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

5. Response to legal problems



Print chapter

Predicting strategy in response to legal problems


This section describes the variables associated with the strategies people use to try to resolve their legal problems. Two binary multilevel logistic regression models were fitted to the Australian data to examine whether various demographic and problem characteristics were independent predictors of the strategy used to try to resolve legal problems. The first regression examined the likelihood of taking action to resolve legal problems. It was based on all problems and compared problems resulting in no action to problems resulting in some type of action (i.e. seeking advice or handling without advice). The second regression examined the likelihood of seeking advice when action was taken to resolve problems. It was based only on problems resulting in some type of action and compared seeking advice to handling the problem without advice. The following demographic and problem characteristics were tested as possible predictors of strategy in each model: gender, age, Indigenous status, disability status, education, employment status, family status, housing type, main income, main language, remoteness of residential area, problem recency and problem group.(6)

Thus, the regressions reveal the types of problems and the demographic groups that had lower levels of taking action and seeking advice. While regression analysis can be used to show where relationships exist, it cannot explain any relationships. As already noted, although failure to take action or seek advice may often reflect unmet legal need, it may sometimes be appropriate. Nonetheless, the regressions on strategy help to signal the types of problems and demographic groups that may particularly benefit from initiatives that facilitate appropriate responses to legal problems. For example, education strategies that better signpost the pathways available for legal resolution may be particularly useful for the types of problems and the demographic groups with low levels of taking action or seeking advice. In addition, the provision of more streamlined resolution processes in some of these cases may be warranted.

Table 5.7 provides a summary of the regression results on strategy for Australia. Problem recency, problem group, gender, age, disability status, education, employment status, family status and main language were significant independent predictors of the type of strategy used in response to legal problems in one or both regressions. For both regressions, problem group was the strongest predictor. Indigenous status, housing type, main income and remoteness were not significant predictors of strategy in either regression.

Table 5.7: Regression summary — strategy in response to legal problems, Australia

SIGNIFICANT VARIABLES
Variable Categories compared
Taking actiona
Odds ratiob
Seeking advice
Odds ratioc
Problem recency 7+ months | 6 months
1.3
1.6
Problem group Accidents | mean
0.6
1.8
Consumer | mean
-
0.2
Credit/debt | mean
0.7
0.4
Crime | mean
0.6
2.6
Employment | mean
1.3
1.2
Family | mean
2.2
1.8
Government | mean
1.1
0.4
Health | mean
0.7
1.4
Housing | mean
1.3
0.6
Money | mean
2.6
-
Personal injury | mean
-
4.1
Rights | mean
0.7
1.2
Gender Female | male
1.4
1.2
Age 1517 | 65+
-
0.4
1824 | 65+
1.3
0.6
2534 | 65+
1.4
0.8
3544 | 65+
1.4
-
4554 | 65+
1.4
-
5564 | 65+
-
-
Disability status Disability | no disability
1.4
1.2
Education <Year 12 | post-school
0.7
0.9
Year 12 | post-school
0.8
0.9
Employment status Unemployed | other
0.8
0.8
Family status Single parent | other
-
1.2
Main language Non-English | English
0.5
0.7
NON-SIGNIFICANT VARIABLES
Indigenous status, family status, housing type, main income, remoteness
Indigenous status, housing type,
main income, remoteness

a I.e. seeking advice or handling without advice.
b An odds ratio (OR)>1.0 indicates that the first category had significantly higher odds of taking action (than taking no action) compared to the second category. An OR<1.0 indicates that the first category had significantly lower odds. The size of the OR indicates the strength of the relationship. E.g. OR=2.0 means that the odds for the first category were twice those for the second category. OR=0.5 means that the odds for the first category were half those for the second category, or, in other words, that the odds for the second category were twice those (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.
c An OR>1.0 indicates that the first category had significantly higher odds of seeking advice (than handling without advice) compared to the second category. An OR<1.0 indicates that the first category had signficantly lower odds. The size of the OR indicates the strength of the relationship. ‘-’ indicates that the comparison was not significant.
Note: N=19 056 problems for regression on taking action. Data were missing for 332 problems. N=15 579 problems where took action for regression on seeking advice. Data were missing for 313 problems.

The results of the two regressions on strategy are further described in the sections below.(7) These regression results are accompanied by the relevant unprocessed percentages. The percentages are based on all problems.

Legal problem characteristics

The recency of legal problems was related to the strategies used. The odds of taking action were significantly higher for problems that had persisted for at least seven months (1.3) than for more recent problems (see Table 5.7). The odds of seeking advice when action was taken were also higher for more persistent problems (1.6). Respondents took no action for 15.3 per cent of problems that had persisted for at least seven months compared to 21.6 per cent of more recent problems (see Figure 5.6). Advice was sought for 57.3 per cent of problems that had persisted for at least seven months but only 43.9 per cent of more recent problems.

Figure 5.6: Strategy in response to legal problems by problem recency, Australia

R Reference category for problem recency in the regressions.
* Significant difference (p<0.05) for problem recency in the regression on taking action.
^ Significant difference (p<0.05) for problem recency in the regression on seeking advice.
Note: N=19 137 problems. Data were missing for 251 problems.

In addition, the regression results indicated that the strategy used in response to legal problems was significantly associated with the type of problem. In fact, problem group was the strongest predictor of both taking action and seeking advice (see Table 5.7).

As Table 5.7 shows, the odds of taking action were significantly lower for accidents (0.6), crime (0.6), health (0.7) and rights (0.7) problems than for all problems on average. When action was taken, these problems resulted in significantly higher odds of seeking advice than average (1.8, 2.6, 1.4 and 1.2, respectively). No action was taken for 20.8–26.8 per cent of these problems compared to 18.3 per cent on average (see Figure 5.7). Advice was sought for 55.1–63.8 per cent of these problems compared to 51.1 per cent on average.
The odds of taking action for consumer problems were not significantly different from those for all problems on average (see Table 5.7). When action was taken, however, consumer problems resulted in significantly lower odds of seeking advice than average (0.2), with 24.0 per cent of these problems resulting in seeking advice compared to 51.1 per cent on average (see Figure 5.7).

Figure 5.7: Strategy in response to legal problems by problem group, Australia

R Reference category for problem group in the regression was the mean of all problems.
* Significant difference (p<0.05) between this problem group and the mean of all problems in the regression on taking action.
^ Significant difference (p<0.05) between this problem group and the mean of all problems in the regression on seeking advice.
Note: N=19 142 problems. Data were missing for 246 problems.

The odds of taking action were significantly lower for credit/debt problems (0.7) than for all problems on average (see Table 5.7). When action was taken for these problems, they also resulted in significantly lower odds of seeking advice than average (0.4). No action was taken for 21.8 per cent of credit/debt problems compared to 18.3 per cent on average, and only 38.6 per cent of these problems resulted in seeking advice compared to 51.1 per cent on average (see Figure 5.7).

The odds of taking action were higher for employment and family problems (1.3 and 2.2, respectively) than for all problems on average. When action was taken, these problems also resulted in significantly higher odds of seeking advice than average (1.2 and 1.8, respectively; see Table 5.7). Only 14.5 per cent of employment problems and 6.5 per cent of family problems resulted in taking no action compared to the average of 18.3 per cent, while 60.8 and 78.5 per cent, respectively, of these problems resulted in seeking advice compared to 51.1 per cent on average (see Figure 5.7).

Compared to all problems on average, government problems resulted in higher odds of taking action (1.1) and lower odds of seeking advice when action was taken (0.4; see Table 5.7). Only 15.2 per cent of government problems resulted in taking no action compared to 18.3 per cent on average, and only 42.9 per cent of these problems resulted in seeking advice compared to 51.1 per cent on average (see Figure 5.7).

Similarly, housing problems resulted in higher odds of taking action (1.3) and lower odds of seeking advice when action was taken (0.6; see Table 5.7). Consistent with the high odds of taking action, only 12.9 per cent of housing problems resulted in taking no action compared to 18.3 per cent on average (see Figure 5.7). However, the low level of seeking advice for housing problems is less obvious from the percentages, suggesting that it becomes evident once the influences of other characteristics are also considered.

The odds of taking action were higher for money problems (2.6) than for all problems on average (see Table 5.7). However, respondents were no more likely to seek advice when they took action for money problems than for all problems on average. Only 6.6 per cent of money problems resulted in taking no action compared to 18.3 per cent on average (see Figure 5.7).

Although personal injury problems resulted in average odds of taking action, when action was taken, however, they resulted in significantly higher than average odds of seeking advice (4.1; see Table 5.7). In total, 72.2 per cent of personal injury problems resulted in seeking advice compared to 51.1 per cent on average (see Figure 5.7).

Demographic variables

The regressions on strategy revealed that some demographic groups were less likely to take action or seek advice to resolve their legal problems, even after the characteristics of the problem (i.e. recency and problem group) were taken into account. As already noted, however, problem group had a stronger effect on strategy than any of the demographic variables.
Main language was the strongest significant demographic predictor of taking action (see Table 5.7). Age, education, gender, disability status and employment status were also significant demographic predictors and had similar strengths of association with taking action. Compared to their counterparts, the following demographic groups had significantly lower odds of taking action:
    • people whose main language was not English
    • people aged 65 years or over (versus 18–54 year olds)
    • people with low education levels (versus those with post-school qualifications)
    • males
    • people without a disability
    • people who had been unemployed.

In descending order of strength, the demographic predictors of seeking advice were age, main language, employment status, disability status, family status, gender and education (see Table 5.7). Compared to their counterparts, the following demographic groups had significantly lower odds of seeking advice when action was taken:
    • 15–34 year olds (versus those aged 65 years or over)
    • people whose main language was not English
    • people who had been unemployed
    • people without a disability
    • people who were not single parents
    • males
    • people with low education levels (versus those with post-school qualifications).

The significant odds ratios from the regressions in Table 5.7 generally reveal a similar picture to the percentages in Table 5.8.(8) Compared to males, females had significantly higher odds of taking action (1.4) and higher odds of seeking advice when action was taken (1.2; see Table 5.7). Females took no action for only 15.1 per cent of problems compared to 21.3 per cent for males (see Table 5.8). Also, females sought advice for 54.6 per cent of problems compared to 47.8 per cent for males.


Table 5.8: Strategy in response to legal problems by each demographic variable, Australia

Demographic variable
Category
Strategy
All problems
Sought advice
Handled without advice
Took no action
%
%
%
%
N
Gender Female
54.6
30.3
15.1
100.0
9 309
*^
MaleR
47.8
30.9
21.3
100.0
9 833
Total
51.1
30.6
18.3
100.0
19 142
Age 15–17
36.5
32.0
31.4
100.0
812
^
18–24
43.1
35.3
21.6
100.0
2 664
*^
25–34
50.1
32.7
17.3
100.0
3 840
*^
35–44
55
29
16
100.0
4 186
*
45–54
55.9
28.2
15.9
100.0
3 515
*
55–64
54.6
27.2
18.2
100.0
2 451
65+R
48.3
32
19.7
100.0
1 674
Total
51.1
30.6
18.3
100.0
19 142
Indigenous statusIndigenous
48.8
36.5
14.7
100.0
386
OtherR
51.2
30.5
18.3
100.0
18 756
Total
51.1
30.6
18.3
100.0
19 142
Disability statusDisability
57.3
27.1
15.7
100.0
5 037
*^
No disabilityR
48.9
31.9
19.2
100.0
14 105
Total
51.1
30.6
18.3
100.0
19 142
Education <Year 12
49.8
28.2
22
100.0
5 087
*^
Year 12
47.9
32.4
19.6
100.0
3 680
*
Post-schoolR
53
31.1
15.9
100.0
10 294
Total
51.1
30.6
18.3
100.0
19 061
Employment statusUnemployed
44.9
33.3
21.8
100.0
2 860
*^
OtherR
52.2
30.2
17.6
100.0
16 282
Total
51.1
30.6
18.3
100.0
19 142
Family statusSingle parent
61.6
23.0
15.4
100.0
2 183
^
OtherR
49.8
31.6
18.6
100.0
16 959
Total
51.1
30.6
18.3
100.0
19 142
Housing typeDisadvantaged
55.3
28.0
16.7
100.0
1 584
OtherR
50.7
30.9
18.4
100.0
17 558
Total
51.1
30.6
18.3
100.0
19 142
Main incomeGovernment payments
53.7
28.4
17.9
100.0
4 814
OtherR
50.2
31.4
18.4
100.0
14 328
Total
51.1
30.6
18.3
100.0
19 142
Main languageNon-English
36.9
33.3
29.7
100.0
1 034
*^
EnglishR
51.9
30.5
17.6
100.0
18 108
Total
51.1
30.6
18.3
100.0
19 142
Remoteness Remote
49
31.3
19.7
100.0
461
Regional
53.6
29.7
16.7
100.0
5 682
Major cityR
50.1
31.0
18.9
100.0
12 999
Total
50.3
30.4
19.2
100.0
19 142
R Reference category for this demographic variable in the regression.
* Significant difference (p<0.05) between this category and the reference category for this demographic variable in the regression on taking action.
^ Significant difference (p<0.05) between this category and the reference category for this demographic variable in the regression on seeking advice.
Note: N=19 061 problems for education and N=19 142 problems for other demographic variables. Data were missing where totals are less than 19 388.

Respondents aged 18–54 years had higher odds of taking action compared to those aged 65 years or over (1.3–1.4; see Table 5.7). The percentages for 25–54 year olds were consistent with these odds (see Table 5.8). They took no action for 15.9–17.3 per cent of problems compared to 19.7 per cent for the oldest age group. However, the higher odds of taking action for 18–24 year olds compared to the oldest group were not reflected in the percentages. Thus, this higher level of taking action for 18–24 year olds becomes evident once the influence of other demographic and problem characteristics are also appropriately taken into account in the regression.

When action was taken, respondents aged 15–34 years had lower odds of seeking advice compared to those aged 65 years or over (0.4–0.8; see Table 5.7). Consistent with these odds, 15–24 year olds had lower percentages of seeking advice (36.5–43.1%) compared to the oldest group (48.3%; see Table 5.8). However, the lower level of seeking advice for 25–34 year olds compared to the oldest group was not obvious from the percentages, suggesting that this effect becomes evident once the influence of other demographic and problem characteristics are also considered.

Compared to others, respondents with a disability had significantly higher odds of taking action (1.4) and significantly higher odds of seeking advice when action was taken (1.2; see Table 5.7). Respondents with a disability took no action for 15.7 per cent of problems compared to 19.2 per cent for those without a disability (see Table 5.8). When action was taken, respondents with a disability sought advice for 57.3 per cent of problems compared to 48.9 per cent for other respondents.

Respondents with low education levels had lower odds of taking action (0.7–0.8) and lower odds of seeking advice when action was taken (0.9) than those with post-school qualifications (see Table 5.7). Respondents with low education levels took no action for 19.6–22.0 per cent of problems compared to 15.9 per cent for those with post-school qualifications (see Table 5.8). In addition, those with low education levels sought advice for 47.9–49.8 per cent of problems compared to 53.0 per cent for respondents with post-school qualifications.

Similarly, respondents who had been unemployed had lower odds of taking action (0.8) and lower odds of seeking advice when action was taken (0.8) than other respondents (see Table 5.7). No action was taken for 21.8 per cent of problems by respondents who had been unemployed compared to 17.6 per cent by other respondents (see Table 5.8). Respondents who had been unemployed sought advice for 44.9 per cent of problems compared to 52.2 per cent for other respondents.

Family status was unrelated to the odds of taking action (see Table 5.7). However, single parents had significantly higher odds of seeking advice when action was taken (1.2) than other respondents. Single parents sought advice for 61.6 per cent of problems compared to 49.8 per cent for other respondents (see Table 5.8).

Respondents whose main language was not English had significantly lower odds of taking action (0.5) and significantly lower odds of seeking advice when action was taken (0.7) than other respondents (see Table 5.7). People whose main language was not English took no action for 29.7 per cent of problems compared to 17.6 per cent for other respondents (see Table 5.8). Also, respondents whose main language was not English sought advice for 36.9 per cent of problems compared to 51.9 per cent for other respondents.

Consistency of strategy in response to legal problems

According to the regressions, respondents who had done nothing in response to a previous legal problem were significantly less likely than others to take action for subsequent legal problems (see random effects in Appendix Table A5.2). However, of the respondents who had taken action for an earlier legal problem, those who had sought advice were no more likely than those who had handled the problem without advice to seek advice for new legal problems (see random effects in Appendix Table A5.3).

6. See Chapter 2, ‘Method: Multivariate analyses’ section, and Appendix Tables A2.8 and A2.9 (models 5a and 6a) for further details.

7. See Appendix Tables A5.2 and A5.3 for the full results of these regressions.

8. The percentages and the regression on taking action are based on all problems, whereas the regression on seeking advice is based only on problems where action was taken.

  


CLOSE
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