2011) who examined the effects of proximity to social and health service providers for parolees in California. Their findings showed that more social and health services were located in communities where Hispanic and black parolees resided. However, even though more services in these areas existed, accessibility was still limited compared to other areas (Hipp et al., 2009, 2011). In an attempt to explain this apparently paradoxical finding, Hipp et al. (2009, 2011) suggested that demand for services may be higher in areas with a greater concentration of ex-prisoners. However, this contradicts earlier findings by Small and McDermott (2006) who found that areas characterised by disadvantage, have fewer organisational resources than do affluent areas This is important to consider in an Australian context considering that areas that are geographically isolated are also more often characterised by extreme levels of disadvantage (Weatherburn & Lind, 2001).

Scholarly work examining the effects of geographical factors on the risk of reincarceration after release from prison for Indigenous and non-Indigenous people is rare. With studies such as Cunningham's (2007) that highlight the relationship between an individual's residential geographical location and reoffending amongst young people, it would be reasonable to suspect that the same relationship may exist for adults released from prison. Furthermore, with studies such as Hipp et al. (2009, 2011), and Small and McDermott (2006), who both demonstrate that a relationship exists between accessibility to services and crime, this is an area that requires further investigation. Accordingly, in a large cohort of adults released from prisons in Queensland, Australia, we sought to answer the following questions:

Is risk of reincarceration for Indigenous and non-Indigenous people who reside in city/urban communities different to those who reside in rural/remote/very remote communities?

When controlling for demographics and prior criminal history, do ecological risk factors such as community disadvantage and remoteness increase or decrease Indigenous and non-Indigenous peoples’ risk of reincarceration?

Methods

Data sources

The data utilised in this study were drawn from two sources. First, survey data from a larger research project called the Passport study were linked with administrative data from Queensland Corrective Services (QCS).² Surveys were conducted between 2008 and 2010 with 1325 incarcerated adults within six weeks of anticipated release from custody (i.e. baseline), and at approximately 1, 3, and 6 months post-release. For the purposes of this study, we have only drawn on baseline survey data. Surveys typically took between 60 and 90 min to complete. Prison records were used to identify potential participants. Only sentenced prisoners were eligible for inclusion, due to the uncertainty of release dates for prisoners held on remand (Kinner et al., 2013). The baseline survey collected self-reported information regarding participants’ demographic characteristics, and prior criminal history (i.e. prior juvenile detention and most recent offence imprisoned for). QCS data included information on participants’ prior adult imprisonment dating back to 2006, age at index release from prison, and dates of reincarceration for new offences and parole breaches until 31 December 2013.

Second, using participants’ self-reported postcodes of the community they expected to return to, we accessed population data from the 2006 Australian Census to determine participants’ community measures of disadvantage and remoteness. For this study, participants who were released more than eight weeks after completing a baseline survey, and who we had any missing administrative data, were excluded from the study. In total, 1238 participants were included in the analyses presented here.

Variables

All control variables included in the study have empirical and/or theoretical importance to understanding the role that an individual's community plays in their risk of reincarceration post-release. Using best practice procedures for defining Indigenous status (i.e. Aboriginal, Torres Strait Islander, and/or South Sea Islander, vs. not³), participants who self-identified as belonging to at least one of these groups in either the baseline survey or in QCS records were defined as Indigenous (Dugbaza et al., 2012). Besides Indigenous status, other control variables that we included in our study were common demographics, criminal history, community disadvantage, and remoteness. Table 1 provides a full description and statistics on missing data for the common demographics and participants’ criminal history. The operationalisation of the ecological factors is explained below.

Table 1.

Description of variables and number of participants with missing data.

Variable	Description (number with missing data out of full sample)
Demographic
Gender	Identified as male (0). No = 0; Yes = 1
Age	Age at time of release. QCS data (0).
Criminal history
Adult imprisonment	Number of prior adult prison sentences categorised as; first time prisoner (no prior adult prison sentence) = 0; one to four prior adult prison sentences = 1; 5 or more prior adult prison sentences = 2 (QCS data)
Violent offence	Imprisoned for a violent offence (0). No = 0; Yes = 1 (QCS data)
Property offence	Imprisoned for a property offence (0). No = 0; Yes = 1 (QCS data)
Drug offence	Imprisoned for a drug offence (0). No = 0; Yes = 1 (QCS data)
Traffic/public orderoffence	Imprisoned for a traffic public order offence (0). No = 0; Yes = 1 (QCS data)
Juvenile detention	Number of prior juvenile detentions (7) categorised as; no prior juvenile detentions = 0; 1–4 juvenile detentions = 1; 5 or more juvenile detentions = 2 (QCS data)
Ecological factors
Community disadvantage	Using SEIFAs IRSD index (ABS, 2008) level of community disadvantage categorized as; low level of disadvantage = 0; moderate level of disadvantage = 1; high level of disadvantage = 2. (10)
Remoteness	Using the ARIA remoteness categorised as; City/Urban = 0; Rural/Remote/Very Remote = 1.

ARIA: Accessibility/Remoteness Index of Australia; QCS: Queensland Corrective Services; IRSD: Index of Relative Socio-Economic Disadvantage; SEIFA: socio-economic indexes for areas.

Community disadvantage

People who reside in areas of high disadvantage are considered to have limited access to personal and social resources, thus limiting their ability to actively participate in society (ABS, 2008; Weatherburn & Lind, 2001). A key concern for providing services to communities is, how best to access and service areas of disadvantage. Using the Index of Relative Socio-Economic Disadvantage (IRSD) and participants’ self-reported postcodes of expected residence, we were able to determine whether the community to which participants returned post-release was an area of high, moderate, or low disadvantage. In Australia, between the 2006 and 2011 census, the IRSD was the preferred way to measure a broad range of disadvantages (i.e. social disadvantage) instead of a specific measure of disadvantage (i.e. a person's income status) when examining communities’ level of disadvantage (ABS, 2008).⁴ The 2006 IRSD was calculated using a combination of census population variables that were known to indicate disadvantage (ABS, 2008). A full list of the variables used by the Australian Bureau of Statistics to calculate the IRSD is provided in Table 2. The decile scoring method of the IRSD was used to assign a level of disadvantage to each participant's home community. A decile score of 1 to 10 was given for each community (ABS, 2008). A community with a score of 1 is ranked in the bottom 10% of communities for IRSD, thus indicating high levels of disadvantage. In contrast, communities with a score of 10 are ranked in the top 10% of communities for IRSD, indicating low levels of disadvantage (ABS, 2008). For the analyses presented here, this measure was collapsed into a categorical variable with the following categories: (a) high disadvantage = 2 (i.e. communities with a score of 1 to 3); (b) moderate disadvantage = 1 (i.e. communities with a score of 4 to 7); (c) low disadvantage = 0 (communities with a score of 8 to 10).

Table 2.

Variables from the 2006 Australian census were used to calculate the Index of Relative Socio-Economic Disadvantage (IRSD).

Variables included in the IRSD
% of occupied private dwellings with no internet connection
% of employed people classified as Labourers
% of people aged 15 years and over with no post-school qualifications
% of people with stated annual household equivalised income between $13,000 and $20,799
% of households renting from Government or Community organisations
% of people (in the labour force) unemployed
% of one parent families with dependent offspring only
% of Households paying rent less than $120.00 per week (excluding $0 per week)
% of people under the age of 70 who have a long-term health condition/disability and need assistance with core activities
% of occupied private dwellings with no car
% of people who identified themselves as being of Aboriginal and/or Torres Strait Islander origin
% of occupied private dwellings requiring one or more extra bedrooms (based on Canadian National Occupancy Standard)
% of people aged 15 years and over who are separated or divorced
% of employed people classified as machinery operators and drivers
% of people aged 15 years and over who did not go to school
% of employed people classified as Low Skill Community and Personal Service Workers
% of people who do not speak English well

Source: Information paper: An introduction to socio-economic indexes for areas (SEIFA), 2006: ABS.

Remoteness

To calculate the level of remoteness of participants’ community to which they returned we used their postcode to assign a score of remoteness as per the Accessibility/Remoteness Index of Australia (ARIA) (Trewin, 2006). The ARIA measures how remote a person's community is based on the distance of road travel between their community and the nearest town centre (Trewin, 2006). As such, we used road distance as a proxy measure of remoteness, while the population size of the town centre is a proxy that is used for ‘choice’ and ‘availability’ of services (i.e. the larger the town population the more services and variety of services are believed to be available) (Trewin, 2006). Each area assigned to a postcode received a score between 0 and 15, with 15 labelled as very remote, and 0 labelled as highly accessible. Next, we collapsed this measure into a dichotomous variable that was coded City/Urban = 0 and Rural/Remote/Very Remote = 1.

Reincarceration

Using administrative data from QCS we were able to measure participants’ time to reincarceration over the maximum five year follow-up period (i.e. 2008–2013). Reincarceration was operationalised as occurring when a participant was returned to prison for a new criminal offence or a parole breach before the end of the follow-up period. Reincarceration as an outcome measure in re-entry research as opposed to rearrest is recommended to avoid problems of overestimating recidivism, thus reducing the probability of a type one error (Ulmer, 2001). It is a conservative measure of recidivism and has been employed in prior research (Payne, 2007; Ryan et al., 2019; Stahler et al., 2013).

Analyses and results

First, descriptive analyses were conducted for each variable included in the study using Indigenous status as a grouping variable (see Table 1). Of the 1238 people included in the study, 935 (76%) were non-Indigenous and 303 (24%) were Indigenous. Of the 935 non-Indigenous people, 816 (87%) resided in city/urban communities and 114 (13%) lived in rural/remote communities. Whereas, of the 303 Indigenous people, 179 (60%) were from city/urban communities, and 121 (40%) resided in rural/remote communities. Of the four groups, 72.6% of Indigenous people from city/urban communities were reincarcerated within five years compared to, 69.4% of Indigenous people from rural/remote communities, 49.1% of non-Indigenous rural/remote people, and 47.9% of non-Indigenous city/urban people.

Next, we examined the time to reincarceration for the four groups. Interestingly, we found that individuals who resided in rural/remote communities took longer to be returned to prison than those who lived in city/urban communities. Results indicated that for those who were reincarcerated, non-Indigenous people from urban communities returned to prison the fastest (M = 372.28 days: standard deviation (SD) = 363.94), followed by Indigenous people from urban communities (M = 387.15 days: SD = 383.94). Of those who were from rural/remote communities’ non-Indigenous people were the next fastest to be reincarcerated (M = 401.61 days: SD = 407.96). Finally, Indigenous people from rural/remote communities who were reincarcerated took the longest amount of time to return to prison (M = 434.45 days: SD = 368.82).

To understand whether Indigenous people in urban and rural/remote communities are more at risk of reincarceration than non-Indigenous people in urban and rural/remote communities the data was split by remoteness and a bivariate Cox proportional regression analysis was conducted using Indigenous status as the independent variable. Results identified that Indigenous compared to non-Indigenous people were significantly more at risk of reincarceration in both the urban and remote/rural groups. Indigenous urban people were 86% more at risk of reincarceration than non-Indigenous urban people, and rural/remote Indigenous people had a 66% increased risk of reincarceration post-release compared to rural/remote non-Indigenous people (see Table 3). To examine whether there was a statistical difference between the two groups we conducted a Kaplan Meier survival analysis with a Log Rank test (see Figure 1). The results of the Log Rank test indicate a significant difference in the risk of reincarceration between the two groups (p≤.001). To begin to understand why there is a significant difference in risk of reincarceration between the two groups it is important to investigate this further by examining whether differences in risk of reincarceration for Indigenous and non-Indigenous people can be partially explained by their demographic, criminal history, and ecological risk factors. To do this we next conducted two multivariate Cox proportional hazard regression models to examine Indigenous and non-Indigenous peoples’ risk factors (i.e. demographics, criminal history, and ecological factors) for reincarceration.

Figure 1.

Indigenous and non-Indigenous peoples’ hazard risk for reincarceration by urban and rural/remote location post-release.

Table 3.

Risk of reincarceration by Indigenous status and remoteness.

			95% CI
Group	Variable	HR	LL	UL	p
Urban	Indigenous	1.862	1.526	2.272	≤.001
Rural/remote	Indigenous	1.659	1.181	2.330	.004

CI: confidence interval; HR: hazard ratio; LL: lower level; UL: upper level.

We identified four significant risk factors of reincarceration for non-Indigenous people, and three for Indigenous people (see Table 4). For non-Indigenous people, prior adult imprisonment of one to four times (hazard ratio (HR) = 2.12, 95% C.I: 1.66, 2.72, p<.001), prior adult imprisonment 5 or more times (HR = 3.10, 95% confidence interval (CI): 2.35, 4.09, p<.001), being most recently incarcerated for a violent offence (HR = 1.34, 95% CI: 1.04, 1.73, p<.027), and being most recently imprisoned for a property offence (HR = 1.36, 95% CI: 1.03, 1.80, p<.032) were all associated with a significant increase in risk of reincarceration. For Indigenous people, being male (HR = 1.87, 95% CI: 1.33, 2.63, p<.001), prior adult imprisonment of five or more times (HR = 1.89, 95% CI: 1.27, 2.82, p<.002), and prior juvenile detention of five or more times (HR = 2.07, 95% CI: 1.37, 3.12, p<.001) were all associated with a significant increase in risk of reincarceration.

Table 4.

Reincarceration hazard by Indigenous status (multivariate).

	Non-Indigenous (n = 935)				Indigenous(n = 303)
	95% CI				95% CI
Variables	HR	LL	UL	P	HR	LL	UL	P
Gender (male)	.97	.75	1.26	.813	1.87	1.33	2.63	.000
Age	.96	.95	.97	.000	.97	.95	.99	.000
Prior adult imprisonment (first imprisonment)				.000				.007
1–4 prior imprisonments	2.12	1.66	2.72	.000	1.45	.97	2.15	.069
5+ prior imprisonments	3.10	2.35	4.09	.000	1.89	1.27	2.82	.002
Violent offence	1.34	1.04	1.73	.027	.98	.63	1.51	.919
Property offence	1.36	1.03	1.80	.032	1.15	.70	1.88	.581
Drug offence	.83	.50	1.38	.463	.76	.23	2.57	.664
Juvenile detention (none)				.378				.002
1–4 prior juvenile detention	1.19	.93	1.52	.165	1.14	.83	1.56	.420
5+ prior juvenile detention	1.09	.76	1.57	.642	2.07	1.37	3.12	.001
Community disadvantage (low)				.405				.028
High disadvantage	1.19	.92	1.53	.190	.59	.38	.90	.014
Moderate disadvantage	1.08	.84	1.38	.539	.56	.36	.88	.011
Remoteness	1.07	.80	1.42	.670	.93	.69	1.25	.612

The bold values indicates statistically significant at .05 or below.

CI: confidence interval; HR, hazard ratio; LL: lower level; UL: upper level.

We also identified one protective factor against reincarceration for non-Indigenous and two protective factors for Indigenous people. Older age at release from prison reduced the risk of reincarceration for both non-Indigenous people (HR = 0.96, 95% CI: 0.95, 0.97, p<.001) and Indigenous people (HR = 0.97, 95% CI: 0.95, 0.99, p<.001). To our surprise, high community disadvantage (HR = 0.59, 95% CI: 0.38, 0.90, p<.014), and moderate community disadvantage (HR = 0.56, 95% CI: 0.36, 0.88, p<.011) were all associated with a significant reduction in risk of reincarceration for Indigenous people only (see Table 4).

Discussion

The aim of our research was to identify if Indigenous and non-Indigenous peoples’ risk of reincarceration differed for those who resided in city/urban communities compared to rural/remote communities and to examine the effects that ecological factors such as community disadvantage and remoteness have on Indigenous and non-Indigenous peoples’ risk of reincarceration after controlling for demographics and prior criminal history.

As expected, Indigenous people who resided in city/urban and rural/remote communities had a significantly higher risk of reincarceration post-release than non-Indigenous people who lived in city/urban and rural/remote communities. However, to our surprise, after we controlled for demographics and criminal history we examined whether community disadvantage effected Indigenous and non-Indigenous peoples’ risk of reincarceration and despite previous studies finding a relationship between communities characterised by high disadvantage and reoffending/reincarceration, we found no association between communities with moderate to high disadvantage and reincarceration for non-Indigenous people. Instead, a negative association between communities with moderate to high disadvantage and reincarceration was identified for Indigenous people. Specifically, results indicated that Indigenous people who resided in highly disadvantaged communities had a 39% reduced risk of reincarceration, and those who resided in a moderately disadvantaged community had a 43% reduced risk of reincarceration. These findings suggest that community disadvantage may be a protective factor for Indigenous people which goes against all literature. To make any sense of these findings the operationalisation of community disadvantage that we employed in this study needs to be scrutinised in conjunction with findings from the social disorganisation literature.

Communities may share common characteristics. Specifically, Indigenous communities have shared historical experiences stemming from colonisation, enforced government policies, culture, and social structure (i.e. kinship) that shape and influence Indigenous peoples’ relationships and interactions in ways that unique to Indigenous people. While communities that have high levels of disadvantage and racial heterogeneity are considered to result in ‘disorganised’ communities that have minimal social interactions, low levels of trust and limited collective efficacy due to a lack of shared values and beliefs resulting in a community's inability to enact informal social controls that prevent the formation of sub-cultural antisocial behaviour (Bursik & Grasmick, 1993; Osgood & Chambers, 2000; Pratt & Cullen, 2005; Sampson & Groves, 1989; Sampson et al., 1997), Australian rural research that examined two Indigenous communities highlighted that community social factors were of more importance to understanding Indigenous offending behaviour than socio-economic factors (Jobes et al., 2004, 2005). Specifically, Jobes et al. (2004) found that the presence of strong extended family social groups and dense acquaintanceship (i.e. kinship) were strong predictors of informal relationships that positively influenced community members’ behaviour, thus preventing the formation of antisocial tendencies and the stigmatisation of offenders.

Considering communities with a dense acquaintanceship (i.e. kinship) has a greater capacity to foster informal social control and thus prevent crime (Jobes et al., 2004, 2005), it may be possible that the measurement of community disadvantage employed in this study (i.e. the IRSD index) is not measuring disadvantage for Indigenous people, but the presence of a dense acquaintanceship in their community. One of the measures taken from the census population data that the Australian Bureau of Statistics used to create the IRSD scale is the total percentage of people who reside in the community that identifies as either Aboriginal and/or Torres Strait Islander (ABS, 2008). Thereby, employing a scale that is calculated using a total measure of Indigenous people – measure of ‘dense acquaintanceship’ – the measure of community disadvantage employed may not in fact be measuring disadvantage when applied to Indigenous people, but may be measuring the presence of kinship.

We do not claim to understand the importance of the kinship system among Indigenous communities and cannot speak of the lived experiences of Indigenous people. The kinship system found in Indigenous communities is not well understood by people outside the kinship system and is a system that needs to be examined and not ignored by researchers to understand whether it is a protective factor against crime for Indigenous people. It should be noted that it has been reported that the kinship system found in Indigenous communities is considered to be central to the identity of what it is to be Indigenous (Macdonald, 1986, 2011; Peterson, 2010; Peterson & Taylor, 2003; Sullivan, 2012), thus making it important for policy, practice and legislation that we develop a better understanding of the mechanisms of the kinship system, and whether these can help reduce an Indigenous person's risk of reincarceration. The kinship system has been described as a system of embedded dense sociality deeply rooted in a relational ontology that acknowledges a connection between all people and the world around them (Kelly, 2014; Peterson, 2010; Warburton & Chambers, 2007). Furthermore, it is reported that the kinship system is a powerful governing structure that is founded on thousands of years of Indigenous culture and customary laws that outlines one's obligations and expectations that are central to Indigenous identity and knowing one's place in the community (Kelly, 2014; Northern Territory Law Reform Committee, 2003; Peterson, 2010; Warburton & Chambers, 2007).

Furthermore, contrary to previous findings in the literature (Cunningham, 2007), after we controlled for demographics, criminal history, and community disadvantage no relationship between remoteness and reincarceration was detected for both Indigenous and non-Indigenous people. This is a surprising finding considering the importance that treatment services and social support services are believed to play in a person's successful reintegration from prison to the community (Willis, 2008), and the belief that people who reside in rural and remote communities may be disadvantaged compared to those who reside in city/urban communities due to having limited or no access to, and limited or no choice of treatment services and social support services (Little et al., 2018; McLachlan et al., 2013; Weatherburn, 2014). It is interesting to note that even though no significant relationship was found, the HR is in the expected direction for non-Indigenous people (i.e. living in rural/remote communities would increase the risk of reincarceration), whereas for Indigenous people, the HR is in the opposite direction (i.e. living in rural/remote communities would decrease the risk of reincarceration) to what would be expected considering previous research. Given the bivariate findings suggested there was a relationship between Indigenous status, remoteness and reincarceration, it may be possible that this relationship is being masked by one or multiple other risk factors (i.e. demographics, prior criminal history, community disadvantage, and accessed services) that were controlled for in the multivariate model.

Implications and future research

Considering our findings raise some concerning questions about the IRSD scale's reliability, especially in relation to measuring disadvantages for Indigenous people, this potentially has some serious implications for policy and practice. The IRSD scale is a commonly used scale in research and government reports that are believed to summarise information about the economic and social conditions of people who reside in an area so a measure of relative disadvantage can be determined (ABS, 2006). Decisions to provide programs, resources, and funding to areas are informed by these measures, but what if the measures of relative disadvantage are not reflecting a community's true measure of disadvantage? This might result in communities missing out on much-needed funding, resources, and programs because the IRSD scale being relied on is not accurately capturing the community's level of disadvantage. This was a major concern expressed by state governments in a 2015 Commonwealth Grants Commission (CGC) working paper that was looking at appropriate ways to assess Australia's Indigenous population (CGC, 2015). Specifically, the Victorian, Queensland and Tasmanian State Governments stated that it would be ‘very problematic’ to assume that an Indigenous community in one state (i.e. Victoria) that has a decile measure of 1 (i.e. indicating high levels of disadvantage) and an Indigenous community in another state (i.e. Queensland) that has the same decile measure of 1 experience the same level of disadvantage. The problem of extrapolating aggregate measures of community disadvantage becomes more complex when comparing urban communities with rural/remote communities.

Besides the question of whether the IRSD scale is measuring disadvantage or the presence of dense acquaintanceship for Indigenous people, the question must be asked as to whether the IRSD scale can accurately measure the relative disadvantage that is experienced by rural/remote communities when compared to urban/city communities. Currently, the way the IRSD scale is calculated very remote Indigenous communities in outback Western Australia, the NT and Queensland's Cape York Region receive the same measurement of a disadvantage as do some city suburban communities such as Inala, and Acacia Ridge in Brisbane Queensland, and Redfern in Sydney, New South Wales. Employing a scale that measures the level of disadvantage in these communities to be the same is dismissive of the highly concentrated levels of the disadvantage of Indigenous and non-Indigenous people experience who reside in Australia's very remote communities.

Finally, we must consider whether the IRSD scale can measure disadvantages for Indigenous people. In general, Australia's Indigenous people experience a greater degree of a socio-economic disadvantage than non-Indigenous people (Biddle, 2011). Whilst all the variables that are included in the calculation of the IRSD scale have been informed by previous literature that examines factors of disadvantage, it may be possible that there are variables included in the scale that may not be as relevant or have a different meaning for Indigenous people compared to non-Indigenous people (CGC, 2015). As such, CAEPR designed an Indigenous Relative Socio-economic Outcomes (IRSEO) Index to measure socio-economic disadvantage specifically for Indigenous communities (Biddle, 2011). However, state governments have expressed concerns regarding the use of the IRSEO and the potential for policy contamination given, at the time, almost half (approximately 48%) of the Indigenous population in the bottom quintile for disadvantage lived in the NT (CGC, 2015). It is hoped that our findings further advance the conversation about how we measure community disadvantage in Australia, and our standardised approach to measuring disadvantage truly reflects disadvantage for Indigenous people and people who live in rural remote communities.

Therefore, we see three areas for future research. First, given that we have raised a question regarding the reliability of the IRSD scale when being used with Indigenous people, future research may wish to examine the validity of the IRSD scale when measuring disadvantages for Indigenous people. This we feel is critically important considering how often the IRSD scale is used in research and government reports, as well as by the CGC, an Australian government department that is responsible for providing recommendations on the distribution of Goods and Services Tax revenue in Australia (CGC, 2021). Second, future research might wish to explore the difference in disadvantage between urban/city communities and rural/remote communities to develop a more accurate measure of disadvantage that better reflects the level of a concentrated disadvantage being experienced in some communities. Finally, to better understand what disadvantage is for Indigenous people, whether it is different to non-Indigenous people, and how disadvantage effects their risk of reincarceration, future researchers need to talk with Indigenous people from city/urban and rural/remote communities and learn about their lived experiences with disadvantage to determine whether the standardised measure of disadvantage accurately captures what disadvantage is and the level of disadvantage that is being experienced by Indigenous people.

Conclusion

As far as we know, very few studies have examined whether differences in ecological factors such as community disadvantage and remoteness (i.e. geographical location) effect Indigenous and non-Indigenous peoples’ risk of reincarceration. When we began this study, we expected to find similar results to previous studies that examined disadvantages, however, to our surprise this was not the case. Considering, we found an interactive effect between community disadvantage, Indigenous status and risk of reincarceration, our findings provide further evidence that when designing and implementing an intervention, treatment and re-entry programs for ex-prisoners, it is not enough to tailor these programs to focus on the individual offender, their immediate family, and their social environment in a culturally appropriate manner, they also need to consider an individual's local environmental needs. As such, we arguably need to move away from programs and offender management strategies that adopt a one-size-fits-all approach that makes a basic assumption that all Indigenous and non-Indigenous people are the same, regardless of contextual factors, to a more tailored approach that considers the individual, family, social, local environment and cultural needs of a person, and that they are designed and implemented in conjunction with grass-root level organisations and/or groups from prisoners’ communities to which they will be returning post-release.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Health and Medical Research Council (NHMRC grant numbers ACPP409966 and APP1078168).

ORCID iD

Nicole R Ryan

Notes

References

Allard

(2010, December). Understanding and preventing Indigenous offending. Indigenous Justice Clearinghouse. Brief No. 9. https://apo.org.au/sites/default/files/resource-files/2011-01/apo-nid23648.pdf

Allard

Chrzanowski

Stewart

(2012). Targeting crime prevention to reduce offending: Identifying communities that generate chronic and costly offenders. Trends and Issues in Crime and Criminal Justice, 445, 1–8. Canberra: Australian Institute of Criminology. https://www.aic.gov.au/publications/tandi/tandi445

Australian Bureau of Statistics (2008). Socio-economic indexes for areas (SEIFA): Technical paper. ABS. Cat. No. 2039.0.55.001). Canberra: ABS.

Australian Bureau of Statistics (2016). Census of population and housing: Socio-economic indexes for areas (SEIFA), Australia, 2016. Cat. No. 2033.0.55.001. Canberra: ABS.

Australian Bureau of Statistics (2018). Corrective services, Australia, June quarter 2018. Cat. No. 4512.0. Canberra: ABS.

ABS (2020). Prisoners in Australia, 2020. Cat. No. 4517.0. Canberra: ABS.

Bernie

Messerschmidt

(1995). Criminology. Harcourt Brace.

Biddle

(2011). CAEPR indigenous population project, 2011 Census papers. Paper 13, Socioeconomic outcomes. Centre for Aboriginal Economic Policy Research, Australian National University (ANU). https://caepr.cass.anu.edu.au/sites/default/files/docs/2011CensusPaper_13_Socioeconomic_Outcomes_1.pdf

Bursik

R. J.

Jr Grasmick

H. G.

(1993). Economic deprivation and neighborhood crime rates, 1960–1980. Law & Society Review, 27(2), 263. https://doi.org/10.2307/3053937

10.

Commonwealth Grants Commission (CGC) (2015). Appropriately assessing the indigenous population. Staff working paper, CJC. Australian Federal Government, Canberra. Retrieved from: https://www.cgc.gov.au/sites/default/files/2021-11/R2015%20-%20Appropriately%20Assessing%20the%20Indigenous%20Population.pdf

11.

Cunneen

Allison

Schwartz

(2014). Access to justice for aboriginal people in the northern territory. Australian Journal of Social Issues, 49(2), 219–240. https://doi.org/10.1002/j.1839-4655.2014.tb00309.x

12.

Cunningham

(2007). Pre-court diversion in the Northern Territory: Impact on juvenile reoffending. Trends & Issues in Crime and Criminal Justice, No. 339.

13.

Dugbaza

Scott

McKeawn

(2012). National best practice guidelines for Indigenous data linkage activities relating to Aboriginal and Torres Strait Islander People. AIHW.

14.

Eitle

D’Alessio

S. J.

Stolzenberg

(2006). Economic segregation, race, and homicide. Social Science Quarterly, 87(3), 638–657. https://doi.org/10.1111/j.1540-6237.2006.00401.x

15.

Giles

Whale

(2016). Models of the impacts of prison education and training on welfare dependence and recidivism: Western Australia. Report to Australian Institute of Criminology and Phase 3 Report. Centre for Innovative Practice, Edith Cowan University.

16.

Grunwald

Lockwood

Harris

Mennis

(2010). Influences of neighborhood context, individual history and parenting behavior on recidivism among juvenile offenders. Journal of Youth and Adolescence, 39, 1067–1079. https://doi.org/10.1007/s10964-010-9518-5

17.

Hipp

Jannetta

Shah

Turner

(2009). Parolees’ physical closeness to health service providers: A study of California parolees. Health & Place, 15(3), 679–688. https://doi.org/10.1016/j.healthplace.2008.10.006

18.

Hipp

Jannetta

Shah

Turner

(2011). Parolees’ physical closeness to social services: A study of California parolees. Crime & Delinquency, 57(1), 102–129. https://doi.org/10.1177/0011128708322856

19.

Huebner

B. M.

Varano

S. P.

Bynum

T. S.

(2007). Gangs, guns, and drugs: Recidivsm among serious offenders. Criminology and Public Policy, 6(2), 187–221. https://doi.org/10.1111/j.1745-9133.2007.00429.x

20.

Jobes

P. C.

Barclay

Weinand

Donnermeyer

J. F.

(2004). A structural analysis of social disorganization and crime in rural communities in Australia. Australian & New Zealand Journal of Criminology, 37(1), 114–140. https://doi.org/10.1375/acri.37.1.114

21.

Jobes

P. C.

Donnermeyer

J. F.

Barclay

(2005). A tale of two towns: Social structure, integration and crime in rural New South Wales. Sociologia Ruralis, 45(3), 224–244. https://doi.org/10.1111/j.1467-9523.2005.00302.x

22.

Kelly

(2014). The intersection of Aboriginal customary law with the NT criminal justice system: The road not taken? Conference presentation at NTBA conference, Dili. https://scholar.google.com.au/scholar?hl=en&as_sdt=0%2C5&q=The+intersection+of+Aboriginal+customary+law+with+the+NT+criminal+justice+system%3A+the+road+not+taken%3F++&btnG=#d=gs_cit&u=%2Fscholar%3Fq%3Dinfo%3AMeO0lF0e0AoJ%3Ascholar.google.com%2F%26output%3Dcite%26scirp%3D0%26hl%3Den

23.

Kinner

S. A.

Lennox

Williams

G. W.

Carrol

Quinn

Boyle

Alati

(2013). Randomised controlled trial of a service brokerage intervention for ex-prisoners in Australia. Contemporary Clinical Trials, 36(1), 198–206. https://doi.org/10.1016/jcct.2013.07.001.

24.

Kubrin

C. E.

Squires

G. D.

Stewart

E. A.

(2007). Neighborhoods, race, and recidivism: The community-reoffending nexus and its implications for African-Americans. SAGE Race Relations Abstracts, 32, 7–37. https://doi.org/10.1177/0307920107073250

25.

Kubrin

C. E.

Stewart

E. A.

(2006). Predicting who reoffends: The neglected role of neighborhood context in recidivism studies. Criminology, 44, 165–197. https://doi.org/10.1111/j.1745-9125.2006.00046.x

26.

Land

K. C.

McCall

P. L.

Cohen

L. E.

(1990). Structural covariate of homicide rates: Are there invariances across time and social space? American Journal of Sociology, 95, 922–963. https://doi.org/10.1086/229381

27.

Law Council of Australia (2018, August). The Justice Project Final Report – Part 1: Aboriginal and Torres Strait Islander People. https://www.lawcouncil.asn.au/files/web-pdf/Justice%20Project/Final%20Report/Aboriginal%20and%20Torres%20Strait%20Islander%20People%20%28Part%201%29.pdf

28.

Little

Stewart

Ryan

. (2018). Restorative justice conferencing: Not a panacea for the over-representation of Australia’s Indigenous youth in the criminal justice system. International Journal of Offender Therapy & Comparative Criminology, 1–24. https://doi.org/10.1177/0306624X18764524

29.

Macdonald

(1986). ‘The Koori Way’: The dynamics of cultural distinctiveness in settled Australia [PhD Thesis, University of Sydney].

30.

Macdonald

(2011). The secularisation of riverine law in south-east Australia. Anthropological Forum, 21(3), 307–328. https://doi.org/10.1080/00664677.2011.617720

31.

McCausland

Vivian

(2010). Why do some aboriginal communities have lower crime rates than others? A pilot study. Australian & New Zealand Journal of Criminology, 43(2), 301–332. https://doi.org/10.1375/acri.43.2.301

32.

McLachlan

Gilfillan

Gordon

(2013, July). Deep and persistent disadvantage in Australia. Productivity Commission Staff Working Paper. Australian Federal Government, Canberra. https://www.pc.gov.au/research/supporting/deep-persistent-disadvantage/deep-persistent-disadvantage.pdf

33.

Northern Territory Law Reform Committee (2003). Report of the Committee of Inquiry into Aboriginal Customary Law: Report on Aboriginal Customary Law. Northern Territory Government. https://www.indigenousjustice.gov.au/resources/report-of-the-committee-of-inquiry-into-aboriginal-customary-law-report-on-aboriginal-customary-law/

34.

Osgood

D. W.

Chambers

J. M.

(2000). Social disorganization outside the metropolis: An analysis of rural youth violence. Criminology, 38(1), 81–116. https://doi.org/10.1111/j.1745-9125.2000.tb00884.x

35.

Payne

(2007). Recidivism in Australia: Findings and future research (No. 80). Australian Institute of Criminology.

36.

Peterson

(2010). Other people’s lives: Secular assimilation, culture and ungovernability. In Altman

J. C.

Hinkson

(Eds.), Culture crisis: Anthropology and politics in aboriginal Australia (pp. 248–258). UNSW Press.

37.

Peterson

Taylor

(2003). The modernising of the indigenous domestic moral economy, kinship, accumulation and household composition. The Asia Pacific Journal of Anthropology, 4(1&2), 105–122. https://doi.org/10.1080/14442210310001706407

38.

Pratt

T. C.

Cullen

F. T.

(2005). Assessing macro-level predictors and theories of crime: A meta-analysis. Crime and Justice, 32, 373–450. https://doi.org/10.1086/655357

39.

Queensland Government (2017, March 5). Corrective Services Act [QLD] 2006. Parliamentary Counsel, Queensland Government. https://www.legislation.qld.gov.au/view/pdf/inforce/2017-03-05/act-2006-029

40.

Rose

D. R.

Clear

T. R.

(1998). Incarceration, social capital, and crime: Implications for social disorganisation theory. Criminology, 36(3), 441–480. https://doi.org/10.1111/j.1745-9125.1998.tb01255.x

41.

Ryan

Ackerman

Bond

Ready

Kinner

S. A.

(2019). Prison life and prior social experiences: Understanding their importance for indigenous Peoples’ Re-entry outcomes. The British Journal of Criminology, 59(1), 1–21. https://doi.org/10.1093/bjc/azy027

42.

Ryan

Ackerman

Kinner

. (2020). Indigeneity, prisoner visitation, & reincarceration: how does prisoner visitation effect risk of reincarceration for Indigenous & Non-Indigenous people. The British Journal of Criminology, 60(4), 1056–1079. https://doi.org/10.1093/bjc/azaa009

43.

Sampson

R. J.

Groves

W. B.

(1989). Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 94(4), 774–802. https://doi.org/10.1086/229068

44.

Sampson

Laub

. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

45.

Sampson

R. J.

Morenoff

J. D.

Gannon-Rowley

(2002). Assessing “neighborhood effects”: Social processes and new directions in research. Annual Review of Sociology, 28, 443–478. https://doi.org/10.1146/annurev.soc.28.110601.141114

46.

Sampson

R. J.

Raudenbush

S. W.

Earls

(1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science (New York, N.Y.), 277(5328), 918–924. https://doi.org/10.1126/science.277.5328.918

47.

Shaw

C. R.

McKay

H. D.

(1942). Juvenile delinquency and urban areas. University of Chicago Press.

48.

Small

M. L.

McDermott

(2006). The presence of organisational resources in poor urban neighborhoods: An analysis of average and contextual effects. Social Forces, 84(3), 1697–1724. https://doi.org/10.1353/sof.2006.0067

49.

Stahler

G. J.

Mennis

Belenko

Welsh

W. N.

Hiller

M. L.

Zajac

(2013). Predicting recidivism for released state prison offenders: examining the influence of individual and neighbourhood characteristics and spatial contagion on the likelihood of reincarceration. Criminal Justice and Behavior, 40(6), 690–711. https://doi.org/10.1177/0093854812469609

50.

Trewin

(2006). Statistical geography volume 1 – Australian standard geographical classification (ASGC). ABS Cat. No. 1216.0). Canberra: Australian Bureau of Statistics.

51.

Ulmer

J. T.

(2001). Intermediate sanctions: A comparative analysis of the probability and severity of recidivism. Sociological Inquiry, 71(2), 164–193. https://doi.org/10.1111/j.1475-682X.2001.tb01107.x

52.

Warburton

Chambers

(2007). Older indigenous Australians: Their integral role in culture and community. Australasian Journal on Ageing, 26(1), 3–7. https://doi.org/10.1111/j.1741-6612.2007.00213.x

53.

Weatherburn

(2010). The effect of prison on adult re-offending. Contemporary Issues in Crime and Justice. No. 143.

54.

Weatherburn

(2014). Arresting incarceration: Pathways out of indigenous imprisonment. Aboriginal Studies Press.

55.

Weatherburn

Lind

(2001). Delinquent-prone communities. Cambridge University Press.

56.

Wehrman

M. M.

(2010). Race, concentrated disadvantage, and recidivism: A test of interaction effects. Journal of Criminal Justice, 38(4), 538–544. https://doi.org/10.1016/j.jcrimjus.2010.04.024

57.

Willis

(2008). Reintegration of Indigenous prisoners: Key findings. Trends & Issues in Crime and Criminal Justice Series No. 364. Canberra: Australian Institute of Criminology.

58.

Wokstrom

P. H.

Treiber

(2016). Social disadvantage and crime: A criminological puzzle. American Behavioral Scientist, 60(10), 1232–1259. https://doi.org/10.1177/0002764216643134

59.

Wundersitz

(2010). Indigenous perpetrators of violence: Prevalence and risk factors for offending. Research and Public Policy Series No. 105. Australian Institute of Criminology.