Choral singing and prison inmates: influences of performing in a prison choir.

The purpose of these two experiments was to compare well-being measurements between a group of prison inmates singing in a choir and prison inmates not singing in a choir. Using the Friedman Well-Being Scale (FWBS) as the dependent measure, measurements were taken before and after performances of two prison-based choirs: (a) an inmate only choir (n = 10) that performed in the correctional facility (experiment one) and (b) an inmatevolunteer choir (n = 48) that performed outside the correctional facility (experiment two). Results indicated no significant differences between experimental and control groups (n = 10) in composite well-being scores in both experiments. In experiment two, there were significant differences between experimental and control groups on four subscales: emotional stability, sociability, happiness, and joviality. A content analysis of weekly written responses of participants in the inmate-only choir suggested a tendency toward: (a) negative responses during containment, (b) positive choir-related responses at the final two rehearsals, and (c) overall choral experience reflections related to a sense of well-being. Implications for choral music education and suggestions for further research were examined.

Over the course of a year, 13.5 million people are incarcerated in U.S. prisons and jails (Gibbons & Katzenbach, 2006). On the last day of 2004, almost 7 million people were either on parole, on probation, or in jail or prison. That figure represents 3.2% of all U.S. residents, or 1 in every 31 adults (Bureau of Justice Statistics, 2005).

Inmates in U.S. correctional facilities generally have few opportunities for self-expression and are not being prepared to live outside the prison setting (Kupers, 1996). According to Sykes (1958), the pains of incarceration are broader than the loss of physical freedoms. In addition to the immediate stresses of loneliness, boredom, thwarted goals, and discomfort, Sykes reports that inmates tend to see themselves as lacking value and worth. Liebling (1999) indicates that inmate suicide, the rate of which is four times higher than that for non-prisoners, is related to psychological distress and coping issues.

In his book Inside: Life behind Bars in America (2006), Michael G. Santos argues that American correctional facilities succeed as temporary human warehouses, but fail at correcting or rehabilitating inmates:

   If the end goal is to warehouse human beings, then the American
   prison system is a costly but effective design. On the other hand,
   if the goal is to prepare people to live as law-abiding,
   contributing citizens, then objective data suggest that our prison
   system is a stellar example of failure, ripe for reconsideration.
   (p. xxiii)

Santos, himself an inmate, goes on to suggest that prison-based education programs, when permitted and funded, can prepare prisoners to succeed in and contribute to society upon their release.

Brewster (1983) investigated outcomes and characteristics of arts-based inmate education. He reported a 74.20% favorable parole outcome rate for inmates who participated in an arts-based educational program in California in contrast to 49.50% favorable rate for those who did not participate. Reduced inmate incident rates were reported for 75 to 81% of the prisoners who had participated in the arts-based programs. Researchers have similarly examined music education programs in correctional settings. According to Mathiti (2002), the participants of a Diversion in Music Education (DIME) percussion program for adjucated youth had a 9.09% recidivism rate six months after participation, which dropped to 0% after one year of participation. Elliot (1981) established effective treatments for adult male inmates' emotional, developmental, and social problems through participation in an instrumental music education program.

Research has indicated that group singing experiences can enhance selfeesteem, social connections, and trust (Anshel & Kipper, 1988; Peters, 1985). Bailey and Davidson (2003) reported benefits for homeless men (N = 7) participating in group singing. Specifically they suggested four emergent themes: (a) emotional health, (b) social interaction and reconnection through performance, (c) group process, and (d) mental stimulation.

Specific to correctional settings, various historical references to prison choir singing indicate choral participation may be a means of character development, moral education, or rehabilitation. Hickson (1838), for example, noted after observing a Rotterdam prison choir that "music may be regarded as a great moral engine, which, when wisely directed can produce the most beneficial results" (p. 3). Van de Wall (1936) described therapeutic uses of music in adolescent and adult reformatory programs. He viewed singing as a practical activity that promotes feelings of belonging, loyalty, and shifts destructive habits into constructive ones.

Four empirical studies indicated that choral singing experiences benefit individuals in correctional settings. Nelson (1997) reported that adjucated adolescent males (N = 21) participating in a choral program experienced shifts in three areas of self-perception: (a) status, (b) co-musical benefits, and (c) inner rewards. Silber (2005) qualitatively investigated an adult female Israeli prison choir (N = 7) and reported the choir community presented an alternative to the criminal context and allowed inmates to listen, form new bonds, and accept criticism. Richmiller (1992) surveyed former male adult prison choir participants (n = 17) and former prison staff members (n=10) with respect to their assessments of a prison choir experience 29 years following a three-year pilot-program (1963-1966). All staff and choir participants agreed that participation in the prison choir was positive. A former staff member wrote, "A choir teaches that through cooperation with other people and through persistent and hard work the participants can produce harmony and inter-dependence and enjoyment for themselves and others" (p. 49). Cohen (2005) indicated that participating in a joint inmate-volunteer choir provides opportunities for positive transformative change.

This investigation focuses on prison inmates' well-being in relation to choral singing in two different settings. Its purpose is to compare well-being measurements between inmates who did not sing in a choir and: (a) inmates who sang in a prisoner-only choir and (b) inmates and community volunteers who sang in a joint prisoner-volunteer choir.

Definitions

P. H. Friedman (personal communication, 2005) has defined well-being as "feeling well about yourself and your life." An inmate is simply a person serving time in a correctional facility for a crime or misdemeanor. Volunteer singers are those who sing with the prison choir by attending rehearsals either in the prison or outside of the prison and perform with the inmates at public concerts.

This investigation was conducted in two phases. Each phase is reported separately.

Phase 1: Inmates

Experiment 1

The purpose of experiment one was to compare well-being measurements of therapeutic community inmate singers (TCIS) (n = 10) who participated and performed in a prisoner-only choir performance to a control group of inmates who were not in the choir (n = 10). To that end the following research question guided this phase of the study: Are there differences in well-being measurements between the TCIS and the control group before and after a choral performance at the prison facility?

Participants

All participants for experiment one were incarcerated males (N = 20) enrolled in a nine-month substance abuse treatment community. The 80-bed minimum security correctional facility was housed on the same grounds as a state mental hospital and did not have a razor wire fence surrounding the property. Between pre and post data collection, four of the initial fourteen control group participants were relocated to a higher security unit of the prison system, leaving ten participants in the control group through completion of the study.

The therapeutic community program's goals focused on changing participants' addictive behavior through cognitive restructuring. The correctional facility required inmates to follow specific rules including how to sign up for group activities. They announced the opportunity to join and posted a sign-up sheet. Therefore, the participants were self-selected male inmates from the same community. The TCIS's age range was 23-60 with a mean age of 38.30 and median age of 36.50. The control group's age range was 22-44 with a mean age of 34.50 and a median age of 32.00.

Method

The supporting university's institutional review board, the correctional facility, and the founder and conductor of the joint inmate-volunteer prison choir granted permission for this research project. The researcher participated in an annual volunteer training session at the correctional facility and obtained a volunteer identification card.

Measurement Tool. The Friedman Well-Being Scale (FWBS), a twenty-item semantic differential scale, served as the dependent measure. It measured composite well-being and five subscales: (a) emotional stability, (b) sociability, (c) joviality, (d) self-esteem, and (e) happiness (Friedman, 1994). Fleenor (2001) reported that the scale's internal reliability for composite well-being and for each of the subscales ranged from 0.92 to 0.98, and that the external validity correlation score was 0.61. It took fewer than three minutes to administer the FWBS in this study. The scale was appropriate for administering twice in one day as a pre-post assessment. See Table 1 for specific subscale external validation correlation scores. A predetermined alpha level of 0.05 was employed to indicate significance for both experiments.

Procedure. Weekly ninety-minute rehearsals (N = 9) each followed the same schedule: (a) physical and vocal warm-ups, (b) solfege practice, (c) pitch matching and choral blend practice, (d) literature practice for upcoming concerts, and (e) a closing song, "May the Gift of Human Kindness." Inmates learned songs through a combination of rote teaching, introduction to score reading, listening to recordings, and repetition. Data for experiment one were collected before and after the prison choir's first performance.

The prison choir performed two selections, "Keep Your Lamps" and "May the Gift of Human Kindness", at the end of the therapeutic community's graduation ceremony. Approximately 100 inmates, staff, and approved guests were in the audience.

Results

The research question for this first phase of the study asked whether differences exist in FWBS well-being scores between the TCIS and a control group of inmate non-choristers before and after a prison choir performance. A repeat measures analysis of variance mixed design showed no significant difference between control and experimental groups' composite well-being scores (F (1, 18) = .038, p = 0.848). However, there was a significant increase between pre-and post composite well-being scores for both groups (F (1, 18) = 6.080, p = 0.024). There was no significant difference between groups on all subscales of the FWBS, yet both groups did show a significant increase between pre-and post measurements on the subscales for joviality, (F (1, 18) = 9.889, p = 0.006) and emotional stability, (F (1, 18) = 5.761, p = 0.027). See Table 2 for a summary of the mean and standard deviation scores.

Nominal Data 

To supplement the FWBS measurements, the following nominal data were also collected from the experimental group. Over a nine-week time frame participants were asked to write a response to the question: "How are you feeling today?" After the public concert, each inmate wrote his reaction to the performance experience. At the final rehearsal, inmates wrote responses to their overall choral experience.

Two research questions guided the nominal data collection: (a) Will the experimental group's self-reported responses to the weekly question, "How are you feeling today?", be choir-related or non choir-related, and will these responses have a positive or negative connotation? (b) Will the experimental group's self-reported responses to prompts about their reactions to performing a concert and their overall choral experience relate to their well-being, according to Ryff and Singer's (1996) dimensions of well-being? Ryff and Singer's dimensions of well-being were: (a) self-acceptance: holding positive attitude toward oneself, (b) positive relations with others: having strong feelings of affection and empathy for all and being capable of a strong identification with others, (c) personal growth: continually developing as opposed to achieving a fixed state, and (d) autonomy: showing resistance to enculturation and autonomous functioning.

Nominal Data Findings

Written responses (N = 96) across nine weeks ("How are you feeling today?") were analyzed by two researchers who coded responses into two mutually exclusive categories (Krippendorf, 1980): (a) choir-related or (b) non choir-related. Within those two categories, responses were additionally coded as either positive or negative. Inter-rater reliability was 100%. Examples of positive choir related responses included, "For the last hour I have been looking forward to rehearsal tonight," and "Hopeful and looking forward to our performances." Examples of positive non-choir responses included, "I'm content," and "I feel peaceful, joyous within myself. I have 34 days left until my release." The one negative choir-related response related to an inmate's insecurity about how his scar from nose surgery would look during the performance. Examples of negative non-choir responses included, "Concern. Worry about my family," and "Stressed, overwhelmed." See Table 3 for a summary of these results.

The second nominal data research question posed was: Will the experimental group's self-reported responses relate to their well-being when describing their choral experiences? Written responses (N = 27) to questions presented following the first public performance ("What are your reactions to participating in this concert?") and at the final rehearsal ("Describe your overall reactions to this prison choir experience.") were coded independently as either: (a) corresponding to one or more of Ryff and Singer's (1996) dimensions of well-being or (b) not corresponding to any of Ryff and Singer's dimensions of well-being. Inter-rater reliability was 100%. All responses were deemed to correspond to Ryff and Singer's dimensions of well-being. Some responses demonstrated a degree of autonomous functioning: "After Sunday's concert I feel so different. I feel that I can be successful out in society," and "The reward of the performances and the satisfaction of achieving a goal such as this are all positives that I will carry with me for the rest of my life." Other responses related to self-acceptance: "I felt really good cause I haven't ever did nothing like that before in my life [sic]."

The following responses were examples of positive relationships with others: "You have made me see there are wonderful people out there other than druggies. That will be the kind of people I want in my life from here on out." "Going away to perform at both churches was a blessing to intermingle with the people of society," and "My heart was filled with great joy as I had the pleasure of being around such positive individuals and great musicians." The nominal data indicated that this choral experience helped inmates to develop interpersonal skills.

Phase 2: Inmates and Volunteers

Experiment 2

The purpose of the second experiment was to compare FWBS measurements of a larger experimental group (n = 48) with the same control group (n = 10) as experiment one, both before and after a public performance of a joint inmate-volunteer choir. The research question guiding this experiment was: Are there differences in well-being measurements between the experimental and control groups before and after this public performance according to the FWBS?

Participants

The second experiment included four groups (N = 58): (a) the same TCIS as experiment one (n = 10), (b) the same control group as experiment one (n = 10), (c) male general population inmate singers (GPIS) (n = 13), and (d) male volunteer singers (VS) (n = 25) from the surrounding community. All choristers, comprising groups (a), (c), and (d) (n = 48), performed in the joint inmate-volunteer choir.

The GPIS were housed at a different correctional facility than TCIS. The GPIS facility was a 628-bed minimum security unit on 85 acres with razor wire fences surrounding the unit. The GPIS age range was 19 to 53 years with a mean age of 37.85 and a median age of 38.00. The volunteer singers' age range was 23 to78 years with a mean age of 57.12 and a median age of 61.00.

Method

The GPIS rehearsed twice weekly in their correctional facility's chapel with the founder and director of the chorus. The researcher attended and assisted these rehearsals twice monthly. Nine of the VS traveled monthly to the correctional facility to rehearse with these inmates, and all VS rehearsed monthly outside of the prison. The whole chorus (TCIS, GPIS, and VS) did not rehearse together until the day of the concert.

The choristers completed the FWBS before and after a joint inmate-volunteer public choral performance at a church in a metropolitan city. The control group completed the FWBS at their correctional facility on the same day both before and after the concert. These men did not travel to the concert venue. The audience size was approximately 400 and consisted of inmates' families and friends, volunteer singers' families and friends, and members of the community. The theme of the concert was "This Little Light" and selections included Schubert's "Holy, Holy Holy," J.S. Bach's "Break Forth O Beauteous Heavenly Light," Cherubini's "Graduale and Sanctus from Requiem in C Minor," Mendelssohn's "The Righteous Living Forever," the chorus and aria from Handel's "The Creation," "Santa Lucia," "O Sole Mio," "This Little Light of Mine," "Set Down, Servant," Clarke's "The Blind Plowman," and Peter Yarrow's "Light One Candle." Inmates recited narrations to introduce these selections. Toward the end of the concert the chorus led an audience sing-along. A predetermined alpha level of .05 was employed to indicate significance.

Results

A repeat measures analysis of variance mixed design showed no significant difference between control and experimental groups in amount of improvement for composite well-being (F (1, 18) = 3.277, p = 0.087), however there was a significant increase from pre to post scores among all four groups (F (1, 18) = 11.872, p = 0.003). A significant difference between control and experimental groups occurred in four subscales: (a) sociability (F (1, 18) = 11.872, p = 0.003), (b) joviality (F (1, 56) = 11.484, p = 0.001), (c) emotional stability (F (1, 56) = 18.475, p < 0.001), and (d) happiness (F (1, 56) = 6.233, p = 0.016). See Table 4 for a summary of the mean and standard deviation scores.

Discussion

The primary purpose of this study was to investigate well-being measurements of prison choristers in comparison to inmates who were not in the chorus. Well-being is a complex construct and difficult to measure quantitatively. Nevertheless, quantitative and nominal data from this investigation indicate that choral singing may enhance inmate singers' well-being.

The social elements of the public concert appeared to play an important role in inmates' perceived well-being. In the second experiment the inmate choristers left their respective correctional facilities to attend the public concert where they had opportunities to converse with volunteers and audience members. After the concert, the inmates received accolades from audience members while standing in a reception line. The evening concluded with inmates and volunteers sharing a home-cooked meal. These social experiences contrasted dramatically with daily interactions between inmates and correctional staff at the facility. Moreover, the simple fact that the inmates left their prison facilities to perform in public may have played a role in the increased well-being scores.

Nominal data reflected how the social element of the public concert related to the inmates' positive experiences. For example, "The warmth of the people after the concert was overwhelming," and "It makes me feel that people overlooked my incarceration." From the researcher's perception, it was clear that these men were eager and happy for this opportunity, although some initially appeared nervous to interact with people outside the correctional facility. For example, some inmates tended to interact exclusively with other inmates unless volunteers encouraged them to intermingle.

Examples of nominal data relating specifically to joviality and happiness included the following responses from inmates: "[The experience] uplifted my spirit throughout this week," "The most wonderful time I had in about 8 years," and "It touched my heart so much that I almost cried three times." Such reflections indicate that performing in the public concert was highly meaningful to these inmates.

The GPIS had the highest composite well-being scores after the concert. This group rehearsed twice weekly while the TCIS only rehearsed once weekly and VS rehearsed only once monthly. The GPIS may have been more familiar with the choral selections, may have felt more prepared for the choral concert, and may have developed a stronger relationship to the conductor than the other two groups. These factors may relate to the mean differences among the three groups' scores.

The higher subscale measurements in experiment two (sociability, happiness, emotional stability, and joviality) demonstrated three of Bailey and Davidson's (2003) characteristics of homeless men participating in a choir: (a) social interaction and reconnection through performance, (b) group process, and (c) emotional health. Social interaction occurred at all rehearsals and performances. Enjoyment was an outcome of social reactions as was evident in written reflections: "It was fun singing with all my brothers," and "Words cannot explain how much I enjoyed that day." Such data correspond to the significant difference between experimental and control groups in the subscale happiness. Additionally, inmate singers' nominal data demonstrated happiness as can be observed in responses such as: "I had a swell time," and "This is how I felt about the concert: ecstatic, elated, excited, joyful, grateful."

The data from this study indicate that participation in choral singing performances, particularly performing outside of the correctional facility, benefit inmates' perceived well-being. The need for improved inmate rehabilitation warrants more research on both short-term and long-term outcomes of participating in and performing in prison choirs. Data-based prison choir research is sparse. The following issues need to be examined: inmate singers' measurements of self-esteem, hostility control, or disciplinary infractions. Collecting pre and post measurements of well-being collected at each rehearsal rather than only at a performance may provide a broader understanding of inmates' well-being in relationship to choral singing. Prison choir participants' family members may offer a different perspective on how singing in a prison choir influences inmates' behavior. Investigating short-term and long-term effects of participating in the choir may direct prison choir conductors as they pursue this line of work to shape their programs to be most beneficial to inmates and society. Because there is potential to improve the lives of inmates, the safety of correctional facilities, and the society at large, further research in a variety of aspects of prison choruses is warranted.

References

Anshel, A. & Kipper, D. (1988). The influence of group singing on trust and cooperation. Journal of Music Therapy, 25(3): 145-155.

Cohen, M. L. (2005, June). Exploration of inmate and volunteer choral experiences among a prison-based choir in the U.S. Midwest. Phenomenon of Singing, International Symposium, St. John's, Newfoundland.

Bailey, B. A. & Davidson, J. (2003). Amateur group singing as a therapeutic instrument. Nordic Journal of Music Therapy, 12(1): 18-33.

Brewster, L. (1983). A cost benefit analysis of the California Department of Corrections Arts in Corrections Program. Santa Cruz, CA: William James Association.

Bureau of Justice Statistics (2005). Prison and jail inmates at midyear 2004. Retrieved September 5, 2005 from http://www.ojp.usdoj.gov/bjs/abstract/pjim04.htm

Dunphy, K. (1999). A creative arts performance program for incarcerated women: The arts of transforming shame. The Arts in Psychotherapy, 26(1), 35-43.

Elliot, T. G. (1981). A study of the psychology of a non-verbal methodology inherent in the teaching of instrumental music as observed in a program for adult offenders. (Doctoral dissertation, Boston University School for the Arts, 1981), Dissertation Abstracts International, 42, 2352.

Fleenor, J. (2001). Test review of the Friedman Well-Being Scale. In B. S. Plake & J. C. Impara (Eds.), The fourteenth mental measurements yearbook.

Friedman, P. (1994). Friedman well-being scale and professional manual. Redwood City, CA: Mind Garden.

Friedman, P. (2005). Personal communication.

Gibbons, J. J. & Katzenbach, N. B. (2006). Confronting confinement: A report of the Commission on Safety and Abuse in American Prisons. Retrieved November 20, 2006 from http://prisoncommission.org

Hickson, W. E. (1838). The use of singing as a part of the moral discipline of schools. Lecture before the members of the Sunday School Union in the theatre of the Mechanics' Institution, Southampton Buildings, Chancery Lane, May 29, 1838.

Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Newbury Park, CA: Sage.

Kupers. T. A. (1996). Trauma and its sequelae in male prisoners: Effects of confinement, overcrowding, and diminished services. American Journal of Orthopsychiatry, 66(2), 189-196.

Liebling, A. (1999). Prison suicide and prison coping. In M. Tonry (Series Ed.) & J. Petersilia (Vol. Ed.). Crime and justice: A review of research: Vol. 26. Prisons (pp. 283-359). Chicago: University of Chicago Press.

Mathiti, V. (2002, July). An evaluation of a diversion into music education (DIME) programme for at-risk youth: Implications for programme development. Paper presented at the 14th International Congress on Child Abuse and Neglect, Denver, Colorado.

Nelson, D. L. (1997). High-risk adolescent males, self-efficacy, and choral performance: An investigation. (Doctoral dissertation, Arizona State University, 1997), Dissertation Abstracts International, 58, 03A.

Peters, M. L. (1985). Music in education: A pilot study investigating improvements in academic achievement and self-esteem through participation in an elementary school boy's choir. Unpublished master's thesis, Allegheny College, Meadville, Pennsylvania.

Richmiller, M. G. (1992). Study of the residual effects of music education experiences of a prison choir, twenty-nine years after participation. Unpublished master's thesis, Southeast Missouri State University, Cape Girardeau, Missouri.

Ryff, C. D. & Singer, B. (1996). Psychological well-being: Meaning, measurement, and implications for psychotherapy research. Psychotherapy and Psychosomatics, 65, 14-23.

Santos, M. G. (2006). Inside: Life behind bars in America. New York: St. Martin's Press.

Silber, L. (2005). Bars behind bars: The impact of a women's prison choir on social harmony. Music Education Research, 7(2), 251-271.

Sykes, G. (1958). The society of captives. Princeton, NJ: Princeton University Press.

Van de Wall, W. (1936). Music in institutions. New York: Russell Sage Foundation.

Biographical Sketches

MARY L. COHEN is Assistant Professor of Music Education at the University of Iowa. She holds B.M.E., M.M.E. and Ph.D. degrees from the University of Kansas. Dr. Cohen presented at the 2006 International Correctional Education Association Conference in Anaheim, California.

Table 1. Subscale External Validation Correlation Scores

Subscale              Correlation Scale                    Score

Emotional stability   Goldberg Emotional Stability         0.82
                      Watson and Clark Affective Balance   0.69
                      Watson-Self-Assurance                0.65

Self-esteem           Goldberg's Extroversion              0.70
                      Snyder-Hope                          0.68
                      Watson and Clark Self Assurance      0.66

Sociability           Goldberg Extroversion                0.60
                      Morey's Warmth                       0.57
                      Costa & McCrae's Warmth              0.56

Joviality             Watson & Clark-Jovial                0.70
                      Northwestern-CWBT (Well-being)       0.68
                      Costa & McCrae Positive Emotions     0.65

Happiness             Watson & Clark-Jovial                0.73
                      Northwestern-CWBT (Well-being)       0.71
                      Watson & Clark Affective Balance     0.65

Table 3. Summary of "How Am I Feeling?" Responses (N=89)

Week         1   2   3   4   5   6   7   8   9   Totals

Category

Non-choir    8   7   3   4   6   2   5   5   4   44
  positive
Non-choir    2   3   5   4   2   1   1   3   2   23
  negative
Choir        0   1   1   2   2   3   2   4   6   21
  positive
Choir        0   0   0   0   0   0   1   0   0   1
  negative

Table 4. Mean and Standard Deviation Scores for Experiment 2

Group              TCIS (n=10     GPIS (n=13)     VS (n=25)

                   M       SD     M       SD      M       SD

Pre Well-Being     59.60   7.69   56.31   10.55   57.80   6.84
Post Well-Being    67.00   5.66   63.69    7.86   63.48   5.85
Pre Social         56.10   6.64   54.31    6.64   56.48   5.94
Post Social        63.70   4.00   60.85    7.67   60.32   6.39
Pre Self-esteem    54.50   8.71   52.00   11.17   52.36   7.62
Post Self-esteem   63.70   5.50   61.15    5.18   58.16   7.27
Pre Joviality      55.70   8.17   54.92    6.69   54.92   8.52
Post Joviality     64.20   6.27   61.50    6.54   60.88   6.83
Pre Emotional      60.40   6.54   57.85    6.69   59.00   8.52
  Stability
Post Emotional     66.00   3.86   63.23    8.84   62.96   4.82
  Stability
Pre Happiness      61.50   7.98   58.15   12.10   58.80   7.20
Post Happiness     56.00   1.58   63.92    5.96   62.96   5.41

Group              Control (n=10)

                     M       SD

Pre Well-Being       55.70  11.83
Post Well-Being      58.70  10.76
Pre Social           56.70  8.14
Post Social          55.80  10.09
Pre Self-esteem      52.80  11.36
Post Self-esteem     54.80  10.31
Pre Joviality        56.60  15.83
Post Joviality       54.60  9.26
Pre Emotional        56.20  12.96
  Stability
Post Emotional       59.50  9.76
  Stability
Pre Happiness        54.00  8.88
Post Happiness       56.00  11.5

Note. TCIS=Therapeutic Community Inmate Singers, GPIS=General
Population Inmate Singers, VS=Volunteer Singers