Community service
7.1 Background
Community service was introduced in 1981. An offender convicted of an imprisonable offence may be sentenced to between 20 and 200 hours of community service, if the offender consents and if a sponsor is available to supervise the offender and provide appropriate work.
As the offender is not as closely supervised by Community Probation staff as they are for periodic detention or supervision, community service is appropriate for a different type of offender and is (with community programme) considered the community-based sentence of first resort for suitable offenders.
This chapter examines which statistical factors most influence the use of community service sentences, looking first at the variables individually and then at their combined effects using multivariate modelling techniques. Changes in the use of community service are analysed in the final section of this chapter.
7.2 Single variable analysis of current factors influencing sentencing
Overall, 10% of the offenders who had a proved case in 1995 received a community service sentence.
Despite being a potential alternative to imprisonment, the community service sentence is most used for offences of relatively low seriousness, which rarely result in imprisonment (Figure 7.1a). Community service is rarely imposed for offences with a high seriousness score. This accords with Hall's comments on community service (Hall 1998, page D/468):
'The sentence of community service would appear most appropriate where the gravity of the offence and the public interest do not require a custodial sentence and where there is no apparent need for continued supervision by a probation officer.'
'The sentence is also appropriate in cases where the offender lacks the means to pay a fine.'
There is relatively little relationship between the number of charges proved in the current case and the probability of a community service sentence, although offenders with a small number of charges have a slightly higher probability.
Figure 7.1: Percentage of offenders receiving a community service sentence in 1995 for selected variables
Criminal history characteristics also show the opposite relationship with community service to that found for imprisonment and periodic detention. A higher probability of community service is associated with offenders who have a small number of previous cases, especially where the past offences are not serious and the rate of conviction has been low or the most recent offence is several years ago.
No single value of any statistical factor led to a high probability of a community service sentence. The highest percentage use of community service was found for females. Twenty two percent of proved cases involving a female defendant resulted in a community service sentence, compared to 8% for males.
Youth offenders have a low probability of receiving a community service sentence (<1%). Community service is more often imposed on 17-19 year olds (14%) than people in their 20s (9%). A higher proportion (11%) of Māori and Pacific peoples receive a community service sentence than Pakeha/Other offenders (9%).
The percentage of offenders receiving a community service sentence is higher if the most recent sentence prior to the current case was also community service (23%), than if the previous sentence was a prison sentence (3%), periodic detention (4%), community programme (12%), supervision (10%) or a monetary penalty (10%).
The offence types most likely to result in a community service sentence are property offences (12% of proved cases), traffic offences (14%), drug offences (9%), minor offences against the person (9%), and other offences against justice (6%). All other offence groups have a less than a 5% probability of receiving community service.
7.3 Multivariate analysis of current factors influencing sentencing
7.3.1 The fit and accuracy of the 1995 model
The logistic regression models for community service 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 and a non-significant residual (unexplained variation) term. The model not only fits the data it was developed on, but fits equally well to other data. Thus, the results of the test phase (Figure 7.2) show that the model developed using one half of the 1995 data (the 'fitted data') fits equally well to the other, unseen half of the 1995 data (the 'test data').
The community service model differs from the prison model in that the results are only close to the ideal line up to a probability of about 0.4 and there were very few offenders whose probability of receiving a community service sentence was greater than 0.5. Thus, there were no factors or combination of statistical factors that adequately predicted a very high probability of receiving a community service sentence. This is perhaps not surprising, given that community-based sentences are targeted at an intermediate group of offenders (those who have neither a high probability of imprisonment nor a high probability of receiving a fine or no sentence).
Also, contextual factors relevant to the specific offence and offender, but not measured by the statistical factors available for this study, presumably make a substantial contribution to the sentencing decision. Hall (1998, page D/468) makes the following comment on community service:
'The nature of the sentence is such that the criteria for determining the suitability of the sentence arise from the person of the offender and his or her circumstances, with rather less emphasis on the seriousness of the offending.'
Figure 7.2: Plot of the predicted probability versus the actual proportion of offenders receiving community service, 1995 fitted and test data
7.3.2 Results of the 1995 model
The gender and ethnicity of the offender are amongst the most significant variables in the logistic regression model for community service (Table 7.1). Women are more than twice as likely as men to get a community service sentence, once the effects of other factors have been accounted for. Māori and Pacific offenders both have a higher probability of receiving a community service sentence than Pakeha/Other offenders with similar current offence and criminal history characteristics.
Table 7.1: Logistic regression model of the probability of community service, 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'.
The reasons for the disproportionate use of community service for these groups cannot be determined using the available statistical data. The circumstances of offending may differ systematically by gender and ethnicity, or the perceived benefits of a community-based sentence may differ. If, as is argued in section 7.4, the community service sentence has increasingly become an alternative to a fine rather than an alternative to imprisonment, its disproportionate use for women, Māori and Pacific peoples may also reflect the ability of these groups to pay fines.
Offenders aged 17-19 and 30 plus have a slightly higher probability of community service than 20-29 year olds, but, as for other sentence types, youth offenders have an extremely low relative probability of community service. The lower use of community-based sentences for youth offenders is not surprising, given the availability of youth supervision and community work orders as alternative sentences.
The current offence type and the most recent sentence served are also very significant factors in the model. Traffic offenders have a high probability of receiving a community service sentence relative to property offences, but all other offence types an odds ratio less than one. Previous community service sentences increase the probability of a community service sentence, whereas previous periodic detention and prison sentences have the opposite effect. A previous breach of any community-based sentence also lowers the odds of receiving community service.
Low to moderate levels of current and previous offending increase the probability of a community service sentence, but these are generally amongst the less significant variables in the model. The relative likelihood of a community service sentence is increased by having 2-4 current charges or an offence seriousness in the low to moderate seriousness range (>1-20 or >20-180), but decreased by having a high seriousness offence, a high rate of conviction, or a short gap between the previous and the current offence. A guilty plea appears to increase the probability of receiving community service.
7.4 Changes in the use of community service
In contrast to the relatively small changes in the percentage of proven cases resulting in imprisonment, there have been very substantial changes in the use of community service over the last decade and a half (Figure 7.3). In 1982, 2.8% of all proved cases involving imprisonable offences resulted in a community service sentence, compared to 9.4% in 1997. The increased use of community service occurred over a relatively short time period (1988-1992), with a peak use at 12.4% of cases in 1992. More than four times as many offenders received a community service sentence in 1997 as in 1982.
Figure 7.3: The percentage of proved cases resulting in a community service sentence for imprisonable offences, 1982-1997
A change in the use of a particular sentence may be due to one or both of the following: (i) a change in the type of offence or offender being dealt with by the courts or (ii) a change in sentencing practice (such that offenders with the same characteristics are now more or less likely to receive a particular sentence than in the past).
Since 1983, there have been significant increases in the average seriousness of offences and the number of persistent offenders dealt with by the courts (section 3.2). As community service is more frequently used for imprisonable offences of relatively low seriousness and for offenders with relative few previous cases, the increasing use of community service is unlikely to be due to changes in the type of offence and offender.
To test the extent of change in sentencing practice, the logistic model developed using 1995 data was used to predict the probability of receiving community service for offenders in earlier years. This means that an offender in 1983 who has the same criminal history and current case characteristics as a person in 1995 would have the same predicted probability of community service. The predicted probabilities are then compared to the actual proportion of people receiving community service in each year. If no change in sentencing practice has occurred, then the predicted probability should be the same as the actual proportion receiving community service.
The very dramatic change in sentencing practice between the 1980s and the 1990s is illustrated in Figure 7.4. An offender sentenced in the 1990s was far more likely to receive community service than a person with the same statistical characteristics in the 1980s. For example, an offender whose characteristics give him or her a 30% predicted probability of receiving community service (based on the 1995 model) had only an 11% probability of actually receiving a community service sentence in 1983 and only 6% in 1987.
Sentencing practice does not appear to have changed between 1991 and 1995, suggesting that the slight drop in the use of community service over this period is due to a change in the type of offence or offender being sentenced. For example, the number of traffic offences prosecuted decreased between 1991 and 1995 and around half of the community service sentences imposed are for traffic offences.
Figure 7.4: Comparison among years of the actual and predicted probabilities of community service, with predictions made using the 1995 model
Overall, 3.0% of the offenders in 1983 received a community service sentence, compared to 2.4% in 1987, 10.8% in 1991 and 10.4% in 1995 (Table 7.2). The major change in sentencing practice occurred between 1987 and 1991, as indicated by the large disparity between the actual percentage receiving community service in 1987 (2.4%) and the predicted percentage (10.7%) based on 1995 sentencing practice.
Changes due to statistical factors appear to have had almost no impact on the use of community service. Had there been no change in sentencing practice, the predicted probability of community service would have been almost the same in 1983 (10.7%) as in 1995 (10.3%). Changes due to statistical factors do appear to explain the slight decrease in the use of community service between 1991 and 1995.
Table 7.2: Actual percentage receiving community service for each variable and year and predicted percentage based on 1995 sentencing practice
Note: The predicted percentage is the percentage of offenders who would have received the sentence had the sentencing practices of 1995 been applied, as predicted by the 1995 logistic model. The difference between the predicted percentage for 1995 and other years indicates the proportion of the total change due to changes in statistical factors (e.g. the increase in average seriousness) while the difference between the actual and predicted percentage for each year indicates the proportion of total change due to changes in sentencing practice.
The most dramatic changes in the use of community service have occurred for offences of low seriousness and offenders with a limited criminal history (Figure 7.5-1a). In 1983, the offences most likely to result in community service were those in the moderate seriousness range (offences with a seriousness score of about 100-200). By 1987 the use of community service was consistently low across most seriousness groups. In contrast, in the 1990s, offences of low seriousness (20 or less; e.g. theft, common assault, cannabis use, disorderly behaviour, most traffic offences) were the most likely to receive community service. These low seriousness offences were three times as likely to receive community service in the 1990s than in the 1980s. These changes cannot be explained by changes in the type of offences and offenders being dealt with by the courts, as indicated by the predicted percentages (Figure 7.5-1b and Table 7.2).
Figure 7.5: Actual percentage receiving community service and predicted percentage based on 1995 sentencing practice, by offence seriousness and previous conviction history, 1983-1995
The use of community service was much greater in the 1990s than in the 1980s for offenders with any number of previous cases, but the disparity is most marked for offenders with no or only a few previous cases (Figure 7.5-2a). For example, first offenders were six times more likely to receive a community service sentence in 1995 than in 1983. These changes cannot be explained by changes in statistical factors, as indicated by the predicted percentages (Figure 7.5-2b).
As indicated in Table 7.2, the increase in the use of community service has occurred across almost all categories of offences and offenders, with the exception of the most serious offences and youth offenders. Offences with a seriousness score of 180-365 or more than 365 were less likely to receive a community service sentence in 1995 compared to 1983.
The increase in the use of community service for less serious offences and offenders suggests community service is now being used as an alternative to a monetary penalty, rather than primarily as an alternative to imprisonment. Certainly the increased use of community service is far greater than could be explained by the decrease in the use of imprisonment (section 4.4), whereas the community service trend does correspond to the significant decrease in the use of monetary penalties (section 10.4).
