ContentJust Search pageLJF site navigationLeft navigation links
LJF Logo
Publications sectionJustice Awards sectionResearch sectionGrants sectionPlain language law sectionNetworks section
Just Search
Research Report: Legal Australia-Wide Survey: Legal need in Australia
cover image

Legal Australia-Wide Survey: Legal need in Australia  ( 2012 )  Cite this report

3. Prevalence of legal problems

Print chapter

Predicting prevalence of multiple legal problems

A multilevel zero-truncated Poisson regression was conducted on the subgroup of Australian respondents who experienced legal problems. The regression examined whether the demographic characteristics of respondents predicted the number of legal problems experienced among those who experienced at least one legal problem. That is, the regression examined whether certain demographic groups experienced a greater number of legal problems or ‘multiple’ legal problems.(15) This regression also used the same set of demographic predictors as the Australian regression on overall prevalence.(16)

A summary of the results of this regression on the prevalence of multiple legal problems is provided in Table 3.8.(17) Age, disability status and housing type were the strongest significant predictors of the prevalence of multiple legal problems. In descending order of strength, family status, employment status, Indigenous status, education, gender and remoteness of residential area were also significant predictors. Thus, when compared to their counterparts, people aged 15–64 years, people with a disability, people who had lived in disadvantaged housing, single parents, people who had been unemployed, Indigenous people, people with post-school qualifications (rather than those who had finished only Year 12), males and people living in regional (rather than major city) areas were significantly more likely to experience multiple legal problems.

Table 3.8: Regression summary — prevalence of multiple legal problems, Australia

VariableCategories compared
Incident rate ratioa
GenderFemale | male
Age15–17 | 65+
18–24 | 65+
25–34 | 65+
35–44 | 65+
45–54 | 65+
55–64 | 65+
Indigenous statusIndigenous | other
Disability statusDisability | no disability
Education<Year 12 | post-school
Year 12 | post-school
Employment statusUnemployed | other
Family statusSingle parent | other
Housing typeDisadvantaged | other
RemotenessRemote | major city
Regional | major city
NON-SIGNIFICANT VARIABLES main income, main language

a An incident rate ratio (IRR)>1.0 indicates that the first category had a significantly higher rate of experiencing legal problems than the second category. An IRR<1.0 indicates that the first category had a significantly lower rate. The size of the IRR indicates the strength of the relationship. E.g. IRR=2.0 means that the incident rate for the first category was twice that for the second category. IRR=0.5 means that the incident rate for the first category was half that for the second category, or, in other words, that the incident rate for the second category was twice that (i.e. 1/0.5=2.0) for the first category. See Appendix A2, ‘Data analysis: Significance and strength of predictors’ section for further details. ‘-’ indicates that the comparison was not significant.
Note: N=10 244 respondents with problems. Data were missing for 45 respondents.

The results for this regression were also similar to those for overall prevalence (see Table 3.5). In both regressions, age and disability status were the strongest significant predictors, and gender, education, employment status, family status and housing type were also significant, with the results being in the same direction. Hence, the regressions suggest that many of the demographic groups that are vulnerable to experiencing legal problems overall tend to be the same groups that, when they do experience legal problems, are vulnerable to experiencing multiple legal problems.

However, there were some differences between the two regression models. Indigenous status was a significant predictor for multiple legal problems, but not legal problems overall. Main language was not a significant predictor for multiple legal problems but was significant for legal problems overall. In addition, although remoteness of residential area was significant in both regressions, the direction of the results was inconsistent. Compared to people living in major city areas, those living in regional areas had lower odds of experiencing a legal problem overall but higher odds of multiple legal problems when they did experience legal problems.

15. For convenience, the term ‘multiple legal problems’ is used instead of ‘a greater number of legal problems’ throughout the report when discussing the results of this regression.

16. See Chapter 2, ‘Method: Multivariate analyses’ section, and Appendix Tables A2.8 and A2.9 (model 3) for further details.

17. See Appendix Table A3.3 for the full results of this regression.


Coumarelos, C, Macourt, D, People, J, MacDonald, HM, Wei, Z, Iriana, R & Ramsey, S 2012, Legal Australia-Wide Survey: legal need in Australia, Law and Justice Foundation of NSW, Sydney