Whilst Nuttal's study was successful in the context of prediction studies in general, it was still a long way from being 100% accurate. There is a ceiling on how accurate a prediction instrument can be because a predictor can only take into account factors that are measurable at the time that the predictor is applied. In the case of re-conviction, factors such as attitude to authority and motivation to remain within the law, which may have predictive power, are almost impossible to measure, especially if an offender believes that these factors will influence their release date.
Other factors that play a significant role in determining whether an offender remains within the law only come into play upon their release. These 'environmental factors' include such things as a supportive family, a law-abiding social group, and a regular job. Davies (1969) examined the role of such environmental factors in a study(1) that was part of the Home Office Probation Research Programme. Whilst Davies' intention was not to predict re-offending, he recorded a 'stress score' for those who participated in his study.
This stress score was based on a number of environmental factors that he speculated would have a detrimental effect on the probationer's chances of success. The stress score had considerable predictive power, achieving an M. C. R. of 0. 24. This is significant because the stress score was not the most powerful instrument that could be derived from the data, and several of the factors included proved to have little predictive power, weakening the instrument as a whole. The relationship between the stress score and the failure rate can be found in fig.
3. The findings of Lohman et al in the San Francisco Project seem to support Davies' finding that environmental factors have a strong influence on the chances of avoiding re-conviction. Lohman et al found that parolees who had successfully completed their parole attributed their success to environmental factors. It is important not to read too much into these findings, as Nuttal and Lohman were not explicitly investigating the role of environmental factors. 1 – Davies, M. 1969. Probationers in their Social Environment. London: H. M. Stationary Office.
Simon (1971) argues that such 'environmental' factors, as well as a small sample, were a factor in the failure of his study. One other possible reason for his failure that he does not discuss is the age of his chosen sample. The average age of an offender is 20(1), this would seem to indicate that most crimes are committed by young people. Clearly not all these individuals go on to become career criminals. It therefore follows that the 18-21 age range is critical for young offenders because at this time they either settle down to a normal life, or continue as recidivists.
The work of Davies seems to indicate that environmental factors are significant in determining whether an individual adopts a life of crime. The work of Simon on the other hand seems to show that the probationer's personal or criminal history has relatively little impact whether the individual will go on to re-offend. Nuttal et al conducted their study on adult prison inmates, and their results show that an offender's personal or criminal history are powerful predictors of re-offending.
A possible explanation for this is that, whilst environmental factors are significant for young probationers, they are relatively insignificant for adult offenders. If detrimental environmental factors are a significant factor in leading an individual to a life of crime, then by adulthood those vague environmental factors will be evidenced in the individuals criminal record. This could explain why Nuttal's study was so successful. However such a comparison between the studies of Simon and Nuttal is complicated by the differences in method. Therefore, as a theory, this can be no more than informed speculation.
A study in which identical measures of both personal and criminal history, and environmental factors, are applied to both young probationers and adult criminals would help to establish the relationship between environmental factors and age. In conclusion then I would argue that it is possible to predict whether adult offenders will re-offend with a fair degree of success, simply by analysing their personal and criminal history. When attempting to predict whether young people will be re-convicted however, it is important to attempt to include some measure of the environmental factors that the individual will face when they are released.