Prevalence of legal problems
Prevalence of legal problems overall
The LAW Survey findings reiterate that legal problems are ubiquitous. Within the one-year period examined, approximately half of the respondents in each jurisdiction experienced a legal problem. The prevalence rate in Australia as a whole was 50 per cent. As reported in Chapter 3, this prevalence rate translates to an estimated 8 513 000 people aged 15 years or over in the Australian population experiencing a legal problem within a one-year period.
The LAW Survey used a general population sample. The present prevalence rates of around 50 per cent were apparently higher than those of most other general population surveys in the UK, other parts of Europe and New Zealand, which have typically fallen below 40 per cent (19–51%; Dignan 2006; Genn 1999; Genn & Paterson 2001; Gramatikov 2008; HKDOJ 2008; Ignite Research 2006; Maxwell et al. 1999; Murayama 2007; Pleasence 2006). However, the present prevalence rates were appreciably lower than those of many of the surveys of disadvantaged samples, including the NSWLNS (69%; Coumarelos et al. 2006) and many US surveys. Most US surveys have produced prevalence rates above 40 per cent, with about half being above 60 per cent (33–87%; AAJC 2009; ABA 1994; CSRA 2003; Dale 2009; GKA 2006, 2008; LASNSC 2005; LSNJ 2009; Miller & Srivastava 2002; Schulman 2003, 2007; TALS 2004; Task Force 2003). Thus, the prevalence rates across surveys have been generally consistent with the level of disadvantage across samples. Populational differences unrelated to disadvantage (e.g. differences in culture, attitudes or geography) and jurisdictional differences (e.g. in legal service provision and redress) may also have contributed to differences in prevalence.
However, the prevalence rates obtained across studies may reflect methodological differences instead of, or in addition to, real differences due to disadvantage, other populational factors or jurisdictional factors. First, the higher prevalence for the LAW Survey compared to other general population surveys may be partly due to its more lenient triviality threshold, which did not filter out less serious, easy-to-solve legal problems, its fairly broad coverage of legal problems(4)
and the greater anonymity afforded by its use of telephone rather than face-to-face interviews(5)
(cf. Currie 2007b; Dignan 2006; Genn 1999; Genn & Paterson 2001; Gramatikov 2008; HKDOJ 2008; Ignite Research 2006; Pleasence 2006; Pleasence et al. 2010). Second, various methodological differences are also consistent with the lower prevalence rates in general population surveys, including the LAW Survey, when compared to disadvantaged sample surveys such as the NSWLNS and US surveys. The NSWLNS used an especially lenient triviality threshold, capturing non-problematic legal events as well as easy-to-solve problems. The US surveys captured problems experienced by entire households rather than problems experienced by one individual within each household. Furthermore, the NSWLNS and US surveys used telephone rather than face-to-face interviews, which may have contributed to higher reporting when compared to the overseas general population surveys. However, the reference periods used across studies cannot explain the variation in prevalence rates. The studies with the higher prevalence rates, such as the present survey, the NSWLNS and the US surveys, had reference periods of one year, which were shorter, not longer, than the 2–5 years used by the overseas general population surveys.
Although the LAW Survey revealed prevalence rates that were close to 50 per cent in all jurisdictions (47–55%), there were nonetheless significant albeit modest differences between these rates. Compared to average, the rates for the Northern Territory (55%) and Western Australia (52%) were significantly higher, while the rates for South Australia (47%) and Victoria (48%) were significantly lower.(6)
Although the reason for the somewhat elevated prevalence in Western Australia is unclear, the higher prevalence in the Northern Territory is consistent with the higher level of disadvantage in this jurisdiction (e.g. ABS 2008c). Regression analyses revealed that only some of the differences in state/territory prevalence rates are likely to be due to differences in demographic compositions.(7)
Other differences between jurisdictions may also have influenced the prevalence rates, such as differences in culture, attitudes, systems of law, legal services or social services.
Prevalence of different types of legal problems
The consumer (18–22% of respondents), crime (13–23%), housing (10–13%) and government (8–12%) problem groups were typically the most common problem groups in most jurisdictions. In Australia as a whole, the most common problem groups were the consumer (21% of respondents), crime (14%), housing (12%) and government (11%) problem groups. Notably, the Northern Territory was the only jurisdiction where crime was the most common problem group, and the percentage of Northern Territory respondents experiencing crime problems (23%) appeared to be higher than the percentages in the other jurisdictions (13–19%). This Northern Territory finding is largely in keeping with official data showing high recorded offender rates and high crime victimisation rates in this jurisdiction (ABS 2009f, 2011d). Given the high Indigenous population in the Northern Territory, this finding is also consistent with the considerable overrepresentation of Indigenous people in the criminal justice system (Cunneen & Schwartz 2008; Steering Committee for the Review of Government Service Provision (SCRGSP) 2007). Personal injury, employment and credit/debt problems were also reported by sizeable proportions of LAW Survey respondents (5–8%) in all jurisdictions.
The LAW Survey findings on the prevalence of different types of legal problems are largely consistent with overseas surveys. For example, past surveys have often reported high rates of consumer and neighbours problems (ABA 1994; AFLSE 2007; Coumarelos et al. 2006; Currie 2007b; Dale 2000, 2005, 2007, 2009; Dignan 2006; GKA 2006, 2008; Gramatikov 2008; HKDOJ 2008; Ignite Research 2006; LSNJ 2009; Murayama 2007; Pleasence 2006; Pleasence et al. 2010; Schulman 2003, 2007; van Velthoven & Klein Haarhuis 2010). Most of the housing problems for the LAW Survey were neighbours problems. Employment and money/debt problems have also been fairly frequent across surveys (ABA 1994; AFLSE 2007; Coumarelos et al. 2006; Currie 2007b; Dale 2000, 2005, 2007, 2009; Dignan 2006; GKA 2006, 2008; Gramatikov 2008; HKDOJ 2008; Ignite Research 2006; LSNJ 2009; Murayama 2007; Pleasence 2006; Pleasence et al. 2010; Schulman 2003; van Velthoven & Klein Haarhuis 2010; van Velthoven & ter Voert 2004). In addition, although many past surveys have not focused or have not reported on crime problems, the present high rate of crime problems is consistent with that for the NSWLNS (Coumarelos et al. 2006).
Furthermore, the present very low rates of legal problems concerning mental health and immigration replicate past findings (e.g. ABA 1994; AFLSE 2007; Coumarelos et al. 2006; Currie 2007b; Dale 2005, 2009; Dignan 2006; GKA 2006, 2008; Ignite Research 2006; Pleasence 2006; Pleasence et al. 2010; Schulman 2003, 2007). These rates tended to be under one per cent across jurisdictions in the present study.
The LAW Survey findings are in keeping with the notion that the incidence of different problem types reflects the defining circumstances necessary for different legal problems to arise. For example, there is considerable opportunity for consumer problems to arise, because consumer transactions are a routine activity for most people. However, there is only limited opportunity for immigration, citizenship or residency problems to arise, because relatively few people change their country of abode (Dignan 2006; Pleasence 2006).
Prevalence of multiple legal problems
The LAW Survey also reinforces existing evidence that the experience of multiple legal problems is common. In each jurisdiction, roughly one-third of respondents reported at least two legal problems in the 12-month period, with roughly one-quarter reporting problems that fell into different problem groups.
The present results confirm earlier findings that a minority of people are particularly vulnerable to multiple legal problems (e.g. Coumarelos et al. 2006). In each jurisdiction, approximately one-tenth of respondents accounted for around two-thirds of the problems reported. Some legal needs surveys have demonstrated an ‘additive effect’ of legal problems, whereby experiencing a legal problem increases the likelihood of experiencing an additional problem, with vulnerability continuing to increase as more problems are experienced (Currie 2007b; Gramatikov 2008; Pleasence 2006; Pleasence et al. 2004c). Pleasence et al. (2004c, p. 107) maintained that:
vulnerability to justiciable problems is … cumulative. Each time a person experiences a problem the likelihood of experiencing an additional problem increases; not just as a consequence of initial vulnerability; but also as a consequence of the increased vulnerability brought about by the impact of initial problems.
Clustering of legal problems
The LAW Survey also supports past findings that the types of legal problems that people experience in combination are unlikely to be purely random. Cluster analyses resulted in considerable consistency in the legal problem groups that co-occurred across jurisdictions. In most jurisdictions, the following problem groups tended to occur in combination:
- the consumer, crime, government and housing problem groups
- the credit/debt, family and money problem groups
- the employment, health, personal injury and rights problem groups.(8)
Although problem groups tended to co-occur in these three combinations, there was some variation across jurisdictions. First, some elements of these combinations were missing in some jurisdictions (see Table 9.1). Second, there was variation in whether or not each combination co-occurred with other problems. That is, in some jurisdictions, a given combination formed a whole, stand-alone cluster that was unrelated to other problems, while in other jurisdictions it formed a subcluster within a broader cluster of co-occurring problems. In Table 9.1, stand-alone clusters are marked with ‘*’, while subclusters are marked with ‘^’.(9)
Table 9.1: Summary — clustering of problem groups, each jurisdiction
* Denotes a whole cluster.
^ Denotes subclusters of a larger cluster. For example, in NSW, the subcluster comprising consumer, crime, government and housing problems joined with the subcluster comprising credit/debt and money problems to form one cluster.
Note: N=20 716 respondents. Problem groups shown in brackets were not part of the subclusters shown but fell within the same larger cluster as these subclusters. For example, in Victoria, family problems were part of neither the ‘consumer, crime, government and housing’ subcluster nor the ‘credit/debt and money’ subcluster but were part of the same larger cluster as these two subclusters. Shading indicates problem groups that were common within clusters/subclusters across jurisdictions.
The first combination, comprising the consumer, crime, government and housing problem groups, was particularly consistent, with all four of these problem groups combining to form either clusters or subclusters in all jurisdictions apart from the Northern Territory. In Australia as a whole, the first cluster was dominated by these four problem groups but also included the money problem group.
The second combination, comprising the credit/debt, family and money problem groups, was dominated by ‘economic and family’ issues. Most jurisdictions had a cluster or subcluster that included at least two of these three problem groups. Note that the money problem group, by definition, included both problems with economic aspects and problems related to family relationships. Business, investment, wills, estates and power of attorney issues were categorised within this problem group. In Australia as a whole, credit/debt and family problems formed a separate cluster, but money problems were not part of this cluster.
The third combination, comprising the employment, health, personal injury and rights problem groups, was dominated by ‘rights and injury/health’ issues. Most jurisdictions had a cluster or subcluster that included at least two of these four problem groups. Note that work-related rights issues were categorised within the employment problem group, while rights issues unrelated to work were categorised within the rights problem group. In Australia as a whole, employment, health, personal injury and rights problems formed a separate cluster.
As noted earlier, the co-occurrence of certain legal problems suggests the possibility that these problems may be connected in some way, because, for example:
- one of these problems may directly cause or trigger another
- these problems may arise from similar or identical defining circumstances
- certain individuals may be vulnerable to experiencing these types of problems (Pleasence et al. 2004c).
Past studies have not produced identical legal problem clusters. Nonetheless, like the present study, they have usually found clustering of ‘family’ and ‘economic’ issues, although these issues have not necessarily formed a single cluster (Coumarelos et al. 2006; Currie 2007b; Pleasence 2006; Pleasence et al. 2004c, 2010). For example, the NSWLNS and the CSJS both produced separate ‘family’ and ‘economic’ clusters (Coumarelos et al. 2006; Pleasence 2006; Pleasence et al. 2010). In addition, the present clustering of rights and health problems is somewhat similar to the clustering of discrimination and clinical negligence problems in the 2004 and 2006–2009 CSJS (Pleasence et al. 2010).
Different clustering patterns across studies are perhaps unsurprising, given their methodological differences. First, different populations may have differing vulnerabilities for certain groups of legal problems.(10)
Second, the coverage, definition and categorisation of legal problems may affect clustering. For example, the CSJS and NSWLNS categorised domestic violence issues into a separate problem type, which clustered with other relationship breakdown issues to form a ‘family’ cluster (Coumarelos et al. 2006; Pleasence 2006; Pleasence et al. 2010). In contrast, other past studies and the present study did not isolate domestic violence issues into a separate problem type but subsumed them within a broader category of problems (cf. Currie 2007b; Genn 1999; Genn & Paterson 2001; Pleasence 2006). Thus, these studies could not examine which problems specifically cluster with ‘domestic violence’ per se. Third, the reference period used may have an impact. For example, clusters reflecting the long-term consequences of certain legal problems may be less evident with shorter reference periods. Finally, differences in the triviality thresholds used to capture legal problems may also affect clustering. For example, observation of the connections between problems may depend on the proportion of substantial problems captured.
Past evidence has suggested that some types of legal problems are especially likely to trigger or directly cause further problems. In particular, the evidence has been consistent with family, injury and employment problems often preceding and triggering money and debt problems (Currie 2007b; Genn 1999; Pleasence 2006). Trigger effects can have dramatic impacts on people’s lives. They can lead to a cascade of spiralling problems and downward mobility in a variety of life circumstances (Coumarelos et al. 2006; Currie 2007b; Genn 1999; Pleasence 2006; Pleasence et al. 2004c). Pleasence (2006) argued that the trigger effect of family problems such as divorce, separation and domestic violence is not surprising, given that they can lead to substantial changes in life circumstances, such as financial hardship, poorer housing, employment problems, difficulties as a single parent, and dependence on maintenance, child support and welfare benefits. The LAW Survey findings that family problems sometimes clustered with economic problems are consistent with the possibility that family problems may sometimes trigger legal problems with a financial impact.(11)
Pleasence (2006) and Genn (1999) argued that personal injury problems can also have a dramatic impact on life circumstances, often as a result of causing unemployment, which in turn can lead to financial hardship. The Paths to justice surveys and the CSJS produced broad clusters including employment, personal injury, money and consumer problems (Pleasence 2006; Pleasence et al. 2010). The present results are partially consistent with these findings, given that personal injury and employment problems clustered in most jurisdictions.(12)
However, the possible impact of personal injury and unemployment on financial hardship was evident only in Queensland and the Northern Territory, where employment and credit/debt problems clustered together. The shorter reference period in the present study compared to the UK studies may have been insufficient for the full impact of personal injury and employment problems on financial circumstances to become evident in all jurisdictions. Furthermore, the full impact of employment problems on financial hardship may occur more quickly for more disadvantaged people who have fewer savings and assets (Saunders et al. 2007). The present Northern Territory and Queensland results are consistent with this possibility. The top 20 most disadvantaged areas in Australia are remote areas within these jurisdictions, and, in addition, the Northern Territory is the most disadvantaged jurisdiction overall (e.g. ABS 2008c).
As discussed earlier, the co-occurrence of legal problems does not necessarily imply a meaningful connection between these problems in all cases. Legal problems may sometimes co-occur by chance — that is, without a connection due to trigger effects, defining circumstances or personal vulnerabilities. In particular, problem types that occur frequently in the population have more opportunity to coincide by chance. Consistent with this possibility, the first cluster or subcluster in the present study typically included the four most prevalent legal problem groups — that is, the consumer, crime, government and housing problem groups.(13)
In addition to clustering with each other, these four problem groups had a sizeable degree of co-occurrence with all problem groups across jurisdictions (see Table 4.9 in each LAW Survey report). Notably, past studies have often similarly found that one of the main clusters that emerges is dominated by high-frequency problem types.(14)
The chance co-occurrence of high-frequency legal problems may be more evident in studies using shorter reference periods, such as the present study and the Coumarelos et al. (2006) study. First, there is less opportunity with shorter periods to capture all of the legal problems that are likely to result from long-term causal effects. Second, shorter reference periods are likely to provide more accurate capture of legal problems that are high volume but ‘minor’, given that memory decay over longer periods tends to result in less salient problems being forgotten (Lynn et al. 2005; Pleasence et al. 2009; Sudman & Bradburn 1973; Tourangeau et al. 2000). Thus, both of these possibilities may work towards shorter reference periods producing greater visibility of high-volume minor problems that tend to coincide by chance rather than due to some sort of meaningful connection. Nonetheless, the possible chance coincidence of some legal problems in no way negates the potential difficulty that some people face when confronted with multiple problems. The present study, like past studies, makes clear that the occurrence of multiple legal problems is a common experience.
4. The LAW Survey, like its predecessor, the NSWLNS, captured a broad range of problems extending to those related to owning a business, wills and estates, and general crime. Coumarelos et al. (2006) and Pleasence (2006) have noted that the broader coverage of problems by some surveys (e.g. by the NSWLNS and some US surveys compared to the UK surveys) may contribute to differences in prevalence.
5. The use of internet questionnaires in the Dutch survey (van Velthoven & ter Voert 2004), which employed a non-random sample, may similarly have afforded considerable anonymity that contributed to the high prevalence rate (67%).
6. x2=38.82, F7,144629=5.35, p=0.000. See Appendix Figure A9.1.
7. The Australian regression model on overall prevalence (see Table 3.5 in the Australian LAW Survey report) was re-run with the addition of state/territory as a potential predictor variable or ‘fixed effect’. See Appendix Tables A2.8 and A2.9 (model 1b) for further details and Appendix Table A9.2 for the full results. The chi-square test examined prevalence given the states’/territories’ actual demographic profiles. In contrast, the regression estimated what the prevalence levels would be if states/territories had identical profiles on the demographic variables examined in the model. The regression showed significant differences in prevalence between states/territories after the demographic variables had been taken into account, so it is unlikely that the differences in prevalence are due solely to differences on these demographic variables. Compared to average, the odds of legal problems were lower in South Australia and the ACT, but higher in the Northern Territory.
8. Across jurisdictions, accidents problems did not cluster strongly with other problem groups. Although accidents problems clustered with personal injury problems in the ACT, this relationship was weak. The finding that accidents problems tended to occur independently of other problems may reflect that, by definition, the accidents problem group consisted exclusively of injury-free motor vehicle accidents, which were typically reported as being minor problems (see Table 3.3). It is plausible that such typically minor accidents are largely chance events that are not often causally linked to other problems. Motor vehicle injuries were categorised within the personal injury problem group.
9. In addition, problem groups shown in brackets in Table 9.1 were not part of the subclusters shown but fell within the same larger cluster as these subclusters.
10. The demographic groups in the present study that were especially vulnerable to experiencing particular types of problems are discussed in the next section, ‘Predicting prevalence of legal problems’.
11. The strongest associations of family problems with credit/debt and/or money problems occurred in Australia as a whole, South Australia, Western Australia and Tasmania.
12. Except in the Northern Territory and the ACT, personal injury and employment problems were part of the same cluster.
13. The consumer, crime, government and housing problem groups were the four most prevalent problem groups in all jurisdictions apart from the ACT where they comprised four of the five most prevalent problem groups. These four problem groups dominated the first cluster/subcluster in all jurisdictions except the Northern Territory. The Northern Territory results were nonetheless similar: consumer, crime and housing problems dominated the first subcluster, and government problems still fell within the same broader cluster, although they fell within a different subcluster.
14. The broad cluster in the Coumarelos et al. (2006) study was dominated by the four most prevalent problem types (i.e. general crime, housing, consumer and government problems), although it also included other problem types. One cluster in the Canadian study (Currie 2007a) exclusively comprised the three most frequent problem types (i.e. consumer, employment and debt problems). The broad cluster based on the data from the Paths to justice study included the two most frequent problem types (i.e. consumer and money problems) as well as other problem types (cf. Genn 1999; Pleasence 2006). The CSJS included a broad or economic cluster that was dominated by the four most prevalent problem types (i.e. consumer, neighbours, money/debt and employment problems), although it also included other problem types (Pleasence 2006; Pleasence et al. 2010).