Community programme
8.1 Background
This sentence was introduced as 'community care' in 1985 and renamed 'community programme' in 1993. As for community service, the offender must have committed an imprisonable offence and must consent to the sentence, and a suitable programme must be available. The sentence can involve attending a programme to meet the medical, social, therapeutic or educational needs of the offender or being placed in the care of a community group or person that can provide appropriate support or a positive influence.
This chapter examines which statistical factors most influence the use of community programme sentences, looking first at the variables individually and then at their combined effects using multivariate modelling techniques. Changes in the use of community programme are documented in the final section of this chapter.
8.2 Single variable analysis of current factors influencing sentencing
The community programme sentence is relatively rarely used. Overall only 1.1% of the proved cases in 1995 resulted in a community programme sentence. The probability of receiving a community programme sentence is higher (2-4%) for offences in the seriousness range 20 to 365 and where several charges have been proved against the offender (Figure 8.1).
Offenders with a long or serious previous offending record are more likely to receive a community programme sentence, especially those with a high rate of conviction (i.e. offenders who are convicted very frequently).
The percentage of offenders receiving a community programme sentence is higher if the most recent sentence prior to the current case was also a community programme sentence (8.2%), than if the previous sentence was a prison sentence (2.3%), periodic detention (1.6%), community service (1.8%), supervision (2.2%) or a monetary penalty (0.8%).
Almost 2% of Māori and Pacific offenders received a community programme sentence, compared to 1% of Pakeha/Other offenders. Slightly more women (1.3%) than men (0.9%) received a community programme sentence.
Community programme sentences are much more likely to be imposed for domestic violence offences (4.8%) and serious offences against the person (2.7%) than for other offence types (0.5-1.5%).
Figure 8.1: Percentage of offenders receiving a community programme sentence in 1995 for selected variables
8.3 Multivariate analysis of current factors influencing sentencing
8.3.1 The fit and accuracy of the 1995 model
The logistic regression models for community programme achieved a significant overall fit to both the full and half 1995 data, as indicated by log likelihood ratios significant at the 0.0001 level of probability. However, the residual (unexplained variation) term of the full 1995 model was significant, indicating that the results should be interpreted with caution, due to the possibility of systematic error in the model's predictions.
Moreover, the community programme model produced predictions only over a very narrow range of probability - very few offenders had even a 10% predicted probability of receiving a community programme sentence (Figure 8.2). This finding is perhaps not surprising, given the very low use of community programme and the importance of contextual factors relevant to the specific offence and offender, but not measured by the statistical factors available for this study. For example, Hall (1998, page D/592) makes the following comment about the targeting of community programme:
'The sentence is particularly appropriate where offending is associated with some social or personal deprivation or condition that may be alleviated by community-based support. Where the offender is lacking day to day living and social skills, and is in need of social education, the Court should consider imposing this sentence.'
Figure 8.2: Plot of the predicted probability versus the actual proportion of offenders receiving community programme, 1995 fitted and test data
Given that the overall probability of receiving a community programme sentence is 1%, the model's ability to identify offenders with up to a 10% probability of receiving a community programme sentence still represents a ten-fold predictive improvement over what could be achieved by a random assignment of probability and is double the maximum discrimination based on any single statistical factor. Nevertheless, the model's usefulness for prediction is very limited, and the following results are therefore restricted to general statements of the type of statistical factors that may have a significant influence on the use of community programme.
8.3.2 Results of the 1995 model
Some comment on the factors contributing to the probability of receiving community programme are worth noting, even though the usefulness of the model is limited.
The results of the 1995 logistic model suggest that community programme sentences are targeted at violent offenders, particularly those who commit domestic violence (Table 8.1). An offender being sentenced for domestic violence has almost a five times higher relative risk of receiving a community programme sentence relative to a property offender (although, as noted in the previous section, the use of community programme is rare for all offences, peaking at only 4.8% for domestic violence). Other serious and minor offences against the person (mainly violent offences) also appear to have a higher likelihood of receiving a community programme sentence, as do breaches of periodic detention.
As for periodic detention, community programme tends to be targeted at people who have committed moderately serious offences or several charges. If their most recent sentence was another community programme sentence or a supervision sentence, the offender also has an increased probability of receiving a community programme sentence. Other criminal history characteristics appear to have relatively little effect on sentencing to community programme.
Māori and Pacific peoples appear be around twice as likely to receive a community programme sentence as Pakeha/Other offenders, once other factors have been taken into account. This is as expected, given that one of the objectives of the sentence was to provide an alternative sentence that could accommodate culturally-appropriate programmes. As noted in Hall (1998, page D/593):
'Community care will generally be appropriate for young offenders who belong to an ethnic or cultural minority, where members of that group are willing to attempt to instil some sense of responsibility and respect for cultural values in the offender.'
Women, 17-19 year olds and people aged 30 or over also appear to have a higher relative risk of receiving a community programme sentence.
Table 8.1: Logistic regression model for community programme, 1995
Note: An odds ratio of >1.0 indicates a high relative risk (i.e. more likely to receive this sentence than the reference group). The most significant variable (highest Wald Chi-square, lowest probability P), is rank '1'.
8.4 Changes in the use of community programme
The community programme (formerly community care) sentence was introduced by the Criminal Justice Act 1985. Since 1986 the use of community programme has fluctuated, reaching a peak in 1992 and decreasing markedly since then (Figure 8.3). At no point has community programme accounted for more than 1.5% of all proved cases for imprisonable offences.
Figure 8.3: The percentage of proved cases resulting in community programme for imprisonable offences, 1982-1997
The very low use of the community programme sentence (and hence the small number of sentences in the data used for this study), the fluctuations in its use, and the fact that it was not introduced until 1985 means that changes in sentencing practice cannot be accurately assessed. The decreasing use of community programme in the 1990s would seem to indicate a change in sentencing practice, given that an increase in use might have been expected due to the rapid growth of the number of cases involving violent offences (especially domestic violence) to reach a peak in 1995.
