Stopping Violent Offending in New Zealand: Is Treatment an Option?

Berry, Steve, New Zealand Journal of Psychology

A quasi-experimental design using survival analysis and other measures showed that a community-based, residential treatment programme for violent offenders could significantly impact on reconviction patterns of a group of men previously convicted of serious violent offences. The results are important because of the paucity of data attesting to treatment success with serious violent offenders and, more specifically, with offenders who are Maori. The implications of these results and of the data collected on non-completers are discussed.

The notion that nothing works in treating offenders has been debunked (Cullen & Gendreau, 1989). However, the literature concerning the efficacy of specifically treating violent offenders is limited. For instance, Polaschek and Collie's (in press) review found few methodologically sound outcome studies to guide the development of violent offender programmes. In fact they located just four studies that reported violent recidivism outcome data on treated, generally violent offenders. Other authors (e.g., Berkowitz, 1993; Meloy, 1995; Polaschek & Reynolds, 2001; Quinsey, Harris, Rice & Cormier, 1998) have concluded that there is equivocal evidence of treatment efficacy with adult violent offenders and that new treatment technologies are needed.

The social and financial costs of violent offending are significant (Polaschek & Dixon, 2001). Violent offending is a particular problem for New Zealand, highlighted by Spier (2001), who reported that, although convictions for violence have shown a decreasing trend to the year 2000, the number of convictions for violent offences in 2000 was still 53% greater than the figure in 1991. Violent offending ranks as the most serious type of crime, attracting relatively much higher terms of imprisonment than other crimes (Spier, 2001). A report by the Department of Corrections (2002) reveals that imprisonment rates (per 100 000 population) in New Zealand (150) are high compared to other similar jurisdictions such as Canada (120), Australia (109), England and Wales (122) and Scotland (115). In 2000 the 2132 jailed violent offenders represented 27% of the 7931 people imprisoned in New Zealand's jails (the second largest grouping after property offenders). Re-offending is an issue because, in the two years after reconviction in 1995, 21% of violent offenders were reconvicted for a violent offence (Department of Corrections, 2002).

Maori make up just under 15% of the New Zealand population (Statistics New Zealand, 2001) but are heavily over-represented in convictions statistics. Maori offenders accounted for approximately 45% of all convictions for violent offences in 2000; Europeans accounted for about 37% and Pacific Islanders for approximately 15% of convictions for violence (Spier, 2001). The proportion of the prison population that is indigenous in New Zealand is 51%, considerably higher than comparison countries such as Canada (17%) and Australia (19.8%). Further adding to the dismay concerning these statistics for Maori is international research, such as that of Canadian researcher Zellerer (1994), that describes poor outcomes of treatment given to "aboriginals" (a term used by Zellerer). Zellerer concludes that non-native counsellors have difficulty helping even highly motivated clients.

One New Zealand initiative designed to address violent offending is the Montgomery House Violence Prevention Programme. The programme began in 1987 as a residential treatment option that employed social learning methodology in the treatment of groups of men who repeatedly commit serious violent offences. An uncontrolled evaluation, by Polaschek and Dixon (2001), using data generated by residents of the first "functional phase" (1987-1990) of the Montgomery House programme, suggested grounds for optimism that the programme was having a beneficial effect. Conviction rates for all those who started the programme were lowered in the post-programme phase and there were also some changes in the desired direction on psychometric measures.

Despite the promising early results presented by Dixon and Polaschek (1992a, 1992b) significant problems in the operation and integrity of the programme emerged from 1990 onwards (Hahn, 1993). The integrity of the programme deteriorated to such an extent that it was recommended that the programme should be closed or completely overhauled (Hahn, 1993). Decision-makers chose the latter option and assigned me the task of assisting with the restructuring of the programme. The problems I encountered as I commenced the restructuring task, were documented (Berry, 1994). These included; absence of integrated and updated treatment manuals; poorly defined and undocumented offender assessment and programme integrity monitoring processes; the programme had a poor reputation with referral agents; staff numbers had been depleted; the focus of the programme had shifted from preventing violent offending; resources to undertake the programme were considered inadequate and few operational guidelines were available. These difficulties were systematically addressed (Berry, 1995) and in 1995 the programme was re-opened.

The promising results noted by Dixon and Polaschek (1992a, 1992b), and later Polaschek and Dixon (2001), were contrasted by the more methodologically sophisticated longer-term study of Behrnes (1996). In his evaluation Behrnes used a pre-treatment/post-treatment quasi-experimental design with matched assignment to the control condition and almost 6 years of follow-up. Behrnes found that there were no significant differences between the completer group and the control group on any of a range of outcome measures following intervention. The Behmes study drew from the sample studied by Dixon and Polaschek (1992a, 1992b) and Polaschek and Dixon (2001).

This paper describes the programme as it ran during 1995 and 1996, and presents reconviction statistics comparing controls with a sample of violent offenders who started during that time. In contrast to the early evaluation provided by Polaschek and Dixon (2001) this study utilises a subject-to-subject matched control design. Additionally, this study is intended to add to the knowledge gained by the Behrnes study in that it reports results from participants who attended a better resourced and modified version of the Montgomery House programme.

Method

Programme Description

Montgomery House is a large house located in suburban …

Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Stopping Violent Offending in New Zealand: Is Treatment an Option?. Contributors: Berry, Steve - Author. Journal title: New Zealand Journal of Psychology. Volume: 32. Issue: 2 Publication date: December 2003. Page number: 92+. © 1998 New Zealand Psychological Society. COPYRIGHT 2003 Gale Group.

This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.