Using Theory to Understand the Barriers to Engagement in Group Offending Behavior Programs

Abstract

Noncompletion of group offending behavior programs is a common problem, indicating barriers to engagement. While existing theoretical models have accounted for determinants of motivation, little focus has been directed towards barriers to engagement. The authors developed the program engagement theory (PET) which not only accounts for the determinants of engagement and the engagement process, it also considers the barriers to engagement. Interviews and session observations were used to collect data from 23 program facilitators and 28 offenders, which were analyzed using grounded theory. The barriers to engagement were classified as program and referral factors (uninformative referrals, offense-focused programs, rigid and abstract content, didactic delivery, and homework), facilitator characteristics (lack of control: contentious and nonassertive), and group member characteristics (unmotivated, pre-contemplative, and blaming others and young, chaotic, and disruptive). Suggestions as to the design and facilitation of group offending behavior programs, and facilitator training and supervision to overcome barriers to engagement are proposed.

Keywords

engagement programs barriers offender facilitator

Barriers to Engagement in Group Offending Behavior Programs

Noncompletion, attrition, or dropout of group offending behavior programs (GOBPs) indicates the presence of barriers to engagement. There has been a general consensus that the success of GOBPs designed to target offenders’ criminogenic needs and risk, irrespective of the type of programs or treatment settings (Drieschner & Verschuur, 2010), is dependent on engagement (McMurran & Ward, 2010; Scott & King, 2007). Therefore barriers to engagement, indicated by noncompletion and attrition, lead to poor program outcomes. Attrition has been related to recidivism among domestic violence offenders (Gondolf, 2002), sexual offenders (Miner & Dwyer, 1995), and parents perpetrating child abuse (Harder, 2005). Furthermore, non-completers of cognitive skills programs have been identified as at higher risk of reoffending than untreated offenders (McMurran & McCulloch, 2007).

Noncompletion of GOBPs is unfortunately not uncommon. McMurran and Theodosi (2007) found that, on average, 15% of institutional samples and 45% of community samples of offenders attending cognitive skills programs did not complete them. Similarly, Daly and Pelowski (2000) reported dropout rates of between 50% and 70% from domestic violence programs, while Gondolf and Foster (1991) reported dropout rates of domestic violence perpetrators as high as 92%. The high dropout rates and related negative program outcomes indicate the need for a clear understanding of the barriers to engagement.

Researchers have proposed models explaining determinants of engagement, such as the integral model of treatment motivation (Drieschner, Lammers, & van der Staak, 2004) and the multifactor offender readiness model (MORM; Ward, Day, Howells, & Birgden, 2004). The integral model of treatment motivation details internal determinants of motivation (e.g., problem recognition) and external factors (e.g., treatment process) which then dictate engagement (Drieschner & Boomsma, 2008). The MORM includes a broader spectrum of individual factors (e.g., cognitive strategies) and contextual factors (e.g., mandated/self-referred) that comprise treatment readiness, which then facilitate engagement (McMurran & Ward, 2010). Both the integral model of treatment motivation and the MORM focus on what determines engagement but do not account for the barriers to engagement. Treatment responsivity formulations have accounted for “obstacles” to engagement, but Ward et al. (2004) have argued that the notion of obstacles is not as useful as focusing on “readiness,” because obstacles represent the absence of readiness factors. However, barriers to engagement might still need to be considered separately from readiness factors, as it may be the case that not all barriers represent the absence or opposite of readiness factors, and vice versa. Furthermore, the nature and content of the treatment program might determine which factors can be considered as indicators of treatment readiness or barriers to engagement. For example, problem recognition might only be a relevant internal determinant of motivation (Drieschner & Boomsma, 2008) if this is considered a prerequisite for the program, as opposed to a treatment target, or not considered as particularly relevant to the program (Bowen, et al., 2018). Even if an offender is considered “treatment ready” at the start of a program, there may be program factors and even factors relating to the facilitator that arise during the course of treatment that lead the offender to disengage, and these factors should be considered barriers to engagement. In a qualitative meta-analysis, Sturgess, Woodhams, and Tonkin (2016) reported a perceived lack of self-efficacy, negative perceptions of treatment, and a lack of perceived control as offenders’ reasons for noncompletion. These findings indicate that individual factors (e.g., perceived lack of self-efficacy) might be impacted by program-related factors (e.g., negative perceptions of treatment) and that cumulatively this represents barriers to engagement.

As there are a number of different types of GOBPs delivered across the United Kingdom, it is important to consider all the factors that constitute determinants of engagement, factors that are part of the engagement process, and those that constitute barriers to engagement. The authors of the current article developed the program engagement theory (PET), which explains both offenders’ (group members’) and facilitators’ engagement in GOBPs. The PET accounts for engagement determinants, the process of engagement, and the barriers to engagement. According to the PET, the process of engagement represents all the efforts clients make during the course of treatment (both within and between sessions) towards the achievement of changes (treatment outcomes). Group members’ engagement is defined as moving on and is integral to their perceptions of program-related changes. Facilitators’ engagement was defined by their abilities to engage group members in moving on. The detailed process of engagement according to the PET has been reported elsewhere (Holdsworth et al., 2017), and a report of the determinants of engagement according to the PET are forthcoming. The focus of this article is on the barriers to the engagement of both group members and facilitators. The aim is for researchers and practitioners to more fully understand the barriers to engagement, which may result in the all too common high non-completion rates, and how to reduce these barriers and thereby increase program effectiveness.

Method

A brief overview of the participants, data collection, and analysis is provided here but a more comprehensive overview of the methodology for developing the PET is detailed elsewhere (Holdsworth et al., 2017).

Participants

Four U.K. Probation Trusts varying in size and locality took part in this research. From these four Trusts, 23 GOBP facilitators (15 females and eight males) and 28 program completers (19 males and nine females) volunteered to take part in this research. Of those volunteering to take part, 17 facilitators (11 females and six males) and nine offenders (six males and three females) agreed to be interviewed and seven facilitators (five females and two males) and 23 offenders (14 males and nine females) agreed to be observed. The offenders ranged from low to high risk, and comprised generally violent offenders, domestic violence offenders, and sexual offenders. Willis (2018) has argued that labels such as “offenders” represent an ethical issue that helping professionals frequently overlook. Researchers may also be guilty of this; therefore, henceforth, offenders are referred to as group members as it was the authors’ contention during this research that referring to individuals as offenders needlessly perpetuates their identities as such. Facilitators were program tutors with 12 months to 15 years experience of delivering at least one type of accredited GOBP. Participating facilitators from one of the Probation Trusts (eight females and four males) also had 1-year experience of delivering nonaccredited brief solution-focused programs.

Design and Data Collection

A constructivist grounded theory methodology was employed for data collection and analysis, following the guidelines set out by Charmaz (2006). Steps were taken to ensure that the methodology adhered to the guidelines proposed by Shenton (2004) for achieving research trustworthiness. Combining interviews with observations provided the researchers with the opportunity to compare accounts of program experiences with observations of programs, and then verify and clarify the data obtained from both methods (Miles & Huberman, 1994).

A semistructured interview schedule was designed based on Spradley’s (1979) ethnographic approach that employs grand tour questions, such as “Can you talk me through your experience of a session when you felt you were really involved?” and mini tour questions, such as “When you say you ‘worked through the role-play,’ what would I have seen you doing?” (see the appendix for the interview schedule). The 26 interviews, which were audio-recorded and transcribed verbatim, were each approximately 1 hr.

Eight program sessions were observed and audio-recorded. The programs included the following: Thinking Skills Program (TSP) group (19-session program targeting self-control, social problem-solving, and positive relationships) and a Women’s Group Program (solution-focused program focusing on personal goal work and personal skills and strengths). With guidance from Cotton, Stokes, and Cotton (2010) and Robson (2002), a nonparticipatory, informal method of observation was employed.

All interview and observational data were collected over several weeks and then collated as one data set prior to analysis. The data were collected by the lead author who has applied research experience in the evaluation and development of GOBPs.

Data Analysis

Constructivist grounded theory is an inductive, exploratory methodology specifically designed for theory development. The key assumption is that meanings correspond with real-world phenomena, but are constructed and interpreted (Charmaz, 2006). The intention of this research was for the resulting theory to explain the process of engagement. This engagement process is reported elsewhere (citation required) but essential to this process are the barriers, which are accounted for in the PET and are the focus of this article.

The data analysis followed the guidance of Charmaz (2006) on the use of initial coding (early observations) and focused coding (more salient aspects of the data). Focused codes were compared for refinement which led to the development of conceptual categories; these were then employed to analyze further data as a method of theoretical sampling (Charmaz, 2006). This process contributed to axial coding, which revealed relationships between conceptual categories and subcategories, and the properties and dimensions of subcategories (Charmaz, 2006). All coding was carried out by the first author, but regular meetings were held among the team of authors, who also have experience in the evaluation and development of GOBPs, to develop and refine the emerging theory.

Data Sources and Codes

Extracts of data are presented in the results along with a code to describe the data source. The code begins with participant information: F or M (female or male participant) and FA or GM (facilitator or group member), followed by program information: TSP (19-session program targeting self-control, social problem-solving, and positive relationships); IDAP (Integrated Domestic Abuse Program—27-session program targeting respect, accountability and honesty, negotiation and fairness); DIDS (Drink-Impaired Drivers’ Program—16-session program targeting attitudes towards the use of alcohol, patterns of drinking, and related behavior); SOTP (Sexual Offender Treatment Program—38-session program targeting relationship skills, attachment style deficits, and victim empathy); ART (Aggression Replacement Therapy—program targeting aggression and anger); SF (Solution-Focused Program, including a Women’s Group Program—10-session program targeting skills and strengths); and GEN (reference to programs in general). As an example, an extract accompanied by the code (M GM TSP 11) denotes data from a male group member referring to the TSP.

Results and Discussion

The aim of GOBPs is behavioral change and according to the PET, group members’ engagement is analogous to the change process, while facilitators’ engagement comprises the work involved in facilitating this process. Therefore, the barriers identified within the PET can be seen as factors that prevent group members from making program-related changes, and that prevent GOBPs from being facilitated in a way that promotes engagement and change.

The barriers were classified into three categories: (a) program and referral factors, (b) facilitator characteristics and behaviors (factors perceived by group members as barriers to their own engagement), and (c) group member characteristics and behaviors (factors perceived by facilitators as barriers to their own engagement in facilitating GOBPs). These factors appear to be highly interrelated (see Figure 1).

Figure 1.

Barriers classified as program and referral factors, facilitator characteristics and behaviors, and group member characteristics and behaviors.

Program and Referral Factors

Most of the barriers to facilitators’ and group members’ engagement originated from program and referral factors that had a pervasive, negative impact on group work. Referral factors in particular made it difficult right from the start for group members to settle into programs. Program factors directly impacted facilitators’ engagement in their work, which ultimately had a negative influence on group members’ engagement. The program and referral factors causing barriers to engagement included uninformative referrals, offense-focused programs, rigid content, abstract content, and a didactic delivery.

Uninformative referrals

The most commonly reported issue causing a barrier to the first stage of engagement, getting started, was uninformative referrals. This appeared to contribute towards group members’ feelings of ambivalence about attending a program that led to resistance and frustration as perceived by facilitators (see Figure 1). Five of the nine group members interviewed referred to a lack of information about the program which left them feeling unprepared. In the following excerpt, the participant had been expecting to go to prison, an entirely different outcome to being referred on to a program.

I had all my bags packed waiting to go, and they said the only thing we can put, but they [crown court] didn’t explain it to me because there weren’t much chance of me getting it. (M GM TSP 33)

I don’t feel like there’s a lot of information. (F GM SF 39)

A lack of information in one instance was also perceived alongside what seems to have been a sense of abruptness and lack of care at the point of referral.

. . . the judge was quite snappy, you’re doing it or you’re not doing it, your choice if you don’t wanna do it—“stuff you” basically. (F GM DID 23)

Another participant reported similar experience in relation to a court referral.

Courts, “it is a Thinking Skills Program, it helps you out.” That was it. I didn’t get any information or have no choice to do it. I had to do that or go in jail. (M GM TSP 35)

The participants seemed to experience coercion—a lack of autonomy and choice. The perception of choice in motivational interviewing (a method for facilitating intrinsic motivation) is conducive to engagement (Neighbors, Walker, Roffman, Mbilinyi, & Edleson, 2008) and therefore, the absence of perceived choice serves as a barrier to engagement.

As well as a lack of program information in referrals, the absence of any information about the group may have caused apprehension.

My vision of it was gonna be sitting in the room with a lot of guys, with skinheads and Doc Marten boots, combats and stuff like that . . . it was a big shock. (M GM IDAP 22)

. . . I was quite shocked at how normal all the people appeared to be. (M GM SOTP 25)

There appears to be evidence of stereotyping of both group members convicted of domestic violence and sex offending as the extracts relate to different GOBPs. According to self-categorisation theory (Turner, Oakes, Haslam, & McGarty, 1994), these participants had not socially identified themselves as being a member of this group of offenders, which is indicated by the “shock” along with the stereotyping of offenders. Seeing the other group members as “normal” may have helped them to identify with other members in the group, but their initial preconceptions may have posed barriers to their initial engagement.

Group members’ early impressions of not only the program but also the group are clearly important to their “treatment readiness” (Ward et al., 2004) and early engagement, and subsequently the lack of information provided at the referral stage seems to diminish readiness and make engagement challenging, thereby generating greater work for facilitators early on in the delivery of GOBPs.

One or two individuals were not engaging at the start, I think because they didn’t have an understanding of how it was going to be relevant to them, they kind of put a barrier up. (F FA GEN 4)

Volitional factors have been conceptualized as internal determinants of treatment readiness (Ward et al., 2004), and although they might prove to be scarce among group members mandated to GOBPs, group members still need to perceive the therapeutic value of choice (Miller, 1987) in relation to attending a program. Effective collaboration at the referral stage has been identified as essential to serving substance abuse clients in the broadest possible context (Center for Substance Abuse Treatment, 2000), but this logic should apply to all offenders referred to GOBPs.

Offense-focused programs

There were three features of offense-focused programs that created barriers to engagement: challenging group members (facilitators’ perspectives), which then lead to the need for self-disclosures (group members’ perspectives), and third-party information. However, even just talking about the general topic of the group members’ offenses proved difficult as this put facilitators in the position of having to discuss subjects they were uncomfortable with.

. . . you know I don’t even talk about sex with my friends, and then I’m having to talk about it with men I don’t even know. (F FA IDAP 11)

Challenging group members

The offense-focused programs were regarded by facilitators as putting them in an uncomfortable position of challenging group members about their behavior, who were in turn uncomfortable at being challenged about their behavior and having to make disclosures.

I think other guys don’t particularly like having to think about it [offending behavior] and being put on the spot. (M FA SOTP 13)

This discomfort for both parties may be due to instinctively wanting to avoid group members feeling shame or guilt. Research has indicated a clear link between shame and the tendency to externalize blame and anger (Harper & Arias, 2004; Harper & Arias, 2004; Paulhus, Robins, Trzesniewski, & Tracy, 2004), and between guilt and anger, depression, and anxiety (Spice, Viljoen, Douglas, & Hart, 2015). In a group format, these responses to shame or guilt may only be exacerbated and prevent the group members from engaging. Marshall et al. (2003) have argued that a warm and motivational style is more conducive to engagement than a challenging approach. Facilitators of solution-focused programs saw challenging group members about their behavior as incongruent with their work because offending behavior is not discussed in these programs.

. . . in the old days the sex offender treatment programs were very challenging for people and I thought it turned people into their shells, so it’s more about drawing them out. (M FA SF 9)

The need for self-disclosures

As a result of open challenges from facilitators, group members revealed the negative impact of hearing other group members making self-disclosures, and then having to make their own disclosures.

I’ve gone into some of the meetings and I’m head shot because I’m thinking; shit—if I hadn’t been like that then my life would be different now. That’s hard. (M GM IDAP 11)

. . . when you’re in the room with guys that are similar in offense, you’ve got to talk about it . . . the danger is it does come along with a lot of regrets. (M GM IDAP 21)

Coming to terms with the past during programs may bring about an emotional cost that orients group members in the past, inhibiting them from moving on and thinking about the future.

It touched a sore point in me that made me very sad. (M GM IDAP 21)

Strong emotions can prevent change if therapists do not acknowledge these emotions through reflective listening (Center for Substance Abuse Treatment, 1999). Consequently, requiring self-disclosures in programs might be a barrier to change, unless facilitators are able and have the resources to support group members through the process and prevent it negatively impacting engagement.

Third-party information

A feature of the accredited domestic abuse program (IDAP) was that facilitators received information from women’s safety workers about group members’ ongoing behavior at home in between program sessions. For some facilitators, having this information posed as a barrier to engagement because it created confusion and frustration at not being able to use it.

. . . and all the stuff that was coming in, sometimes, you know you have something in your head, where did that, how did I know that? Do I know that because it has come from the women’s safety worker? And then again, you know part of the victims, doesn’t mean to say they are perfect and right, so they might be lying, you know, so yeah, I have got this information . . . I can’t really use it, so sometimes, I would rather not know. (F FA IDAP 6)

Being privy to information from the women’s safety worker, being unsure of its truthfulness, and working with only one of the parties appear to be problematic. Researchers have argued that conjoint interventions may present with treatment advantages (Stith, Rosen, McCollum, & Thomsen, 2004), particularly for intervening in situational couple violence (Kelly & Johnson, 2008). However, where programs are not conjoint, the introduction of third-party information appears to be counterproductive to facilitators’ engagement in their work with the group, and distracting if programs are solution focused rather than offense focused.

Rigid and abstract content

Building engagement: Improvising and personalizing treatment frameworks was the cornerstone of facilitators’ work but this was negatively impacted by the rigid content of programs. Facilitators and group members were aware that sometimes facilitators were waiting for the right answer which can create ruptures to the developing alliance (Ackerman & Hilsenroth, 2003). In a meta-analysis by Hoogsteder et al. (2015), the researchers found that few cognitive behavioral interventions were tailored to an individual and argued that group therapy might not be appropriate in every situation. However, it seemed in the current study that it was the content that constrained facilitators from being able to work autonomously and flexibly, to improvise and explore, which were important means of building engagement.

For me that’s having the rigid few minutes for this, five minutes for that, ten minutes for that, watching the clock and that to me is very difficult. (F FA GEN 8)

The rigid and prescriptive nature of programs was construed by facilitators as features of accredited programs versus the nonaccredited programs, which were construed as more flexible in allowing facilitators to be “inventive.”

So you have to be a bit inventive, you can do that with non-accredited programs but you can’t do that with accredited programs. (M FA SF 9)

. . . not being disparaging about accredited programs but they’re very fixed and set in what they do. (M FA GEN 12)

The fixedness of accredited programs also seemed to prevent facilitators working in the way that they felt they were best at, therapeutically and responsively.

. . . my background was in kind of therapeutic groups as well so we’d key into what was going on in the room there and then at that minute. That’s what I’m better at, that’s what I enjoy more. (M FA GEN 12)

There was evidence of some awkwardness as group member participants attempted to decipher facilitators’ delivery of abstract content.

. . . some things felt very repetitive erm and he [facilitator] was doing his best to get it across but it did feel like he was a bit like a broken record at times, erm, and a few of the words and wordings, we’re a bunch of guys—we’ll understand lay man’s terms as if you, stuff like “objectiveness,” it’s like . . . what’s it in real terms? (M GM SF 42)

The repetition referred seems to reflect the facilitator’s failed efforts to get a program message across, quite possibly leading to frustration for the group members and the facilitator. A lack of realism in program content may make work overly problematic for facilitators. According to the PET (citation blinded), when facilitators personalized treatment frameworks, making abstract content more concrete and relevant, group members were engaged because they could personally relate to the content in a way that enabled them to see the changes they could make. The absence of personalization or content in “lay man’s” terms appears to prevent engagement.

Didactic delivery

Facilitators regarded a didactic delivery or a reliance on text-based materials that required group members to read as a barrier to engagement because it may have reflected group members’ earlier negative experiences of school.

. . . because if they are the person at school who didn’t like reading stuff off the board, so that can be quite uncomfortable. (F FA GEN 6)

Facilitators saw that the type of group members they worked with typically did not like reading or writing, yet programs have always remained text-based, requiring group members to participate in ways that they are uncomfortable with.

. . . these are people that don’t like to sit and read things, they don’t like writing, so a program that consists mostly of written work and reading just, in terms of that particular client group . . . I mean, we are still on this dull old road, of workbooks and even homework you know. (F FA TSP 10)

A didactic delivery may not only have been uninspiring for group members, but facilitators too. Facilitators perceived an inconsistency between a didactic delivery and what they saw their job as entailing.

When you’re doing a power point presentation to someone over a long period of time and you know that it’s not going to be interactive, but it’s about giving information to people, I don’t work well with that . . . I’m talking at people and I see people, gradually their lids are actually going, or looking at their watches, stifling yawns, and that is like “blimey I’m not doing my job here” and that’s kind of instant feedback that I’m not doing my job. (M FA GEN 12)

Text-based materials and PowerPoint proved particularly problematic for one group member, who found it difficult to engage because he was unable to read some of it.

. . . some of the questions I mean I couldn’t really answer yeah, and they didn’t have them printed in large format so I couldn’t read them either . . . less power-point presentations because I can’t bloody see them anyway and they don’t say much. (M GM SOTP 25)

A didactic delivery, along with rigid and abstract program content, are all factors relating to the responsivity principle (tailoring the intervention to the learning style, motivation, abilities, and strengths of the offender), which has been argued to have the potential to interfere with the effective delivery of programs (Jung & Dowker, 2016). Jung and Dowker (2016) have argued for screening of responsivity barriers to maximize treatment gains, regardless of offense type.

Homework

Within the solution-focused programs, homework involved group members practicing solution behaviors between sessions. Facilitators then explored with group members the helpfulness of these behaviors. Consequently, homework formed a functional basis of subsequent sessions, whereas in other programs homework constituted worksheets to check learning. Group members had only negative perceptions about these homework sheets, which seemed to remind them of school.

I know it is like, you know, it is classed as homework and you think of school straight away. (M GM TSP 33)

Some researchers have employed homework as a proxy for engagement (McCarthy & Duggan, 2010), and it is intuitive that program engagement should exist between as well as within sessions to reflect group members’ efforts to change, but with the exception of the solution-focused programs, homework appeared to be at best an indication of compliance, and at worst a barrier to engagement.

I do it because I have been asked to do it, I wouldn’t choose to do it but because I have been asked to do it, I do it. (F GM DID 23)

I do it because I have to. (M GM TSP 35)

An observation of group members’ responses to homework in accredited programs supported the interpretation of homework sheets as a barrier to engagement.

Is this homework? Has she just given us f****** homework? (M GM TSP 30)

Facilitator Characteristics and Behaviors

While most of the facilitator characteristics posing barriers to engagement stemmed from program and referral factors, group members perceived a number of independent characteristics of facilitators that were barriers to their engagement: lack of control—contentious and nonassertive.

Lack of control

Weak engagement ensued from perceiving facilitators as having a lack of control over the group. This lack of control was construed as causing sessions to become chaotic and difficult. The lack of control appeared to be characterised by either a contentious or nonassertive approach to facilitating the group.

Contentious

A contentious delivery of content was attributed as responsible for generating heated discussions that made group members feel angry and uncomfortable.

I think it was about sexual awareness or something like that there was . . . it just got all blown out of proportion and the whole two and a half hours turned into arguments about rape and things like that, and it was very uncomfortable for a lot of people in there . . . all the barriers, the barriers went up with everyone, but I think how it was put over, it’s what put their barriers up. (M GM IDAP 22)

The reference to other group members’ “barriers” going up indicates how obvious the evidence of the negative impact the facilitators’ lack of control was. The participant attributed the reason for these barriers going up to the lack of control in how the content was delivered, as opposed to the potentially contentious content itself.

It just got railroaded into something completely else. I think it was put over quite aggressive and you . . . the dynamics in that group wasn’t great anyway, there was a lot of guys in there who . . . look I shouldn’t be here this is . . . no one was taking anything on board and then basically be told you’re raping people in the group. (M GM IDAP 22)

From the participant’s perspective, the group members were already resistant; hence, he evaluated the situation and deemed the way the facilitator was delivering the session as only making matters worse by enhancing resistance.

Nonassertive

A female group member revealed a sense of frustration at perceiving facilitators as lacking control over the group, but because of a lack of assertiveness as opposed to a contentious delivery style.

They need to be a bit more, I don’t know, they need to sort of stand up and say, “right!” Be more forward, much stronger with their voice vocally, they’re in charge, let them be heard. (F GM SF 39)

The participant also admitted her frustration from seeing facilitators not controlling how much time other group members were using up to make self-disclosures.

. . . not talking when somebody else is talking but somebody can talk for 15 minutes and you sit there and you think; you know I’m not being nasty, that’s nice and she’s sharing or whatever but that’s 15 minutes out of the hour and a half, not being spiteful just . . . maybe there should be an egg timer. (F GM SF 39)

What emerged was that group members regard the time spent within sessions as important and that facilitators need to control the session to ensure fairness of time allocated to group members, and that a lack of this creates a barrier to moving on. Relating to the group leader (facilitator) has been established as an important measure of group engagement (MacGowan and Newman, 2005), and how group members perceive facilitators’ leadership style may be an important factor.

Group Member Characteristics and Behaviors

There were a number of distinct characteristics and behaviors of group members that facilitators found problematic to their own engagement in their work, and thereby the engagement of group members within sessions. These were group members who were perceived as unmotivated, pre-contemplative, and blaming others; young, chaotic, and disruptive; and manipulative and deceitful.

Unmotivated, pre-contemplative, and blaming others

Facilitators defined group members who were unmotivated as particularly problematic to facilitating group engagement because this had a pervasive influence on the other members.

. . . you might have a group who are not so motivated and if you’ve got more of a majority of them not being motivated they bring the others down. (F FA GEN 11)

Facilitators reported difficulties if they perceived a lack of motivation and needed to see some willingness of group members to facilitate engagement.

If I have a group who are clearly telling me they don’t want to be there, they are not interested, the only reason they are there is to avoid going to prison, they don’t want to learn this crap, you know, when you get all of that I, I struggle to function, I can’t, I mean, I do think the people at least have to have a willingness to give it a go. (F FA TSP 10)

The concept of “treatment readiness” including treatment motivation is an important one in relation to maximizing engagement in programs (McMurran & Ward, 2010), but McMurran (2002) has argued that motivation to change should not be seen as a treatment requirement, but as a treatment target. However, there seems to be evidence that a lack of treatment readiness prevented facilitators “functioning,” and that overcoming resistance was not part of the program or considered by facilitators as part of their work. Therefore, group members needed to be at the right stage of change before facilitators could engage them.

. . . and I just think, it’s very simple—don’t give us the people that are just pre-contemplative. (F FA TSP 10)

Coupled with being pre-contemplative was group members’ lack of insight and personal responsibility. Facilitators reported finding it difficult to work with group members who externalized responsibility for their offending behavior.

. . . and they don’t see what they need to do to change, so all their examples are of what other people do wrong rather than their own. (M FA GEN 9)

One facilitator expressed disappointment in having to invest time in group members who preferred to blame others than make an effort to change.

To me it’s disappointing because I feel that my time could have been better spent working with someone that actually wanted to change rather than someone who’s quite you know, happy to sort of go along blaming other people for their mistake. (M FA SF 7)

The perception that group members should already be contemplating change and internalising responsibility before facilitators could work with them seems to have led to frustration and helplessness at having to work with group members who are not at that stage. McMurran (2002) has argued that it is important to reinterpret offenders’ “unmotivated statements” (e.g., “I didn’t do it” could be reinterpreted as “I’m too ashamed to admit it”). Such reinterpretations may be more conducive to developing engagement but facilitators may not always be inclined to reinterpret these statements if they regard their work as beginning only when offenders are motivated to change.

In a group context, blaming attitudes can have pervasive damage on the rest of the group, which facilitators have to work hard to minimize.

They all were blaming their partner . . . and I think we got to the stage where we thought, yeah, we’re just losing the group here . . . and we just had to erm, end the session short, rather than pushing on with it because we felt that had we gone with it, you know, I think we would have lost the erm, the group. (M FA SF 7)

Ending the session short seemed to be perceived by the participant as the necessary course of action to avoid “losing the group.” Externalizing blame might be a prime example of an attitude that fuels groupthink in a GOBP context that is detrimental to engagement. Groupthink however could function to have the opposite effect and encourage engagement if is consistent with the program aims, because working as a group was found to be central to the engagement process in GOBPs (Holdsworth, Bowen, Brown, & Howat, 2017).

Young, chaotic, and disruptive

Young, chaotic group members were identified by some facilitators as difficult to engage because it was challenging to find “the hook” that was fundamental to building engagement. According to the PET (citation blinded), “hooks” are a mechanism for developing a working alliance, but bigger priorities for some group members were difficult to compete with.

I think the chaotic offenders, I think the offenders that come in and you know they have nowhere to live and they have got no job, and maybe they are on drugs. (F FA GEN 4)

An apparent dilemma was that facilitators felt they had to remove some members from the group if their disruptive behavior was having an adverse influence on the “good ones.”

And we are removing people from the group that shouldn’t be there, if it is at the detriment to the three good ones, the three ones, who would be good, if the others weren’t there, but are too scared to take part, because they going to have to see each other out of the group, and they don’t want the piss taking out of them. (F FA GEN 6)

Facilitators seem to perceive that they are in a challenging position of having to make judgments about who the “good ones” are, and then protecting their engagement by removing disruptive group members. On one hand, this may preserve the engagement of some, but the disruptive group members remain untreated and possibly at a high risk of reoffending (McMurran & Theodosi, 2007).

Manipulative and deceitful

Perceptions of group members as manipulative and deceitful were directly related to third-party information from women’s safety workers on domestic violence programs, when it conflicted with group members’ accounts.

Just real liars, but again that’s because, someone who I know is lying, and complete denial all the way through, I find that really difficult . . . Because where I have got my information from I can’t share that, and if someone is lying, you know they are lying, but you can’t say, I know you are lying . . . I think that’s the thing I find hardest to work with. (F A IDAP 6)

The knowledge of deceit because of third-party information seems to feed into a perception that, in general, domestic violence offenders are manipulative and deceitful.

I often find with a lot of domestic violence offenders or people that are quite instrumental in the way they think, that can be quite manipulative, the program kind of almost goes over their head in a way because they’re quite egocentric. (M FA SF 9)

Suspicion of deceit may have had a detrimental impact on the facilitator’s ability to develop a strong working alliance, but it seems to be that it is the knowledge of deceit, stemming from third-party information that may lead to a general perception of group members in a domestic violence program as being manipulative and deceitful, and this knowledge is a barrier to engagement.

Summary

The majority of barriers to the engagement of both group members and facilitators in GOBPs were related to program referrals, how programs were delivered, and program content. Uninformative referrals left group members apprehensive about programs. Offense-focused programs created three particular barriers to engagement: (a) facilitators had to challenge group members, which facilitators regarded as pushing group members in the opposite direction of engagement; (b) knowledge from women’s safety workers (third-party information), which prevented facilitators’ engagement and seemed to feed into their perception that domestic violence offenders are generally manipulative and deceitful; and (c) group members having to make self-disclosures, which led to some group members experiencing emotional distress that prevented engagement. Programs with rigid and abstract content led to awkward communications of program concepts from facilitators, which in turn frustrated group members. It prevented facilitators from working autonomously and flexibly with group members, which was important to building engagement. A didactic delivery was counter to the engagement of group members and facilitators, because of a distracting reliance on reading and writing. Homework in relation to accredited programs was met with resistance and at best indicated only compliance, not engagement. Group members’ perceptions of facilitators as lacking in control—who were either contentious, which promoted group member anger and resistance, or nonassertive, which promoted chaos and frustration—were barriers to their engagement. Facilitators’ perceptions of group members as unmotivated, pre-contemplative, and blaming others led to frustration, which negatively impacted their ability to “function” in their role. Developing a working alliance with young, chaotic, and disruptive group members was challenging, sometimes causing facilitators to remove these group members to preserve the engagement of the rest of the group. Facilitators’ knowledge of third-party information from the women’s safety worker led to facilitators perceiving group members on domestic violence programs as manipulative and deceitful, making it difficult to work with them.

Implications for Research and Practice

In the interests of maximizing GOBP effectiveness, it is essential to consider the range of factors that predict treatment readiness, but it is also necessary to consider all the factors that may impact readiness, and at different points during the course of treatment, constitute barriers to engagement. This study has demonstrated that barriers to engagement are not limited to representing the absence of readiness factors. For example, offense-focused programs involving facilitators challenging group members and third-party information from women’s safety workers are barriers to engagement, but they do not constitute the absence of treatment readiness factors. The danger of assuming barriers to engagement are only those that represent the absence or opposite of readiness factorsis is that the specific areas in which programs and program facilitation can be improved to maximise engagement are neglected. Furthermore, offenders in need of GOBPs will be at various stages of change and treatment readiness at the point they could be referred to a program and in relation to the risk of reoffending, it is unethical and irresponsible to withhold programs until they are motivated or “treatment ready.” This has a number of implications in relation to program referrals, program content and facilitation, and facilitator training.

Program referrals are essential to paving the way for engagement, or lack thereof. Group members who are unprepared or who perceive coercion rather than choice in relation to programs are less likely to engage. It is important to incorporate within referrals (a) information about other group members, (b) the therapeutic value of choice (Miller, 1987) in attending GOBPs, and (c) an effective collaboration at the referral stage (Center for Substance Abuse Treatment, 2000).

Offense-focused programs create three main problems for engagement. First, challenging group members about their offending behavior is not only likely to yield resistance but it also makes it very difficult for facilitators to build the necessary working alliances with offenders for program engagement. Links between shame and the tendency to externalize blame and anger (Harper & Arias, 2004; Paulhus et al., 2004) have been established, but research needs to focus on the impact exposing shame and guilt has in a GOBP format. Second, there was evidence that third-party information from women’s safety workers led to a negative regard for group members attending domestic violence programs as manipulative and deceitful, which was a barrier to the development of a necessary working alliance. Third, the need for self-disclosures in GOBPs brings about emotional turmoil that if left undetected and unsupported causes a barrier to engagement. Solution-focused programs sidestep these issues, but if providers are obligated to deliver offense-focused programs, pretreatment sessions targeting motivation should be incorporated. Furthermore, facilitators need training and support in how to fulfil these program requirements and overcome the associated barriers, while build a strong working alliance, which is necessary for engagement.

Rigid and abstract content, and a didactic delivery were barriers to engagement. The responsivity principle can interfere with the effective delivery of programs (Jung & Dowker, 2016), and as this is a prevalent issue across different types of offenders (Wormith & Olver, 2002). A solution might be for researchers to design programs that can be flexibly adapted by facilitators within treatment, according to the different responsivity levels they encounter among group members, but given the manualized and standardised approach to most GOBPs, research is needed to investigate the potential for more flexible and adaptable programs.

Homework is frequently treated as a proxy for engagement in offender rehabilitation (Holdsworth, Bowen, Brown, & Howat, 2014), but this study has shown that it is a barrier to engagement. Researchers should investigate how homework might be best framed and embedded within GOBPs so that it enhances rather than prevents engagement. It may need to have a more functional role in engaging group members in the process of change in between sessions, offering opportunities for experimenting with program-related changes that are personally relevant, so that group members can experience the direct benefits of completing it.

GOBP facilitators are expected to work therapeutically with the most difficult to engage clients. Facilitators’ aggression and conversely a lack of assertiveness had a negative impact on engagement and therefore achieving a middle ground is important. The current study indicates the need for facilitators to demonstrate control and fairness in attention to each group member. Perceiving equity is important for group cohesion, which is in turn important for group climate and engagement (Moos, 1994; Wilson et al., 2008). Therefore facilitators need training not only in program delivery but also and more importantly in therapeutic skill development, particularly in how to interpret motivation (or an apparent lack thereof) and how to maximize engagement. Program providers should also consider ongoing supervision as an important step in strengthening therapeutic skills and maintaining facilitators’ engagement in their work in facilitating GOBPs.

Facilitators found it difficult to work with unmotivated and pre-contemplative offenders. Researchers and practitioners need to work on how programs can accommodate group members at various stages of change, and how to target motivation and contemplation of change, rather than require it (McMurran & Ward, 2010). Facilitators also found domestic violence offenders to be generally manipulative and deceitful, because of third-party information from the women’s safety worker. Researchers and practitioners need to closely examine the pros and cons of communicating reports from the women’s safety worker to facilitators, with particular attention to the possible costs to facilitators’ efforts in building necessary working alliances with domestic violence offenders for program engagement.

Limitations

There are four particular limitations to this research which should be taken into account. First, the lead author alone collected the data which may have resulted in selective attention to the data captured. Second, while steps were taken to ensure that the methodology adhered to the guidelines proposed by Shenton (2004) for achieving research trustworthiness, one aspect that might not be entirely fulfilled was transferability (equivalent to generalizability). The barriers to engagement in GOBPs are embedded within its context, and therefore these barriers can only be fully understood within the context of the GOBPs investigated in this research. Third, demographic factors such as age of the participants were not obtained. As these factors are likely to impact perceptions of barriers to engagement, inclusion of this information would have likely altered some of the findings of the study. Finally, the sample did not include group members who had dropped out of GOBPs. Offenders who drop out of treatment are difficult to engage in research but the absence of this data must be considered when interpreting the findings. However, the contribution of group members who had dropped out, depending on when they dropped out, would more likely highlight barriers to attendance, which is only one proxy for engagement.

Conclusion

Most of the barriers to engagement were rooted in program and referral factors, indicating that barriers or obstacles to treatment are not limited to the absence or opposite of internal or individual readiness factors. To date, GOBPs have not been designed to maximise or even account for group members’ engagement, or how facilitators need to work to maximise engagement. Fortunately, this can be rectified by a clear understanding of the barriers to engagement to create programs that overcome these barriers, maximize engagement, and thereby maximize program efficacy. Offense-focused programs feature requirements (challenging group members, third-party information, and the need for self-disclosures) that are counter to engaging group members in the process of change, and force facilitators to work in ways that prevent a working alliance. A move towards solution-focused programs may sidestep many of the barriers to engagement. Furthermore, programs need to move away from rigid, abstract content, and a didactic delivery. Facilitators need to work flexibly with group members so that they are able to personalize treatment and communicate program aims more effectively, according to the responsivity of the group members they have in front of them. To reduce the risks of reoffending, program providers have an ethical and practical responsibility to work with group members who seem unmotivated, pre-contemplative, and chaotic. These individuals are challenging to engage, but are still in need of treatment. Providers should therefore ensure facilitators receive therapeutic skills training and ongoing supervision to support them in their challenging work. This will provide facilitators with greater autonomy and confidence in their role, which should be considered the prerequisites for the successful facilitation of GOBPs and engaging group members in the process of change.

Footnotes

Appendix

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) received no financial support for the research, authorship, and/or publication of this article.

References

Ackerman

S. J.

Hilsenroth

M. J.

(2003). A review of therapist characteristics and techniques positively impacting the therapeutic alliance. Clinical Psychology Review, 23, 1-33.

Bowen

Walker

Holdsworth

(2018). Applying a Strengths-Based Psychoeducational Model of Rehabilitation to the Treatment of Intimate Partner Violence: Program Theory and Logic Model. International Journal of Offender Therapy and Comparative Criminology. doi:10.1177/0306624X18798223

Center for Substance Abuse Treatment. (1999). From precontemplation to contemplation: Building readiness. In Enhancing motivation for change in substance abuse treatment (p. 57). Rockville, MD: U.S. Department of Health and Human Services.

Center for Substance Abuse Treatment. (2000). Effective referrals and collaborations. In Integrating substance abuse treatment and vocational services (Treatment Improvement Protocol (TIP) Series, No. 38). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Charmaz

(2006). Constructing grounded theory: A practical guide through qualitative analysis. London, England: Sage.

Cotton

D. R. E.

Stokes

Cotton

P. A.

(2010). Using observational methods to research the student experience. Journal of Geography in Higher Education, 34, 463-473.

Daly

J. E.

Pelowski

(2000). Predictors of dropout among men who batter: A review of studies with implications for research and practice. Violence and Victims, 15, 137-160.

Drieschner

K. H.

Boomsma

(2008). The Treatment Engagement Rating scale (TER) for forensic outpatient treatment: Description, psychometric properties, and norms. Psychology, Crime & Law, 14, 299-315.

Drieschner

K. H.

Lammers

S. M. M.

van der Staak

C. P. F.

(2004). Treatment motivation: An attempt for clarification of an ambiguous concept. Clinical Psychology Review, 23, 1115-1137.

10.

Drieschner

K. H.

Verschuur

(2010). Treatment engagement as a predictor of premature treatment termination and treatment outcome in a correctional outpatient sample. Criminal Behaviour and Mental Health, 20, 86-99. doi:10.1002/cbm.757

11.

Gondolf

E. W.

(2002). Batterer intervention systems: Issues, outcomes, and recommendations. Thousand Oaks, CA: Sage.

12.

Gondolf

E. W.

Foster

R. A.

(1991). Pre-program attrition in batterer programs. Journal of Family Violence, 6, 337-349.

13.

Harder

(2005). Prevention of child abuse and neglect: An evaluation of a home visitation parent aide program using recidivism data. Research on Social Work Practice, 15, 246-256.

14.

Harper

F. W. K.

Arias

(2004). The role of shame in predicting adult anger and depressive symptoms among victims of child psychological maltreatment. Journal of Family Violence, 19, 367-375.

15.

Holdsworth

Bowen

Brown

Howat

(2014). Offender engagement in group programs and associations with offender characteristics and treatment factors: A review. Aggression and Violent Behavior, 19, 102-121.

16.

Holdsworth

Bowen

Brown

Howat

(2017). The development of a program engagement theory for group offending behavior programs. International Journal of Offender Therapy and Comparative Criminology, 61, 1479-1499.

17.

Hoogsteder

L. M.

Stams

G. J. J. M.

Figge

M. A.

Changoe

van Horn

J. E.

Hendriks

Wissink

I. B.

(2015). A meta-analysis of the effectiveness of individually oriented cognitive behavioral treatment (CBT) for severe aggressive behavior in adolescents. The Journal of Forensic Psychiatry & Psychology, 26, 22-37.

18.

Jung

Dowker

B. A.

(2016). Responsivity factors among offenders. Journal of Offender Rehabilitation, 55, 148-167.

19.

Kelly

J. B.

Johnson

M. P.

(2008). Differentiation among types of intimate partner violence: Research update and implications for interventions. Family Court Review, 46, 476-499.

20.

Macgowan

M. J.

Newman

F. L.

(2005). Factor structure of the group engagement measure. Social Work Research, 29, 107-118.

21.

Marshall

W. L.

Serran

G. A.

Fernandez

Y. M.

Mulloy

Mann

R. E.

Thornton

(2003). Therapist characteristics in the treatment of sexual offenders: Tentative data on their relationship with indices of behaviour change. Journal of Sexual Aggression, 9, 25-30.

22.

McCarthy

Duggan

(2010). Engagement in a medium secure personality disorder service: A comparative study of psychological functioning and offending outcomes. Criminal Behaviour and Mental Health, 20, 112-128.

23.

McMurran

(Ed.). (2002). Motivating offenders to change. A guide to enhancing engagement in therapy. Chichester, UK: John Wiley.

24.

McMurran

McCulloch

(2007). Why don’t offenders complete treatment? Prisoners’ reasons for non-completion of a cognitive skills programme. Psychology, Crime & Law, 13, 345-354.

25.

McMurran

Theodosi

(2007). Is treatment non-completion associated with increased reconviction over no treatment? Psychology, Crime & Law, 13, 333-343.

26.

McMurran

Ward

(2010). Treatment readiness, treatment engagement and behaviour change. Criminal Behaviour and Mental Health, 20, 75-85.

27.

Miles

M. B.

Huberman

A. M.

(1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.

28.

Miller

W. R.

(1987). Motivation for treatment: A review with special emphasis on alcoholism. In Cordray

D. S.

Lipsey

M. W.

(Eds.), Evaluation studies review annual, (Vol. 11, pp. 602-625). Thousand Oaks, CA: Sage.

29.

Miner

M. H.

Dwyer

S. M.

(1995). Analysis of dropouts from outpatient sex offender treatment. Journal of Psychology & Human Sexuality, 7, 77-93.

30.

Moos

R. H.

(1994). Group environment scale manual: Development, applications and research (3rd ed.). Mountain View, CA: Consulting Psychologists Press.

31.

Neighbors

Walker

D. D.

Roffman

R. A.

Mbilinyi

L. F.

Edleson

J. L.

(2008). Self-determination theory and motivational interviewing: Complementary models to elicit voluntary engagement by partner-abusive men. American Journal of Family Therapy, 36, 126-136.

32.

Paulhus

D. L.

Robins

R. W.

Trzesniewski

K. H.

Tracy

J. L.

(2004). Two replicable suppressor situations in personality research. Multivariate Behavioral Research, 39, 303-328.

33.

Robson

(2002). Real world research (2nd ed.). Oxford, UK: Blackwell.

34.

Scott

K. L.

King

C. B.

(2007). Resistance, reluctance, and readiness in perpetrators of abuse against women and children. Trauma, Violence & Abuse, 8, 401-417.

35.

Shenton

A. K.

(2004). Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22, 63-75.

36.

Spice

Viljoen

J. L.

Douglas

K. S.

Hart

S. D.

(2015). Remorse, psychopathology, and psychopathy among adolescent offenders. Law and Human Behavior, 39, 451-462.

37.

Spradley

(1979). The ethnographic interview (1st ed.). Long Grove, IL: Waveland Press.

38.

Stith

S. M.

Rosen

K. H.

McCollum

E. E.

Thomsen

C. J.

(2004). Treating intimate partner violence within intact couple relationships: Outcomes of multi-couple versus individual couple therapy. Journal of Marital & Family Therapy, 30, 305-318.

39.

Sturgess

Woodhams

Tonkin

(2016). Treatment engagement from the perspective of the offender. International Journal of Offender Therapy and Comparative Criminology, 60, 1873-1896.

40.

Turner

J. C.

Oakes

P. J.

Haslam

S. A.

McGarty

(1994). Self and collective: Cognition and social context. Personality and Social Psychology Bulletin, 20, 454-463.

41.

Ward

Day

Howells

Birgden

(2004). The multifactor offender readiness model. Aggression and Violent Behavior, 9, 645-673.

42.

Willis

G. M.

(2018). Why call someone by what we don’t want them to be? The ethics of labeling in forensic/correctional psychology. Psychology, Crime & Law, 24, 727-743.

43.

Wilson

P. A.

Hansen

N. B.

Tarakeshwar

Neufeld

Kochman

Sikkema

K. J.

(2008). Scale development of a measure to assess community-based and clinical intervention group environments. Journal of Community Psychology, 36, 271-288.

44.

Wormith

J. S.

Olver

M. E.

(2002). Offender treatment attrition and its relationship with risk, responsivity and recidivism. Criminal Justice and Behavior, 29, 447-471.