Evaluation of a Community-Based Sex Offender Treatment Program Using a Good Lives Model Approach

Abstract

In this study the authors assessed a Good Lives model (GLM) approach to sex offender treatment and compare it to a standard Relapse Prevention program. The comparisons examined (a) attrition rates, (b) treatment change in areas targeted in treatment and achievement of a posttreatment treated profile, and (c) views of offenders and facilitators. There were no differences in the attrition rates or the rates of treatment change between the two programs, indicating that they were equally effective at retaining participants and achieving change on areas targeted within treatment. Both facilitators and program participants reported the Good Lives approach module’s impact in a positive, future-focused manner. In contrast, those who attended the Relapse Prevention module did not report their perceptions and motivations in a manner that was focused on the positives in their future as frequently as those who attended the module with the Good Lives model approach.

Keywords

relapse prevention sexual offender treatment treatment outcome

Introduction

The successful treatment of sexual offenders is understandably an important goal, due to its implications for society in general and for the individual offender. To reduce the likelihood of future offending, it is integral that the treatment being offered is the most effective available. Mounting evidence suggests that programs adhering to a risk-need-responsivity (RNR) approach are effective for reducing recidivism in sexual offenders (Hanson, Bourgon, Helmus, & Hodgson, 2009). According to this approach, the highest-intensity treatment should be offered to the highest-risk offenders (risk principle) and treatment should target criminogenic need factors (the need principle), that is, characteristics that are potentially changeable and have a demonstrated relationship with recidivism and are offered in a manner that maximizes the likelihood that the individual benefits from it (i.e., matching treatment to their abilities and motivation—responsivity principle; Andrews & Bonta, 2007). Alongside the drive for providing effective treatment, there is increasing recognition that treatment should be offered in a manner that highlights the inherent value and rights of the individual offender (Ward, Gannon, & Birgden, 2007; Ward & Stewart, 2003) and the importance of offering treatment in a manner that is appealing to the individual to increase his or her engagement (Ward, Mann, & Gannon, 2007). Such positive approaches have the potential to complement and perhaps improve upon the RNR approach (Ward, Mann, et al., 2007; Wilson & Yates, 2009).

This movement toward a more positively oriented (as opposed to risk management or avoidance oriented) treatment approach holds a number of benefits (Marshall et al., 2005), particularly with regard to the “Good Lives model” rehabilitation framework (GLM; Ward, 2002; Ward, Mann, et al., 2007; Ward & Stewart, 2003). According to GLM, all individuals seek a set of primary “goods,” including states of mind, personal characteristics, and experiences, which are sought for their own inherent value. It is theorized that sexual offending arises as a result of an attempt to obtain these goods in an inappropriate manner, out of frustration at being unable to achieve these goods or out of an imbalance between the goods so that some goods are prioritized over others (e.g., sexual gratification over emotional intimacy). Hence, treatment should be about instilling knowledge, skills, and competence to lead successful and satisfying lives, which should be incompatible with offending. Avoidance-focused Relapse Prevention (RP) approach has been the dominant model internationally until relatively recently when more positively focused rehabilitation frameworks such as the GLM emerged (e.g., Ward, 2002).

In line with the ideas of increasing motivation for treatment, some have suggested that treatment effectiveness may be maximized by using a more positive approach to treatment, focusing on an individual’s strengths, enhancing skills, instilling prosocial attitudes, and increasing the client’s self-worth rather than approaching treatment from negative or avoidance terms (Fernandez, 2006; Marshall et al., 2005). Indeed, Mann, Webster, Schofield, and Marshall (2004) found that therapists perceived the sexual offenders in their treatment groups to be more genuinely motivated to live life without offending by the end of treatment and using a more positive, goal-oriented approach compared to using avoidance treatment goals. Craig, Browne, and Beech (2008) noted that psychological characteristics such as an internal locus of control, self-efficacy, and motivation are important individual “strengths” that may act to decrease an individual’s chances of committing a new sexual offense. Hence, positive approaches such as the GLM that aim to instill strengths can potentially help in reducing offending.

Furthermore, the theoretical support for the GLM approach is promising (see for detailed descriptions, Ward & Maruna, 2007; Ward, Polaschek, & Beech, 2006). It is a coherent, clearly articulated approach that can accommodate aspects of both the RNR and the positive psychology approaches, each of which are well established and respected (Ward & Maruna, 2007). The external consistency of the GLM is highlighted by the fact that the primary goods outlined in the model converge with those derived from a wide array of different disciplines (Ward et al., 2006). Using a GLM approach is likely to increase engagement and motivation while also addressing risk management (Ward & Maruna, 2007).

It should be noted that a number of criticisms of the GLM have been expressed. More traditionally, treatment has been approached using the principles of RNR, which have a strong empirical basis (see Andrews, Bonta, & Wormith, 2006, for a review) and are used as a guide in providing effective treatment. One criticism of the GLM is its theoretical focus in the absence of empirical support (Bonta & Andrews, 2003). It has also been noted that in criminal justice settings, which are often limited in resources, it is important to focus on factors in treatment that have been empirically demonstrated to reduce recidivism (Ogloff & Davis, 2004). Concern is also expressed that if efforts are directed toward addressing issues that may not be directly related to reoffense (i.e., self-esteem) at the expense of focusing on empirically supported risk factors, this may actually increase the likelihood of reoffending (e.g., increasing pro-offending attitudes; Ogloff & Wong, 2003). In fact, Ogloff and Davis go so far as to say that they do not believe correctional services system should implement changes in line with the theoretical positions of the GLM until it has evidence supporting its use.

Others have argued that the GLM and the more traditional focus on RNR should coexist as the two are not mutually exclusive (Ward, Mann, et al., 2007, 2006). It has been suggested that the GLM may converge with the principles of RNR via responsivity issues (Ogloff & Davis, 2004; Ward & Stewart, 2003). This would allow recognition of empirically supported principles for administering treatment (i.e., RNR) while also highlighting the benefits of recognizing the individual’s inherent striving for a meaningful and fulfilling life (i.e., the GLM). Ward and colleagues (2007) note that psychological well-being should play a role in rehabilitation programs, alongside risk management. They note that treatment with sex offenders should have the twin focus of promoting goods and managing or reducing risk.

In light of the limited empirical, but sound theoretical support for the GLM, the aim of this research is to examine the short-term effectiveness of the GLM-derived module (referred to as the Better Lives module) in the Northumbria Sex Offender Groupwork Program (N-SOGP) in the north of England. This is examined in terms of attrition rates, treatment change in areas targeted in treatment and achievement of posttreatment treated profiles (assessed by psychometric measures), and the views of facilitators and participants on the program. The GLM is expected to retain more participants than the RP module because it will be perceived as a more engaging and motivating module to attend. It is also expected that the two programs may differ in terms of treatment change on relevant treatment targets (i.e., pro-offending attitudes, socioaffective functioning, and relapse skills). Specifically, the RP component places greater emphasis on specifically addressing these issues; therefore, those who attended this program may show more improvements in these areas. Finally, it is expected that both the offenders and facilitators will have more positive perceptions of the GLM–derived module due to its more optimistic, goal-oriented approach.

Study Objectives

Specifically a number of research questions are examined:

Are there differences in attrition rate between the two different modules (i.e., BL and RP)?

Are there differences between those who attended the two different modules in terms of treatment change in pro-offending attitudes, socioaffective functioning, relapse skills, and their overall posttreatment treated profile?

What are the facilitators’ experiences and thoughts about running the revised version of the N-SOGP (including the Good Lives component) compared to the standard RP module?

How do the perceptions of the participants differ, depending on which module they attended?

Method

Setting

In England and Wales, the National Offender Management Service (NOMS) is an executive agency of the Ministry of Justice that enables the effective treatment delivery for prisons and probation to protect the public and reduce reoffending. NOMS oversees geographical areas that each encompass a number of probation areas. One of these geographical areas is Northumbria in the northeast of England. The N-SOGP is offered by 12 Probation Areas in the United Kingdom, and consists of two components, which last for 180 hr in total. The first is the Core module, which is attended by all medium- and high-risk and deviance, offenders, who are either on license (i.e., serving part of a custodial sentence in the community) or subject to a community order (i.e., serving all of their sentence in the community). This Core module comprises four blocks, which look at offense-specific areas such as victim empathy, problem solving, and cognitive distortions. The second module offered was previously the Relapse Prevention (RP) module, but this has recently been replaced with the new Better Lives (BL) module, which is derived from the GLM rehabilitation theory. In line with the risk principle, low-risk men only attend the second component (either the BL or RP module), whereas medium- or high-risk men have first attended the Core module described above before attending the RP or BL module. The second module (RP or BL) lasted for 36 hr. The aim of Module 2 (whether it be RP or BL) was to teach the offender alternative ways of behaving and skills to employ should he or she feel vulnerable or likely to reoffend in the future. Although the RP module concentrates on the standard avoidance goals approach, the BL module focused more on approach goals.

In its broadest sense, the BL module aimed to steer the client toward the acquisition and possible enhancement of the following “goods”: (a) life (i.e., healthy living and a high level of personal functioning), (b) knowledge acquisition, (c) achievements both in work and play (including the ability to be good at something), (d) excellence in agency (i.e., being in control and the ability to get things accomplished), (e) inner peace (i.e., lack of stress and inner tension/ emotional disregulation), (f) friendship (including intimate, romantic, and family relationships), (g) community (i.e., involvement with others beyond intimate/family relationships), (h) spirituality (in its broadest sense of finding meaning and purpose in life), (i) happiness, and (j) creativity. This module comprised three main sections as outlined in the program manual for the N-SOGP Relapse Prevention Community Better Lives (NOMS, 2007). Sessions began with a news round that allowed facilitators to check motivation and allowed for discussion of any issues that had arisen since the session prior and ended with open spaces to reflect on the offenders’ thoughts and feelings following the session.

The first section (Sessions 1-4) aimed to build motivation and acquaint offenders with the treatment approach. It also took an audit of the goods to be acquired for each individual and possible obstacles (i.e., dynamic risk factors) that may interfere with acquiring these goods. In particular, the first session served as an introduction to the module and important concepts, such as the wheel of life. This was used to illustrate the different goods people seek in life, with each good represented by a section of the wheel, and to highlight that people should be aiming for a wheel that is well rounded. Other exercises used in the first session included work on the concept of being prepared for the future and anticipating challenges that may arise, possible consequences of not being prepared, and an exercise to generate ideas of what sort of support will be needed to achieve their goods. In Session 2, the participants presented their offense accounts and were introduced to the Good Life plan, which they were told would eventually be comprised of their final wheel of life at the end of treatment, and their list of obstacles. Sessions 3 and 4 involved the presentations of individual’s wheel-of-life exercises and lists of obstacles. Links were explicitly made between the obstacles and personal treatment targets. In these sessions, care was taken to illustrate how the obstacles discussed could be grouped into four main areas that are identified as being linked to sexual offending (i.e., dynamic risk domains: sexual interests, attitudes supportive of offending, management of relationships, and self-management).

The second section (Sessions 5-11) focused on skills practice in relation to meeting individual needs in prosocial ways and managing obstacles. Specifically, the fifth session focused on the obstacles identified for each person. This involved the individual enacting a role-play scenario representing one of their biggest obstacles with consideration of how he or she would have responded to the obstacle in the past and how he or she would address it in the future. The sixth session had a psychoeducational component, with a focus on increasing awareness of relationship problems and developing intimacy, followed by skills practice related to developing love, friendship, and intimacy. Session 7 focused on attitudes supportive of sexual offending through identification of offense-supportive and appropriate sexual attitudes and skills practice by using sexual attitude cards for which each individual has to identify whether the card represents a healthy or unhealthy attitude. Session 8 focused on how to monitor sexual thoughts and attitudes, ways to change unhealthy sexual thoughts and attitudes, consideration of how attitudes are developed, development of realistic expectations about a healthy prosocial sex life, and consideration of how to overcome deviant fantasies and urges. The ninth session focused on identification of and skills practice for expressing and managing emotions. The tenth session introduced the use of a decision matrix to assist in problem solving by reminding the participants to consider the consequences of their actions, followed by a role play to practice dealing with any obstacles related to self-management problems. Session 11 highlighted the importance of support networks and included role plays of the participants disclosing their offenses to relevant others.

The last section (Session 12) involved presenting a Good Life plan, an action plan for maintaining change that drew upon the learning from the module and the formation of future lifestyle goals and targets.

Participants

The participants were men attending the nationally accredited N-SOGP (N = 777) on either the new BL module (n = 76) or the previous RP module (n = 701) in one of six probation areas: Cheshire, Durham, Humberside, Merseyside, Northumbria, or South Yorkshire. The mean risk score for the sample was 1.9 (SD = 0.8) on the Risk Matrix 2000, indicating an overall medium risk level for the sample. In terms of the gender of the victim of the index sex offense, 63.2% (n = 491) were female, 12.6% (n = 98) were male, and 24.1% (n = 187) had victims of both genders, and this information was missing for one offender. Most of the victims were under age 16 (89.1%, n = 692), with the remainder having adult victims. The majority of the victims were unrelated (74.3%, n = 578). In terms of ethnicity, the majority of the offenders were White (96.0%, n = 746), 1.4% Asian (n = 11), 0.4% Black (n = 3), 0.6% “Other” ethnic minority group, and 1.7% (n = 13) were of unknown ethnicity. Attrition information was provided by treatment managers or offender managers for a subset of 269 participants. From this subsample, the attrition rate was 1.5% (n = 4).

Interviews were conducted with 20 participants with 5 of those having attended the RP module and 15 having attended the BL module. Initially, offender managers in the identified areas were contacted to request the participation of the offenders on behalf of the researchers. The offender managers who responded indicated that there were 46 individuals who had completed either the RP or BL module and were still available to be interviewed. Of these 46 participants, 43.5% (n = 20) agreed to be interviewed. The goal was to have equal numbers of participants from each module, but the number of available RP participants was limited because many of those who had completed the RP module before the BL module was implemented had completed their sentences and so were not able to be contacted. The 15 individuals from the BL module who took part in the interviews were those who were willing to take part. All of the participants who were interviewed were White British males.

Interviews were also conducted with 11 facilitators. These facilitators were identified by treatment managers in the area who put forward the names of people who were willing to be interviewed. As these individuals were put forward to us, we are unsure how many of the remaining facilitators were unwilling to take part. The facilitators interviewed all had experience with the BL module. Some of them had also facilitated the RP module and so were in a position to compare the two modules.

Measures

The following measures comprise a psychometric test battery that is routinely collected for sex offender treatment programs and collated centrally by NOMS. A database containing the psychometric results for those who had attended the N-SOGP, either the RP or BL, was provided for examination in this study.

Self-Esteem Scale (Thornton, 1989; Webster, Mann, Thornton, & Wakeling, 2007)

This is an eight-item questionnaire devised by Thornton in which clients answer true or false to questions regarding how they feel about themselves. Thornton (personal communication) reports the scale has high internal reliability (α = .80). Beech (1998) reported a test–retest reliability of .75 in a group of 40 untreated sexual offenders. A higher score on this measure indicates higher self-esteem.

UCLA Loneliness Scale (Russell, Peplau, & Cutrona, 1980)

The UCLA (University of California, Los Angeles) Emotional Loneliness Scale is a 20-item scale designed to detect variations in loneliness that occur in everyday life. The authors report high internal consistency (α =. 94). Beech (1998) reported the test–retest reliability to be .70 in a group of 44 treated child molesters. In terms of scoring, the higher the score, the more emotionally lonely an individual is reporting him or herself to be.

Interpersonal Reactivity Inventory (IRI; Davis, 1980)

The IRI is a 28-item scale that measures general empathy in terms of cognitive and emotional components. It measures four dimensions: perspective taking, which measures the ability to take the cognitive perspective of another; the empathic concern, which measures feelings of compassion and concern; the fantasy scale, which measures the ability to identify with fictional characters; and the personal distress scale, which measures the individual’s ability to self-regulate in coping with negative feelings. Davis (1980) reported an internal reliability of .78 and a test–retest reliability of .68 in 56 adult males over a 2.5-month period. Only the distress scale was used here.

The Social Response Inventory (SRI; Keltner, Marshall & Marshall, 1981)

This measure is a self-report questionnaire that measures assertive behavior in a variety of social situations. The questionnaire asks the respondent what he or she will do in a given social situation (intention) and tests his or her knowledge as to what he or she believes he or she should do. There are 22 situations that have a range of responses that demonstrate varying levels of assertiveness ranging from (–2) extremely underassertive to (+2) extremely overassertive. Two scores can be derived from this measure: overassertiveness and underassertiveness. In this study the underassertiveness score was used. The higher the score on underassertiveness, the higher the individual’s self-reported tendency to react under assertively in social situations. Marshall found that when three independent judges were asked to rank each set by alternatives along an assertiveness dimension there was 100% concordance in their ratings. Beech (1998) found the test–retest reliability to be .80 over a 7-month period in a group of 44 treated child molesters.

Nowicki–Strickland Locus of Control (Nowicki, 1988)

This is a 40-item questionnaire that measures “the extent to which subjects feel that events are contingent on their behaviour and the extent to which they feel events are controlled externally” (Nowicki, 1988). Nowicki and Duke (1974) reported the test–retest reliability to be .83 in 158 participants over a 6-week period. They also reported the internal consistency of this scale to be .69. The higher the score, the more the individual believes that he or she, and events in his or her life, are externally controlled.

Victim Empathy Scale (Beckett & Fisher, 1994)

This is a 28-item scale devised by Beckett and Fisher (1994) to gauge sex offenders’ views of the impact of their offending on their victim. This questionnaire is filled out on their most typical victim or a general scenario. The scale has high internal reliability (α = .90). Beech (1998) found this scale to have a test–retest reliability of .95 in 46 untreated sexual offenders. Beech also found the scale to have an internal consistency of .89 in 140 untreated child molesters. The higher the score on this measure, the less able the individual is to demonstrate appropriate identification with the victim’s point of view.

Beliefs About Children Scale (BACS; Beckett, 1987)

This is a 30-item scale developed to measure sex offenders’ distorted beliefs about children’s sexuality. It yields a Cognitive Distortions scale score and an Emotional Congruence scale score. Twelve Lie items are also embedded in the questionnaire.

The Cognitive Distortion scale is a 15-item scale designed to assess an individual’s beliefs about children and their sexuality. Questions are scored on a 4-point scale and include items such as “Children can lead adults on.” The scale has high internal reliability (α = .90). Beech (1998) reported the test–retest reliability to be .77 in 45 untreated child molesters.

The Emotional Congruence scale is a 15-item scale designed to measure the extent to which individuals can understand, relate to, and identify with what they believe to be the thoughts, feelings, and concerns of children. Beech (1998) found the test–retest reliability of this measure to be .63 in 45 untreated sexual offenders. The higher the score, the higher the individual’s self-reported level of emotional congruence.

Relapse Prevention Questionnaire (Beckett, Fisher, Mann, & Thornton, 1997)

The Relapse Prevention Questionnaire is a 12-item questionnaire that assesses an individual’s awareness of their risk situations and associated thoughts and feelings and their strategies to deal with these. Each response is scored from 0 to 2 according to how well the offender’s response meets certain conditions (e.g., ability to identify coping strategies) for that question with higher scores representing better responses. Test–retest reliability was not available for this measure.

Procedure

The attrition rates and treatment change information were compared between the BL and RP modules. Within the currently running or recently completed groups, offenders’ and facilitators’ perceptions of the two modules were investigated using semistructured interviews.

Data sources. Data was collected from a number of different sources: Psychometric data

Psychometric data for all individuals who attend sex offender treatment are collated centrally. A database of 777 individuals who attended the N-SOGP was provided, which included demographic information, psychometric scores, treatment module details, and each individual’s calculated treated change and posttreatment treated profile. The measures were combined into scores indicating whether the participants had achieved an overall posttreatment treated profile on the pro-offending attitude measures (i.e., BACS Cognitive Distortions and Emotional Congruence With Children scales, and Victim Empathy Scale), socioaffective measures (i.e., Self-Esteem Scale, UCLA Loneliness Scale, SRI Underassertiveness, IRI Personal Distress, and Nowicki–Strickland Locus of Control) and relapse prevention skills (i.e., Relapse Prevention Questionnaire). For details on how the treated profiles were calculated, see Mandeville-Norden, Beech, and Middleton (2010).

File review

Information related to recent treatment attendance and treatment attrition was provided by offender managers.

Interviews

Semistructured interviews were conducted by the first three authors with the offenders and facilitators upon completion of the module to ascertain their perceptions of the module. See Appendices A and B for the interview questions. Interrater reliability for the qualitative analysis was conducted by the fourth author. The kappa statistic was 0.67, which indicates a “substantial” level of reliability (Landis & Koch, 1977).

Analytic Approaches Used in the Study

Psychometric data were collected pre- and post-treatment and are collated and stored in a central location. An algorithm, based on change from pre- to post-treatment and how far the individual’s psychometric scores deviate from the scores of a nonoffender, is applied to these data to determine whether the individual demonstrated treatment change (yes/no) on the treatment targets of pro-offending attitudes, socioaffective functioning, and relapse skills, as well as whether or not their overall treated profile indicated change (see Mandeville-Norden et al., 2010, for details of the procedures for calculating these scores). The percentage of participants who demonstrate treatment change was compared between the two modules.

Quantitative and qualitative methods were used. Attrition rates and differences in treated profiles were examined using chi-square analysis. Content analysis was used to examine the qualitative data using inductive category development as per the procedure outlined by Mayring (2000). Responses to interview questions were coded into “themes.” The percentage of clients whose responses fell into each theme was calculated, and comparisons were made between the two modules.

Results

The results are reported under the following headings: Attrition, Treatment Change, Facilitators’ Perceptions, and Participants’ Perceptions, including comparison of RP and BL participants, and BL participants’ perceptions.

Attrition Rates

Attrition information was available for 182 participants who attended the RP module and 87 who attended the BL module. The attrition rates between the groups did not differ significantly—χ²(1) = 0.58, ns—with an attrition rate of 1% for the RP module and 2% for the BL module.

Treatment Change

Information on treatment change was available for 643 offenders, with some missing information depending on whether all the psychometrics needed to determine the posttreatment profile had been completed by the individual. There were no significant differences between the modules for the proportion of individuals who had achieved treatment change for any of the following: For pro-offending attitudes, 70.4% demonstrated treatment change for the RP module (n = 366 of 520) and 69% for the BL module (n = 46 of 67), χ² (1) = 0.08, ns; for socioaffective functioning, 60% of the RP (n = 341 of 567) participants and 66% of the BL (n = 50 of 76) participants demonstrated treatment change, χ²(1) = 0.90, ns; for relapse skills, 66.5% of RP (n = 276 of 415) participants and 71.7% of BL (n = 38 of 53) participants demonstrated treatment change, χ²(1) = 0.57, ns; and for the overall posttreatment treated profile, 53% of RP (n = 197 of 368) participants and 55% of BL (n = 26 of 47) participants achieved a treated profile, χ²(1) = 0.54, ns.

Interviews

Responses to each interview question were not available for every participant interviewed, due to difficulties in understanding or hearing some responses on the recordings.

Facilitators’ perceptions

Eleven facilitators were interviewed about their thoughts on the BL module. The results of the analyses are shown below. These were individuals who had facilitated the BL module, but a number of them also had experience running the RP module so they could comment on both modules and the differences between them.

Overall thoughts on the module

Most (10 of 11) facilitators noted that they and/or the group members liked the positive and future-focused elements as exemplified by the following quote:

I think that the program has a good forward thinking feel about it and I like the fact that in the talking about the goods, we’re emphasizing to the men they’re not different to any other human being but the same as any other human being and everybody wants to pursue these goods and that they’re not . . . for once, they are not being treated as someone who is . . . who are a lot different. So I think it has that forward looking positive appeal to it.

Some (7 of 11) did not think the module was appropriate for everyone. In particular, they did not feel the module would be appropriate for high-risk or unmotivated men.

I think it’s very good to work with men who have made a lot of progress and who are quite motivated towards change but I think it’s not so good with men who are not that motivated and I don’t think it’s good for men who are risky . . . there seems to be things, the section on managing sort of sexual risk factors like managing sexual fantasy, there was an awful lot [covered] in that session and there was some elements of the [previous] program, which was managing sexual urges and the problem of immediate gratification and so on, that were given much more time in the [previous] program and I think for some men, particularly men who are risky, they are really important areas that were kind of cramped together and not given a lot of time [in the BL module].

Some of the participants also commented that it did not appear appropriate for those men who had not done any previous treatment because they did not have the skills and terminology to be at the same level as the rest of the group members. Many (8 of 11) thought it was missing important elements; in particular 60% of the facilitators thought that the BL module was missing a sexual offending/risk component, highlighted by the following quotes:

I think it’s good to have a positive, looking forward program and I think, I think the men get into that. As I say, it’s the bit about their sexual offending sometimes gets lost in all of that positive bit, so it would be good to have that in there.

I think it’s a good program but I think, if it’s compared to the previous Relapse Prevention program I think the previous Relapse Prevention program had some strengths that weren’t included in this program. Well, my thoughts are that generally it has a lot of good aspects. I quite liked some aspects of the old Relapse Prevention program and you know, in some ways I think something that would combine the best qualities of both might have suited some of the offenders better.

Influence of the module on their motivation

Many (8 of 11) facilitators thought the module appeared to have a positive influence on the men’s motivation, as exemplified by the following:

The change to this program, I think, in kind of an unspoken kind of a way, it has a different feel to it right from the start, you know it kind of . . . I mean relapse already did it, it was always a different kind of a course and you’d come out of the intense self-analysis into something a bit different. I think it [BL] is almost metaphorically like throwing open a pair of doors and stepping out into the light. I think it has that effect on some of them and you can see some guys they are looking at the wheel¹ and stuff and think, you know, thinking “ah this really makes sense to me.”

I think it is motivational in the sense that it’s, again, the positive nature, it’s not de-motivating or depressing, you know? It is motivating because, it you know, the same way that motivational interviewing is . . . you know what I mean?. . . something about “it is possible to go forward, you have permission to go forward, you are capable of going forward, and having some of the nice things other people have.”

How Better Lives differed from other programs

Many (8 of 11) of the participants made positive comments in terms of how BL compared to other sex offender treatment programs. The following quotes highlight this perspective:

Overall it’s probably an improvement on the previous one, but it kind of, lacks a little bit of focus on risk issues, I would say.

It’s very different to the core program but absolutely rightly so. I think quite a lot of practitioners struggle with this idea about moving, about focusing on treatment and talking about treatment with the men and not kind of being focused on “this is a risky issue, therefore you’re a bad person, you must do something about it,” and kind of keeping this idea that this is about them developing, their development, managing their risks and understanding their thoughts and processes. So I think its, its very different from the core which does tend to look at “What have you done?” and “Why did you do that?” and, if that was wrong therefore what’s right? So it’s a positive move away from that. I think the right and wrong can be left for the offender managers . . . and there is a good balance now between the core and the Better Lives.

I’ve run the old RP programs on a few occasions. I think it’s very different from that. I think the old program was very much about, for me . . . what you shouldn’t do . . . yeah . . . things to avoid, strategies to deal with risk, not that there is anything wrong with that, but really talking all the time about quite negative aspects of behavior . . . whereas this program is framed a lot more in sense of positives, about looking at, . . . helping them think about themselves, how they . . . about others around them, how they run their lives in a more general sense, how that actually contributes to risk management. I think the model that “Better Lives” uses is also more accessible for people who are using it, by which I mean the clients. The old RP model, I think is harder to grasp for the people I’ve worked with than this one would be.

Although there were concerns and suggestions noted about the BL module, no one reported thinking that other programs were superior.

Participants’ perceptions

The perceptions of participants who completed the BL module were examined. The mean time from the end of treatment to the interview date was 11.2 (SD = 4.8) months for those who attended the RP module and 2.9 (SD = 3.5) months for those who attended the BL module. These were first contrasted with a small number of participants who completed the older RP module. Then the responses of the BL participants were examined in more depth.

Comparison of RP and BL

Areas in which differences might be expected between those who attended the RP and BL modules were examined. It should be noted that the comparisons are based on only a small number of responses for the RP group (n = 5), so statistical analyses were not completed. Although efforts were made to ensure each participant was interviewed in the same manner, some participants were not asked or did not answer some of the questions or did not provide coherent answers to the questions. Other questions were asked only as follow-up questions in some early interviews but were found useful and so were added to the interview schedule for the remaining interviews.

Overall thoughts (positive or negative)

The responses of the majority of people on each of the modules generally indicated that they thought the module was positive or had positive aspects (BL: 83%, 10 of 12; RP: 80%, 4 of 5). Reponses for both included that they had a better understanding of themselves and that they had gained knowledge in terms of how to prevent reoffending. More of the BL group (27%, 3 of 12) compared to the RP group (0%, 0 of 5) noted that they liked that the module was positive and focused on the future. A small percentage who attended each module noted negative aspects, including one person from each module noting that they thought it was boring because they had learned it all before.

Areas for improvement

Many of those in the BL group (64%, 9 of 14) reported that they did not think there were any areas of the module that needed to be improved, whereas all of those from the RP group identified at least one area for improvement. Individuals from both groups did not feel people with mixed levels of knowledge (i.e., those who had not taken previous courses and those who had) and commitment should be included in the same group. One individual on the RP group felt the group should have been more forward looking. This is highlighted in the following quote:

I didn’t really enjoy the RP as much as I enjoyed the main group . . . because I thought that the RP element was . . . it should have been more forward looking. It should have been more about the “you” that you want to be, not the “you” that you were. It shouldn’t have taken you back over the bad aspects of you. . . . I want to look at positive things now. I want to look forward to, you know? . . . to sort of think more positively about the future. I’m fed up of going back. We’ve dealt with that, you know? I thought relapse prevention should have been more about the future than about the past. You know, you lose your house, you lose your job, you lose your wife, you lose your children, you lose everything and you know . . . well, I certainly did. . . . You’re alone in the world and you are struggling to survive and you got “this” problem as well. So I thought, I got enough, I need to think more positively. If I go . . . if you start going back, you know, one of the attitudes I think leads to offending is “I don’t give a toss,” “I really don’t care” . . . I have a suicidal history and if I get into that state then we really could be in problems.

Changes in understanding themselves

Those who attended the RP module reported more frequently that they improved their understanding of themselves as related to their offending (80% [6 of 13], compared to 46% [4 of 5] for those who attended the BL module). Those who attended the BL module reported more frequently that they had a better understanding of the positive aspects of themselves, rather than improved understanding specifically related to their offending (61% [8 of 13] for BL, compared to 20% [1 of 5] for RP).

Changes in thoughts and attitudes

Those who attended the RP module reported more frequently that their thoughts and attitudes had changed in a way that they were better able to manage themselves or their reoffending (80% [4 of 5] for RP, compared to 27% [4 of 15] of those who attended the BL module). Those who attended the BL module reported more frequently that their thoughts and attitudes about themselves or the future were more positive (47% [7 of 15], compared to 20% [1 of 5] in the RP module).

BL participants’ perceptions

The participants who attended the BL module were interviewed using a structured interview for their thoughts on the module. The main responses are discussed here.

Overall thoughts on the module

Most participants (10 of 12) reported their overall thoughts of the module in positive terms, using descriptions such as “enlightening,” “informative and worthwhile,” and “deep, useful, and enjoyable.” Three of the 12 participants specifically noted that they appreciated that the module was positive and future focused, exemplified by the following quote:

For me it was actually, going from doing the core program to do [Better Lives], was a complete change around. Whereas the Core [module] was looking at what actually happened previously, offending etc., [Better Lives] was a more positive look on the future basically, sort of, more along the lines of opening doors. Well, basically just, showing the path and well, how to walk down it.

Some also noted that they felt that they had gained a better understanding of themselves (3 of 12), and others noted that they had gained a better understanding of how to stop offending (2 of 12). Some individuals made negative comments about the module (3 of 12). One individual reported that

It was ok. I think some people got a lot more out of it than I did because there were some people who hadn’t done the main NSOGP course which I did and umm, I really learned, most of my knowledge is from the NSOGP . . . but yeah, it was ok.

Another reported, “I found it boring. I’ve done it all in jail, it was a similar, basically all the same thing. It was boring, that’s what I thought.”

Impact of the module

One individual reported that the module did not have an impact on him because he had learned it all before. Some participants (4 of 15) responded in a manner reflecting that the most important aspect of the module for them was that it allowed them to see a positive future exemplified by the following quote:>

I think it helped us see that I had a future because at the time I was feeling . . . . I didn’t have any self-esteem at all. I was really down, self-harming, a lot of things. I was just . . . it helped to see that I’ve got reasons to live and that I can, that I can be a good person.

Some participants’ responses (5 of 15) highlighted that they had gained insight into their offending and/or how to prevent future offending. An example of such a response was

From start to finish it was all very important because it’s . . . it’s sharp, its something which once the surface is scratched and you get deeper and deeper and deeper into your problem and into yourself, its quite upsetting and suddenly coming around to . . . you gotta climb a big hill here and its been preparing for what you gotta go through, to reach that goal.

Participants’ responses (4 of 15) also reflected that they had felt one of the biggest impacts the module had was self-improvement and self-awareness. One participant reported:

It’s had a big impact on me to be honest. Yeah, I’ve learnt more about myself in the 12 weeks [of the BL module] than I did all the time I was in prison. I did 12 ½years in prison and I did the core program in prison. But, umm this has taught a lot about myself.

Areas for improvement

There were no consistent responses in terms of improvements that the men would suggest. In fact, the majority of participants suggested that there were no areas that needed improvement (9 of 14). Some suggestions that were made by single participants were as follows: more role plays, more breaks because it is difficult to remember all of the material covered in 90 min, to “liven it up,” more challenging about your future by the facilitators, that perhaps the groups could be longer when there are a lot of participants,² that people should be informed that “everyone shows resistance at first but that their walls will come down eventually,” and that people with mixed level of commitment should not be in the same group because it brings down the people who are motivated and committed.

Impact of the module on thoughts of the future

Some individuals noted that they did not think that the module had any impact on their thoughts of the future because they had learned the material in prison or because they had already been offense free for many years and so they did not think that the module could help them (3 of 14). A number of participants reported that they felt the module helped address their problems related to offending and helped to prevent future offending (7 of 14). An example of such a response is

It was upsetting because during, leading up to me offending, during my offending I was very self-centered, a very selfish person, ignorance of a number of things. During, during that program, its more self-awareness, don’t be ignorant to what’s around you, what have you got in your tool bag now that’s gonna get you to the top of that mountain, lets go through these, lets check these. It was fascinating.

Some participants’ responses (5 of 14) reflected that they were more positive and hopeful for the future. One participant responded,

Just helped to see that I have a future and that I wasn’t a bad person for what I’ve done, some people may disagree but . . . I mean . . . well I am in a way in the sense of what I did, but in the whole.

Better understanding of self

Some participants (6 of 13) reported that they had a better understanding of themselves as it related to offending, exemplified by the following:

I have a better understanding of myself and of why I offended. Also, now I would say I’m a better person now, or I’d like to think I am anyway. So I would say better understanding as of why I’ve done it and what I can do to rectify it. I have got a better future now so. . . . [This was also counted as a response indicating a more positive perception of self/future.]

Many of the participants (8 of 13) also indicated that they have a more positive perception of themselves/their futures. An example of this is

I can see my own sort of short falls, the areas I can do with improving on, no matter how minor. But at the same time I can also see what areas are also stronger.”

Anticipated “positives” in the future

Two participants (2 of 13) noted that their future is positive because they have a better understanding related to offending. The majority of individuals’ responses (11 of 13) indicated that their futures were generally positive in that they are better people now or have better/more positive futures. Examples of such responses are the following:

Well, the positives in my future. . . it’s got to be, I have perception of what life should be. Life should be tranquil, life should be loving and life should be happy and life should be contentment. They were not in my life before. They have come into my life now and I’ve got, I’m not very good at words, but I got this burning desire for these things to come into my life and it would be good for [wife], it would be good for me.

Well at the minute, just that I can have a life, that I can have a family and friends and a good job, just have a life again.

Discussion

The present study examined the effectiveness of the N-SOGP Better Lives module, which was derived from a GLM rehabilitation framework. Neither the overall attrition rate nor the percentage of participants who achieved treatment change or a posttreatment treated profile differed between the two groups. The BL group was viewed as more positive and future oriented in several aspects by both the facilitators and the participants. The facilitators offered a number of suggestions in terms of changes that they would recommend. Comparisons between those who attended the old RP module and those who attended the BL modules revealed that there are aspects that are more positive or future focused, and less offense focused, about the BL module. However, participants who attended the BL module still reported that they had gained a better understanding of how to avoid future offending as well.

In terms of the attrition rate, the two modules did not differ. It is likely that this is a result of many of the men being required to attend treatment. There are negative consequences (e.g., return to custody) for the men if they leave the group; so many would feel compelled to remain, even if they did not find it beneficial. This is reflected in the overall low attrition rates for both the RP and BL module (1% and 2%, respectively). In spite of the requirement to attend treatment, there are some people who drop out of treatment (i.e., these numbers were low in those we had information for, but not zero); thus, it was still felt to be appropriate to examine a possible difference between the two modules.

Although the facilitators highlighted concerns they had (i.e., that there was not enough focus on risk) and offered suggestions about the BL module, no one noted that he or she thought other modules were superior. The majority of facilitators viewed the module in a positive manner and thought that it was motivating. A number of them noted that, although there were positive aspects to the module, they did not think it was appropriate for all men. Several highlighted difficulties when running the module with people who have had no previous treatment or those who are higher risk. They also noted that some useful elements from the old RP module were removed unnecessarily.

This is consistent with some of the criticisms of the GLM that have been expressed. Particularly, Ogloff and Davis (2004) noted their concerns about failing to focus on risk factors. It has been highlighted that a critical therapeutic task in the GLM rehabilitation framework is finding the balance between promoting good and reducing risk (Ward, Mann, et al., 2007). Some of the facilitators reported that they felt the balance had tipped more in the direction of promoting goods to the detriment of work on reducing risk. However, newer implementations of this module (i.e., on the UK’s Community-SOGP and Thames Valley-SOGP) include a better blend of the GLM and risk, with more attention given to risk than was included in the N-SOGP (Elizabeth Hayes, personal communication, November 14, 2008). As well, the men who attended the module felt that it had helped them manage their risk or increased their insight into how to avoid reoffending so they did not appear to feel they were missing anything. When asked for suggestions, none of the men mentioned that there should have been a greater focus on risk.

It should also be noted that although elements of the N-SOGP BL module were consistent with the GLM framework, it does not represent an entirely GLM consistent program. A comprehensive GLM framework is focused both on promoting goods and managing/ reducing risk (Ward, Mann, et al., 2007). Risk factors represent omissions or distortions in the conditions needed to implement a realistic Good Lives plan (Ward, Mann, et al., 2007). However, Ward and colleagues highlight that it is crucial that the therapist finds the right balance between approach goals such as promoting goods and the avoidance goals of reducing risk, noting that “erring on the side of either goal can result in disastrous social and personal consequences for the therapist and offender ” (p. 92). For instance, too much focus on promoting goods could result in a happy, but still dangerous offender. Based on the comments of the facilitators, it would seem this balance was not quite right in the N-SOGP BL intervention, but this has been addressed in newer implementations. Individuals who are considering or are in early days of implementing a GLM intervention may find it useful to pay particular attention to finding this delicate balance. Ward, Mann, and colleagues also note the importance of the therapist adopting a constructive and humanistic approach to treatment, in which they respect the offenders for their capacity to change and see them as fellow human beings seeking a common set of goals. This is something that was noted as a positive of the BL program by the facilitators interviewed.

Furthermore, using a GLM framework would entail assessing client’s heavily weighted goods and problems in their previous life plan, with a focus on those problems that contributed to sexual offending. The GLM treatment plan then focuses on attaining these highly valued goods and addressing the problems that previously led to offending. In the BL module, assessment and treatment were not entirely consistent with a GLM approach, as the group members were encouraged to have a roughly equal balance of all of the goods rather than determining which were the most heavily weighted for them. However, the BL module did focus on identifying problems in the offenders’ previous plans through discussions of obstacles (i.e., dynamic risk factors) to attaining their goods in the past and obstacles anticipated in the future. Each cluster of obstacles (i.e., dynamic risk domains, namely, management of relationships, attitudes supportive of sex offending, sexual interest, and self-management) had a session devoted to it along with the opportunity for skills practice related to the most significant obstacles. It would seem from the results described here that these practitioners also support the use of the RNR and GLM rehabilitations theories in a complementary manner, as is suggested by Ward, Mann, and colleagues (2007). Offenders who present the highest risk should still be offered the most extensive treatment. In addressing criminogenic need, this can be done in a way that is still compatible with supporting the offender to manage their risk level while also meeting their needs and achieving their goals in prosocial ways, thus allowing them to strive for a better life that is incompatible with offending. Taking such an approach would also address responsivity because offenders are more likely to engage with treatment that focuses on positives and overcoming obstacles as opposed to having an avoidance focus.

Participants who attended either module generally reported on their respective modules in a positive manner. Many of the individuals who attended the BL module reported that they did not think that the module needed any changes. Conversely, all of the men who attended the RP module identified something that they would suggest changing. Men who attended the RP module more frequently discussed their motivation for doing future personal work and their improved understanding of themselves in terms of how to avoid future offending, whereas for those who attended the BL module, it was more about a desire to better themselves or have better lives. Those who attended the RP module reported more frequently that their thoughts and attitudes had changed in a way that they were better able to manage themselves or their risk of reoffending. Alternatively, participants who attended the BL module reported more frequently that their thoughts and attitudes about themselves or the future were more positive. It appears that the more positive future focus of the BL module influenced the participants’ motivations for future personal work as well as their understanding of themselves and their thoughts and attitudes about the future. This was consistent with what would be expected from a GLM-derived approach.

It should be noted that this positive focus of the BL program did not occur at the detriment of treatment progress on important treatment targets. Particularly, distorted attitudes and socioaffective functioning have been previously identified as criminogenic need factors that have a relationship with recidivism (e.g., Thornton, 2002). The proportion of individuals who demonstrated treatment change and achieved a treated profile in these areas did not differ between the two groups. Furthermore, the groups did not differ in terms of change in relapse prevention skills, which in particular would have been expected to be higher in the RP group. That the groups did not differ in this respect suggests that the BL program is providing many of the skills acquired on the RP module but doing so in a more positive and motivating manner. This suggests that the BL module was equally effective in a number of areas as measured by psychometric measures, as well as being viewed as a more positive approach and likely to instill optimism about the future, according to the participants and facilitators.

The findings discussed highlight a number of the GLM’s strengths. One of the key strengths is the focus on the individual’s potential for a positive future, and this was noted by both the participants and facilitators. Another noted strength is that rather than focusing on reducing maladaptive behaviors, the focus is on enhancing or building on prosocial capacities (McGuire, 2004; Ward, Mann, et al., 2007). Many of the participants noted that they felt they had improved themselves and were better people now. An additional strength of the GLM is its ability to be more motivating for the participants than the more avoidance-focused approaches used previously, without sacrificing improvement on relevant treatment targets. The module appeared to have the motivating impact on the participants that it intended in that many participants felt positive and optimistic about their futures.

Limitations and Future Directions

There were a number of limitations to the present study. One limitation was that there were few men who completed the old RP module who were still on license and available (and willing) to be interviewed. The intention was to interview approximately equal numbers of men. However, because the RP module has been replaced with the newer BL module, many of the men were no longer available to be interviewed. It should be highlighted that the very small numbers of participants who were available to be interviewed threatens the generalizability of the findings to the sample as a whole. Furthermore, as attrition information was not available for the whole sample, this limits the conclusions that can be drawn about between-group differences in attrition rates.

An additional limitation was the short amount of time since the completion of the module. Given that the BL module is still relatively new, there has not been a sufficient amount of time to examine recidivism data. Due to the low base rate of sexual offending, there is little value in looking at recidivism rates as an outcome with less than 5 years follow-up. Given that reduced sexual offending is often the primary outcome of interest for sexual offender treatment, this is an area that should be explored in the future.

Conclusions

On the whole, both the facilitators and the treatment participants reported their thoughts on the BL module in a positive, future-focused manner. None of the facilitators reported that the previous RP module was superior. Some useful suggestions were provided by the facilitators in terms of the importance of considering risk and that efforts should be made to ensure the group members have similar levels of motivation and knowledge. The participants who attended the BL module and the RP module were equally likely to achieve treatment change and an overall treated profile on a several important treatment targets. However, those who attended the RP program did not report their perceptions and motivations in a manner that was future focused as frequently as those who attended the program with the GLM approach. These findings suggest the value of the Good Lives approach implemented alongside consideration of risk reduction and management.

Footnotes

Appendix A

Appendix B

Acknowledgements

We are grateful to Ruth Mann, Karl Williams, Elizabeth Hayes, and Anita McLeod for their guidance on the development of this research and to Claire Sweeney for her assistance with the research. We would also like to thank Rebecca Mandeville-Norden and Janine Rakestow for providing the psychometric data. Thanks are also due to the offender managers and treatment staff for all their assistance facilitating the data collection for this study. We would also like to thank Tony Ward for his comments on this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Offender Management Service of the UK’s Ministry of Justice.

Notes

Bios

Leigh Harkins is a registered psychologist with the UK’s Health Professions Council and a forensic psychologist with the British Psychological Society. Dr. Harkins is a lecturer at the Centre for Forensic and Family Psychology at the University of Birmingham. She has published a number of reports, papers, and book chapters, primarily in the area of sexual offender treatment effectiveness. Leigh also has experience working in forensic practice settings in the UK and Canada. Her research interests include sexual offending, and group aggression.

Vanja E. Flak recently earned a PhD at the University of Birmingham in Forensic Psychology. Her research interests include the assessment of sexual interest. Specifically she has looked into forensic assessment of child sex offenders sexual interest by the use of a computerized measure termed the Rapid Serial Visual Presentation procedure. She has published book chapters and peer-reviewed journal articles.

Anthony R. Beech is Head of the Centre for Forensic and Criminological Psychology at the University of Birmingham, UK. He has authored over 140 peer-reviewed articles, 35 book chapters and six books in the area of forensic science/criminal justice. His particular areas of research interests are: risk assessment of offenders; reducing online exploitation of children; and increasing psychotherapeutic effectiveness of the treatment given to offenders. In 2009 he received the Significant Achievement Award from ATSA. In 2009 he was the recipient of the Senior Award from the Division of Forensic Psychology, British Psychological Society, for a significant lifetime contribution to forensic psychology.

Jessica Woodhams is a forensic psychologist who works at the University of Birmingham, UK, where she leads postgraduate courses in clinical criminology. She conducts operational research on sexual offending for UK and overseas police forces and for other government bodies. Specifically she researches multiple perpetrator rape, consistency in serial sex offending, rape stereotypes and rape victim behaviour.

References

Andrews

D. A.

Bonta

(2007). The psychology of criminal conduct (4th ed.). Cincinnati, OH: Anderson.

Andrews

D. A.

Bonta

Wormith

J. S.

(2006). The recent past and near future of risk and/or need assessment. Crime and Delinquency, 52, 7-27.

Beckett

R. C.

(1987). The children and sex questionnaire (Available from Richard Beckett, Room FF39, The Oxford Clinic, Littlemore Health Centre, Sanford Rd., Littlemore, Oxford). Oxford, UK: Author.

Beckett

R. C.

Fisher

(1994, November). Assessing victim empathy: A new measure. Paper presented at the 13th Conference of the Association for the Treatment of Sexual Abusers (ATSA), San Francisco, CA.

Beckett

R. C.

Fisher

Mann

R. E.

Thornton

(1997). The relapse prevention questionnaire and interview. In Eldridge

(Ed.), The rapists guide for maintaining change: Relapse prevention manual for adult male perpetrators of child sexual abuse. Thousand Oaks, CA: SAGE.

Beech

A. R.

(1998). A psychometric typology of child abusers. International Journal of Offender Therapy and Comparative Criminology, 42, 319-339.

Bonta

Andrews

D. A.

(2003). A commentary on Ward and Stewart’s model of human needs. Psychology, Crime, and Law, 9, 215-218.

Craig

L. A.

Browne

K. D.

Beech

A. R.

(2008). Assessing risk in sex offenders: A practitioner’s guide. New York, NY: John Wiley.

Davis

M. H.

(1980). A multi-dimensional approach to individual differences in empathy. JSAS Catalogue of Selected Documents in Psychology, 10, 85.

10.

Fernandez

Y. M.

(2006). Focusing on the positive and avoiding negativity in sexual offender treatment. In Marshall

W. L.

Fernandez

Y. M.

Marshall

L. E.

Serran

G .E.

(Eds.), Sexual offender treatment: Controversial issues (pp. 187-197). New York, NY: John Wiley.

11.

Keltner

A. A.

Marshall

P. G.

Marshall

W. L.

(1981). Measurement and correlation of assertiveness and social fear in a prison population. Corrective and Social Psychiatry, 27, 41-47.

12.

Landis

J. R.

Koch

G. G.

(1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159-174.

13.

Mandeville-Norden

Beech

A. R.

Middleton

(2010). The development of the Sex Offender Psychometric Scoring System (SOPSSys) for use in the Probation Service. Probation Journal, 53, 71-76.

14.

Mann

R. E.

Webster

S. D.

Schofield

Marshall

W. L.

(2004). Approach versus avoidance goals with sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 16, 65-75.

15.

Marshall

W. L.

Ward

Mann

R. E.

Moulden

Fernandez

Y. M.

Serran

Marshall

L. E.

(2005). Working positively with sexual offenders: Maximizing the effectiveness of treatment. Journal of Interpersonal Violence, 20, 1096-1114.

16.

McGuire

(2004). Commentary: Promising answers, and the next generation of questions. Psychology, Crime, and Law, 10, 335-345.

17.

National Offender Management Service. (2007). N-SOGP Relapse Prevention Community Better Lives programme manual. Unpublished (Available from Ruth Mann, Rehabilitation Services Group, National Offender Management Service, Clive House, 70 Pettyfrance, London, SW1H 9EX, UK).

18.

Nowicki

(1988). Adult Nowicki–Strickland Internal-External Locus of Control Scale. In A .C. Salter, Treating child sex offenders and their victims: A practical guide (pp. 310-313). London: SAGE.

19.

Nowicki

Duke

M. P.

(1974). A locus of control scale for non-college as well as college adults. Journal of Personality Assessment, 38, 136-137.

20.

Ogloff

J .R. P.

Davis

M. R.

(2004). Advances in offender assessment and rehabilitation: Contributions of the risk-needs-responsivity approach. Psychology, Crime, and Law, 10, 229-242.

21.

Russell

Peplau

L. A.

Cutrona

C. A.

(1980). The Revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39, 472-480.

22.

Thornton

(1989). Self-esteem Scale. Unpublished manuscript (Available from David Thornton, the Sand Ridge Secure Treatment Center, Mauston, WI).

23.

Thornton

(2002). Constructing and testing a framework for dynamic risk assessment. Sexual Abuse: A Journal of Research and Treatment, 141, 139-153.

24.

Ward

(2002). Good lives and the rehabilitation of offenders: Promises and problems. Aggression and Violent Behavior, 7, 513-528.

25.

Ward

Gannon

T. A.

Birgden

(2007). Human rights and the treatment of sex offenders. Sexual Abuse: A Journal of Research and Treatment, 19, 195-216.

26.

Ward

Mann

Gannon

T. A.

(2007). The Good Lives model of rehabilitation: Clinical implications. Aggression and Violent Behavior, 12, 87-107.

27.

Ward

Maruna

(2007). Rehabilitation: Beyond the risk paradigm. London: Routledge.

28.

Ward

Polaschek

Beech

A. R.

(2006). Theories of sexual offending. New York, NY: John Wiley.

29.

Ward

Stewart

C. A.

(2003). The treatment of sex offenders: Risk management and the good lives model. Professional Psychology: Research and Practice, 34, 353-360.

30.

Webster

S. D.

Mann

R. E.

Thornton

Wakeling

H. C.

(2007). Further validation of the short Self-Esteem Scale with sexual offenders. Legal and Criminological Psychology, 12, 207-216.

31.

Wilson

R. J.

Yates

P. M.

(2009). Effective interventions and the Good Lives model: Maximizing treatment gains for sexual offenders. Aggression and Violent Behavior, 14, 157-161.