Evaluation of a Music Intervention in a Swedish Prison Setting: A pilot project within Skådebanan’s “Culture for Life”-project

(1)

Degree Project in Criminology Malmö University

30 Credits, Two-year Master Faculty of Health and Society

EVALUATION OF A MUSIC

INTERVENTION IN A SWEDISH PRISON SETTING

A PILOT PROJECT WITHIN SKÅDEBANAN’S

“CULTURE FOR LIFE”-PROJECT

LAUREN GHIJS

(2)

EVALUATION OF A MUSIC

INTERVENTION IN A SWEDISH PRISON SETTING

A PILOT PROJECT WITHIN SKÅDEBANAN’S

“CULTURE FOR LIFE”-PROJECT

LAUREN GHIJS

Ghijs, L. Evaluation of a Music Intervention in a Swedish Prison Setting. A pilot project within Skådebanan’s “Culture for Life”-project. Degree project in

Criminology 30 Credits. Malmö University: Faculty of Health and Society, Department of Criminology, 2021.

Background. The Swedish cultural organization Skådebanan conducted a pilot prison-based music intervention at a department of Kristianstad’s prison

institution as part of their “Culture for Life”-project, aiming to bring culture closer to Swedish inmates and to contribute to a criminal-free life for this group.

Aim. The present study aims to conduct a half-way evaluation of Skådebanan’s music intervention. More specifically, it was explored (1) whether the intervention affected participating inmates’ sense of coherence, (2) what attitudes inmates had towards the project, (3) how inmates commented on the music intervention, and (4) what expectations and experiences the project leader had of the project.

Methods. Before and after intervention, a survey containing three parts (Sense of Coherence scale, Likert-scale questions on attitudes, open questions on

expectations/experiences of the project) was distributed to the 27 inmates staying at the particular department. The project leader was interviewed before and after he conducted the intervention. Quantitative analysis was performed but limited due to low sample size, qualitative data were subjected to thematical analysis.

Results. Minor indications were found that inmates who participated in the intervention had a more positive change in their sense of coherence than those who did not. Inmates held positive attitudes towards the project. Both inmates and project leader noticed positive impacts of the intervention on the participants.

Conclusion. Skådebanan’s music intervention may function as a catalyst for change and indirectly contribute to inmates’ desistance from crime. However, no firm conclusions can be drawn and further research is needed to explore potential effects of this musical initiative.

Keywords: desistance from crime, evaluation, music, positive criminology, prison- based programme, sense of coherence, Sweden.

Total word count: 8.861

(3)

ACKNOWLEDGEMENTS

I would like to express my appreciation for the helpful feedback and insights of my supervisor Claes Andersson, which truly lifted this work to a higher level.

Secondly, I am grateful for Skådebanan offering me the opportunity to be part of their project and for ‘Christian’, the project leader, his help and participation in the research. I would also like to thank the inmates who took part in this study.

Lastly, I am thankful for the mental support I received from family members, friends and my partner in particular. Thank you all for making this work possible.

(4)

INTRODUCTION

The main task of prisons is to protect society by reducing criminal recidivism (Legislative Analyst’s Office, 2017). Prison programmes aim to contribute to this by fostering rehabilitation and can take various forms, for example therapy, educational programmes, professional training activities, sports and cultural activities (Maculan et al., 2013) and often target particular groups (Day et al., 2006; Heseltine et al., 2011). To validate prison programmes’ quality and effectiveness in terms of whether they contribute to factors that in their turn reduce recidivism, it is of importance to subject them to evaluation (Gendreau, 1996; Leschied et al., 2001). This can provide crucial information to the prison institution on what programmes to implement and to possible third parties on how to organize these.

The present study evaluates a prison-based music intervention conducted by the cultural non-governmental organization Skådebanan at a Swedish prison in Spring 2021. The evaluation may play a role in the further development of the present and other prison-based interventions organized by Skådebanan.

Positive criminology

Correctional approaches such as educational prison programmes or in general approaches aiming to develop positive identity and life meaning, are based on positive criminology (Ronel & Elisha, 2011). This concept was pioneered by aforementioned authors and is defined as a perspective incorporating existing criminological theories and models that focus on offenders’ encounters with positive-experienced influences distancing them from crime. These influences can be based on personal traits or be brought about by (in)formal therapy programs or through stressing positive social aspects (Ronel & Elisha, 2011).

Positive criminology can thus provide a framework in conducting research on prison programmes. Ronel and Segev (2014) for instance stated that:

Criminological research might benefit from focusing their attention on, for example, various criminal justice innovations and the manner in which these innovations can be experienced positively, as well as how that positive experience might play a role in the desistance process. (p.1392)

Elisha and colleagues (2013) for example already found support for the ideas of positive criminology in their qualitative research on how incarcerated sex offenders experience their change process during their prison stay.

Desistance from crime

Desistance from crime is a research area within positive criminology (Ronel &

Elisha, 2011) that specifically focuses on understanding (ex-)offenders’ reasons for and processes of ceasing to commit crimes over time (Maruna, 2001).

Important here is to consider the transition from criminal to noncriminal

behaviour not as a single event, but as a complex process (Carlsson, 2012; Laub &

Sampson, 2003; Maruna, 2001) that multiple pathways (e.g. ageing, stable employment, attachment to conventional persons, a transformation of identity) can lead to (Laub & Sampson, 2001). Maruna (2001) found that many offenders credited the empowering effects of external parties (e.g. relatives, partner,

(7)

children, an institutional authority, or other people who believed in them) for changing their way of life.

Antonovsky’s Sense of Coherence

In Ronel and Elisha’s (2011) discussion of positive criminology, Antonovsky’s salutogenic model (1979, 1987) is often referred to. This is no coincidence since positive criminology is seen as an adaption of the perspective of positive

psychology (Ronel & Segev, 2014) and the salutogenic framework has been

connected to this same perspective (Huss & Samson, 2018; Joseph & Sagy, 2017).

With his salutogenic model, Antonovsky (1979, 1987) argued for a focus on factors, pathways and causes of positive health within health research to complement the knowledge on prevention and treatment of negative health aspects (the pathogenic approach). When applying the salutogenic model to criminology, the focus would lie on what factors are enabling an offender’s successful re-entry into society (Zajenkowska et al., 2021). Antonovsky (1979, 1987) furthermore emphasized the importance of how one manages the

unavoidable stressors that life carries and used the concept of sense of coherence (SOC) to refer to this coping capacity, defined as:

A global life orientation that expresses the extent to which one has a

pervasive, enduring though dynamic feeling of confidence that one’s internal and external environments are predictable and that there is a high probability that things will work out as well as can reasonably be expected. (Antonovsky, 1979, p.10)

SOC thus exists of (1) a cognitive aspect (comprehensibility), (2) a behavioural component (manageability) and a motivational aspect (meaningfulness)

(Antonovsky, 1979). People with a high SOC are less likely to feel pressured by tense experiences and are more likely to have a high quality of life (Eriksson &

Lindström, 2005). In criminological research, Ristkari and colleagues (2009) for example found that a low SOC was associated with higher levels of criminality.

Interestingly, in some research evaluating prison-based interventions focused on rehabilitation, the SOC-concept has been used to measure changes in the inmates and indications were found that the programmes contributed to an increase in the inmates’ SOC, what might be beneficial for their behaviour (e.g. Berman, 2004;

Lindblom et al., 2017). As for arts-based programmes, a study on art therapy in a health setting suggests that the use of arts includes mechanisms enhancing comprehensibility, manageability and meaningfulness (Huss & Samson, 2018).

Arts-based prison programmes

Arts-based prison programmes are projects, workshops or courses offered in criminal justice institutions with a focus on various art forms, such as visual art, poetry, drama, dance and music (Justice Arts Coalition, n.d.). These kinds of programmes usually focus on empowerment and thus fit the positive criminology framework and can be linked to desistance of crime.

In a literature review exploring arts-based prison projects, McNeill and colleagues (2011) presented six themes crucial for these projects’ effectiveness and

contribution to the process of desistance: identity and diversity, motivation and hope, self-determination and respect, professional and personal supporters,

(8)

personal and social strengths and resources and social capital. According to Henley (2015), who explored the relationship between musical learning and desistance from criminal behaviour, all these themes can also be found in the literature specifically on music projects in criminal justice settings.

Multiple authors (Caulfield et al., 2016; Cheliotis & Jordanoska, 2016; Henley et al., 2012; Hughes, 2005; McNeill et al., 2011) have emphasized that arts-based prison programmes only indirectly contribute to desistance from crime. Here, it is important to look at the distinction between primary and secondary desistance.

Primary desistance refers to any crime-free gap in a criminal career course, while secondary desistance implies changes in one’s role or identity, that lead to

perceiving oneself as a ‘changed person’ or non-offender (Maruna & Farrall, 2004). Cheliotis and Jordanoska (2016) extended the concept of secondary desistance to ‘soft’ conditions, referring to “an array of other ways in which arts- based programmes may indirectly contribute to desistance from crime” (p.26), arguing that their emergence may assist in achieving primary desistance. Since it is rather to these ‘soft’ conditions associated with recidivism (e.g. motivation to change, personal and social skills) that arts-based prison interventions contribute (Cheliotis & Jordanoska, 2016; McNeill et al., 2011), such projects can more accurately be seen as catalysts for change (Caulfield et al., 2016; Henley et al., 2012). Regarding music programmes specifically, Henley (2015) suggested that musical learning and developing key attributes that inspire desistance emerge from shared learning processes.

Hughes (2005) conducted a review of research literature, practice and theory on the use of arts in the criminal justice sector and concluded that evidence suggests that participatory arts in such setting have impact on neurological, cognitive, affective and behavioural levels. She also theorized that different art forms could have different influences on offenders: music could possibly be more connected to communication skills, while creative writing and storytelling might be more linked to exploring moral issues and naming experiences for example.

More specifically, manifold positive impacts (and elements contributing to these) of arts-based prison programmes on inmates have been found. A non-exhaustive overview is presented in Table 1, where studies that conducted research on music- based programmes are marked to indicate which effects were also found for this specific art form. Following Cheliotis and Jordanoska (2016), all these impacts should be understood as ‘soft’ conditions that can indirectly contribute to desistance from crime.

(9)

Table 1. Arts-based prison programmes’ positive impacts on inmates and elements contributing to these impacts (non-exhaustive overview).

Studies Positive impact on inmates’…

a Mood Ca ulfield et a l., 2016*; Gussa k, 2007

b Self-confidence a nd self-esteem Ba in & Eisen-Ma rtin, 2017*; Brewster, 2010;

Brewster, 2014; Ca ulfield et a l., 2016*; Chen &

Ha nniba l, 2019*; Henley, 2015*; Henley et a l., 2012*; Wilson et a l., 2009*

c Feelings of a nger, stress, depression, a nxiety or boredom

Ca ulfield et a l., 2016*; Chen & Ha nniba l, 2019*;

Gold et a l., 2014*; Gussa k, 2007 d Feelings of sa fety a nd

mea ningfulness

Brewster, 2010; Ca ulfield et a l., 2016*

e Socia l skills Brewster, 2014; Ca ulfield et a l., 2016*; Gussa k, 2007; Koch et a l., 2015; Wilkinson & Ca ulfield, 2017; Wilson et a l., 2009*

f Communica tion a nd colla bora tion skills

Ca ulfield et a l., 2016*; Henley et a l., 2012*;

Wilkinson & Ca ulfield, 2017; Wilson et a l., 2009*

g Motiva tion to lea rn, succeed or pa rticipa te in other a ctivities

Anderson & Overy, 2010*; Brewster, 2014;

Ca ulfield et a l., 2016*; Henley, 2015*; Wilkinson &

Ca ulfield, 2017; Wilson et a l., 2009*

h Rela tionships with sta ff, peers a nd fa mily

Brewster, 2010; Brewster, 2014; Henley, 2015*;

Wilkinson & Ca ulfield, 2017; Wilson et a l., 2009*

i Coping skills a nd emotion regula tion

Brewster, 2014; Henley et a l., 2012*; Willia ms &

Ta ylor, 2004

j Development of positive identity Henley, 2015*; Willia ms & Ta ylor, 2004 Elements contributing to impacts

a Feeling of mea ningfulness, responsibility a nd freedom

Ca ulfield et a l., 2016*

b ‘Humanisation’ Ca ulfield et a l., 2016*; Cheliotis & Jorda noska , 2016 c Experience of respectful a nd

positive rela tionships

Ca ulfield et a l., 2016*; Winder et a l., 2015*

d Genuine interest a nd desire for a rtistica l expression

Gussa k, 2007

e Feeling that one’s voice matters Ba in & Eisen-Ma rtin, 2017*,

f Enga ged project lea der, role model Brewster, 2014; Cheliotis & Jorda noska , 2016;

Henley, 2015*; Tett et a l., 2012 g Unthrea tening environment free of

pressure

Wilson et a l., 2009*

* Indica tes study (pa rtly) focused on a music-ba sed project specifica lly.

Skådebanan’s project “Culture for Life”¹

In 2020, the Swedish cultural non-profit organization Skådebanan initiated, with the support of Postkodstiftelsen, the project “Culture for Life”: an arts-based project in different settings within the Swedish Prison and Probation Services². According to the project plan, written by Göran Sarring from Skådebanan (2021), the overall aim is to bring culture to inmates, a group with very limited access to cultural activities, in the hope they would find strength and motivation to leave their criminal life-styles and identities by discovering their own (creative) possibilities. The project thus embraces the ideas from positive

psychology/criminology.

As a start, “Culture for Life” includes three pilot projects with different focus in the geographic regions of Norrbotten (theatre), Dalarna (bibliotherapy) and Skåne (music). All three projects will be conducted twice and evaluated each time by

1 Tra nsla ted from the origina l na me “Kultur för Livet”.

2 In Sweden known a s ‘Krimina lvå rden’.

(10)

university students from Umeå, Gävle and Malmö. When the pilot projects have taken place, Skådebanan intends to create a new, bigger cultural project in settings within the Prison and Probation Services in eight Swedish regions. Thus,

Skådebanan hopes that the evaluations will provide insights on how the projects affect the participating inmates, benefitting the development and execution of future projects.

The music-based project in Skåne takes place in the prison institution of Kristianstad, with a musician from Skådebanan as project leader. It has been conducted once in Spring 2021 and will be offered again in Fall 2021, with places for six inmates each time. This study evaluates the Spring edition and thus

functions as the half-way evaluation of the music project.

AIM

The overall objective of the present study is to evaluate effects of the music intervention of “Culture for Life” offered by the organization Skådebanan to inmates at the prison in Kristianstad (Sweden) during Spring 2021. Specific research questions are:

1. Did sense of coherence (SOC) develop in a positive direction in inmates participating in the music intervention compared to non-participating controls?

2. What were the overall attitudes towards the music intervention and did attitudes develop in a positive direction in inmates participating in the music intervention compared to non-participating controls?

3. How did inmates comment on the music intervention, and did comments differ between pre- and post-assessment and between inmates participating in the music intervention and non-participating controls?

4. What expectations and experiences were expressed by the project leader?

METHODS

Setting

Skådebanan’s music project was conducted at the prison of Kristianstad, an institution for sex-offenders housing 115 inmates divided over five departments.

The intervention was offered in a low-security department housing 27 inmates.

All inmates are males and the time spent at the specific department is usually maximum 1.5 years. At the time of the intervention, the age range was

approximately 20-70 years and most served their first sentence, lasting around 2-4 years in total.

Music intervention

The project leader responsible for conducting the music intervention is a professional musician with no previous experience in working with a prison population. The music intervention included 8 weekly sessions of 45 minutes with

(11)

the following content: (1) inventory of musical goals and ways of

communication³, (2-7) creating music, (8) recording. One week after the final session, the project leader delivered a CD for each inmate with the recording they made. Additionally, the project leader provided the inmates with contact

information of himself (specially created email address) and of possible

interesting parties they could get in touch with after their release if they wished to proceed with music.

Research design

A twofold design was used to collect data on two levels (see Figure 1). Inmates, both project participants and non-participants, at the department responded to a survey before and after the intervention period (pretest-posttest nonequivalent groups design). Similarly, the project leader who held the group sessions was interviewed before and after the intervention period.

Figure 1. Research design.

Participants

To clarify the inmates’ group division and research participation, a flowchart is presented (Figure 2).

Figure 2. Flowchart of the inmates’ group division and research participation.

Boxes with thick borders indicate the final groups included in analysis.

3 The project lea der ha d experience with IPU tools, which a re developed by Institutet Personlig Utveckling (the Institute for Persona l Development) a nd used to understa nd underlying fa ctors of one’s behaviour in organizational contexts (Institutet för Personlig Utveckling, n.d.). He used some elements from this fra mework a s a lea d for getting to know the inma tes’ driving forces a nd communica tions styles.

(12)

Skådebanan placed an announcement of the music intervention at the department, offering participation to all 27 inmates, of which 5 signed up. Before

participation, these interested inmates were shortly interviewed by the project leader to secure (a) stay at the department during the project period, and (b)

motives for participation. All 5 were included, being the intervention group. Then, all inmates were asked to respond to pre-and post-intervention surveys. Response rates were 9/27 (±33%) and 7/27 (±26%) for pre- and post-assessment,

respectively. It should be noted that the department chief estimated a change of 5- 10 inmates in between the two assessments, but stated that the total department population always was 27 (therefore, nout = nin).

Procedure

At both pre- and post-assessment, informed consent and surveys were distributed to the inmates with the help of staff at the department. Due to security reasons, hand-out and collection could not be done by the author herself. To guide the process of distribution and collection however, a document with instructions for the staff as well as one for the department chief were provided (see Appendices A and B). The inmates received the surveys within one week before and af ter the intervention and had three days to respond before the surveys were collected by the staff. Further, surveys were available in Swedish and English. By default, the Swedish surveys were handed out, together with the informational letter that stated that an English version was available upon request. The prison staff were also asked to verbally inform the inmates about the English alternative during hand-out. In both assessments, all English surveys were returned empty and separate from the collected surveys, indicating they had not been necessary.

After informed consent, the project leader was interviewed within two days before and after he conducted the intervention. The interviews were in English and took place in a calm environment. The durations of both interviews were

approximately one hour, which together resulted in 2 hours and 5 minutes of audio material. This material was then transcribed before analysis.

Measures

The surveys distributed to the inmates in pre- and post-assessment are shown⁴ in Appendices C and D, and include the following content:

(1) Sense of coherence. Sense of coherence (SOC) was assessed at both pre- and post-assessment by using a 13-item scale (SOC-13, Antonovsky, 1987). The scale measures how one experiences life and consists of three components:

comprehensibility (5 items), manageability (4 items) and meaningfulness (4 items). All items are answered on a 7-point scale, where 1 and 7 indicate more extreme feelings towards the statements and questions. Summed end scores range from 13-91 for total SOC, and for the subscales from 5-35 (comprehensibility) and 4-28 (manageability and meaningfulness). Higher scores indicate a stronger SOC (Antonovsky, 1987). Overall, the scale’s reliability, validity and cross- cultural application have been positively evaluated (Antonovsky, 1993; Eriksson

& Lindström, 2005, Eriksson & Mittelmark, 2017; Olsson et al., 2009). The internal consistency of SOC-13 is considered good, with Cronbach’s alpha scores ranging from 0.70 to 0.92 (Eriksson & Lindström, 2005). In a Swedish population specifically, Lindmark and colleagues (2010) found the Cronbach’s alpha for the

4 Due to copy-right, it is not a llowed to publish the full SOC-sca le. Therefore, only 3 questions (nr.1, 6, 13) a re shown in the Appendices, others a re left bla nk.

(13)

SOC-13 scale to be 0.86 for the total scale and 0.85, 0.80 and 0.84 for

comprehensibility, manageability and meaningfulness, respectively. In the present study, a Cronbach’s alpha value of 0.46 (pre-assessment) and 0.74 (post-

assessment ) was found for the total scale. For the three subscales, Cronbach’s alpha pre- and post-values of 0.45 and 0.66 for comprehensibility, 0.73 and 0.53 for manageability, and 0.73 and 0.71 for meaningfulness were found.

(2) Attitudes towards the project. Both pre- and post-assessment included three questions assessing attitudes towards the project regarding meaningfulness, helpfulness and recommendability. Each question was answered on a 9-point Likert scale ranging from 1 (negative) to 9 (positive). The questions were

formulated by the author, but inspired by the credibility/expectancy questionnaire of Devilly and Borkovec (2000).

(3) Comments on the music intervention. The survey included open-ended questions formulated by the author on expectations (pre-assessment) and

experiences (post-assessment) of the music intervention. At post assessment, the questions slightly differed for the intervention and control group since they are thought to have different experiences. Participants in the intervention group were asked about their satisfaction with the project, experiences and thoughts

(strengths, elements for improvement, method) of the project, while inmates in the control group were asked about their impressions of the project and whether their interest for participation changed.

For the interviews with the project leader, an interview guide was developed by the author (See Appendix E). Focus of these interviews was methods used and thoughts about the project including expectations and experiences of impacts of the intervention.

Analysis

Quantitative data

The software used for statistical analysis was IBM SPSS Statistics, Version 27.

Two quantitative analyses were done in order to answer the research questions regarding the inmates’ SOC (1) and attitudes towards the project (2). The following description applies to both analyses.

First, 2 respondents from the control group who filled in the pre-assessment only were excluded⁵. Further, within both intervention and control group a distinction was made between respondents who completed both assessments and those who answered one assessment only, resulting in four groups being subjected to

analysis. Then, means and standard deviations, change scores (difference between pre-and post-assessment), and proportions representing the amount of respondents with a positive, neutral or negative change were calculated for all groups. Due to small sample size (see Figure 2), no statistical tests were performed to look at between- and within-group differences regarding mean scores and change scores.

Alternatively, Fisher's Exact test (two-tailed) was used to analyze whether

aforementioned proportions differed between intervention and control group (who completed both assessments), where p < .05 was regarded as statistically

5 These respondents indica ted tha t they ha d wished to pa rticipa te in the music intervention.

Therefore, their a nswers might ha ve been closer to those of the intervention group. It wa s opted to a na lyze them a s a sepa ra te group, but the la ck of post-a ssessments a nd missing items in the pre- a ssessment for one of them, led to the decision to exclude them from the qua ntita tive a nalysis.

(14)

significant. The proportions formed the dependent outcome, while intervention or control was the fixed variable.

Qualitative data

Inmates’ answers on the open questions as well as the transcribed interviews with the project leader were analysed in NVivo (Release 1.0, March 2020), a software program for qualitative and mixed-methods research. Two separate thematical analyses were done in order to answer the research questions considering the inmates’ comments on the music intervention (3) and the project leader’s expectations and experiences (4). Note that here, answers from all inmates who filled in open questions were included without distinguishing between who completed both or only one assessment, resulting in two groups: intervention and control group.

Thematical analysis is a method that allows to identify, analyse and report themes in data (Braun & Clarke, 2006). According to Spencer and colleagues (2013), there are four steps a researcher can adopt when doing a thematical analysis. First, it is important to get familiar with the data, which was achieved by typing out the inmates’ answers on the open questions, transcribing the interviews, and then reading through all the collected material. Then, preliminary thematical frameworks were created, which in this study was done according to the open questions and interview guide. Thirdly, those frameworks were used to sort and code the textual material. The coherences of the data extracts were reviewed and the frameworks further refined as the final step of the thematical analysis (Spencer et al., 2013).

For the comments of the inmates, a final framework with following three categories was found adequate to present the findings: (a) reasons to (not) participate, (b) impacts of the music intervention and (c) thoughts on the project (strengths, elements for improvement and method). The results are presented for pre- and post-assessment separately within intervention and control group each.

The data from the interviews with the project leader are shown within the categories of (a) impacts on inmates and (b) impacts of prison setting on project.

In the presentation, a distinction is made between expectations (pre-assessment) and experiences (post-assessment).

ETHICS

This study was approved by the Ethics Council at the Faculty of Health and Society at Malmö University (application number: HS2021 #17, see Appendix F).

An application was handed in because of three reasons. Firstly, a prison population is regarded as a vulnerable research group because of a power imbalance and limited autonomy inherent to the environment the inmates find themselves in (Field et al., 2019; Hayes, 2006). Secondly, that the research participants have been convicted for crime is considered as sensitive data according to the Swedish Law on Ethical Review of Research involving Human Subjects (SFS 2003:460). Thirdly, information on the inmates’ SOC is collected, a concept that has been linked to mental health (Olsson et al., 2006), thus also considered as sensitive data (SFS 2003:460).

(15)

Above matters were handled through providing clear information and asking for consent, i.e. informed consent, as demanded by the previous mentioned Swedish law (SFS 2003:460). The informational letters for both pre- and post-assessment (see Appendix G and H) used simple, clear language in function of readability.

Even two versions of the informational letters were provided: one version that presented an overview (front side of the paper) and one version with continuous text (back side of the paper). The overviews were added because both the project leader and the prison department chief recommended to have a shorter version to increase the chance of the inmates reading it. It was also clearly stated that inmates’ decision to participate in the evaluation research would not affect their stay in prison and they could always withdraw their participation by telling the staff, who would then inform the author. The inmates were guaranteed anonymity by pseudonymization of their four initials⁶ to a number when digitalizing the data and by anonymous reporting of the results. The informational letters also included information on how the data would be handled and where the study would be published. There were no requirements to sign an informed consent form to further guarantee anonymity. Instead, the informational letters stated that one agrees to participate by filling in and returning the survey. To ensure anonymity towards the prison staff, the surveys were handed out in big white envelopes that could be sealed before returning.

For both interviews, the project leader received the informational letter

beforehand that stated that participation was voluntarily and could be withdrawn at any moment, and provided information on how the data would be handled, reported and published. Before the start of the interviews, the author went through the informational letter and asked to sign the informed consent form.

For both the inmates and the project leader, the risk of harm because of research participation was estimated as very low.

RESULTS

Inmates’ Sense of Coherence

Table 2 shows data collected on the assessment of sense of coherence (SOC). Pre- and post-assessment scores, change scores, and proportions (the number of

participants with a positive/neutral/negative development) are presented for inmates who completed either both or one assessment for both intervention and control group. Only between-group differences in proportions were tested with Fisher’s Exact test (FET) and are described; for other comparisons, tendencies are shortly outlined.

6 The inma tes were a sked to fill in the two first letters of their first a nd la st na me with the only purpose to connect the pre- a nd post-a ssessment. This wa s sta ted in both the informa tiona l letter a nd the survey itself where they ha d to fill in these four letters. No other persona l da ta were obta ined.

(16)

Table 2. Group means, change scores and proportions for SOC-scale and subscales.

Both a ssessments completed One a ssessment completed Control

group n=2

Intervention group

n=3

Control group

n=2

Intervention group

n=2 SOC Tota l Score Pre 50.00 (1.41) 43.33 (6.03) 63.50 (3.54) -

Post 55.00 (15.56) 54.67 (14.74) - 48.00 (7.07)

Cha nge 5.00 (14.14) 11.33 (8.96) - -

Prop.^a 1/0/1 3/0/0 - -

I. Comprehensibility Pre 16.00 (0.00) 15.00 (7.00) 22.50 (2.12) - Post 17.00 (7.07) 21.67 (7.77) - 14.00 (1.41)

Cha nge 1.00 (7.07) 6.67 (1.15) - -

Prop. 1/0/1 3/0/0 - -

II. Ma na geability Pre 19.50 (6.36) 11.33 (1.53) 15.50 (4.95) - Post 19.50 (0.71) 13.67 (4.16) - 17.50 (2.12)

Cha nge 0.00 (5.66) 2.33 (3.79) - -

Prop. 1/0/1 2/0/1 - -

III. Mea ningfulness Pre 14.50 (7.78) 17.00 (1.73) 25.50 (3.54) - Post 18.50 (9.19) 19.33 (2.89) - 16.50 (7.78)

Cha nge 4.00 (1.41) 2.33 (4.62) - -

Prop. 2/0/0 2/0/1 - -

All va lues for Pre, Post, a nd Cha nge a re presented a s Mea n (Sta nda rd Devia tion). Da shes mea n no da ta were obta ined or could be ca lcula ted. Between-group differences in proportions were tested with Fisher’s exact test, no significant results at p <.05 were found.

aProp. refers to proportion, i.e. a mount of people who ha d a positive/no/nega tive cha nge (+/0/ -).

Looking at the inmates who answered both assessments, those from the control group show higher scores (except for meaningfulness) at the pre-assessment than the inmates who received the intervention. The same differences cannot be observed in the post-assessment, where the intervention group shows similar or higher means for total SOC and comprehensibility. The change scores indicate that inmates from the intervention had a more positive development in their SOC, except regarding meaningfulness. Although no significant differences were found, more positive proportions of change are observed in the intervention group

compared to the control group for SOC (p=.4, FET) and for comprehensibility and manageability (for both p=1, FET), but not for meaningfulness (p=1, FET).

Additionally, inmates in the control group who completed only pre-assessment have almost exclusively higher scores than other respondents, while inmates in the intervention group who only responded to the post-assessment have mostly lower scores compared to all other participants.

Inmates’ attitudes towards the music intervention

Table 3 shows scores on attitudes towards the music intervention in terms of meaningfulness⁷, helpfulness, and recommendability. Pre- and post-assessment scores, change scores, and proportions (the number of participants with a positive/neutral/negative development) are presented for the intervention group, the control group and the group as a whole. Here too, only between-group

7 Note tha t the concept here refers to one specific question tha t wa s a sked to the inma tes, not to one of the sub-sca les of the SOC-sca le.

(17)

differences in proportions were tested (FET) and are presented; other observable trends are shortly described.

Table 3. Group means, change scores and proportions for attitudes towards project.

Both a ssessments completed

One a ssessment

completed Tota l Control

group n=1

Intervention group

n=3

Control group

n=1

Intervention group

n=2 n=5/6/4^b Mea ningfulness Pre 7.00 (-) 7.67 (1.53) 9.00 (-) - 7.80 (1.30)

Post 8.00 (-) 7.67 (2.31) - 9.00 (0.00) 8.17 (1.60) Cha nge 1.00 (-) 0.00 (1.00) - - 0.25 (0.96)

Prop.^a 1/0/0 1/1/1 - - 2/1/1

Helpfulness Pre 5.00 (-) 7.33 (2.08) 9.00 (-) - 7.20 (2.05) Post 7.00 (-) 8.00 (1.73) - 9.00 (0.00) 8.17 (1.33) Cha nge 2.00 (-) 0.67 (0.58) - - 1.00 (0.82)

Prop. 1/0/0 2/1/0 - - 3/1/0

Recommendability Pre 5.00 (-) 6.67 (2.52) 1.00 (-) - 5.20 (3.03) Post 6.00 (-) 7.67 (2.31) - 9.00 (0.00) 7.83 (1.84) Cha nge 1.00 (-) 1.00 (1.00) - - 1.00 (0.82)

Prop. 1/0/0 2/1/0 - - 3/1/0

All va lues for Pre, Post, a nd Cha nge a re presented a s M ea n (Sta nda rd Devia tion). Da shes mea n no da ta were obta ined or could be ca lcula ted. Between-group differences in proportions were tested with Fisher’s exact test, no significant results at p <.05 were found.

aProp. refers to proportion, i.e. a mount of people who ha d a positive/no/nega tive cha nge (+/0/ -).

bN for Tota l is 5 for the pre-a ssessment, 6 for the post-a ssessment a nd 4 for cha nge scores a nd proportions.

The numbers for participants who completed both assessments show somewhat more positive attitudes in the intervention group compared to the control group at both pre- and post-assessment. Change scores are more positive in the control group. No significant between-group differences regarding proportions of change could be identified (for all three questions p=1, FET).

In comparison to those who completed both assessments, scores are generally higher in those who responded to one assessment, except for recommendability where the single respondent in the control group provided a low score.

Looking at the total, all research participants hold positive attitudes towards the music intervention regarding all three concepts, with slightly higher scores in the post-assessment, indicating an improvement in attitudes towards the project.

Meaningfulness and helpfulness of the project were assessed somewhat better than recommendability.

Inmates’ comments on the music intervention

An overview of the thematical analyses of the inmates’ comments are presented in Tables 4-6 according to three categories. A division between pre- and post-

assessment in intervention and control group each is made to compare between and within groups.

(18)

Reasons to (not) participate

Reasons inmates gave for (not) wanting to participate in the music intervention are shown in Table 4. Within the intervention group, musical reasons such as wanting to develop musical skills, record a song and artistic freedom were given, but music was also mentioned as a possibility for support or coping with

emotions. Besides, positive atmosphere, to have variation in activities and spend the prison time in a meaningful way were motivations. The inmates were not asked for their motivation in the post-assessment, but all five answered on another question that they would participate again in the music project.

In the control group, someone expressed a lack of interest in music as a reason to not participate. Others stated they preferred to focus on other plans or treatment.

Two inmates of the control group in the preassessment for example did have interest in music but would not participate because of other plans or practicalities.

In the post-assessment, respondents from the control group indicated they were still not interested if there would be a next opportunity.

There were 7 out of 27 (±26%) inmates who expressed interest in participating in the music intervention and above findings suggest that the specific art form of music is a reason for both wanting and not wanting to participate. Besides, some regard the intervention as an interesting alternative to their usual day schedule, while others preferred to focus on other activities/programmes available.

Table 4. Overview of thematical analysis on inmates’ reasons to (not) participate.

Intervention group Control group Pre (n=3) Post (n=5) Pre (n=3 [2]^a) Post (n=2)

Musica l rea sons 2 - 1 [2] 1

Music a s support 1 -

Music a s coping 1 -

Positive a tmosphere 1 -

Va ria tion in a ctivities 1 -

Mea ningfulness 1 -

Other pla ns/focus - 1 [2]

Numbers represent the a mount of respondents whose a nswers fit the theme. Bla nk spa ces equa l a value of zero. Dashes indicate ‘not applicable’ because the respondents did not receive the question.

a Numbers between squa re bra ckets represent the respondents who expressed interest in pa rticipa ting in the music intervention but eventua lly did not.

Impacts of music intervention

Table 5 presents an overview of the inmates’ expectations and experiences of the impacts of the project. Within the intervention group, musical development and achievement (e.g. produce something that is good enough to publish) and personal growth (increase in self-esteem, becoming a stronger person) emerge as themes in both assessments. So does mood improvement (feeling happier and more positive), named by one person in the pre-assessment, and by four in post-assessment.

Music as support and meaningfulness were not named again, but four new

positive themes arise with the intervention as a get-away from one’s thoughts and prison life, and gaining new insights as most prevalent unexpected impacts.

(19)

The inmates from the control group expected that the intervention would help the project participants in their musical development and offer them a positive

atmosphere, meaningfulness and a get-away (break boredom and isolation).

When in the post-assessment asked about how they thought the ones who participated were impacted, the same answers (except for get-away) were given.

Moreover, some new themes emerge, for example mood improvement and variation in activities. Inmates from the control group were also asked about the project’s influence on themselves (answers not shown in Table): one enjoyed the positive energy the project leader spread, one said not to be influenced at all.

Looking at above results, it seems that the music intervention fulfilled most expectations and even exceeded them, indicated by new themes emerging in the post-assessment. The findings suggest that inmates who received the intervention mostly felt impacts on oneself personally (e.g. personal growth, mood

improvement, new insights) and on their musical skills, which inmates from the control group partly suspected as well. Remarkably, several intervention

participants put forward personal growth as an impact, whereas respondents from the control group never mentioned this. Also, there is a limited overlap in answers between both groups, which could mean that no elaborate communication took place between inmates about impacts of the project.

Table 5. Overview of thematical analysis on inmates’ expectations and experiences of impacts of music intervention.

Intervention group Control group Pre (n=3) Post (n=5) Pre (n=3 [2]^a) Post (n=2)

Musica l development 2 2 [2] 1

Music a s support 1

Persona l growth 2 3

Positive a tmosphere [1] 1

Mea ningfulness 3 1 1

Mood improvement 1 4 1

Get-a wa y 2 1

Insights 2

Stimula tion 1 1

Crea tivity 1

Va ria tion in a ctivities 1

Sa tisfa ction 1

Numbers represent the a mount of respondents whose a nswers fit the theme. Bla nk spa ces equa l a va lue of zero.

Thoughts on project

In the post-assessment, inmates from the intervention group only were asked about their thoughts on the strengths, elements for improvement and method of the project (see Table 6). Contributing to personal growth and mood improvement were both mentioned by multiple inmates as strengths. Six other themes emerge from their answers here, amongst others get-away and help to leave criminal ways. Regarding what could be improved, four out of five named issues that can be linked to implementation, such as the need for more time, sessions and budget.

(20)

Only two answers were given about the method, stating that the project leader was great and that everything was good.

What was regarded as strengths of the intervention, mostly relates to the personal impacts the inmates experienced. One inmate made a direct connection with offending. Further, statements about what could be improved indicate that the inmates find it important to have such projects offered in the prison institution.

Table 6. Overview of thematical analysis on inmates’ thoughts on project.

Intervention group Control group Pre (n=3) Post (n=5) Pre (n=3 [2]^a) Post (n=2) Strengths

Persona l growth - 3 - -

Mood improvement - 2 - -

Get-a wa y - 1 - -

Stimula tion - 1 - -

Opportunity to crea te - 1 - -

Approa ch project lea der - 1 - -

Frequency - 1 - -

Help people lea ve crimina l wa ys - 1 - -

Room for improvement

Implementa tion - 4 - -

Offer other culture forms - 1 - -

Method

Approa ch project lea der - 1 - -

All wa s good - 1 - -

Numbers represent the a mount of respondents whose a nswers fit the theme. Da shes indica te ‘not applicable’ because the respondents did not receive the question.

Overall emerging themes

When looking at all three categories, the themes that appear at least twice are musical development, music as support, personal growth, mood improvement, stimulation, positive atmosphere, variation in activities and the music sessions as a get-away. Worth notifying is that meaningfulness was mentioned several times in the pre-assessment mostly by the intervention group, while it was mentioned only once in post-assessment by an inmate from the control group as a possible effect on the ones who did participate. Overall, differences between both groups can be noticed: answers from the intervention group are often more person-related and concrete than those of the control group.

Project leader’s expectations and experiences

Table 7 presents an overview of the thematical analysis on the project leader’s expectations and experiences of how inmates were impacted and how the prison setting influenced the project, which were expressed before and after the music intervention was conducted.

(21)

Table 7. Overview of thematical analysis on interviews with project leader.

Expecta tions Experiences Influence on inmates

Intervention group

Musica l development X X

Culture a s support X X

Music a s coping X X

Persona l growth X X

Colla bora tion skills X

Communica tion skills X

Stimula tion X X

Insights X X

Get-a wa y X X

Help people lea ve crimina l wa ys X X

Wish for culture room X

Control group

Development interest for culture X X

Positivity a bout project X X

Impact of prison setting

Infra structure X X

Schedule X

Security mea sures X

Impacts on inmates

Most of the expectations the project leader had about the impacts of the intervention were also experienced by him during the sessions through observation and conversations with the participating inmates. He noticed for example that inmates developed and improved their musical skills (e.g. song- writing) and some showed a greater belief in themselves. Moreover, he thought the music intervention succeeded in its intentions of contributing to inmates’ view of culture as a form of support (something positive to get strength out of) and music as a way of coping with emotions and difficult situations. He also believed that the project would and did help some inmates leaving criminal ways (e.g. by providing them with potential contacts for when they are released, which can possibly lead to supportive friends), which can be illustrated by the following quote: “That you feel when you come out that you're not coming out from two doors to the open but you come out from two doors but into another one”. This could however not be observed, as this is a rather long-term effect and the inmates were still in prison at the end of the intervention. Furthermore, as the project leader planned beforehand, he shared insights with the inmates and felt like this helped them to see things in a different perspective.

Since collaborations between inmates were not allowed, no improvement in their collaboration and communicative skills with other inmates could be observed⁸ by the project leader. A new matter brought up in the post-interview was that the

8 However, it ha ppened once tha t two inma tes colla bora ted together, which went well. The project lea der a lso sta ted tha t some inma tes helped or crea ted something for a nother in their ‘own’ session time a s a n a lterna tive for a ‘live’ colla bora tion.

(22)

inmates expressed a wish for a permanent culture room in the prison institution, where different forms of culture could be practiced.

Regarding the inmates who did not receive the music intervention, the project leader thought that the project might awaken an interest for music or culture in some of them, or at least that they would regard Skådebanan’s project as positive.

Afterwards, he stated that he did observe a curiosity in them and he received only positive comments about the music intervention when he met them in the

hallways.

It appears that the project leader’s expectations and experiences of the

intervention’s impacts related to musical aspects as well as to how the inmates would be affected on a more personal and interpersonal level. The project leader stressed inmates’ wish for a culture room multiple times and advocated this himself as well.

Additionally it can be noted that the project leader did not expect that the fact that the inmates had committed sex crimes specifically would affect the project’s influence because he regarded one’s crimes independent of one’s possible interest in culture. Still, he did speculate in both interviews that it might be less likely for this group of offenders to be part of a gang or have a rooted criminal lifestyle and that this could influence their susceptibility for the music intervention as a

motivator for change.

Impacts of prison setting

Beforehand, the project leader expected that the prison setting would hinder the project in some way, but additional obstacles were still experienced. Regarding the infrastructure, concerns were expressed in the pre-interview, more specifically that the room foreseen would not be pleasant or comfortable. These worries turned out justified: as a lunch room was made available for him, this space had to be cleared during lunch, which forced him to pack up everything in between sessions⁹. There was also a lack of privacy as other inmates sometimes disturbed sessions by walking in. Then, two additional constraining factors were

experienced by the project leader. One was the tight day schedule of the prison, which limited the duration of the session and the pace of the music intervention.

Secondly, security measures did not allow musical collaborations between the participating inmates, nor could the project leader hand out material (CDs) to them for practice purposes. The project leader expressed that all these issues limited the music intervention and how he wanted to work.

DISCUSSION

Results

Inmates’ sense of coherence (SOC). Findings suggest a small tendency of more positive changes in inmates who participated in the “Culture for Life”-project’s music intervention on total SOC and on subscales comprehensibility and

manageability, compared with the control group. However, within- and between-

9 Eventua lly, it wa s decided to lock the room during lunch time the la st couple of sessions, a s the gua rds sta rted to understa nd to inconvenience.

(23)

group differences in means and change scores were not tested and differences between groups regarding proportions of inmates’ positive, neutral and negative changes were not found statistically significant. Therefore it cannot be concluded that this more positive tendency observed in the intervention group can be

ascribed to the project in question. Yet, the non-significance of the found

differences does not necessarily mean that Skådebanan’s music intervention does not contribute to inmates’ SOC. Rather, no conclusions should be drawn based on such small sample size.

Previous research with larger sample sizes on the effect of prison-based rehabilitation programmes on Swedish offenders’ SOC for example did find significant changes in SOC between intervention and control groups (Berman, 2004; Lindblom et al., 2017). Naturally, results might differ for arts-based interventions, but no previous research on arts-based prison programmes’ effects on inmates’ SOC specifically could be found to compare with. Huss and Samson (2018) suggested that the use of arts enhances comprehensibility, manageability and meaningfulness, but this research was conducted in a health setting.

Additionally, one should bear in mind that the inmates in the control group also followed their day schedule that included activities that have the goal to

rehabilitate them, therefore it is possible that their reported changes were affected by other prison programmes already in place. Also, the duration and low intensity of the music intervention are likely to play a role in the effects that were (not) brought about.

Inmates’ attitudes towards the music intervention. In general, inmates held very positive attitudes regarding the meaningfulness and helpfulness of the intervention. They were also quite likely to recommend the project to others in their situation. The findings suggests that the actual happening of the music intervention slightly altered the inmates’ attitudes in a positive direction, but it should be borne in mind that this change was not statistically tested. Nor were between-group differences in means and change scores tested, but it was observed that inmates from the intervention group tended to have more positive attitudes in both assessments and that change scores were slightly higher for the inmate in the control group. Possibly, there was less room for improvement in scores for the inmates who participated in the intervention, who had better attitudes from the start, compared to the inmate who did not participate. However, this explanation does not hold when looking at the inmate from the control group who filled in pre-assessment only, who gave the maximum score for both meaningfulness and helpfulness of the project. Lastly, no significant between-differences regarding proportions of inmates’ positive, neutral and negative changes were observed.

Looking at research on music interventions in prisons, no similar assessments on offenders’ attitudes towards music interventions themselves were identified. This is unfortunate because the perspective of those who receive a service or

intervention is valuable in evaluation research (Larsen et al., 1979).

Inmates’ comments on the music intervention. First of all, interest in the music intervention could be regarded as quite good considering that ‘making music’ is a very specific topic. Previous research on prison-based music projects seemingly has not commented the preliminary interest of the offenders, except for Gold and colleagues (2014) who noted that music therapy was well-accepted and found attractive. Importantly, all 5 participants indicated that they would participate again, suggesting that the intervention was a positive experience and beneficial to

(24)

them in some way. Reasons to (not) participate were mainly connected to a(n) (non-)interest in music as an art form itself and to whether they regarded the intervention as a good alternative to their day schedule, which corresponds exactly to what Henley and colleagues (2012) for example found.

Regarding impacts of the project, it is suggested that the project fulfilled and exceeded the expectations and that inmates from the intervention group seemed to give more concrete and personal-related answers than the control group. The eventually experienced impacts are in line with what previous studies on prison- based music programmes have found (see Table 1, Introduction). In this study, mood improvement and personal growth (expressed as self-esteem and -

confidence) came forward as the main themes, especially in the intervention group. These impacts have been reported by several authors as well (Bain &

Eisen-Martin, 2017; Caulfield et al., 2016; Chen & Hannibal, 2019; Henley, 2015;

Henley et al., 2012; Wilson et al., 2009). Also meaningfulness was mentioned in the present study and in previous research, and can be considered as both an effect and a condition contributing to effects (Caulfield et al., 2016). Furthermore, the music intervention as a ‘get-away’ or variation in activities could be linked to research findings on music programmes as breaking boredom (Caulfield et al., 2016). Interestingly, some inmates in this study focused on the musical

development itself, while previous research does not give attention to this aspect at all. This was noted before by Lucas (2013), who opposed the strong, often sole, focus on the therapeutic side of in-prison arts programmes that reduces arts to a

‘means to an end’ and offenders to objects of their art instead of the creating agents of it. Other prevalent themes were stimulation and positive atmosphere.

The latter can be linked to the importance of an unthreatening environment as an element contributing to impacts (Wilson et al., 2009). Also worth mentioning are the comments related to the project leader’s approach (e.g. kindness), since an engaged project leader has been found important as a condition fostering positive effects of music interventions (Henley, 2015; Tett et al., 2012).

Inmates seemed very satisfied with the music intervention itself, as the only comments regarding what could be improved dealt with obstacles (related to the prison institution) for implementation, and no negative comments were given about the method. Considering strengths, answers overlapped with impacts of the intervention, indicating that the inmates viewed being affected by the project as its greatest strength.

Project leader’s expectations and experiences of the music intervention. The expectations and experiences of the project leader regarding impacts of the intervention seemed to correspond to the comments of the inmates to a certain extent: musical development, personal growth, stimulation and the intervention as a ‘get-away’ were all named. A remarkable difference was that the project leader expected that the intervention would contribute to the inmates’ communication and collaborations skills. This was neither mentioned by the inmates, nor

observed by the project leader during the sessions. However, in previous literature such positive impacts have been established (Caulfield et al., 2016; Henley et al., 2012; Wilson et al., 2009). It is likely that these effects were not found here because all sessions were individual and no collaborations were allowed. Further, the project leader argued for implementing a ‘culture room’ for the purpose of practicing diverse arts, which could indeed be beneficial for the inmates, bearing in mind the numerous positive effects that have been linked with arts-based prison

Evaluation of a Music Intervention in a Swedish Prison Setting: A pilot project within Skådebanan’s “Culture for Life”-project

EVALUATION OF A MUSIC

INTERVENTION IN A SWEDISH PRISON SETTING

A PILOT PROJECT WITHIN SKÅDEBANAN’S

“CULTURE FOR LIFE”-PROJECT

LAUREN GHIJS

EVALUATION OF A MUSIC

INTERVENTION IN A SWEDISH PRISON SETTING

A PILOT PROJECT WITHIN SKÅDEBANAN’S

“CULTURE FOR LIFE”-PROJECT

LAUREN GHIJS

ACKNOWLEDGEMENTS

TABLE OF CONTENTS

INTRODUCTION

AIM

METHODS

ETHICS

RESULTS

DISCUSSION