Abstract
The meta-analyses that have to date been published provide no support for the contention that the use of electronic monitoring (EM) in the home as a substitute for the whole or part of a prison sentence might produce any positive effect in relation to reoffending. The few studies that these analyses are based on have a number of shortcomings, however; several of them are very small, and they often fail to provide a complete description of the elements that electronic monitoring programmes include in addition to the monitoring itself. The study presented in this article has produced more positive results however. It focuses on the first 260 individuals to participate in an early release programme that included electronic monitoring in the home. This group was compared with a register-based control group. In addition to electronic monitoring by means of an ankle bracelet, it was obligatory for programme participants to have a daily occupation, which could be arranged by the prison and probation service if necessary, and they were subject to regular sobriety controls. The early release group reoffended to a significantly lesser extent than the control group did. It is not possible, however, to state to what extent this was a result of the electronic monitoring in the home or of the other elements included in the programme. When the group was trichotomised on the basis of levels of prior involvement in crime, it was found that the difference between the early release group and the control group was particularly large among those with intermediate levels of previous criminality.
This is a preview of subscription content, access via your institution.
Notes
The Campbell Collaboration is an international network of researchers that works to produce systematic reviews of research into the effects of measures employed in the school system, in social work, and in the corrections system. More information on the Campbell Collaboration is available on the network’s website: www.campbellcollaboration.org
This could logically be concluded from the selection of studies in the second review. The second review focused on the three studies that where either about back door or front door projects for offenders with a low risk for reconviction.
Intermediate and high risk refers to a reoffending rate of at least 30%.
This study was included in the review because the offenders who had participated in back door programs were regarded as a high-risk group per se. The actual reconviction rates show, however, that they should rather be considered as a low risk group.
The article does not specify how high the level of the reoffending risk was in either the group with a “moderately high risk” or the low-risk group.
In principle, inmates are released on parole once they have served two-thirds of their prison sentence.
A random sample of 50 EMB participants was drawn to be included in the interview. A telephone interview was conducted with 36 of the 50 participants in the sample, using a structured questionnaire. For half of those who were not interviewed, the reason was that it was not possible to reach them. The other half, seven individuals, did not want to participate for one reason or another.
Contrary to this study, the evaluation of the front door programme did not identify any significant differences in reconviction rates between the programme group and the control group.
Since the trial project was intended to focus on a relatively small target group, the government determined that all inmates who fulfilled the requirements for EMB should be given the opportunity to participate in the trial. It was felt that a geographical restriction of the trial project would have led to its being too small to allow conclusions to be drawn as to its effects.
Since this study focused exclusively on inmates serving a prison term of at least 2 years, two variables were excluded, since they became constants in this context. One related to drink-driving as the principal offence in the conviction, and the other related to the length of the sentence. Two further variables, namely whether the current conviction included a sexual offence, and the number of registered theft offences over the past 5 years, no longer made a significant contribution to the model when it was applied to Swedish prison inmates serving sentences of at least 2 years. These were also excluded from the model.
The use of multivariate logistic regression models to analyse the comparability of treatment and control groups has also been applied in other studies [see e.g. Weisburd et al. (1995)]. The way we employed the logistic regression model differed from that in the Weisburd et al. study in three ways, however. The first difference was that we had a population of all prisoners released during the year prior to the introduction of EMB in Sweden. We used this ‘historical’ population when we developed the logistic regression model, thereby minimising the risk that there would be systematic motivational differences between those who had undergone EMB and the control group. The second difference was that we calculated, for each offender in the study, the predicted probability of reoffending, whereas Weisburd et al. instead calculate the predicted probability of going to prison for the criterion offence for each individual. The third difference was that we combined the regression model with a twin matching model.
For a further description of the model see, e.g., Taylor (1999). The OGRS uses eight different variables as predictors for reconviction: offender’s age in years at time of sentence, gender, number of youth custodial sentences, current offence group, age at current conviction, age at first conviction “the Copas rate”, which measures the rate at which offenders are convicted, history of burglary and history of breach of court orders. When the reconviction rate within 1 year of release was assessed on the OGRS, the proportion of correctly predicted cases was 67%.
In cases where there was more than one individual in the population from which the control group had been drawn with the requisite values on both variables, the ‘twin’ included in the project was chosen at random.
The EMB project started on 1 October 2001, with the first inmates completing an EMB period being released on parole at the beginning of November 2001.
The release date refers to the date on which the term in prison was concluded. This was the same for all study participants, irrespective of whether a given inmate was in prison at the time, in the EMB programme, or in some other form of pre-release programme, such as a treatment home for drug or alcohol abuse.
During the first year, 11% of the EMB group and 15% of the control group were reconvicted. During the second year the corresponding values were 9% for the EMB group and 17% for the control group, and, during the third year, approximately 6% of both groups were convicted of new offences.
In principle, all of those who are released on parole after a term of two or more years in prison are placed on probation. This involves the paroled inmate’s being required to maintain regular contact with a probation officer and to follow certain regulations. During this period the released inmate may also be given assistance and support by the probation service. The period during which the parolees are subjected to this type of supervision lasts 1 year, but it may be extended if the former inmate breaches regulations and if the probationary period is longer than a year.
Since two or more individuals with a similar value as regards the risk for reoffending may not be assigned to different risk groups, the three groups into which participants were divided were not exactly the same size.
The conviction that led to the current prison sentence was excluded from the material analysed in this context.
The fact that the Swedish EMB model includes not only electronic monitoring but also several other measures means that it is relatively expensive. The daily cost for a client with EMB is almost as high as the cost of a place in an open prison.
In normal cases, pre-release measures focused on individuals serving a long prison sentence may be introduced up to a year prior to the date of the inmate’s release. If a number of different measures are combined, then the pre-release phase should continue for, at most, 18 months prior to the release date. In special cases pre-release measures may continue for over 2 years.
References
Bonta, J., Wallace-Capretta, S. M., & Rooney, J. (2000). Can electronic monitoring make a difference? An evaluation of three Canadian programs. Crime and Delinquency, 46(2), 61–75.
Brottsförebyggande rådet (2005). Kriminalvårdens särskilda narkotikasatsning – en utvärdering av arbetet under åren 2002–2004. (2005:3). (Stockholm: Fritzes)
Brottsförebyggande rådet (2007). Kriminalstatistik 2006. (Stockholm: Fritzes)
Finn, M. A., & Muirhead-Steves, S. (2002). The effectiveness of electronic monitoring with violent male parolees. Justice Quarterly, 19(2), 293–312.
Kriminalvårdsstyrelsen (2005). Kriminalvård och statistik 2004. (Norrköping: Kriminalvårdsstyrelsen)
Nilsson, A. (2002). Fånge I marginalen—uppväxtvillkor, levnadsförhållanden och återfall I brott bland fångar. Stockholm: Kriminologiska institutionen, Stockholms universitet.
Renzema, M., & Mayo-Wilson, E. (2005). Can electronic monitoring reduce crime for moderate to high-risk offenders? Journal of Experimental Criminology, 2, 215–237.
Sugg, D., Moore, L., & Howard, P. (2001). Electronic monitoring and offending behaviour–reconviction results for the second year of trials of curfew orders, (research finding no. 141). London: Home Office.
Taylor, R (1999). Predicting reconvictions for sexual and violent offences using the revised offender group reconviction scale. Research finding no 104. London: Home Office.
Weisburd, D., Waring, E., & Chayet, E. (1995). Specific deterrence in a sample of offenders convicted of white collar crimes. Criminology, 33(4).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Marklund, F., Holmberg, S. Effects of early release from prison using electronic tagging in Sweden. J Exp Criminol 5, 41–61 (2009). https://doi.org/10.1007/s11292-008-9064-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11292-008-9064-2
Keywords
- Electronic monitoring (EM)
- Evaluation
- House arrest
- House detention
- Offender
- Parole
- Recidivism
- Reoffending
- Tagging
- Intensive supervision by electronic monitoring (ISEM)