Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Breaking with tradition : role development in a prison-based baccalaureate program Clarke, Grant Stewart 1987-08-23

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


831-UBC_1987_A2 C52.pdf [ 7.15MB ]
JSON: 831-1.0055823.json
JSON-LD: 831-1.0055823-ld.json
RDF/XML (Pretty): 831-1.0055823-rdf.xml
RDF/JSON: 831-1.0055823-rdf.json
Turtle: 831-1.0055823-turtle.txt
N-Triples: 831-1.0055823-rdf-ntriples.txt
Original Record: 831-1.0055823-source.json
Full Text

Full Text

BREAKING WITH TRADITION: ROLE DEVELOPMENT IN A PRISON-BASED BACCALAUREATE-PROGRAM by GRANT STEWART CLARKE A DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION in THE FACULTY OF GRADUATE STUDIES (Department of Administrative, Adult and Higher Education) We accept this dissertation as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA February 1987 © Grant Stewart Clarke 1987 - 3 1 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6(3/81) i i Breaking With Tradition: Role Development In A Prison-Based Baccalaureate Program ABSTRACT Prisons are organized to hold and control inmates. Inmates traditionally oppose authority, and the social ecology of prisons resists change-oriented programs. Successful educational programs appear to neutralize certain negative aspects of the social ecology while engaging inmates in setting and working toward pro-social goals. One initiative is the Simon Fraser University prison-based baccalaureate program in the humanities. Inmates in this program appear to develop positive student roles. Explanations for the program's apparent success had not previously examined the interaction between inmates and the social ecology of the program. Previous accounts of the program relied on anecdotal reviews and psychological explanations of inmate development. To bridge this gap, this study was designed to explicate a theoretical model to explain student roles and associated feeling states and expectations, to operationalize it, and to examine relationships with various socio-demographic and carceral variables. Three approaches were used. The first involved formulating the model, drawing on previous studies and experience with inmates in this program, literature about the program, and role theory. A model of role development was posited. It has five stages: (1) Recruitment, (2) Disorientation, (3) Separation, (4) Transition, and (5) Solidarity. The second phase involved operationalizing the model. Seventy written statements were constructed representing inmates' feelings toward prison, and the university program, at each stage of the model. They were judged by five experts in correctional education who strongly concurred in assigning the 70 statements into respective stages. The second phase also involved a card sort of these 70 statements by 33 inmate university students in one prison. They sorted the cards according to: (1) "how I feel now"; (2) "how I used to feel, but not now"; (3) "never felt like this"; and (4) "don't know." For the third phase, data were analyzed using Pearson correlations and ANOVA statistical procedures. The major conclusions which emerged from the study pertained to the three purposes. With regard to the explication of a model of role development, it was concluded that (1) Role theory is an appropriate framework for articulating a model of prison ecology, and (2) Inmates experience five distinct and sequential stages of role development. With regard to the operationalization of the model, it was concluded that (1) Judges found the overall model plausible and workable, (2) Judges were able to reliably discriminate items into stages, and (3) Inmates' responses confirmed intra-stage reliability. With regard to relationships between scores obtained from operationalizing the model and various socio-demographic and prison-related variables, it was concluded that (1) The expected associations were not confirmed, (2) Inmates' forwarding of i v feelings from previous incarcerations supports the Importation model, (3) A counter-intuitive finding (university term by Recruitment) is probably an artifact of previous penitentiary experience, and (4) The university program does foster pro-social role development, thus providing support for the "some things work" position. V TABLE OF CONTENTS ABSTRACT ii LIST OF TABLES viLIST OF FIGURES viiACKNOWLEDGEMENTS ix Chapter 1: INTRODUCTION TO EDUCATION IN PRISONS 1 The Prison Setting 1 Prison Organization 4 Models Of Correctional Practice 7 The Problem Of Effective Programs For Inmates 9 The "Nothing Works" Position 10 The "Some Things Work" Position 1 Correctional Education 13 The Prison University Program 4 Purposes Of The Study 20 Chapter 2: LITERATURE REVIEW 1 Views Of The Inmate Social System 2The Deprivation Model 25 The Importation Model 30 Chapter 3: CONCEPTUAL FRAMEWORK 3General Assumptions Of Role Theory 35 Concepts In Role Theory 40 Criteria For A Model Of Role Development 48 Stages Of The Model 51 Recruitment Stage 3 Disorientation Stage 4 Separation-Alienation Stage 55 Transition-Reframing Stage 6 Solidarity Stage 57 Chapter 4: INSTRUMENT DEVELOPMENT 60 Item Construction And Selection 2 Item Judging Process 64 Inmate And Stage Meanings 7 Reliability Estimates 86 Questions Derived From The Stage Model 8vi TABLE OF CONTENTS (continued) Chapter 5: METHODOLOGY 89 Choice Of Research Setting 8Subject Selection Procedures 91 Administration Of Q-Sorts 2 Carceral And Demographic Information 94 Data Analysis 98 Chapter 6: RESULTS 9 Characteristics Of Participants 9Stage Intercorrelations And Internal Consistency 103 Principal Stage Orientations 106 Relationships Between Stages And Respondent Background ..112 Effects Of Background Variables On Stage Scores 118 Chapter 7: CONCLUSIONS 12The Prison ContextThe Model 123 Explication Of The Model 125 Operationalizing The ModelRelationships With The Model .127 Limitations Of The Study 130 Chapter 8: DISCUSSION 134 Concerning Future Research 13Concerning Prison Administration 138 Concerning Training Of Correctional Educators 139 REFERENCES 141 APPENDICESAppendix A: Seventy Items Listed By Stage 148 Appendix B: Offence Severity Scale 151 Appendix C: Information Sheet 152 Appendix D: Carceral And Demographic Coding Sheet 153 vii LIST OF TABLES 1. Inter-Judge Agreement On 70 Items 66 2. Overall Item Score Means By Role Development Stages ... 72 3. Summary And Contrasts Of Stage Score Meanings 82 4. Reliabilities For Five Stages Of Role Development 86 5. Item Characteristics For Five Carceral Variables 100 6. Severity Of Offences And Previous Federal Term Served .101 7. Occupational And Educational Characteristics- 102 8. Stage Score Intercorrelations And Reliabilities 104 9. Percentages Of Items Describing Present Feelings 109 10. Items Describing Feelings: Present, Past Or Never 111 11.. Stage, Carceral And Demographic Correlations 113 12. Effects Of Enrollment Term On Role Stage Scores 119 13. Carceral And Demographic Effects On Stage Scores 120 14. Effects Of Federal Prison Experience On Stage Scores ..121 viii LIST OF FIGURES 1. Oppositional Social Ecology In The Deprivation Model .. 27 2. Pre- And Post-Incarceration Factors In The Importation Model 31 3. Role Interactions Pertaining To Education In Prisons .. 49 4. Interactions In The Academic Ecology And Stages Of Role Development 52 5. Inter-Judge And Inmate Agreements On Items 68 6. Inmate Responses By Stage And Category 74 7. Inmate Item Responses By Category And Stage 108 ix ACKNOWLEDGEMENTS I would like to thank the Correctional Service of Canada and its representatives for allowing this study to be conducted, and for facilitating my access to relevant records. I am grateful to the 33 men who participated in this study without whom none of this was possible. I am indebted to the Simon Fraser University faculty and administrators who ensure the prison university program works, and who served as expert judges in this study. They are Stephen Duguid, Jacquelyn Nelson, Hendrik Hoekema and Wayne Knights. I particularly appreciate the logistical assistance given me by Dr. Jacquelyn Nelson during the data collection phase. My committee was amicable, supportive, and uncompromisingly rigorous in its scrutiny of my work. To Drs. Roger Boshier, John Collins, Tom Sork, and Stephen Duguid, I extend sincere thanks for seeing me (and the dissertation) through. There were times I didn't think I would finish, ever. I am really grateful to family, friends, and fellow students who buoyed me up over a long haul. My wife, Elaine McCreary, and children have been my best friends through this process. My parents and mother-in-law have given me unfailing support and encouragement. Thank you. 1 CHAPTER 1 INTRODUCTION TO EDUCATION IN PRISONS The Prison Setting Prisons are places where the police and courts send people for breaking criminal laws. Concretely and abstractly, they represent society's power to impose extreme sanctions against its members for violations of legal codes. Moreover, prisons are largely closed to mainstream society. Both keepers and the kept play out a substantial portion of their lives unseen and unknown by the larger public except for occasions when a riot, hostage-taking or murder focuses attention on a prison's inner workings. Society does not seem to want to know more about its prisons. Regulations prevent members of the public from taking a casual look inside. Prisons are isolated from society, largely impermeable to the influence of societal standards while enclosing inmates behind physical and regulatory barriers (Goffman, 1961, 1961a; Grosser, 1968). Inmates are uncritically stereotyped as dangerous; this results in a "hands-off attitude toward the purposes and workings of prisons. Boyanowsky (1977, p. 126) contends that society views an inmate's time spent in prison as confirmation of his criminality and dangerousness "rather than accepting time in prison as evidence of change, cure, or rehabilitation." Prisons and prisoners raise dark and disquieting feelings 2 in the public. Despite recent exceptions in Canada, communities are generally unwilling to have new prisons built in their midst (Scull, 1977), and there is a widespread perception that crime is increasing. This has resulted in public support for "sterner" treatment of offenders. A federal government Task Force (1977) concluded that the Canadian public wanted those convicted of serious crimes to be given longer sentences requiring incarceration. A notable current example of a tougher policy toward certain crimes is an increasing use of imprisonment for drinking-driving offences. Criminal justice policy in Canada has shifted in emphasis from.the treatment and rehabilitation of offenders to the rights of victims of crime and to crime prevention (Ekstedt & Griffith, 1984). Moreover, the notion that prisons can fight crime and its causes has lost support (Task Force, 1977). Ekstedt and Griffith (1984, p. 363) contend that the correctional enterprise "is viewed, and perceives itself, as the 'tail-end' of the criminal justice system." Prisons have not lived up to society's expectations of them as a vehicle for reforming criminally-oriented individuals. Indeed, the opposite seems true. Prisons are viewed as "schools for crime" by public and inmates alike (Thomas & Peterson, 1977; Gosselin, 1982). In the face of this lack of success in "correcting" criminal behaviour, and the public's disenchantment with the system, it may be argued that prisons do little more than function as human warehouses. However, the experience of prison is not benign, nor is it intended to be. Prison authorities and 3 the public seem to feel inmates are sent to prison for punishment, not just as punishment (Hepburn & Stratton, 1977; Ekstedt & Griffith, 1984). Inmates are not supposed to enjoy the time they spend in prison. Cressey's (1960, p. 82) comment that punishment is "an unequivocal function of prisons" is still true today. Thus prison practices tend to be reflexively punitive though no longer in the form of corporal punishment. Instead, prison authorities enforce multiple directives and regulations that govern daily prison life. Grosser contends that the routine of prison life is used as a mechanism of control by authorities: Even eating and sleeping tend to become routinized in the prison. The discipline extended to every aspect of the prisoner's multiple roles leaves little scope for nonregulated or unsupervised activity. Historically, this development was based on the belief that discipline per se is a good and that its enforcement will have benefits beyond the prison walls in the habits inculcated in individual prisoners....The routinization of prison life is also traceable to: (1) the punitive philosophy, which tends to be expressed in over-elaboration of the necessary routines of living so that prisoners 'don't have it too easy'; and (2) the imperatives of any functioning institution which cannot leave time schedules, allocation of resources, etc., to the discretion of each individual (1968, pp. 15-16). Notwithstanding the public's desire to see tougher penalties imposed on serious offenders, most will eventually be released from prison. After a sustained experience of dependency-inducing prison routine (Grosser, 1968), inmates face the formidable task of "making it" on the outside. Therefore, how prisons treat inmates, and how they respond to prison authority, is central to any examination of socio-psychological 4 change. Moreover, treatment and program initiatives that purportedly equip inmates with the cognitive, social, and emotional skills to "make it" should be thoroughly studied because prison organization and practices appear to be a substantial obstacle to achieving these kinds of outcomes. Prison Organization Federal prisons are operated by the Correctional Service of Canada (CSC), under the direction of the Solicitor-General. Prisons are labelled "correctional institutions." The name implies a concern with "correcting" behaviour. The situation is somewhat more complex than implied by the correctional label because prisons have a diversified mandate that includes retribution (punishment), incapacitation, deterrence, and rehabilitation (reformation) (Ekstedt & Griffith, 1984; Cressey, 1960). These functions are typically assigned varying priorities. In a discussion document prepared by the Solicitor-General of Canada (Solicitor-General, 1977), correctional functions were ranked as follows: first, control and custody of offenders; second, humane treatment; and third, sound correctional programs for rehabilitation. A basic dilemma that prison authorities face is how to reconcile controlling and changing inmates. Thomas and Peterson contend that while there is no inherent contradiction in the prison's pursuit of control and change, a problem arises because "prisons are not primarily organized to pursue changes in those they process and, instead, remain committed to the goal of 5 insuring effective custodial control" (1977, pp. 64-65). Notwithstanding formal statements by correctional authorities about goals, a commonly held view is that prison structure has changed little (Thomas & Peterson, 1977). Moreover, the germinal work from the 1960's on prison structure and inmate society by authors like Cressey, Sykes and Messinger have retained their relevance in contemporary corrections literature; they established "benchmarks" to which researchers still refer. There is no evidence to suggest that custody and control promotes positive changes in inmates. Moreover, a widely held view is that prison's dominant concern with control and custody neutralizes possibilities for change (Ekstedt & Griffith, 1984; Thomas & Peterson, 1977; Goffman, 1961a; Cressey, 1960; Conrad, 1983; Murton, 1976). A corollary of "control and custody" in prison is an authoritarian administrative style which relies on coercion (Burns, 1969; Thomas & Peterson, 1977). The concern of all types of prisons for control of inmates encourages the use of coercive power. Burns (1969, p. 153) argued that "prisons appear to form a group of social systems differing in detail, but alike in their fundamental processes." He too perceived force against inmates as a mechanism of power shared by all prison systems. Perhaps the prison's continued reliance on coercive power accounts for part of the controversy over their rehabiltative efforts, despite shifts in contemporary policy and practice toward more humane treatment. In Canadian corrections there 6 have been a number of different treatment approaches aimed at changing the inmate. Treatment and program are usually distinct from custodial staff. Control (custody) is the foremost organizing principle of prisons (Solicitor-General, 1977; Thomas & Peterson, 1977). From guard to warden, custodial personnel are members of a line organization, while treatment and program personnel are not. Cressey (1960) wrote of three personnel hierarchies in prisons, responsible for (a) keeping, (b) using, and (c) serving inmates. Prisons have not provided for the integration of their divergent purposes. Moreover, Cressey viewed the purposes of personnel concerned with keeping and serving inmates as entirely different and partly contradictory. In the give and take of resource allocation within the prison, programs and personnel serving a rehabilitative or change purpose are at a clear disadvantage: On an organizational level the effect of the primacy of custodial or control goals is quite pronounced. The physical structure of the institution, the manner in which available resources are allocated, the rigid organizational hierarchy, lines of communication, the distribution of decision-making power, the routinization of organizational activities, the means by which organizational participants other than inmates are evaluated, and related characteristics and activities of the organization all reflect the dominant concern of the prison (Thomas & Peterson, 1977, p. 37). 7 Models of Correctional Practice Even though custody and control are the prime organizing principles of Canadian correctional institutions, various policies deal with inmate rehabilitation. Ekstedt and Griffiths (1984, pp. 67-73) identify five models of Canadian correctional policy and practice spanning the years 1700 until the present. The period 1700-1938 featured punishment as the primary characteristic of correctional policy and practice, though the type and severity of punitive practices changed over the years. Three models are associated with correctional practice from 1938 until today: Rehabilitation, Reintegration, and Reparation. While promoting other objectives such as treatment, prisons never relinquished punishment as their key organizing principle. The Rehabilitation model characterized the period 1938-1970. Under this model corrections took on the task of reforming the offender and precipitated the development of inmate treatment modalities or models, chiefly the medical model which likened criminality to a disease for which therapeutic treatment was prescribed. The medical model lingers to the present day, but is largely out of favour due to its lack of success in "curing" inmates of their presumed criminality. The other significant program to emerge (though not for the first time) during the tenure of the Rehabilitation model was education and training. Though significant, education and training programs have not received support within the correctional system comparable to that given initiatives based on the medical model. 8 The Reintegration model characterized correctional policy from 1970 to 1978 and, according to Ekstedt and Griffiths (1984), continues to this day to account for much correctional practice. The principal feature of this model was the shift in emphasis from the medical model of treatment to the belief that rehabilitation is best sought in the community, not the prison. Thus, diversion programs, community-based corrections, parole and probation all received more support. Offenders were to be sentenced to prison as a last resort and, in some direct proportion to the kind of criminal behaviour exhibited, not on the basis of presumed criminality. Policy for the period 1978 to the present has moved toward the Reparation model in which the rights of victims are a focus. There is a renewed emphasis on the punishment of offenders who are liable to pay restitution where feasible and accept responsibility for their own rehabilitation (Solicitor-General, 1977). The correctional system will make resources available to inmates for their own betterment, but not claim to be in the business of rehabilitation. This new treatment orientation is known as the "Opportunities" model. Ekstedt and Griffiths (1984) claim that the current model of correctional practice is still Reintegration. It is probably more accurate to say a mixed model is in effect. While community-based corrections and diversion are emphasized, the notion that inmates must assume greater responsibility for their own rehabilitation has gained prominence. Given this emphasis, greater use of reparation is likely. 9 However, in practice, "punishment" of inmates still occurs despite current correctional practices. Whatever the merits or moral arguments for or against punishment, this orientation inevitably increases tension, especially in medium and maximum security prisons. The report on The Role of Federal Corrections in Canada (Task Force, 1977) acknowledged that tension is "paramount" in corrections. Punishment relies on control and coercion. Punishment motivated actions by CSC staff towards inmates are likely to result in greater antipathy between the groups. Furthermore, given the correctional system's passive stance towards rehabilitation, benefits to inmates under the Opportunities model are likely to fall short of official claims. Any prospect of positive change in inmates must be weighed against pressure on inmates to conform to inmate codes that prescribe resistance to authoritarian actions by CSC staff. The Problem of Effective Programs for Inmates There are two divergent viewpoints regarding the efficacy of change (rehabilitation) treatments and programs. One holds that "nothing works"; the other that "some things work." 10 The "Nothing Works" Position Doubts about the efficacy of rehabilitation programs in the prison context gained a foothold in Canadian corrections with the publication of the Ouimet Report (1969). Ouimet noted the lack of consensus among experts on the appropriate theoretical bases for correctional programs, and growing evidence to suggest that community-based treatment might be more effective than prison-based programs. Subsequent federal reports by the Law Reform Commission (1975) and a Task Force (1977) went further than Ouimet in concluding that rehabilitation was not a supportable justification for incarceration. A study that had a significant and lasting effect on Canadian and American corrections was reported by Martinson (1974). He concluded from a survey of 231 evaluations of correctional treatment programs conducted between 1945 and 1967 that "nothing works." He claimed that the reports on rehabilitation treatment and programs failed to demonstrate any appreciable amelioration of recidivism. A number of authors have since argued that the impact of the Martinson report exceeded what was justified by its scientific validity (Ekstedt & Griffith, 1984; Gendreau & Ross, 1979; Cousineau & Plecas, 1982). Nevertheless, Martinson's report seemed to confirm the worst suspicions raised by Ouimet (1969) and others that successful rehabilitation (change) programs are unlikely to be found in prisons. Part of the legacy of the "nothing works" viewpoint is the reparation policy in Canadian corrections (Ekstedt & Griffith, 11 1984) within which the "program opportunities model" is supposed to accomplish three things: It makes the offender responsible for changing his own conduct, it provides Federal Corrections with a realistic goal rather than an unattainable goal of changing the offender's behaviour, and it does not lead the public to believe that Federal Corrections can resolve the problem of crime (Solicitor-General of Canada, 1977, p. 34). The "Some Things Work" Position Others have asserted that, contrary to Martinson's initial claims, there are effective rehabilitation programs. Ross and Fabiano (1983a) claimed some programs have demonstrated success in reducing recidivism rates by 30 to 60 percent over follow-up periods of up to 15 years after program completion. Gendreau and Ross (1979) evaluated published studies of 95 programs treating antisocial behaviours for the period 1973 to 1978 and concluded that several types of correctional treatment programs were successful. Summarized below are five issues which they felt related generally to the failure of so many correctional programs. 1. Reliance on a single method - more positive results occurred by using a combination of methods. 2. Reliance on a single outcome, such as recidivism. Other outcomes such as resolving interpersonal, educational and vocational problems are valid goals too. 3. Interactions and individual differences - success is increased by taking into account how individual traits and treatment settings interact. 1 2 4. Not enough treatment - too few inmates have adequate access to treatment sessions, many correctional environments are a poor context for promoting prosocial behaviours. 5. Lack of interrelation among agencies - fragmented, lack of services to offender. Ross and McKay (1978) evaluated 53 institutionally and community based behaviour modification programs for the period 1965 to 1976. They discovered three factors that distinguished successful from unsuccessful programs. 1. The successful programs were not imposed on the offenders in an authoritarian fashion, but involved them in program planning. 2. In successful programs the target behaviours were not anti-social behaviours. They sought to strengthen prosocial behaviours rather than attempting to reduce the frequency of inappropriate or anti-social acts. 3. They neutralized or mobilized the offender's peer group. (1978, pp. 291-292) Support for the rehabilitative potential of certain programs is being linked increasingly with specific criteria such as those cited above. Education is one area that has attracted interest because of its potential to incorporate the various elements of successful correctional programs. 13 Correctional Education Correctional education confronts problems similar to those faced by other kinds of treatment programs. Educational objectives are thrust up against the reality of prison, an organization embodying practices predicated on control, not change. The Ontario Institute for Studies in Education (OISE) Report (OISE, 1978) to the Solicitor-General of Canada reviewed education and training in Canadian corrections and concluded that participation was voluntary, and thus, on this point, in accord with principles of adult education. Furthermore, the reviewers were impressed by the apparent willingness of inmates to involve themselves in a variety of programs. The OISE reviewers were particularly concerned with possible barriers or inhibitors to participation in prison education and training programs. The correctional system's resistance to recommendations for the improvement of education for inmates is attributable to the importance given control rather than change in setting correctional priorities. Furthermore, Cosman (1981) contends that much education sponsored by the correctional system itself lacks commitment to real attainment. Penitentiary education in Canada has been characterized by a general lack of interest in genuine educational achievement, by inadequate standards of teacher selection and training, by a lack of discrimination in matters of curriculum between the trivial and the important, a lack of discipline and structure, and by a complete lack of educational research (1981, p. 46). 14 Moreover, prison education often admits no vision other than that of fitting the inmate for work (Cosman, 1981; Fawcett, 1983; Knights, 1983). Yet, inmates in federal prisons are primarily adults. Canadian society is fundamentally a "politically based system of human relations, not one aimed solely at economic well-being" (Fawcett, 1983). If inmates are to function in mainstream society, they will require more than the barest utilitarian training or education that characterizes much correctional education. Indeed, Morin (1981 ) argues that education for inmates is essential for their dignity and development. The Prison University Program One Canadian prison-based educational program stands out because it appears to provide a significant educational experience for its inmate-students. An academic, undergraduate program has been in existence for over thirteen years, sponsored first by the University of Victoria, and more recently by Simon Fraser University. The program is based on a conceptual model which assumes that inmates (1) are more decision makers than victims, (2) have cognitive deficits (lack of critical reasoning ability), (3) possess a limited repertoire of social responses and (4) have limited moral reasoning ability (Duguid, 1979; 1981a; 1981b; 1983). These cognitive deficits are thought to contribute to inmates' initial and recurring criminal behaviours. The way to overcome cognitive, social and moral reasoning deficits in inmates is to involve them as learners in 15 an issue-oriented, liberal-arts program within an interactive and democratic community (Duguid, 1980; 1981). Duguid (1981a) observed changes in inmates who participated in the university program, in motivation, attitude, and social interactions. Whetstone (1981), a former student in the program, described the changes he had experienced, from skeptic and cynic to enthusiastic supporter of the university program. Boshier (1983) cited reports by inmates of the positive changes wrought in their lives by the university program. A study on the post-release effects of the university program on 65 men over a two year period concluded that the rate of recidivism for students was 14 per cent compared to 52 per cent for a matched group of non-student prisoners. Though concern has been expressed over the fact that inmate-students in the university program were self-selecting and that the study used a matched rather than randomly derived control group, a great deal of interest was generated by the optimistic results (Ayers et al., 1980). The apparent success of the university program in facilitating significant personal and social change in inmate-students casts it into the "some things work" category. In an external review of this university program, Ross (1980) observed that it had several components in common with other effective correctional intervention programs. He described the program as "multi-faceted", one that emphasized cognitive development and inter-personal problem-solving skills. A key feature of the program was its ability to neutralize the anti-social influences 16 of the offender's peer group while at the same time mobilizing it as a prosocial force. Similarly, in their review of effective prison education programs, Linden and Perry (1982) concluded that intensive programs, ones that provide inmates with an alternative community and peer support, will have the most impact and success. Ross and Fabiano (1983a) maintained that the university program embodied principles of intervention associated with what they label the "Cognitive Model" of crime and delinquency. The cognitive model suggests that the offender needs training, not therapy. He needs to learn not only social skills and vocational skills, but thinking skills, problem-solving skills, and decision-making skills. He needs to develop his social perspective: to go beyond an egocentric view of the world by developing the ability to take the perspective of other people (1983a, p. 7). Ross and Fabiano (1981) distinguished between impersonal and interpersonal cognition. Impersonal cognition encompasses the physical realm including the development of an understanding of causality, time, movement and space. Interpersonal cognition refers to developing an understanding of people, including the ability to interact with them based on an understanding of their perspectives. The cognitive model assumes that many offenders have developmental delays in the acquisition of a number of cognitive skills which are essential to social adaptation. Yet, Ross and Fabiano (1983a) concluded that interventions aimed only at reducing impersonal and interpersonal cognitive deficits in 17 offenders did not necessarily change criminal behaviour, nor lead to better post-release adjustment. Successful correctional interventions attended as well to the processes of program implementation, to the manipulation of the inmate's social environment. Cognitive and other conceptual underpinnings of the university program (Duguid, 1981a; 1981b) provide a mentalistic portrayal of the dynamics of change in inmate-students even though there is general agreement about the importance of the inmate's peer group, the instructor's role as a model of prosocial attitudes and behaviour, and the need to involve inmates in creating and sustaining social processes supporting the program. Yet little attention has been given to describing how interactions occur among individual inmates, their peers, program personnel, and the program content. This study attempts to operationalize these interactions. Boshier (1983) pointed to the need for prison research that operationalized "person" and "environment" variables in interaction with one another. Authors like Duguid (1980; 1981) have acknowledged the importance of the prison environment because it impinges on all facets of inmate experience, including membership in the academic community. The prison environment is able to overpower individuals (Gosselin, 1982; Jackson, 1984; and Zimbardo, 1977). Yet inmates in the university program appear to learn, to develop intellectual and social skills, and to acquire new attitudes. Inmate-students spend more time as prisoners than they do 18 as students. Duguid (1981) observed that the academic community in prison "remains an island in a hostile sea." Inmate-students move between two communities which embody opposing values and which expect different kinds of behaviour from them. Tension results. It may deter the individual from further involvement in the university program or possibly impel him to identify even more with the norms of the academic community. Boshier (1983) and Bell (1982) cited examples of reluctance among inmates to become involved with the program because of institutional disturbances and lingering perceptions among some inmates that the program is "therapy" and therefore in conflict with inmate codes. An inmate-student in a Scottish prison on a twenty year sentence for bank robbery recently wrote: Because of my offence, sentence, and the fact that I was in the notorious security party at Peterhead (sic). It caused quite a stir when I opted for education classes after four years in prison. Normally this would have meant being a creep, joining the "other side", or at the very least sycophantic behaviour in surrendering to the "enemy" (J. Crosbie, personal communication to SFU Program, 1984). There is evidence that inmates who stay involved with the university program accommodate the tension of conflicting expectations coming from the academic and traditional prison communities by developing a dual role or social identity as an "inmate-student" (Duguid, 1981a). The change from "inmate" to "inmate-student" occurs within an "alternative" academic community within the prison, one characterized by peer group support and extensive modelling by university personnel. Inmates appear to develop their student roles gradually rather 19 than instantaneously, as they become involved in the program. The alternative academic community provides them with opportunities to practice student roles, receive feedback and reinforcement, and deal with the personal and social tensions their choices entail. The process of taking on student roles seems to be marked by stages which characterize changing feelings and expectations inmates' hold toward the university program. The study reported here involved investigation of stages of role development inmate-students experience, not just as a cognitive process of acquiring interpersonal skills (Ross & Fabiano, 1981) but also in relation to the authoritarian character of prisons and the competing expectations and norms of significant social groups and individuals within the prison and academic communities. Prisons present formidable obstacles to successful change-oriented programs. Successful initiatives like the SFU university program must accommodate the impact of prison organization on the social ecology of individuals and groups within their domains. The literature relating to the SFU university program lacks a comprehensive explanation of how changes in the prevailing social ecology of the prison are achieved. The study reported here involved development of a theoretically grounded explanation of inmate-students' changing roles and accompanying feelings and expectations toward the university program and prison. 20 Purposes of the Study Inmate-students' interactions with prison and academic environments provided a context for this study, whose purposes were threefold: 1. Explicate a theoretical model to identify student roles (and associated feeling states and expectations) occupied by inmates who participate in the prison university program. 2. Operationalize the model with expert judges and inmates. 3. Examine relationships between scores obtained from operationalizing the model and various socio-demographic and prison-related variables. The purposes were accomplished by (1) utilizing role theory as a framework to develop a model, (2) operationalizing the model with input from expert judges, and (3) conducting a study with inmate-students. Before describing the study procedures it is necessary to review literature pertaining to the social ecology of prisons, particularly the "deprivation" and "importation" views of inmate society and behaviour. 21 CHAPTER 2 LITERATURE REVIEW Successful correctional initiatives such as the SFU university program engage the social ecology of the prison in a number of ways: The program approach is based on a social learning or educational model. Environmental factors which support delinquent behaviour or prevent prosocial adaptation are neutralized. The offender's peer group is neutralized or is re-mobilized as a therapeutic force. Program personnel provide prosocial modelling and reinforcement. Offenders are actively engaged in various aspects of the program. (Ross & Fabiano, 1981; 1983a) This chapter will review salient literature on the origins and structure of prison ecology. Prominently featured in the literature on inmate social systems are discussions of reference groups, typologies of inmate roles and distinctions between "deprivation" and "importation" views of inmate society and behaviour. Views of the Inmate Social System The inmate social system is an important element of prison life, an integral part of prison culture. The inmate social 22 system is the repository of inmate norms and traditions that exist in counterpoint to prison authority. Grosser contends that membership in the social system influences an individual inmate's response to correctional programs. The role of the inmate social system is far more significant and exerts far deeper influence on the personality of its members than is implied by the superficial notion that criminals teach each other bad habits....Treatment orientation and reform have to reckon with the prisoner as a group member and all that this entails (1968, p. 23). Insofar as the prison is an isolated social organization structured around principles of control, discussion of the inmate social system must consider its relationship to prison authorities. Grosser (1968) asserts the relationship between prison authorities and inmates typifies a "ruling caste" and "subordinate caste" respectively. Similarly, Thomas and Peterson (1977) made a distinction between the "haves" (staff) and "have-nots" (inmates). Grosser (1968) elaborated on the notion of caste-like differences between inmates and staff by observing that opportunities for vertical mobility across caste lines do not exist in prison. Thomas and Peterson (1977) contend that, notwithstanding recent challenges to the power of prison authorities, even the most mundane inmate activities remain subject to official dictate. This view of the different statuses for staff and inmates is consistent with a portrayal of prisons as "total" institutions. Goffman succinctly defines their characteristics: The central feature of total institutions can be 23 described as a breakdown of the barriers ordinarily separating.... spheres of life. First, all aspects of life are conducted in the same place and under the same single authority. Second, each phase of the member's daily activity is carried on in the immediate company of a large number of others, all of whom are treated alike and required to do the same thing together. Third, all phases of the day's activities are tightly scheduled, with one activity leading at a prearranged time into the next, and the whole sequence of activities being imposed from above through a system of explicit formal rulings and by a body of officials. Finally, the contents of the various enforced activities are brought together as parts of a single over-all rational plan purportedly designed to fulfill the offical aims of the institution (1961, p. 17). In total institutions, the split between staff and inmates is frequently characterized by hostile stereotyping. Referring to the Canadian federal correctional system, Eichman observed that officials view "...virtually all convicts as more or less dangerous, untrustworthy and/or too immature to function without tight security" (1981, p. 6). This kind of stereotyping has further consequences for the disposition of inmates once they are taken into the correctional system. Once convicted, the Canadian federal offender is twice as likely as his counterparts in the U.S.A. to be classified to a maximum security institution (Eichman, 1981). In addition to holding increasing numbers of prisoners relative to minimum security, maximum security prisons have influenced medium security institutions to acquire similar attributes such as peepholes, increased internal discipline and armed guards (Gosselin, 1982). To the extent that inmates are presented with an authoritarian prison organization and a caste-like, antagonistic schism between themselves and staff, they are likely to make 24 oppositional responses to their imprisonment. Collective inmate responses to the shared fact of their imprisonment are shaped by the existence of an inmate code which defines the prescriptions and proscriptions of behaviour (Thomas & Peterson, 1977). Deviations from the code result in inmate imposed sanctions, so most inmates respect and defer to the code even though they may not actively promote it. Elements of the code include: Opposition to the values of conventional society, and to prison officials. No supportive or nonexploitive liaison with prison officials. Status and prestige accorded to inmates most visibly opposed to the administration. Positive valuing of physical violence and strength, exploitative sex relations. Predatory attitudes toward money and property. Strong emphasis on in-group loyalty and solidarity. Aggressive and exploitative relations with conventionally oriented out-groups (Ohlin, 1956, pp. 28-29). Sykes and Messinger classified the tenets of the inmate code into five major groups: 1. Don't interfere with inmate interests - be loyal to your class - the cons. 2. Don't lose your head - don't argue or quarrel with fellow prisoners. 3. Don't exploit inmates - by means of force, fraud or chicanery. 4. Don't weaken - show courage, maintain integrity in the face of privation. 5. Don't be a sucker - don't accord prestige or 25 respect to prison authorities or the world they stand for, don't adopt values of hard work and submission to routine (1960, pp. 6-9). While there is general agreement about the existence of an inmate code, there is debate over the extent to which it is caused by prison organization or is sufficient to predict the variety and degree of inmate adaptation to incarceration. These issues form the two principal theoretical perspectives of inmate society - the "Deprivation Model" and the "Importation Model." The Deprivation Model The deprivation model has been characterized as a "structural functional perspective" in which an oppositional inmate code and subculture are viewed as a response to prison organization (Thomas & Peterson, 1977). Much of the most important literature on the deprivation model was written in the 1960's and retains its relevance. These authors have been cited in the explication of deprivation and importation models. Proponents of the deprivation model have focused on three areas of concern: 1. Processing and induction procedures that contribute to the "homogenization" of inmates; 2. The problems and deprivations of confinement that are either directly or indirectly a joint product of the prison organization and the position held by inmates within that organization; 3. The collective or subcultural response that inmates make to their common problems (Thomas & Peterson, 1977, p. 48). The processing and induction of inmates and the 26 deprivations of confinement they experience are a highly regulated process of status attribution or degradation (Grosser, 1968; Davis, 1976; Cloward, 1960). Goffman (1961) describes the stages of the degradation process in terms of a tension between status attributions arising from an inmate's civilian world and those imposed within the prison. It is characteristic of inmates that they come to the institution with a "presenting culture"....derived from a home world -- a way of life and a round of activities taken for granted until the point of admission to the institution....Whatever the stability of the recruit's personal organization, it was part of a wider framework lodged in his civil environment — a round of experience that confirmed a tolerable conception of self, and allowed for a set of defensive maneuvers, exercised at his own discretion, for coping with conflicts, discreditings, and failures....The recruit, then, comes into the establishment with a conception of himself made possible by certain stable social arrangements in his home world. Upon entrance, he is immediately stripped of the support provided by these arrangements... he begins a series of abasements, degradations, humiliations, and profanations of self. His self is systematically, if often unintentionally, mortified (1961, pp. 22-23). Once invested with a degraded status of "criminal" and "non-person" (Sarbin & Allen, 1968; Ekstedt & Griffiths, 1984), prisoners experience a host of "pains of imprisonment" including physical separation from society, deprivation of goods and services, deprivations of freedom of movement, heterosexual relationships, autonomy, status and security (Sykes & Messinger, 1960, pp. 13-16). According to the deprivation model, these "facts" of confinement are sufficiently powerful to displace other sources of life experience and history from the outside, leaving prisoners isolated as a group at the bottom of the prison 27 Figure 1. Oppositional Social Ecology In The Deprivation Model organizational hierarchy. The cultural vacuum thus created impels the growth of an inmate society which is oppositional in nature to the formal goals of prison authorities and society at large. Figure 1 shows interactions among and between inmates and prison organization. This diagram does not show all of the dynamics of the deprivation model, but rather that inmates are presumed to oppose any expression of prison authority. In Figure 1, the jagged lines between the arrows (representing separate goals of inmates and authorities) suggest antipathy. 28 The growth of the inmate subculture or society, according to the deprivation model, is best explained as a process of "prisonization" which Clemmer (1958, p. 299) defined as "the taking on in greater or less degree of the folkways, mores, customs, and general culture of the penitentiary." The key feature of the deprivation model's view of prisonization is that virtually all prisoners are likely to become part of the inmate subculture to varying degrees, some more than others according to factors such as sentence length, differences in prison organization, psychological attributes, kind and extent of contact sustained with the outside and so on. Inmate society, with its prescriptions and proscriptions, is the central feature of prison life that all inmates, according to Roebuck (1963), have to engage in to exploit or face being exploited by it in turn. A major area of study on prisoners' assimilation into inmate society looks at the development of social role types which accord differential status to their occupants. Sykes and Messinger (i960) viewed the development of distinct social roles as a function of group cohesion, or inmate solidarity, which mitigates to some extent the "pains of imprisonment". Prisoners enact the various roles, or are assigned a particular role, based on the extent to which they exemplify the ideals of the inmate code, especially the notion of inmate solidarity (Sykes & Messinger, 1960; Kassebaum, Ward & Wilner, 1971). One of the most notable depictions of inmate social role types was devised by Sykes and Messinger (1960). They described 29 a variety of social roles labelled in argot or slang terms. Among them is the "rat" who betrays another prisoner; the "tough" who is aggressive, quarrelsome and fights readily for no obvious cause; the "merchant" who uses manipulation and trickery rather than force to exploit other inmates, usually involving the sale or trade of scarce goods; the "weakling" who shows he cannot cope with the rigors of prison; the "wolf" or "fag" who becomes actively or passively involved in homosexual relationships; the "square John" who is seen to conform to the values and norms of prison authorities; and the "right guy" who is supposed to be the most loyal to other prisoners, the most steadfast in holding to the values of the inmate code in opposition to prison authorities (1960, pp. 9-11). Cloward (1960) proposed that because society denies prisoners means of access to higher status, the prison system recognizes the need to allow inmates to evolve means of restoring status within inmate society. By so doing, prison authorities ensure that potentially disruptive behaviour becomes channelled instead into acquiring status within inmate society by officially tolerated "illegitimate means." The "illegitimate opportunity structures" correspond to the influence associated with the argot social role types. Those prisoners who become upwardly mobile through roles like the "right guy" or "merchant" help to preserve stability in prison so as to maintain their superior status. However interesting, these social types are still undocumented in the universality of their influence on inmates. 30 A number of studies have questioned the utility of social role types because of the difficulty of actually classifying inmates according to the categories (Kassebaum, Ward, & Wilner, 1971; Glaser & Stratton, 1961). Glaser and Stratton (1961) observed that a limitation of social role typologies results from their portrayal of extreme rather than average roles in the inmate social system. In effect, they are abstractions of inmate behaviour perceived to connote interests and consequences in relation to the inmate code and group solidarity. The Importation Model Proponents of the "importation model" do not reject all the claims of the deprivation ^model. The importation model, according to Thomas and Peterson (1977), is an "extension" of the deprivation model and should be seen as complementary rather than contradictory. The principal criticism of the deprivation model is that it is too restrictive in suggesting that inmate society emerges primarily as a direct response to the impact of prison organization. Rather, in arguing for the importation model, Thomas and Peterson note that "... inmates have a past, a future, and, not unimportantly, a present which is not exclusively tied to their position within the prison (1977, p. 56). Advocates of the importation model look to external and intra-institutional variables to analyze prisonization and to seek positive resocialization among inmates. External 31 Figure 2. Pre- And Post-Incarceration Factors In The Importation Model influences take the form of pre-prison (past), present and future experiences. Pre-prison influences include social class of origin and attainment and previous criminal convictions. Present influences are things like the number of letters inmates 32 receive per week and the quality and frequency of contacts they maintain with outside reference groups or individuals. Future influences refer to the concerns and understandings inmates construct around post-release life-chances and opportunities_. External or extra-institutional factors contribute to responses inmates make to prison organization and the inmate code (Thomas, 1973; Thomas & Peterson, 1977; Garabedian, 1963; Sapsford, 1978). Figure 2 depicts the intra and extra-institutional interactions of the importation model. Inmates manifest opposition to authority but the inmate social system (shown as a circle with spaces or gaps in the circumference) is less insular than as shown in Figure 1. Pre- and post-incarceration factors in the outside community are shown (with arrows) to act on inmates, outside of the control exercised by the prison. The importation model allows for both prosocial and antisocial inmate adaptation as outcomes associated with any of these factors which may be under or outside the control of prison authorities. Some researchers have compared the deprivation model with a variant of the importation model, one that evaluates the impact of previously held criminal values on prisoner adaptation (Irwin & Cressey, 1962; Cohen & Taylor, 1972; Kassebaum, Ward, & Wilner, 1971; Akers, Gruniger, & Hayner, 1976; Hepburn & Stratton, 1977). These studies give qualified support for an importation model. Garabedian (1963) explored the influence of length of time served and remaining to be served on the degree of prisonization. While Clemmer (1958) maintained that variations 33 in the degree of prisonization from inmate to inmate were inevitable, the deprivation model implies that prisonization should increase as a function of time served across all social role types. Garabedian (1963) measured the degree of prisonization for three separate groups of inmates in one prison based on time served and time remaining to be served. Inmates in the "early phase" had served less than six months, those in the "middle phase" had served more than six months but had more than six months remaining to be served, and in the "late phase" inmates had less than six months remaining (1963, p. 141). Garabedian found that conformity to staff norms among most inmates in the three groups showed a curvilinear, or U-shaped "adaptive" pattern. Inmates in the early phase were "proportionately twice as likely to conform to staff norms as compared with inmates in the middle period" (1963, p. 142). Moreover, this process of prisonization, that is, movement away from staff norms to inmate code norms, "is reversed as the inmate comes to the end of his prison career" (p. 142). Garabedian's interpretation of the return to conformity with staff norms in the late phase as anticipatory socialization to external groups and values buttresses the claims of the importation model. The utility of the importation model for this study is its explication of factors, beyond the control of prison authorities, that influence inmate adaptation to prison organization and the inmate code. Inmates are adults. They relate to their present situation in a way that reflects their preprison 34 learning experiences, their extra-prison learning experiences, their extra-prison relationships, and their post-prison expectations Most prison inmates do not become completely or uniformly assimilated into the inmate society Resocialization can be efficiently developed even in the context of a custodially oriented maximum-security prison (Thomas, 1973, pp. 20-21). The importation model recognizes that inmates have the opportunity to develop relationships with groups, like the prison university community, outside the direct control of prison authorities. Moreover, such relationships suggest possibilities for developing roles in addition to or instead of those associated with the inmate code. In any case, proponents of the deprivation and importation models concur on the importance of viewing inmates as members of groups and of the saliency of utilizing social role theory to analyze the inmate social system. In Chapter 3, the social ecology of the prison university community will be analyzed using social role theory and a model formulated to explain changes associated with inmate participation in the university program. The model of role development assumes that the prison ecology influences inmate behaviour in the manner proposed in the "deprivation" position and incorporates the "importation" premise that extra-institutional influences determine behaviour as well. 35 CHAPTER 3 CONCEPTUAL FRAMEWORK This chapter outlines a conceptual framework based on role theory, from which a model of role development in inmate-students emerged. Discussion centres on assumptions and major concepts in role theory, criteria of the model, and the model's five stages. General Assumptions of Role Theory Role theory is well-suited to the analysis of prison environments. Within these total institutions the roles for staff and inmates are rigidly prescribed and status is accorded to individuals on the basis of them. In the case of inmates creating "new" roles for themselves as students in the university program, role theory provides a rationale for understanding difficulties arising from the competing demands of "prison" and "education." The all-encompassing role of inmate is challenged by the educational environment with its emphasis on democratic student-instructor and student-student relationships. There are many variations of social role theory, but Sarbin and Allen advanced a version with particular relevance to the SFU prison university program (see also Sarbin & Schiebe, 1983). For them "role" is a theatrical metaphor, "...intended to denote 36 that conduct adheres to certain 'parts' (or positions) rather than to the players who read or recite them" (1968, p. 489). Thus, in a prison context, individuals manifest conduct according to the primary part they play at any given time; inmate, guard, or student. But each player maintains a larger self-identity than is accounted for by any one role. Sarbin and Allen's principal concept is "role enactment", which refers to the overt social conduct of a person in one or more roles. They pose a number of questions relating to role enactment: What are the positions of the others with whom the actor is performing? How effective is the actor in validating the occupancy of his status? What is the contribution of the others to the enactment - do they provide discriminative cues which lead the actor to select another role performance? (1968, p. 490) Role theory, with its focus on role enactment, "...bridges the gap between the individual and the group, between personal history and social organization" (1968, p. 490). They propose three criteria for making inferences about observed or reported social behaviours — the "appropriateness", "propriety", and "convincingness" of the enactment: 1. Is the conduct appropriate to the social position granted to or attained by the actor? That is, do his performances indicate that the actor has taken into account the ecological context in which the behavior occurs? In short, has he selected the right role? 2. Is the enactment proper? That is, does the overt behavior meet the normative standards which serve as valuational criteria for the observer? Is the performance to be evaluated as good or bad? 3. Is the enactment convincing? That is, does the enactment lead the observer to declare unequi vocally that the incumbent is legitimately 37 occupying the position? (1968, p. 490) Sarbin and Allen consider three additional dimensions of role enactments to determine their appropriateness, propriety and convincingness: (1) number of roles, (2) organismic involvement (effort), (3) preemptiveness (time) (1968, p. 491). Number of roles refers to the variety or repertoire of social roles that are not only well-practiced and readily available to the actor, but socially realistic as well. They contend that without a variety of such roles to call on, a person is less well prepared than the skilled role-taker to meet new and stressful life situations. Inmates have been characterized as lacking an adequate range of responses (both cognitive and behavioural) to the economic and social demands of society (Duguid, 1981a; Ross & Fabiano, 1981). Their repertoire of roles has more to do with survival in that specialized environment than with realistic social interaction outside of prison (Davis, 1976; Garabedian, 1963; Roebuck, 1963; Sykes & Messinger, 1960; Thomas & Foster, 1976). Thus, the development and practice of student roles within an academic community in prison implies an increased capacity for inmates to respond with behaviour acceptable to society (Ross & Fabiano, 1983a; Ross, 1980). The student role encompasses a range of sub-roles such as learner, colleague, researcher, group member, helper. If offenders are put in prison not just as punishment, but for more punishment, their involvement in student roles lacks legitimacy because inmates are not, according to this view, 38 supposed to enjoy any aspect of prison. On the other hand, if it is important for inmates to be able to develop socially realistic and acceptable behaviours, then inmate involvement in programs which promote socially positive values of mainstream society is "appropriate" and "proper." A second dimension is the "organismic involvement" or intensity of role enactments. Organismic involvement refers to the degrees of effort and visceral participation in a role enactment by a person. Sarbin and Allen suggest as an example of low involvement a ticket seller in a neighbourhood cinema during a slow period of business (1968, p. 492). At the high end of involvement is the quarterback's role during a championship football game. Great degrees of effort are involved through muscular exertion, autonomic nervous system activation (visceral), and involvement of the self in the role. Sarbin and Allen assert that "even the casual observer can identify characteristics of role enactment along an intensity dimension" (1968, p. 492). The university requires inmates to maintain a minimum standard of performance in order to be allowed to remain in the program. As with students everywhere, effort is involved in meeting and exceeding that standard. Those who perform at or beyond the minimum standard are enacting the student role with propriety. There are other indicators of involvement in the student role. These provide insights into how involved certain inmates want to be. An awards ceremony is held at the end of each term in each of the prisons where the university program operates. 39 At these award ceremonies inmates typically exhibit pride at receiving awards for coursework. Their peers, by responding with unrestrained applause, reinforce the legitimacy of scholarly effort. Instructors talked of the persistence and depth of motivation of their students. During the Matsqui prison riot of 1981 students in the university program dissuaded rioters from burning the university library. To oppose the will of inmate rioters in order to protect an important symbol of the university program demonstrated how involved with the program some inmate-students had become. The awards ceremonies, the Matsqui riot, and the evaluation by university instructors all support conjecture which holds that inmates become convincingly involved with student role enactments. Moreover, a recent survey involving 11 of the 16 instructors in the SFU university program concluded that most instructors viewed inmate-students as highly motivated to engage in education (Duguid, 1985). A third dimension of role enactment is the preemptiveness of roles — "the amount of time a person spends in one role relative to the amount of time he spends in other roles" (1968, p. 496). In prison, opportunities to spend time in a student role are regulated by prison authorities. Furthermore, authorities treat time spent attending educational courses as a work placement for which inmates are paid. But the pay scale for education is lower than for other prison work assignments such as in maintenance, kitchen and laundry. Moreover, inmates in the university program are now required to pay a token fee 40 for each course they take. Yet inmates choose to be students despite financial disincentives. Concepts in Role Theory Sarbin and Allen treat role enactment as a dependent variable with judgements about its appropriateness, propriety and convincingness subject to the influences of a number of independent variables. The first of these independent variables is the concept of role expectations — the beliefs, expectancies, "rights, privileges, duties and obligations of any occupant of a social position in relation to persons occupying other positions in the social structure" (1968, p. 497). As such, role expectations set the normative limits for social behaviour. They operate as personal-psychological and social imperatives to a person's conduct and cognition while enacting a role. Different role expectations and imperatives for conduct arise from inmate and student roles within a prison context, in part because the complementary roles and the people who occupy them are different. For instance, correctional employees enact roles of "keeper" or "guard." Role expectations encompassed in inmates' relationships with them are imbued with the antagonism inherent to the prison system. On the other hand, the role of instructor is complementary to a student role. University instructors are employed by universities, not the correctional service. Role expectations between instructors and students are not encumbered by the traditional suspicion that exists between 41 correctional staff and inmates. Role expectations between occupants of complementary social positions are mutually determined. Thus, Sarbin and Allen contend that conformity to certain role expectations may occur even in the absence of strong commitment to a particular role because of a person's sensitivity to other peoples' reactions (1968, p. 502). They use the concepts of role clarity and role location to further explain the mutually-determined behaviour of persons in complementary social roles. Role clarity refers to the "difference between the optimal amount of information needed about role expectations and the amount actually available to a person" (1968, p. 503). Where a lack of clarity in role expectations exists, a person has difficulty knowing which role enactments are appropriate and what the complementary conduct of others should be. For both inmate and student roles in a prison context, role clarity is prescribed at a fairly explicit level, although conflicts between roles may occur. Role location refers to the inferential process whereby a person's "choice of role follows from the location of self in the social structure, such location being determined conjointly with locating the position of the other" (1968, p. 507). Thus, in the context of prison life, locating the position of a CSC employee as "cop" or "keeper" will lead an inmate to choose a complementary inmate role. On the other hand, locating the position of a university employee as "instructor" creates the possibility of an inmate choosing other than an inmate role, 42 such as "student." Specific role enactments may arise from the folkways or mores associated with a particular social structure. These are called role demands. An inmate experiences intense pressures to conform to an acceptable range of inmate roles or face the consequences. However, there are role demands associated with a student role within an academic community established by the university and these demands help to impel inmates who participate in the program into enacting a convincing student role. Sarbin and Allen maintain that how well a person enacts a particular role depends not only on variables like role expectations, role location, role demands and self-role congruence, but also on those role skills possessed by the person. Role skills "refer to those characteristics possessed by the individual which result in effective and convincing role enactment: aptitude, appropriate experience, and specific training" (1968, p. 514). Accordingly, people may vary widely in the convincingness, appropriateness, and propriety of similar roles because of differential role skills. While role skills are learned largely through socialization experiences of early life, they may be enhanced in later life through appropriate training as in the case of the prison university program. Role skills may be divided into cognitive and motor skills, each of which divides into general and role-specific skills. General cognitive skills include the ability to make accurate inferences in a social situation, to take the role of 43 the other in both cognitive and affective terms, and may include general intellectual ability (1968, p. 515). It is the apparent lack of development in these general cognitive skills among inmates that provides a focus for the university program. Consistent with this view is the notion of eliminating or reducing cognitive "deficits" through appropriate training and education centred on a liberal arts curriculum. Ross and Fabiano's (1981; 1983; 1983a - see Chapter 1) claims concerning developmental delays in inmates' impersonal (i.e. general intelligence) and interpersonal (i.e. social skills) cognition match very closely these aspects of Sarbin and Allen's role theory. General motor skills refer to the body movements, posture, facial expression and tone of .voice required for many role enactments. For the student roles of interest in this study, an average level of motor ability is probably sufficient to allow inmates to fully enact those roles. For the study and enactment of drama a higher level of motor ability is a requisite for full enactment. Concerning self-role congruence, Sarbin and Allen contend that: Social roles are perceived and enacted against the background of the self. The term 'self refers to the inferences the person makes about the referent 'I'. It is a cognitive structure and derives from past experience with other persons and with objects. We define the self as the experience of identity arising from a person's interbehaving with things, body parts, and other persons (1968, pp. 522-523). To the extent that self-role congruence is high, there is a "goodness of fit" that should manifest itself as commitment or 44 involvement in a role whose enactment would then be judged convincing, proper and appropriate. In the absence of self-role congruence, confusion is likely. The person enacting the role experiences a state of tension and cognitive strain. Self-role congruence can be explored in terms of role conflict (interrole and intrarole), a fundamental dynamic in the model of role development in inmate-students. Sarbin and Allen contend that role enactment is an ongoing process involving (1) the role performer, (2) the person in the complementary role, and (3) a third member who observes the process of social interaction (1968, p. 528). The third member is the audience or "the other" or "the generalized other." The audience may be physically or only symbolically present and be a large group, small group or dyad. The audience establishes consensual reality for a role, provides discriminative responses (cues) to guide a performer's role enactment, gives social reinforcement to or withholds it from a role performer, and contributes to the maintenance of role behaviour over time (1968, p. 534). An audience which is only symbolically or cognitively present, but which a person values, and toward which a role enactment is directed, is called a reference group. The greater the congruence between an individual's values and the perceived values of a symbolic group, the more likely that this group will become a reference group for that individual. Several distinct audiences are important for inmates. The two most apparent groups, peers and prison staff, are physically present. Their influence reinforces inmate subculture roles 45 (see Chapter 2). The other audience of interest, the academic community, is both physically and symbolically present. It is physically present in the form of the university staff and fellow students. Equally important, it is symbolically present as society outside of prison but with the potential to become a reference group for inmates who become involved with the program. The significance of this possibility stems from inmates' acknowledgement of values and norms shared with a legitimate segment of society, one lying beyond the scope of usual inmate roles. Sarbin and Allen present a number of ways of looking at complex role phenomena. They discuss multiple roles in terms of either simultaneous or successive enactment. Successive role enactments are best characterized by the maturational process and its associated rites of passage or developmental tasks. Simultaneous enactment of multiple roles may occur when (1) only one role is activated, others being "latent", (2) two or more roles merge, (3) multiple roles alternate within a given period of time, and (4) informal roles occur simultaneously with formal roles, such as in the case of a surgeon who acts the comedian while performing surgery (1968, p. 538). A central feature of prison life for students is simultaneous role enactment and the development of role conflict between inmate and student roles. Interrole conflict is "due to simultaneous occupancy of two or more positions having incompatible role expectations," and intrarole conflict "to contradictory expectations held by two or more groups of 46 relevant others regarding the same role" (1968, p. 540). In general, an individual allocates time and energy among roles according to (1) the individual's norm commitment, (2) estimate of reward or punishment by role partners, and (3) estimate of reactions of a third party or an audience (1968, p. 569). However, role conflict results in cognitive strain which impels the individual to seek a resolution of the condition by any of the following adaptive responses: (1) instrumental acts and rituals, (2) attention deployment, (3) changes in belief system, (4) tranquillizers and releasers, and (5) no or unsuccessful adaptation (1968, p. 541). Although all five types of response may occur within a prison education program, the third response (change in belief system) is the one that has the greatest potential to explain role development among inmate-students. The other four response modes imply no real change in inmates involved with the university program. The dynamics of change in belief system are congruent with "perspective transformation", a process of developing a new perspective in a way that acknowledges differences between one's old and new perspective concerning thought, feeling and will (Mezirow, 1978). This includes an awareness of one's roles, and action to test or practice new roles which incorporate new values. Boyanowsky (1977, p. 116) documents three phases of perspective transformation — (1) separation from old relationships, (2) a distinctive transition period, and (3) incorporation into a new group or relationship. A similar three 47 phases are proposed by Mezirow (1978, p. 105) — (1) alienation from prescribed social roles, (2) reframing or restructuring one's conception of reality and one's place in it, and (3) contractual solidarity within which it becomes possible to participate again in society. These notions of perspective transformation enhance understanding of the process of role development from inmate to inmate-student. Moreover, the concept of perspective transformation also suggests how ex-students may be drawn to utilize new role opportunities outside prison subsequent to their release. The process of perspective transformation implies role conflict to an extent requiring resolution. Also implied is change in inmate belief systems and worldviews. The notion of "contractual solidarity" supports the view that inmate-students in the university program do not just become better educated criminals, but undergo significant positive development in the cognitive and social skills referred to by Ross and Fabiano (1981; 1983; 1983a). Concerning role learning, Sarbin and Allen stress the importance of persons enacting complementary roles as teachers, models and coaches, and the importance of relevant audiences. They differentiate between role learning in childhood of ascribed, and role learning in adulthood of largely achieved roles. Practice of roles is essential to role learning. Thus, by occupying and practicing student roles inmates gradually learn their full and convincing enactment. This gradual role learning process is the focal point for a model of role 48 development. Criteria for a Model of Role Development The preceding section on concepts of role theory served to outline the general assumptions of a framework for a model of in role development among inmate-students. According to Sztompka (1974), a theoretical or conceptual model must also specify its particular assumptions in terms of concrete choices along the dimensions outlined in the general assumptions. Dubin (1978) refers to these concrete dimensions in a theoretical model as (1) units of interaction, (2) laws of interaction, (3) boundaries, (4) system states, (5) propositions, (6) empirical indicators, and (7) hypotheses. Figure 3 depicts inmate-students role enactments and interactions in the prison and academic environments, and in relation to the outside community. Inmate-students enact roles (square symbol) in association with other inmate-students who perform complementary roles (round symbol) such as tutor or study partner, and audience roles (triangle) such as classmates. Role interactions also occur between inmate-students and instructors (triangle — complementary roles; round symbol — audience role) who legitimize their status as students. The academic area is a physically distinct location within prison. Non-student inmates do not normally go there. Nor do correctional staff frequent the area except on business. Consequently, the principal role performers in interaction with one another are instructors and inmate-students. 49 PRISON ENVIRONMENT ROLE INTERACTIONS PRISON ENVIRONMENT UNIVERSITY ADMINISTRATION UNIVERSITY STUDENT BODY THE OUTSIDE COMMUNITY Figure 3. Role Interactions Pertaining To Education In Prison 50 The symbols in Figure 3 represent units, or variables, "whose interactions constitute the subject matter of attention" (Dubin, 1978, p. 7). The units of the model of role development are principally the three sets of actors involved with any role enactment, (1) the role performer, (2) the complementary role performer(s), and (3) the audience(s). The laws of interaction among the units derive from the interrelationships of role expectations, role demands, role location, role skills, role learning, and role conflict associated with inmate and student roles. The laws of interaction provide for role conflict, cognitive strain, and the resolution of that strain in a number of ways. This model proposes that inmate-students' resolution of role conflict and cognitive strain results in a changed set of beliefs and a greater commitment to student roles. The boundaries within which the model is expected to hold are clearly defined by the correctional setting within which any such program is set. Thus, the university program in the four federal prisons in British Columbia are included within the boundaries, as are the students in that program. System states refers to any prominent condition or feature of the correctional system which fosters characteristic and persisting values among inmates and correctional employees. The most salient feature of that system is conflict and tension related to authoritarian prison organization. Inmate-students presumably experience the role conflicts and tensions that come from competing demands of inmate norms and student role demands. 51 The propositions of the model are its five stages and the transitions from one stage to another. These are described in the next section of this chapter. The means by which the five stages were operationalized is reported in Chapter 4. Stages of the Model A five-stage model explains how the social ecology of the academic environment, an alternative community within prisons, fosters cognitive and social development in inmate-students. The academic environments referred to in the model exist within four federal prisons in British Columbia. The general dynamics of prison life, organization and inmate social systems described in Chapters 1 afid 2 pertain to these prisons. The university program is open to inmates who are assessed by university faculty to possess the necessary English language skills. Pre-university qualifying English courses are periodically given for inmates expressing interest in the program but who lack the required language skills. Inmates may be recruited to the university program by other inmates, CSC staff, or university personnel. They may apply for enrolment upon entering a prison that hosts the program, or at some later time. Those who enrol on a full-time basis spend most of their weekdays in the academic area. Prison authorities view time spent in the academic area as a work placement. Every inmate is expected to "work," be it in the kitchen, on maintenance, or in other traditional prison jobs, in the 52 vocational shops, prison industries, or the university program. Pay scales vary for different jobs. Education is among the lowest paying "work" inmates may perform. In cash-short prison life even the smallest expenditure is regarded as important by inmates. Yet, inmate interest in the program remains high. Reasons for their continuing interest and involvement emerge from the model of role development. Figure 4 diagrams the essential interactions within the academic environment and shows the stages of the model of role development in sequence, starting with Recruitment. Descriptions of each stage of the model follow. Figure 4. Interactions In The Academic Ecology And Stages Of Role Development 53 Recruitment Stage Initially, inmate-students role expectations are derived from the inmate code. Cynicism and suspicion about the program are even more likely to predominate among inmates who have been through the prison system before. If they have General Education Diploma (G.E.D.) or vocational courses, they have some role expectations about being a student. However, these expectations may be different from those arising later out of participation in the university program because of the considerable differences in the scope and curricula of the various programs. At this stage, inmate expectations are likely to be instrumentally concerned with "making it in prison." Thus, they may think the university program is worth exploring for its possible strategic value. New recruits to the program have not yet experienced the expectations, cues or reinforcement from those who occupy complementary or audience roles to their student roles. At the very least, they do not anticipate being punished by any group (inmate or staff) for exploring the options the university program may offer. Even students enrolled with no thought of changing or getting something out of the experience must initially comply with the minimum role demands put forward by university instructors in order to maintain a foothold in the program while exploring these options. Thus, they experience student role enactment at some minimal level without necessarily being committed to it. Because of the assumptions and practices of 54 the instructional personnel and the physical encapsulation of the academic area, the enactment of student roles begins to create an awareness of its contrast with prison experience in general. Disorientation Stage Inmates' understanding of role demands and expectations increases as they continue to enact student roles. They become better able to locate roles and accompanying expectations of instructors and classmates. Inmates either continue to adjust their role enactments to the emerging social and cognitive criteria or reject them and drop out, or are required to leave the program. If they continue to adjust expectations and behaviour they also develop cognitive skills through participation in educational experiences (course work, discussions, analysis, projects). Their role skills improve and they become more skillful in their student role enactments. Cognitive development may be intrinsically as well as socially rewarding. Social reinforcement comes in the form of encouragement from instructors and classmates. However, because they do not expect anything important to happen, they may be initially surprised by the changes and rewards associated with their student roles. Moreover, they are disturbed and disoriented because expectations and behaviours deriving from student roles contrast sharply with inmate norms and roles. The contrast creates uncertainty about their primary relationships to the prison population in general which in turn 55 results in cognitive strain. The strain leads them to question the wisdom or "Tightness" of continuing in the university program. They may question the motives behind the program, and though role demands and expectations from instructors are easier to "locate", they may feel unsure about what is really involved and expected of them. Separation-Alienation Stage Inmates who stay involved in the program continue to develop their role and cognitive skills with practice. Social reinforcement by those in complementary roles (eg. instructors) and audience roles (eg. fellow students) also continues. Role conflict is likely to deepen until cognitive strain becomes uncomfortable enough to require resolution. Cognitive strain must be resolved when inmate-students feel that continuing their student roles poses a serious contradiction to the inmate code. Cognitive strain may be caused by interrole and/or intrarole conflict. Interrole conflict occurs when inmates confront the possibility of simultaneously occupying inmate and student roles which appear to have incompatible role expectations. Intrarole conflict occurs when inmates are subjected to the contradictory expectations of several groups of relevant others (instructors, classmates, general inmate population, staff) concerning their student roles. For example, instructors may expect them to take responsibility for their work and development, to show initiative, to practice acceptance and tolerance of other 56 students and their views. However, non-student inmates who support the inmate code expect them to withhold cooperation, to be suspicious, exploitative, and to take no responsibility for their actions. To recapitulate, an adaptive response to an uncomfortable condition of cognitive strain due to role conflict may lead to any of five kinds of outcomes. For those whose adaptive response is a change in their belief system, the chances of a deeper involvement in a student role and achievement of the goals of the program are greatest. Transit ion-Reframing Stage Individuals whose perspectives on student-inmate role conflict undergo transition (ie. a belief system change), experience a lessening of conflict and an increase in the degree of self-role congruence. Not only do those individuals continue their student roles, their commitment to get the most they can out of the program intensifies. Individuals begin to conceive of themselves as legitimately engaged in changing their social identity through the positive effects of student role enactments on their cognitive and social skills. They reconceptualize what is possible and attribute new value to outcomes associated with their participation in student roles. They need not anticipate an indefinite continuation of student roles in an academic setting in order to believe that fundamental and lasting benefits or changes have occurred for them. However, while in prison, and as the opportunity to do so 57 continues, they are likely to persist as students and stay involved with other students and instructors enacting complementary and audience roles. At this stage, inmate-students feel their roles have been validated through acceptance by others in complementary roles (their instructors and classmates). Additionally, other students and instructors in the academic area act as important audiences by giving general social reinforcement and by providing cues from their role expectations which guide the individual's role enactment and contribute to its maintenance over time. Solidarity Stage Inmate-students at this stage of development become stable supporters of the education program and its processes. Thus in the face of destabilizing events such as the Matsqui prison riot, the students who acted as guardians and advocates for the university program and its library exhibited a compelling attachment and positive valuation toward the concrete symbols of their academic community. Their strong identification with their student roles led to a display and test of their solidarity with the program, its personnel, and its property in the face of severe inmate opposition to prison authority and organization. Individuals who resolve cognitive strain and continue in the program become models for new recruits to the program. "Seniors" have a stabilizing effect on the behaviour of newer 58 students. Their own relationships with instructors and program administrators becomes more collegial as they become tutors and facilitators of other inmates' learning. These individuals have a "robust" relationship to the program and are less likely than new recruits to discontinue their student roles when personal disruptions in the prison occur. To reiterate, this chapter presented a model composed of five stages of role development. The model assumed that inmates are socialized into the "positive" higher stages (Transition, Solidarity) through participation in the SFU program and its social ecology. Moreover, a tension (role conflict) is thought to exist for inmates as they move between the distinct ecologies of the prison and academic program. This tension impels inmates into transition to higher stages of role development as they take on the role of student. The relationship among stages implies linearity. Inmates newly recruited to the program, without prior university experience, would be expected to experience Recruitment and Disorientation states first, then Separation, Transition and Solidarity. However, this process of role development may be sudden or gradual depending on the particular characteristics of the prison and academic ecologies, and the cognitive development of the individual. This last point raises the possibility that inmates might "jump" stages (from initial to later) if the transition is very sudden. The model assumes that role development, once it occurs, will remain stable if no significant disruptions occur to the person (eg. transfer to another prison, being placed in 59 solitary) or in the academic ecology (eg. program being closed). Inmates would not be expected to "jump" from a higher stage to one lower. However, as Garabedian's (1963) research demonstrated, inmate social roles are responsive to stages in sentence lifespan, and so may be individual inmates in the SFU program. Student role development, unlike inmate roles, is presumed to provide inmates with cognitive and social skills (and reinforcement) that provide an alternative to prison norms. Chapter 4 outlines the rationale for and steps taken to operationalize the model, provides various estimates of reliability and validity, and proposes several research questions deriving from the model of role development. Chapter 5 describes the methodology employed in the study. 60 CHAPTER 4 INSTRUMENT DEVELOPMENT Chapter 3 described how the model was developed. Before the model could be tested with inmates, it first had to be operationalized, and that process is reported in this chapter. An integral part of the instrument development process involved the use of expert judges. Also reported are selected results obtained from testing the instrument with inmates. Although these results were obtained from the study proper, they were included in this chapter to provide additional information about the plausibility of the model's stages. The unique characteristics of prison environments present special difficulties for researchers. Inmates are often reluctant to participate in research, especially that which involves questionnaires. They express resentment at being "tested" by researchers when they perceive no benefit to themselves and feel that the results of such research could be used against them by prison authorities. Inmate reticence about research necessitated using more nonreactive methods of obtaining information. A nonreactive method is one that reduces (though it cannot eliminate) the risk of deliberate or unintentional response bias by inmates. A crucial component of method is the instrument. A reactive instrument may negate a less-reactive method (eg. using a questionnaire within participant observation). 61 From previous experience (Boshier & Clarke, 1983) it was known that inmates resist paper-pencil survey-type instruments, so this approach was not used. A card sort (Q-sort) was selected as an appropriate instrument to obtain information from inmates, and more likely to secure their participation than written questionnaires. A card sort requires no writing and is gamelike. Q methodology is the name given to a set of psychological, philosophical, statistical ideas and procedures concerned with research on individuals (Stephenson, 1953; Kerlinger, 1973). Q technique involves the rank ordering of items or groups of objects such as words, pictures, verbal statements or written phrases into a number of categories or piles based on some criterion (Kerlinger, 1973). The Q-sorts by which this rank ordering is achieved may be structured or unstructured. Unstructured Q-sorts consist of items of presumed equivalence that refer to one domain, such as attitude, and which are not differentiated in any other way from one another except as the respondent presumes. Structured Q-sorts use items from one or more domains partitioned in one or more ways. For example, role development in inmate-students is a domain of interest. Items representing this domain were partitioned into the five stages of the model. In effect, the propositions of the model were built into the Q-sort. Subsequently, inmates' responses to these items in the Q-sort tested the internal coherence of each stage. Q-sorts are either "forced", requiring sorters to put 62 varying numbers of items into different piles based on a normal or quasi-normal distribution, or "unforced", whereby sorters distribute cards into piles without restrictions (Stephenson, 1953; Kerlinger, 1973). Sorting strategies which force regular distributions are sometimes adopted for statistical convenience. A Q-sort constructed to test the model of inmate-student role development was necessarily structured. Sets of written statements (items) incorporated the distinct characteristics of stages of role development. The sorting exercise was unforced to reduce the risk of deliberate reaction to the constraints of a required distribution of items. This "one-way structured sort" (Kerlinger, 1973) was analogous to a one-way analysis of variance, because the items in the sort were created and classified with reference to one domain -- broadly, stages of role development. Item Construction and Selection Written statements (items) making up the card sort incorporated theoretical propositions of the model. Descriptions of the stages derived from conversations with inmates enrolled in the SFU university program in four federal prisons in British Columbia. The conversations took place during the course of another study on inmates' motives for participation in prison education (Boshier & Clarke, 1983; Boshier, 1983). Numerous statements made by inmates during that study referred to their perceptions of personal change in social identity which they attributed to their involvement in the 63 university program. The propositions of the model and reports made by inmates in the university program formed the basis for constructing statements describing the subjective or "lived" experience of inmates in any of the hypothetical stages of role development. The initial process of generating statements (items) involved only the author. Items were constructed one stage at a time. When no additional statements emerged for a stage, the procedure was repeated with the next stage. Some items were discarded almost immediately because they were redundant or obscure. Four sample items follow. I don't expect much out of this program. I'm confident of my ability to learn. There is more work than I expected. I'm fed up with prison games. The initial item construction process resulted in 60 statements representing the model's five stages. These 60 statements were printed on labels and stuck one to a card. A second person familiar with the prison university program (as a teacher and administrator) evaluated the statements (in no particular order) for clarity, face validity, and content. He became the first of four expert "judges." The intended relationship of the items to the stages were explained to him. He was asked to sort the items into their respective stages and to point out any that should be deleted or modified should their adherence to the model's stages be unclear or contrary to his experience. His ideas for additional items for several stages, 64 notably Solidarity (Stage 5), resulted in further changes. Some items were rewritten, others dropped, and a number of new ones constructed. Seventy statements emerged from this stage of the item refinement process. The 70 items were computer printed on mail labels and then each one was stuck to a blank 3 by 5 inch file card. Multiple decks of file cards were assembled in this fashion. Item Judging Process In addition to the first judge, three more individuals were asked to serve as expert judges; all four were familiar with the specific prison university program. Each of the four had taught for extended periods in the prison university program. Two judges were still engaged in teaching in the program, two others were involved as university administrators. Separate appointments were made with each judge. Two were interviewed in offices at Simon Fraser University. The other two judges were interviewed on Vancouver Island, one at the prison where the study was subsequently conducted, the other in a residence near Victoria. The same procedure was followed with each judge. A brief and general account of the research project was given. The model and the function of the 70 items were explained. Judges looked at a figure (see Figure 4 in Chapter 3) displaying the five sequential stages of role development. Each stage was briefly explained in terms drawn from the conceptual framework (Chapter 3). Judges were invited to ask questions in order to 65 clarify their understanding of the model, its assumptions and its propositions. Each judge was handed a deck of file cards containing the 70 statements with the cards sequenced in no particular order. Five envelopes were placed before them. Each envelope had one of the stage names printed on it. Judges looked at the statements on cards and assigned each one to the stage they felt was most clearly suggested by each statement. They put any statements about which they were unsure into a pile to one side until after they had been through the complete deck of cards. They sorted the remaining cards into stages if they could. Judges then looked over their initial placements and moved any items they wished to other stages. Each judge assigned all the cards to one stage or another. Judges were asked for their observations regarding the item-statements. A few comments indicated that certain statements might be true of more than one stage because inmates had been heard to make similar comments, but (in the view of the judges) for quite different reasons. However, none of these observations prevented them from assigning each of the 70 items to a stage. The piles of cards for each stage were inserted into envelopes marked with the corresponding stage headings. These five smaller envelopes were inserted into a larger envelope which was then sealed and labelled with the judge's name. Judges' placements of 70 items into stages were entered into computer files for storage and statistical analysis. Table 66 1 summarizes the agreement of judges in placing items into stages of the model. With respect to 18 items, the judges showed unanimous agreement in placing them in stages. Fifty-eight of 70 items (83 percent) elicited agreement about their stage placements from three or more judges. Table 1 INTER-JUDGE AGREEMENT ON 70 ITEMS Judges Agreement No. of Items Percent Cumulative % 5/5 18 26 26 4/5 17 24 50 3/5 23 33 83 2/5 12 17 100 Kendall's coefficient of concordance (W) was calculated on inter-judge placements and yielded a coefficient of W=.81 (p<.000l). There was thus a high consensus among judges in assigning each of the 70 items to its respective stage of the model. Appendix A lists the 70 items under the five stages of role development. 67 Inmate and Stage Meanings The face validity of each item was assessed at each stage of item construction and judging. Each judge was asked to comment on the language and expressions used in the statements. Apart from changes to some items made after the initial involvement of one judge, no further changes appeared necessary. If the five stages of role development are plausible, that is, meaningful for inmate-students, they will choose items from each stage to represent how they feel toward the university program. Whereas judges' agreements on item placements within the five stages of role development should be free of bias towards any particular stage, inmate-students' responses may be biased in favour of items that best describe their current feelings toward the university program. Consequently, any particular sample of inmates may validate certain stages more strongly than others. This is the pattern revealed in Figure 5. Figure 5 plots inter-judge consensus and the extent of inmate participants' agreement about which of the 70 items represents how they feel toward the university program. This figure provides two different bases (judges; inmate-students) for assessing congruence within the stages of role development. The vertical axis displays the 70 items (in abbreviated phrasing) in five stages of role development. The horizontal axis indicates percentage agreements in judges' placements of each item into a stage and in inmates' choices of each item as an indicator of "how I feel" about the university program. For example, judges displayed 100 percent consensus in placing item RECRUITMENT STAGE 1. Can't be any uor<« than other programs 2. Don't plan to work too hard in program 3. Maybe can -fraud it awhile 4. Not Interested in any subject 5. Don't care what is taught 6. Don't expect much out of program 7. Curious what program is like 3. Program Is some sort of scam 9. Program lIKe all the others 10. Don't care what 1 study 11. No way program change me 12. Only take courses I prefer 13. Only one subject interested In 14. Not get too involved in program IS. All Instructors the same 16. All courses the same DISORIENTATION STAGE 17. More work than I expected 13. Instructors expect a lot 19. Surprised how Involved Instructors are 20. Don't Know I f like program 21. Program different from I expected 22. Hard to be a student 23. Not sure I should continue 24. Feel too Isolated In academic area 25. Other students more serious than me 26. Haven't figured out program 27. Don't know what to make of program 23. Other students take program seriously 29. Don't know what to do as student 30. Being student Increase pressure SEPARATION STAGE 31. Harder to put up with bullshit 32. Lot of student vs. Inmate role conflict 33. Feel cut off from rest of inmates 34. Feel pressure not to get Involved 35. Less to do with inmates, better for me 36. Starting to handle being student 37. Like less contact with general inmates 38. Found one subject to spend all my time 3?. Fed up with prison games 40. Less Interested in non-school activities TRANSITION STAGE 41. Program is helping me control my life 42. Making changes In my life as student 43. Program has value for me 44. Accept demands of being student 45. Getting something useful from program 46. Different opportunities than before 47. Student status asset in prison 48. Getting something for self In program 49. Student experience helps with problems 50. Lead me to different things 51. Spend much time on student Interests 52. More involved with all aspects 53. Like more courses In subjects 34. Enjoy learning many subjects 55. Best to stay involved in program 56. Being student Is changing things 57. Confident of my ability to learn SOLIDARITY STAGE 58. Feel involved with program 59. Can rely on student community 60. See program as means to express self 61. Feel responsible to support program 62. Best thing to happen in prison 63. Good working relations with instructors 64. Equal to any challenge in program 65. Have something to add to program 66. Can get support from other students 67. More tolerant toward others 6B. UouId defend program to inmates 69. Feel supported by student community 70. Other students want same as me Percentage Agreement 0 10 20 30 40 50 60 70 80 90 100 (•) Judges' agreement - items to each stage (+) Inmates' agreement - Items to "how I feel now" Judges (»)=89% Inmates + • * + • + • *• • * • +• * • * + • * +• • + • Judges>84Z Inmates«15X • • •f * • • » + • • +• * + • + * • + * + • • * * Judges-58Z Inmates-281 + • • • • • » • f +• • * • Judges (»)-62Z Inmates (+)-79% • • •*• • *• • + • • • 4-+ • * + * • • • Judges-60X Inmates"75Z + « » +• » +• » +• • • * + * * +• 0 10 20 30 40 50 60 70 80 90 100 Figure 5. Inter-Judge And Inmate Agreements On Items 69 1 — "can't be any worse than other programs" -- into the Recruitment stage, whereas only about 12 percent of 33 inmates indicated that it described how they feel toward the university program. At the top of each stage are two additional percentages ("J" for judges; "I" for inmates). The judges (J) percentage represents their agreement in assigning all items to a stage. These percentages were calculated by multiplying the number of items in each stage by the number of judges to obtain a value representing the maximum possible agreement per stage. Then the actual number of judges' agreements on the items within a stage was divided by this maximum value to obtain a percentage agreement for each stage. Responses (in percentages) to the 70 items by 33 inmate-students were calculated using SPSS:X MULTRESPONSE. The "I" percentages represent inmate-students' "agreement" about the number of items from each stage that describe how they feel toward the university program. Judges' consensus on items for Recruitment (J=89 percent) and Disorientation (J=84 percent) were higher than for items in Separation (J=58 percent), Transition (J=62 percent) and Solidarity (J=60 percent). Thus, judges strongest validation of stage descriptions occurred with Recruitment and Disorientation. Overall, the judges' consensus in assigning 70 items to five stages of role development was calculated to be W=.83. One element in the judging process concerns the diverse prison experience of the judges. All had experience with students in prisons with security levels ranging from maximum to low-medium. 70 Their judgements were influenced by the different conditions under which the program operates in various prisons. This balanced perspective on the plausibility of states expressed in the items could not be expected of inmates in the program. These different perspectives on the items (overview - judges; subjective - inmates) probably increased the likelihood of obtaining the results reported in Figure 5 — lack of congruence between judges and inmates. About seven percent of inmate-students' responses to the 16 Recruitment items were for "how I feel" toward the university program. Inmate-students' support for items describing their present relationship to the university program increased to 15 percent for the Disorientation stage and 28 percent for the Separation stage. They assigned much higher percentages of items from Transition (1=79 percent) and Solidarity (1=75 percent) to "how I feel" than from the first three stages. Thus, inmate-students' most strongly suscribed to Transition and Solidarity items. These results suggest that this sample of inmate-students felt its relationship to the university program in terms best characterized by the Transition and Solidarity stages of role development. Another way of representing inmate responses depicted in Figure 5 is by calculating mean scores for all items within each stage. Inmate-students sorted the 70 item statements into four categories. When coded "how I feel now" = 3, "how I used to feel" = 2, "never felt like this" and "don't know" = 1, they constitute a three point ordinal or rank-ordered scale. 71 The scale is ordinal because the categories satisfy the ordinal transitivity postulate (Kerlinger, 1973). That is, statements like "a has more of a property than b; b has more of the property than c; therefore a has more of the property than c" pertain to the categories (Kerlinger, 1973, p. 435). In this case, "a" refers to "how I feel now", "b" to "how I used to feel", and "c" to both "never felt like this" and "don't know." The property being measured in each of 70 items is a student's present self-perceived feeling state toward the university program and prison. Feeling states reflect the cognitive and affective development of inmates, as well as the influence of the social ecology of prison and academic program with their interactive and mutually determining social dynamics. States change as students develop their roles. For example, an inmate decides an item (number 26) — "I haven't figured out this program" — is "how I feel now" about the university program. His choice means this item describes his current self-perceived state. Had his response been "how I used to feel", the item would not refer directly to his present state. However, in this case, the item is not without meaning for the student. His past feelings are integral to his lived experience and are a part of his present. Statements assigned to the past may not reflect his current state, but they do provide an orientation to changes he is experiencing. On the other hand, items assigned to "never felt like this" and "don't know" do not correspond at all to the lived (present or past) experience of students. They do not reflect a 72 student's current self-perceived state the way items labelled "how I feel now" or "how I used to feel" do. Therefore, the response scale satisfies the major criterion of ordinality in that it permits a plausible rationale for ranking responses. In calculating means it is recognized that ordinal scales do not necessarily provide equal intervals. On this issue Kerlinger (1973) argues that using interval statistics with ordinal measures is commonplace in educational and psychological research and frequently yields satisfactory results. For the purpose of calculating item and stage score means there does not appear to be a too-serious risk of distortion or error that would result in faulty interpretations of data. Table 2 displays stage score means and standard deviations for the sample of 33 inmate-students. Stage score means as well Table 2 OVERALL ITEM SCORE MEANS BY ROLE DEVELOPMENT STAGES Stage Mean S.D. No. of Items Recruitment 1. 35 .21 16 Disorientation 1. 52 .28 14 Separation 1. 70 .36 10 Transition 2. 64 .38 17 Solidarity 2. 54 .38 13 73 as standard deviations increase from the Recruitment stage (X=1.35, S.D.=.21) to the Transition stage (X=2.64, S.D.=.38). The stage score mean for the Solidarity stage (X=2.54) is slightly lower than its counterpart in the Transition stage. An overall stage score mean is calculated by averaging the 33 inmate-students' responses to the "n" items assigned to that stage. Scores will be high when many inmates designate most of the items in a stage to describe how they "feel now" toward the university program, because all such items are coded "3." A stage score mean will be lower when many inmates in the sample say most items stand for "how I used to feel" (coded 2), "never felt like this" or "don't know" (both coded 1). An interpretation of these stage score means is outlined in Figure 6. The histogram displays percentage responses to items within a stage by category of choice. The vertical axis displays, for each stage of role enactment, stage score means and standard deviations, and the four response categories used in the card sort. The horizontal axis represents percentages of responses by 33 inmate-students to items within each stage. Recruitment's stage score mean is 1.35, the lowest of all the five stages. Its low mean coupled with a standard deviation of .21 indicates that inmates concurred in picking only a few of the 16 Recruitment items as valid descriptors of their present feelings toward the university program. Instead, most items were assigned to "never felt this way" (65.3 percent) and "don't know" (6.8 percent). "High" item scores may be due to values added by a substantial number of items coded "2" ("how I used to Legend 'Now' - Feel this way now 'Past' - Used to feel this way 'Nv' - Never felt this way 'DK' - Don't know SOLIDARITY X=2.54 S.D.=.38 TRANSITION X=2.64 S.D.=.38 SEPARATION X=1.70 S.D.=.36 DISORIENTATION X=1.52 S.D.=.28 RECRUITMENT X=1.35 S.D.=.21 Now Past Nv DK Now Past Nv DK Now Past Nv DK Now Past Nv DK Now Past Nv DK 75.1 % 4.2 % 7.0 % 13, 8 % 79.3 % 5.0 % 8.2 % 7.5 % 27.6 % 14.8 % 10.9 % 46.7 % 15.4 % 21.6 % 8.2 % 54.8 % 6.8 % 21.0 % 65.3 % 6.8 % Figu: .re 6. Inmate Responses By Stage And Category 75 feel" - 21 percent) rather than from a number of items coded "3" ("how I feel now" - 6.8 percent). Therefore, when describing the meaning of high Recruitment scores, readers should keep in mind that a small number of items may typify Recruitment for this sample of inmate-students. Figure 6 shows almost seven percent of inmates' responses indicated that they were "now" in the Recruitment stage. Figure 5 showed which items for each of the stages were chosen most frequently by inmate-students to describe how they feel toward the university program. In general, inmates with high Recruitment stage scores reported feeling curious about the program (item 7), that they only wanted to take a course in a preferred subject (item 12), and thought all instructors were pretty much the same (item 15). Up to one third of the 33 students defined their feelings toward the university program in terms of one or more of these items. A smaller number of inmates with high Recruitment scores (7 to 12 percent of 33 inmates) indicated feeling that the university program couldn't be worse than other prison programs (item 1), they were very sure the program wouldn't change them (item 11), and didn't plan to get too involved with it (item 14). Inmates with low Recruitment scores were more likely than those with high scores to deny or attribute to the past a relationship to the university program based on these 16 items. Thus, Recruitment appears to play a minor role as reported by this sample of inmate-students. Those items associated with high Recruitment scores represent curiosity about the university 76 program tempered by an intention not to be "sucked in" or changed by it, and an uninformed view of curriculum and instructors. The Disorientation stage score mean (X=1.52, S.D.=.28) indicates that most inmate-students chose few of its 14 items to describe their present orientation toward the university program. Figure 6 shows the response rate to be 15.4 percent, well below the 54.8 percent for "never felt like that", and moderately lower than the 21.6 percent for "how I used to feel." Figure 5 shows about 45 percent of the 33 participants reported feeling surprise at how involved instructors were with students (item 19). Those with high Recruitment scores also indicated there was more work than they expected (item 17), that other students were more serious about the program than they (item 28), that the program was different from what they expected (item 21), and that being a student increased the pressure on them (item 30). As with low Recruitment scorers, those with low Disorientation stage scores either consigned these feelings about the program to the past, or more frequently, denied they ever had them. Though the percentage of Disorientation items indicating students' current feelings was more than twice that of Recruitment's, the overall importance of "disorientation" in defining the social ecology of these 33 students appears minor. Separation-Alienation's stage score mean (X=1.70) indicates that students chose a few more of these items than from the preceding two stages to describe their present feelings toward 77 the university program. An increase in the standard deviation from .28 in Disorientation to .36 in Separation indicates a greater degree of differentiation between the "high" and "low" scorers in Separation, and consequently more distinctiveness as a stage of role development compared to Recruitment and Disorientation. Figure 6 shows the response rate for "how I feel now" (27.6 percent) to be higher than that for "how I used to feel" (14.8 percent) and moderately lower than for "never felt like that" (46.7 percent). In general, inmates with high Separation scores chose items to indicate they were at the point of feeling they could "handle being a student" (item 36). Moreover, they were "fed up with prison games" (item 39), wanted even less contact with the general inmate population (items 35, 37) and found it harder to "put up with prison bullshit" (item 31). A smaller proportion of inmates with high Separation stage scores also felt they had found a subject in the curriculum they would like to devote all their time to (item 38), were less interested in activities outside of school (item 40), and felt a substantial conflict between their inmate and student roles (item 32). For the most part, inmates with low Separation scores denied feeling "separation-alienation" or "didn't know" about some of its component items. A few indicated feeling that a number of statements constituting this stage were true for them in the past. Moreover, it is probable that a number of inmate-students with low to average Separation scores did feel, in common with high scorers, that they were at the point where they 78 could "handle being a student" (item 36 - picked as "true" by 70 percent of inmate-students). However, they were less likely than high scorers on Separation to associate "getting a handle" on being a student with a felt need for distancing themselves from other non-student inmates and prison activities. Some low Separation stage scorers were also "fed up with prison games" (item 39 - "true" for 49 percent of inmate-students). Separation-Alienation appears to be a moderately distinct "reality" for inmate-students with high stage score means. The highest of the five stage scores occurs in the Transition stage (X=2.64). Many inmate-students chose a majority (79.3 percent response) of these 17 items to represent their current feeling state toward the university program. Correspondly, much lower response rates characterized Transition items as describing past feelings (5 percent) or non-pertaining statements ("never felt like this" - 8.2 percent; "don't know" -7.5 percent). The standard deviation of .38 coupled with a stage score mean of 2.64 indicates that inmate-students with high stage scores chose virtually all Transition items to describe their feelings toward the university program. In general, high scorers reported feeling the university program was helping them get control over and make changes in their lives (items 41, 42), had value (item 43), and was useful (item 45). In addition, they felt confident of their ability to learn (item 51), enjoyed learning about many different subjects (item 54), felt more accepting of demands made on them as students (item 44), and saw 79 their student status as an asset in the prison (item 47). Inmates with low Transition stage scores designated a number of these items to describe past feelings, and others as not describing their feelings at any time. The high stage scores suggest most inmates shared common elements of a "transition" orientation. However, Figure 5 suggests that respondents with low and average stage scores were less likely than high scoring respondents to report feeling that the program was helping them get control over and make changes in their lives (items 41, 42), that it had value for them (item 43), or that they had different opportunities than before (item 46). In addition, low scoring respondents on Transition were less likely than high scoring respondents to feel like spending as much time as possible on student interests (item 51), or to want to become involved with other aspects of the program. However, most respondents viewed their relationship with the university program in terms of the forward-looking, positive-valuing stance of the Transition stage items. The stage score for Solidarity (X=2.54, S.D.=.38) was marginally smaller than for Transition, but still quite high and the pattern of responses is similar. Many respondents selected most (75.1 percent response) of these 13 items to describe how they felt toward the university program. Again, rates of response to these items by other categories were low -- "how I used to feel" - 4.2 percent; "never felt like this" - 7 percent; "don't know" - 13.8 percent. A standard deviation of .38 indicates that respondents with 80 high scores selected virtually all of the Solidarity items to describe their relationship to the university program. This is because high stage scores approached 3.00, possible only when most Solidarity items were chosen to represent current feeling states. In general, these respondents reported feeling completely involved with the program (item 58), saw it as a means to express themselves (item 60), felt a responsibility to support it (item 61), that it was the best thing to happen for them in prison (item 62), and believed they could add something to it (item 64). In addition, they reported feeling more tolerant of other peoples' views (item 67), equal to any challenge posed by the program (item 64), and able to defend it to other inmates (item 68). They felt good about their working relationships with instructors (item 63) and about support from the student community (item 69). On the other hand, inmates with low Solidarity stage scores were less likely than high scoring respondents to report feeling that other students wanted the same things they did (item 70), that they could rely on the student community (item 59) or that the student community supported them (item 69). In general, respondents with low Solidarity stage scores did not report complete involvement with the program (item 58), did not see it as a means to express themselves (item 60), nor feel it was the best thing to happen to them in prison (item 62). However, the overall majority of respondents viewed their current relationship with the program in strong "solidarity" terms. Table 3 summarizes differences between high and low scores 81 within each stage of role development. A noticeable feature in Table 3 is the polarity crossover in stage meanings between Disorientation and Separation. Item meanings in Recruitment and Disorientation associated with high scores expressed oppositional, reactive states., In contrast, Separation, Transition and Solidarity meanings associated with high scores suggested emerging aspirations, commitment to learning and to the program. The "negative" or "positive" polarity of items is attributable to the item construction process which was deliberately anchored in the the character of different stages of role development proposed in the model. (Chapter 3 provides a full discussion of the model's stages of role development.) To recapitulate the information contained in Figure 6, the distribution of item responses within each stage showed that most students strongly suscribed to current states characterized by feelings associated with role development at the Transition and Solidarity stages. Figure 6 indicated weaker support among students for the three initial stages — Recruitment, Disorientation, and Separation. However, there was no reason to expect that all five stages would equally reflect current states for this or any sample of inmates. Different item sets defined various phenomenologically distinct stages of role development. Moreover, respondents acknowledged the saliency of some items within the initial three stages in describing previously held feelings toward the university program. As such these items had meaning for respondents and, though not representative 82 Table 3 SUMMARY AND CONTRASTS OF STAGE SCORE MEANINGS Stage High Scores Low Scores RECRUITMENT - curious about program - interested in 1 subject only - instructors all the same - program couldn't be worse than others - won't be changed by program - won't get too involved - not curious about program - interested in more than 1 subject - see differences in instructors - not against being changed by program - becoming involved not an issue DISORIENTATION - surprised at level of instruc tor involvement - more work than expected - other students believed more serious - program different than expected - being student increases pressure - accept instructors' involvement - amount of work anticipated - serious about student role - aware of program - pressure not dependent on student role SEPARATION - can handle being a student - fed up with prison games - less contact with other inmates wanted - harder to put up with prison bullshit - less Interested in non-school activities - substantial conlict between inmate and student roles - already managing student role or not yet confident about it -not bothered by prison games - not concerned about contact with non-student inmates - coping with prison - not exclusively interested in school - not feeling much conflict between inmate and student roles TRANSITION - program helping to get control over and change lives - program has value for them - program is useful to them - confident of ability to learn - enjoyed learning about different subjects - accepting of demands made on them as students - student status an asset in prison - program not viewed as resulting in control or change in lives - no particular value to program - program not necessarily useful - not concerned about or confident of ability to learn - not turned on by different subjects - not reconciled to demands made on them as students - student status unimportant or a liability in prison SOLIDARITY - felt completely involved with program - program means to express self - felt responsible to support pro gram - best thing to happen in prison - felt they had something to add to program - more tolerant of others' views - equal to any challenge in program - able to defend program to other inmates - felt good about working relation ships with instructors - confident of support from student community - not deeply involved in program - program instrumental rather than expressive - not responsible for supporting program - unaware of making an impact on program - unaware of any change in tolerance towards others - not prepared to defend program to other non-student inmates - not confident about ability to meet future program requirements - unsure about quality of relationships with instructors - not confident about intentions of fellow students toward self 83 of current feelings, enhanced the plausibility of the theoretical stages of role development nonetheless. A striking feature of Figure 6 is the considerable amount of "denial" shown by respondents toward items within the initial three stages. "Denial" response rate for Recruitment was 65.3 percent "never felt like this", for Disorientation 54.8 percent, and for Separation 46.7 percent. The extent to which this "denial" represented a threat to the validity of the initial three stages warrants consideration. Different assumptions yield different explanations. One assumption is that respondents really did not ever feel more "recruitment", "disorientation", or "separation" than was indicated by responses to "how I feel now" and "how I used to feel." In other words, the "never felt like this" response percentages were a valid reflection of states felt by these respondents within the limits of card sorts performed on 70 items. If, on the other hand, there is concern that inmates did feel more "recruitment", "disorientation" and "separation" than was indicated by responses to "how I feel now" and "how I used to feel," then the response rates for "never felt like this" constituted "denial" of the "truth." Denial can take the forms of deliberate misrepresentation or self-deception. Some items in the three initial stages expressed negative sentiments about the program, prison, other inmates, and respondents' own capacity to learn. Respondents could have deliberately misrepresented the extent to which these more negative aspects defined their relationships to the 84 university program, prison, or other inmates. They might have been loathe to give the program "bad press", fearing that correctional authorities might somehow use this result against the program or its students. There was no sure defence against deliberate manipulation of item responses. Any educational or social science research attempted in prison faces a difficult job in "winning" the cooperation of inmates who distrust the role and purposes of research. Even after inmates consent to participate they remain wary to safeguard their perceived best interests, if necessary at the expense of disclosing what they really think or feel. From their point of view, long after the researcher has gone they will still be struggling to maintain their wellbeing in an often treacherous social environment. Steps were taken to minimize inmates' perceptions of threat arising from the methods of obtaining information, the purposes of the study, and the likely uses of its findings. Each concern shaped the approach taken to this study. Assurances about the confidentiality of information used were given to inmates. There was no evidence that respondents deliberately misrepresented their feelings toward the university program. The participation rate in the study was very high relative to available subjects (see Chapter 5). A second "denial" proposition concerns respondents' self-deception about their "true" feelings. They could have unwittingly denied to themselves some of the more negative features of their relationships to the university program. In 85 an unpleasant prison environment, any "pain" associated with being an inmate-student might seem like "pleasure" relative to other prison roles. Yet, the prospect that inmate-students were out of touch with themselves is counter to the program's focus on cognitive and social development. Inmate-students purportedly become more insightful about themselves in the university program, and thus more discriminating about their true feelings. Finally, there are two explanations not based on the "denial" assumption. One possibility is that the measurement of the constructs in the Recruitment, Disorientation, and Separation stages was flawed, resulting in flawed findings. The development of measures proceeded through a number of careful steps which are outlined earlier in this chapter. Moreover, the judges' concurred on items used to measure Recruitment and Disorientation, and to a lesser degree, Separation. The other explanation takes into account the type of prison environment within which the university program was located. A low-medium security prison like William Head permits inmates more movement and imposes less stringent security measures on them than maximum security institutions. Typically, inmate "opposition" to authority and programs in medium is less pronounced than in maximum security. The relative absence of inmate opposition to the university program at William Head facilitates inmate-students' rapid socialization into their new roles. For these inmate-students, feelings of "recruitment", "disorientation", and "separation" quickly give way to 86 "transition" and "solidarity." Thus, inmate-students' responses represented the "truth" rather than a "denial" of their feelings toward the university program. Reliability Estimates Serviceable estimates of reliability for this study were obtained from Standardized Item Alphas. Table 4 provides scale characteristics. Estimates ranged from moderate (.57) for Table 4 RELIABILITIES FOR FIVE STAGES OF ROLE DEVELOPMENT Stages Mean Stage Score S.D. Observed Range Possible Range No. of Items Reliability Standardized Item Alphas Recruitment 1.35 .21 1.0--1.8 1--3 16 .75 Disorientation 1.52 .28 1.0--2.1 1--3 14 .65 Separation 1.70 .36 1.2--2.4 1--3 10 .57 Transition 2.64 .38 1.6--3.0 1--3 17 .85 Solidarity 2.54 .38 1.5-•3.0 1--3 13 .75 n=33 Separation to high (.85) for Transition. Transition (.85), Solidarity (.75) and Recruitment (.75) measures (items) are sufficiently homogeneous (internally consistent) in accounting for "true" variance in subjects' responses to allow reasonably confident interpretation of subsequent analyses. Greater caution should be exercised in 87 interpreting results for Disorientation (.65) and particularly Separation (.57) due to increased levels of error variance indicated by moderate alpha coefficients. Questions Derived From the Stage Model The creation of an instrument to test the model of role development required the framing of a number of questions. These, along with questions already implicitly put and answered, conclude this chapter. The initial question asked: 1. What are the stages? The stages were first theoretically defined, and subsequently empirically anchored (see Chapter 3). The justification for their existence and sequencing comes from the theoretical framework within which they were derived. Judges endorsed the adequacy of 70 statements used to describe the five stages of role development. They did not test the "truth" or "realness" of the stages except insofar as 70 items corresponded to them. With the five stages of the model operationalized, the question became: 2. Were the five stages of role development, as conceptualized and measured, a fair approximation of inmates' self-perceived feeling states? Discussions concerning the operational definitions of these five stages for inmates occurred earlier in this chapter. The 88 results of additional statistical tests of this question are given in Chapter 6. The stage model of role development suggested that inmates relate to the university program in terms of one dominant stage, rather than through multiple stages. Thus, the question about the representativeness of stages for inmates became: 3. Did inmate-students "reside" within only one stage of role development at a time? This posed the issue of inmate-students' distribution through the five stages based on their responses to the 70 statements in the card sort. Moreover, this question raised a related one: 4. Were the five stages of role development related to other variables? The next chapter describes the method employed in the study and the statistical analyses performed. Discussion of test results regarding the research questions occurs in Chapter 6. 89 CHAPTER 5 METHODOLOGY Choice of Research Setting Several criteria resulted in the selection of a single prison as the setting for the study. First, in recognition of the importance attached by the model to the existence of an academic community within prison, a desirable setting is one in which the greatest opportunity exists for achieving full participation in the study by inmates in that community. Participation by a majority of members affords the best opportunity to understand the social.ecology within which roles are enacted. Secondly, a preferred setting is one in which inmate members of the academic community occupy the full spectrum of academic statuses, from "new" to "senior" students. Thirdly, the setting ought to be located within a prison with a general prison population characterized by a "typical" range and variety of offences. The SFU university program operates in four federal prisons in the lower mainland of British Columbia and on Vancouver Island. The security classifications of the four prisons varies. Canadian federal prisons and other correctional facilities are categorized according to security levels that range from "1" and "2" (minimum security) up to "6" (maximum security) and "7" (super-maximum). Two of the four prisons in which the university program operates are classified at level 90 "3" or low medium security. One of these is heavily populated by inmates convicted of sex-related offences and as such, constitutes something of a special case. Two other prisons are classified at levels "5" and "6" respectively. These latter two, along with the remaining level "3" prison, contain inmates convicted of a broad cross-section of offences within limits set by the CSC for the institutional placement of inmates according to benchmark criteria (e.g. escape risk, violence risk, age) and severity of offence (Appendix B). Thus, high rather than low security prisons contain a large proportion of inmates convicted of major offences such as murder and hostage-taking. One of the three "general" prisons (William Head) in which the university program operated met the "preferred setting" criteria considerably better than the other two (Kent, Matsqui). William Head Institution is a level "3" or low medium security prison located on Vancouver Island. Discussions with university personnel involved with the SFU prison program indicated that inmate-students at William Head would be more likely to participate in the study, and in greater numbers, than inmate-students in the other two higher security prisons. Moreover, the university program at William Head had a larger and more comprehensive range of "new" to "senior" students than either of the other two "general" prisons. University personnel involved with the program at William Head discussed the feasibility of doing the study there and indicated their willingness to facilitate it. Discussions with the assistant wardens for Education and Training and Offender 91 Programs at William Head identified the appropriate steps to obtain approval for the study by the Correctional Service of Canada. Approval for the study followed the submission of a formal research application to the Regional Chief Planning & Analysis (Pacific) of the CSC. In his absence, the Regional Manager, Planning & Administration (Pacific) together with the Research Committee of the CSC (Pacific) reviewed and approved the research proposal. Subject Selection Procedures About 40 inmates were enrolled in university courses at William Head at the time of the study. As many of these inmate-students as possible participated, subject to their own consent. The university co-ordinator at William Head offered to "spread the word" about the project and helped get inmates to listen to an explanation concerning it. The card sorting exercise performed by inmate-students occurred in the first week of October 1984 over three consecutive days. The academic centre at William Head is located in a self-contained building within the prison compound. All meetings with inmates took place within the centre. The university coordinator created a schedule for inmate-students willing to listen to a description of the study. She introduced one or two at a time, giving them this researcher's name and his affiliation with the University of British Columbia, and the fact the study constituted a doctoral dissertation. She mentioned that the study concerned inmates' views of the 92 university program in prison. At this point, the university co-ordinator withdrew. Inmates read a printed form (see Appendix C) which explained the project, outlined the Q technique procedures, affirmed their right to decline or withdraw from the exercise at any time, and assured them of confidentiality regarding their identities. Some inmates asked questions about the purposes to which the research would be put. Quite a few volunteered "testimonial" types of comments regarding the benefits or value of the university program. Inmates generally expressed viewpoints about the program that involved comparisons with other kinds of prison experiences or programs. Of the numerous comments made, only two or three were negative about the program in any way. Thirty-five inmates agreed to do the Q-sort. Three who had been asked by the co-ordinator of the university program to listen to the presentation on the study declined to do so. Two inmates did not turn in their Q-sorts, resulting in 33 completed Q-sorts. Administration of Q-Sorts Each participant received an envelope containing a stack of plain file cards on which the 70 statements were printed. Cards were sequenced in no particular order. Each envelope contained four smaller envelopes, each labelled with a different heading. The four headings were: - How I feel now 93 - How I used to feel, but not now - Never felt like this - Don't know Participants were told the items represented a wide range of viewpoints about the university program. They were instructed to read and sort them into the headings which best represented their feelings about the statements. The sorting procedure was unforced. That is, inmates did not sort cards into a required normal or quasi-normal distribution. They were told to put as many cards as they wanted under each heading, that no heading was "right" or any more preferable than the others. Once the cards were sorted, participants were told to reconsider their selections and to reassign statements to other headings if they wished. Most participants completed the sort within 10 to 15 minutes. A few took longer, up to, about 30 minutes. Upon completion, the stacks of sorted cards were inserted into labelled envelopes which then went into the larger envelope. A number code corresponding to the name of each participant marked the corner of each of the larger envelopes. Participants were thanked for their involvement. 94 Carceral and Demographic Information Information about carceral and demographic characteristics of participants came from the records of the Transfer Board at the Regional Headquarters (Pacific) of the Correctional Service of Canada. Participants' year and term of enrolment in the university program came from university registration forms at William Head. The initial definition of year of enrolment (year 1 to year 4) posed a problem. Inmate-students vary considerably in the number of courses they take per term. Some inmates enrolled for one or two courses per term while others enrolled in four or five when courses were available. Thus, there was wide variation in the number of terms inmates spent in any one year before amassing the required number of credits to advance to the next higher "university year." This study assumed that time spent in the university program provides the opportunity to develop cognitively and socially, to work out role conflicts, and to achieve new stages of role development. New students, at the beginning of their first term in the university program, were less likely than other first year students in second or third terms to have worked out the nuances of the student role and associated expectations of others. Numerous conversations with inmates in this and another study (Boshier & Clarke, 1983) revealed that many inmates new to the program require at least two terms to find out whether or not they are going or want to "make it" in the program. 95 Most first year students were enrolled in first or second terms, with several in third terms. Accurate information about the number of university terms taken by inmate-students currently enrolled in years two, three and four was not readily available, in part because most of them had begun the program at other prisons under a two semester system in contrast to this program's trimester system. Therefore, the minimum number of terms an inmate-student required before advancing to years two, three or four was based on the number of credits required and the maximum possible course load per term. Thus, "year of enrolment" was recoded as "term of enrolment" as follows: 1,2,3,5,7. First year students were coded "1","2" or "3" according to the actual term occupied. Second year students were coded "3", for a minimum of three terms, third year students were coded "5" (minimum terms taken), and fourth year students were coded "7" (minimum terms taken). Carceral and demographic information concerned age, date of most recent sentence, length of sentence, mandatory supervision date (MSD), warrant expiry date (WED), severity of most serious present offence, previous penitentiary (federal) term served, educational level and occupational status at the start of the present sentence. Some of the information came from the Transfer Board's card index, the remaining information from Pententiary Placement Forms in the Board's inmate files. After receiving permission to obtain the information, two partial days at the Regional Headquarters of the CSC were required to gather the data from 96 the cards and forms. The coding schedule used to gather carceral and demographic data is shown as Appendix D. Previous federal penitentiary term served was coded dichotomously, yes or no. "Yes" meant the Penitentiary Placement Form indicated that in addition to the present sentence being served, the inmate in question had also served at least one prior term in a federal prison. Educational level at the start of present sentence was coded according to the actual grade level indicated in the Pen Placement Form. Years 4-12 correspond to public and secondary schooling, 13-15 to partial university including post-secondary diplomas, 16 to university graduation. In some cases the extent of an inmate's post-secondary education had to be inferred from less than precise indications in the records. Concerning previous occupational status, Penitentiary Placement Forms were sometimes explicit, at other times not. Lack of precision in the records was common where individuals had no clearly defined job history or training. The categories used to code occupational status were: - unskilled=1 - skilled=2 - clerical/sales=3 - managerial/administrative=4 - professional/technical=5 Severity of "most serious present offence" was coded according to the Offence Severity Scale (included as Appendix B) employed by the CSC to determine institutional placement of 97 inmates. Offence categories are: Minor=1; Moderate=2; Serious=3; Major=4. Examples of "major" offences are murder, kidnapping, espionage, and violent terrorist activities. "Serious" offences include robbery with violence, violent sexual assaults, arson, and trafficking in dangerous drugs. "Moderate" offences include forgery, break and enter, criminal negligence causing death or bodily harm, theft over 200 dollars, and trafficking in soft drugs. "Minor" offences include common assault, public mischief, criminal negligence not resulting in bodily harm, and possession of a restricted or prohibited weapon. Present offences of participants observed on the Penitentiary Placement Forms were converted to the appropriate categories according to the list of offences contained under "each heading on the Offence Severity Scale. Because several participants had been given life sentences, it was not possible to code Mandatory Supervision or Warrant Expiry dates for them. "Lifers" have neither legally. Setting an approximate length of sentence raised additional difficulties. Technically, the sentence is life. However, many lifers do get out on parole eventually. The issue became one of estimating a minimum period of incarceration before the possibility of parole arose. Records for two of the four lifers in the study indicated they had to serve a minimum of 10 years before being eligible for parole. No such indication was given for the other two lifers. Therefore, an "informed" guess provided a possible parole release date, taking into account the time already served on the sentence, the severity of the offence 98 (eg. second-degree murder versus first-degree murder), and the current conventions governing sentence length for major crimes (eg. 10, 15, or 25 year minimums for major offences). Data Analysis Data from the 70 Q-sort items were coded into four categories as follows: 1 = How I feel now 2 = How I used to feel, but not now 3 = Never felt like that 4 = Don't know The identification numbers assigned to participants upon completion of the Q-sort exercise linked the Q-sort information to their carceral and demographic information. Data records were checked against the original data collection for errors. None were found. Descriptive and inferential methods of analysis were employed using the SPSS:X package of statistical programs. Tests of questions arising from the model (see Chapter 4) and initial interpretations of those results are the topics of Chapter 6. 99 CHAPTER 6 RESULTS This chapter presents findings in five sections. The first describes the participants, the second describes stage score data, the third describes participants' responses across stages, the fourth reports relationships between stage scores and background variables, and the fifth section reports tests of significance on the associations between duration of program enrolment and history of previous incarceration with carceral and demographic covariates on stage of the model. Characteristics of Participants Thirty-five inmates of William Head Institution participated in this study. Two inmates did not complete or return their Q-sorts, resulting in 33 usable cases. As is the case for all but one federal prison, William Head has only male inmates. All respondents were enrolled as students in the university programs administered by Simon Fraser University. Table 5 presents the year of birth (age), year in which the sentence was handed down, length of sentence, warrant expiry date (year), and mandatory supervision date (year) for respondents. Warrant expiry dates and mandatory supervision dates were not given for the five "lifers" in the study, for whom no release dates are are mandated. Table 5 ITEM CHARACTERISTICS FOR FIVE CARCERAL VARIABLES Variables n Mean S.D. Range Age (years) 32 34.3 10.16 22 - 65 Year of sentencing 33 1982 .1 2.53 1974 - 1984 Length of sentence (days) 33 2513 1408 500 - 5475 Warrant expiry date (year) 29 1988 .6 2.82 1985 - 1995 Mandatory supervision date 29 1986 .9 2.54 1984 - 1995 (year) The participants' average age was 34.3 years (year of birth X=1948.66, S.D.=10.16 years) with ages ranging from 22 to 65. The average respondent was sentenced to prison in the early eighties (X=1982.18, S.D.=2.53 years) to a sentence of about 6.8 years (X=2513 days, S.D.=1408). Sentences ranged from 500 to an estimated 5475 days. Barring changes to the regulations governing release on mandatory supervision, and in the absence of earlier parole, the average "non-lifer" respondent will be released late in 1986 (year X=1986.97, S.D.=2.54 years) after serving two thirds of his sentence. The warrant binding him to a prison term, should mandatory supervision or parole be revoked, expires late in 1988 (year X=1988.69, S.D.=2.82 years). Table 6 shows the distribution of respondents by category of offence and previous federal prison term served. Most respondents (51.5 percent) had been convicted of serious 101 Table 6 SEVERITY OF OFFENCES AND PREVIOUS FEDERAL TERM SERVED Variable n Percent Severity of worst present offence Moderate 12 36.4 Serious 17 51.5 Major 4 12.1 Minor 0 0 Previous term served in a federal penitentiary No 20 60.6 Yes • 13 39.4 offences, including armed robbery and drug trafficking, or possession for the purpose of trafficking. The remaining respondents had been convicted of moderate offences (36.4 percent) such as fraud and break and enter or major offences (12.1 percent) such as murder. None of the respondents were in prison because of a conviction for a minor offence. Thirteen respondents (39.4 percent) had served at least one previous term in a federal prison. Table 7 reports the distribution of three characteristics — occupational status; educational level at start of sentence; and university term. Ten respondents (30.3 percent) were classified as unskilled. The remaining occupational groups represented were skilled (24.2 percent), clerical and sales (15.2 percent), professional and technical (21.2 percent), and 1 02 Table 7 OCCUPATIONAL AND EDUCATIONAL CHARACTERISTICS Variables n Percent Cumulative X Occupational status Unskilled 10 30.3 30.3 Skilled 8 24.2 54.5 Clerical & sales 5 15.2 69.7 Managerial & administrative 3 9.1 78.8 Professional & technical 7 21.2 100.0 Years of education at start of sentence 9 1 3.0 3.0 10 4 12.1 15.2 12 14 42.4 57.6 13 6 18.2 75.8 14 5 15.2 90.9 15 2 6.1 97.0 18 1 3.0 100.0 University term first 12 36.4 36.4 second 5 15.2 51.6 third 7 21.2 72.8 fifth 4 12.1 84.9 seventh 5 15.2 100.0 managerial and administrative (9.1 percent). Educational levels at the start of current sentences ranged from 9 to 18 years of schooling. The largest group (42.4 percent) had completed grade 12. Another cluster (42.5 percent) had credit for at least one year of post-secondary education at the start of their sentences. In terms of the prison education program itself, the largest cluster of respondents (36.4 percent) were enrolled in the first term of their university program. Inmates in advanced 103 terms (those in fifth or seventh terms) represented under a third (27.2 percent) of all respondents. The next section reports inter-stage relationships and measures of internal consistency for each stage. Additional support for five distinct stages would result from stages that are more internally cohesive than they are externally related to other stages. Stage Intercorrelations and Internal Consistency This section reports the extent to which the five stages of role enactment, as they were conceptualized and measured (see Chapter 4), approximated inmate-student self-perceived feeling states toward the university program and prison. To test this question, correlations among the stages and measures of internal consistency were compared. Coefficient alpha is a statistic that estimates how tightly items within each stage cohere. Pearson's r shows the strength of association between any two stages. Some degree of intercorrelation was expected because the stages are parts of one conceptual model and share common units of interaction and fundamental concepts. Table 8 displays stage intercorrelation values with coefficient alphas in the diagonal. Alpha values are higher than their counterpart Pearson r's for all stages which means that items comprising each stage had more in common with each other as a set than with sets of items comprising other stages. Of the five hypothetical stages, Transition and Solidarity share the most in common (r=.71, p<.0l) but have 104 Table 8 STAGE SCORE INTERCORRELATIONS AND RELIABILITIES Recruitment Disorientation Separation Transition Solidarity Stages r r r r r Recruitment .73* Disorientation .62 .64* Separation .21 .21 .59* Transition .04 .04 .46 .85* Solidarity -.02 .15 .47 .71 .76* Correlations above .45 are significant of the .01 level for 33 cases. * The entries in the diagonal are coefficient alpha estimates. moderately (.85) and marginally (.76) higher alpha values respectively for internal consistency. Recruitment and Disorientation overlap moderately (r=.62, p<.0l) but also have moderately (.73) and marginally (.64) higher alpha values respectively for internal consistency. Recruitment and Disorientation are both unrelated to Transition or Solidarity. Separation-Alienation, the middle stage, is significantly related to both Transition (r=.46, p<.0l) and Solidarity (r=.47, P<.01) but shows a moderately higher alpha value (.59) than either correlation. So far, these findings suggest that the model, even though it has five putative stages, really has three "themes" roughly defining the "loosening", "restructuring" and "tightening-up" of one's role structure reminiscent of Kelly's (1955) theory of personal constructs. What Recruitment and Disorientation (the "loosening" stages) have in common with each other but not with other stages is an "oppositional" theme or perspective on role interactions within the university program and prison as a 105 whole. Cynicism,. skepticism, confusion and uncertainty are some of the principal dynamics of these stages. Transition and Solidarity (the "tightening-up") exemplify a forward-looking, positive commitment to the university program and the value of education and learning for changing personal opportunities. They have this in common with each other but not with the other stages, except Separation. Certainty, self-confidence and accommodation of inmate-student tensions characterize these stages. Separation-Alienation (the "restructuring") has some common elements with the "tightening-up" stages, principally disengagement (alienation) from the prescriptions and proscriptions of the inmate code. The model suggests that occupants of Separation-Alienation are already moving away from the conventional inmate social system toward the commitment of Transition. What distinguishes Separation from Transition and Solidarity is the growing urgency of cognitive strain and inmate-student role conflict. The resolution of that conflict, as pointed out in the conceptual framework of this study, may lead to a transition stage. The significance of the Separation, Transition, and Solidarity overlap is the implication that inmates with high Separation scores were resolving the conflict of the "restructuring" stage as well as "tightening-up" their role structures in positive terms. From the standpoint of the university, prison authorities, and respondents themselves, this would be a preferred outcome. 1 06 To reiterate the findings of this section, coefficient alpha values indicate that, notwithstanding clear stage intercorrelations, subjects' responses to items provide moderate support for five phenomenologically distinct stages of role development. The moderate stage intercorrelations (Recruitment and Disorientation; Transition and Solidarity) suggest that respondents perceived commonalities between items comprising these stages, making more difficult the task of stage demarcation. However, these stage intercorrelations suggest a "loosening", "restructuring" and "tightening-up" of inmate-students' role structures consistent with the model of five stages of role development. Although support for five psychometrically distinct stages is lessened by the existence of shared variance, the degree to which items cohered within stages based on subjects' and judges' responses (see Chapter 4) justifies the use of five stage scores in further analysis. The next section deals with the extent to which respondents identified with one stage more than another. Principal Stage Orientations The model suggests that an inmate-student, at any given point in time, identifies principally with one of the five stages — Recruitment, Disorientation, Separation, Transition, or Solidarity. The previous section and Chapter 4 showed that all five stages have meaning for these subjects, as determined by their responses to 70 items. This section describes the 1 07 extent to which participants in this study identified with one particular stage. Figure 7 shows how 33 inmate-students sorted 70 items into four response categories. Differences in rates of response to these categories were discussed in Chapter 4. Considered by stage, subjects' designations of items as describing "how I feel now" about the university program increase from Recruitment to Transition. Only 6.8 percent of subjects' responses to Recruitment items described their "present feelings." The proportion of responses in this category more than doubled to 15.4 percent for Disorientation stage items. The proportion increased to 27.6 percent for Separation. It is unlikely that any of these initial three stages constitutes a dominant or principal stage orientation for these particular inmate-students. Inmate-students identified their current feelings toward the university program with a substantially higher proportion of items from Transition (79.3 percent) and Solidarity (75.1 percent) compared to the initial three stages. It is likely that one of these stages is dominant for most of these inmate-students and, with an intercorrelation of .71 between them, the other next most dominant. However, because the proportions in Figure 7 are calculated for all 33 respondents, they do not conclusively settle the issue of dominant stage orientation. Table 9 gives the proportion of items in each stage that individual respondents designated as describing their "present feelings" toward the university program. 108 S w 04 90 80 70 60 50 40 30 20 10 Legend 'NOW' - Feel this way now 'PAST' - Used to feel this way 'NEVER' - Never felt this way 'DON'T KNOW' - Don't know NOW 90 80 70 60 50 40 30 20 10 *4 Figure 7. Inmate Item Responses By Category And Stage In every case but one, the dominant stage is either Transition or Solidarity. Subject 33 designated 100 percent of items from both Transition and Solidarity to indicate his "present feelings." A Transition stage orientation was dominant for 21 and Solidarity for 11 respondents. Percents for dominant Transition stage orientation ranged from 29 to 100 percent; for Solidarity, 46 to 100 percent. In only two cases did the next most dominant stage occur in the initial three stages (subjects 4 and 20). The findings in this section show that, based on the proportion of responses describing "present feelings" toward the university program, 32 of 33 inmates had one dominant stage orientation. However, that numerical fact must be reconsidered in view of the relative similarity in size of most Transition Table 9 PERCENTAGES OF ITEMS DESCRIBING PRESENT FEELINGS Stage Respondent Recruitment Disorientation Separation Transition Solidarity (16 items) (14 items) (10 items) (17 items) (13 items) 33 0 7 60 100 100 32 0 7 40 100 92 31 13 7 20 88 85 30 6 7 10 100 77 29 6 29 20 47 92 28 6 21 20 88 77 27 0 7 30 82 92 26 0 0 40 88 85 25 0 0 30 94 77 24 19 21 50 82 62 23 0 14 30 100 85 22 6 21 20 76 92 21 13 21 20 88 62 20 25 43 40 41 54 19 25 29 10 65 54 18 0 0 30 71 77 17 6 29 50 100 92 16 6 7 70 100 85 15 0 14 30 100 85 14 6 36 20 53 62 13 6 7 10 53 62 12 0 14 60 94 100 11 19 14 40 100 92 10 13 29 20 82 92 9 0 21 10 100 92 8 6 7 40 76 85 7 0 7 20 88 77 6 6 21 10 76 69 5 0 7 10 88 62 4 13 36 20 71_ 31 3 13 14 10 41 .46 2 0 0 0 53 46 1 13 0 20 29 23 Note: Dominant stage is underlined 1 10 and Solidarity percentages (Table 9) compared to those comprising the initial three stages. It appears that all respondents were "tightening-up" their role structures and that for most, the demarcation between Transition and Solidarity feelings, as represented in the 30 items for those two stages, was negligible. Furthermore, the range between the highest and lowest percentages for a dominant Transition or Solidarity orientation suggests that inmates interpret their commitments to the program differently, even when viewed from within a common role orientation. Notwithstanding individual differences in stage of role orientation and item response patterns, 29 of the 30 items most frequently assigned to describe "present feelings" referred to Transition or Solidarity. Table 10 lists the top 10 items for "how I feel now", "how I used to feel", and "never felt like this." The first three items referring to present feelings (1. like more courses in different subjects; 2. confident of my ability to learn; 3. enjoy learning) suggest that above all, these inmate-students valued learning for its own sake and had confidence in themselves as learners. Item four refers to feeling a responsibility to support the program. Items four (feel a responsibility), five (good working relationships), six (more tolerant toward others), and seven (would defend this program) place the respondent in a position of being externally evaluated. They appear to indicate a willingness to be called on to demonstrate support for the program, to practice tolerance of others' views, to "go on record" with other, possibly 111 Table 10 ITEMS DESCRIBING FEELINGS: PRESENT, PAST OR NEVER Response Category Items Present Feelings: "How I Feel Now" 1. 2. I'd like more courses in different subjects. I'm confident of my ability to learn. 3. I enjoy learning about many different subjects. 4. I feel a responsibility to support the program. 5. I have good working relationships with instructors. 6. I feel more tolerant toward other peoples' views. 7. I would defend this program to other inmates. 8. I feel like I'm starting to get something useful out of this program. 9. I'm looking forward to getting something for myself out of this program. 10. I feel like my student experiences will help me deal with problems more effectively. Past Feelings: "How I Used to Feel" 1. 2. I'm curious to see what the program is like. There is more work than I expected. 3. I only want to take courses in the subject I prefer. 4. Other students take this program more seriously than I do. 5. It's getting harder to put up with prison bullshit. 6. I don't know what to make of this program. 7. I find it hard to be a student. 8. I'm surprised at how Involved the instructors are with students. 9. X think all courses are pretty much the same. 10. Being a student increases the pressure on me. Never Felt: "Never Felt Like This" 1. As a student, I feel too cut off from the rest of the inmate population. 2. I feel like I'm too isolated here in the academic area. 3. This program is some sort of scam. 4. Maybe I can fraud it for awhile in this program. 5. I think all courses are pretty much the same. 6. I don't know if I like this program. 7. I feel pressure not to get too involved with student roles. 8. I don't care what is taught in this program. 9. This program is probably like all the others. 10. I don't care what I have to study. 1 12 unsympathetic inmates to defend the program, and to risk having instructors deny the quality of their relationships. These first seven items suggest expressive rather than instrumental concerns and entail the possibility of going against the grain of prevailing inmate norms, the situation of greatest risk for personal safety. Past feelings about the program ranged from curiosity and surprise to acknowledgements that increased pressure was associated with being a student. Inmates in this study for the most part denied feeling cut off from the general population, categorizing the program as a scam, or having an uncritical attitude toward their student roles. This section reported that respondents in this study, though ostensibly identified with a dominant, frequently endorsed a second related stage orientation to describe their "present feelings" toward the university program. Moreover, stages were found to be more internally consistent than intercorrelated with each other. The next step determined whether background and carceral variables were related to stage scores. The following section reports relationships between stage scores and respondents' background characteristics. Relationships Between Stages and Respondent Background So far, inmate responses have been used to calculate stage scores and relationships among stages. The last research question concerned the extent to which stage scores are related to respondents' carceral and demographic characteristics. The 1 1 3 model suggested that certain variables, such as duration of time spent in the university program should relate significantly to respondents' stage scores and result in systematic variance in those scores. Testing the significance of background and carceral variables for stage scores involved correlational and analysis of variance procedures. Table 11 displays Pearson correlations among the five stages with ten background variables. Recruitment correlated Table 11 STAGE, CARCERAL AND DEMOGRAPHIC CORRELATIONS Recruitment Disorientstion Separation Transition Solidarity Variables r r r r r Year of birth .12 .25 .06 .01 .01 Year of sentence -.13 -.14 -.03 -.06 -.35* Length of sentence .10 -.05 -.16 .06 .24 Warrant expiry date .01 -.11 -.16 .07 .11 Mandatory supervision date -.12 -.18 -.28* .09 -.03 Occupational status -.17 -.12 -.22 -.17 -.12 Educational level at start -.14 -.10 -.27 -.06 -.18 of sentence Previous federal tern served .58*** .39** .35* .29* -.02 University term .35* .15 .10 .25 .36* Severity of offence -.05 -.08 -.04 -.10 .19 * p<.05 ** p <.01 *** p<.001 with two background variables: "previous federal term (r=.58, p<.00l) and "university term" (r=.35, p<.02). served" Inmates 1 1 4 with high Recruitment scores were more likely to have served a previous federal prison sentence and to have taken a few more terms of university than were inmates with low Recruitment scores. These results suggested that Recruitment stage orientation, characterized by feelings of cynicism, skepticism and negativity toward the university and prison programs generally, most strongly corresponds to the experience of repeated incarcerations in the federal prison system. This finding is consistent with the view that repeated imprisonment is likely to increase the socialization of prisoners into inmate subcultures that express opposition to authority and to prison programs. The correlation between Recruitment and university term was contrary to what the model implies, that inmates high on Recruitment would more likely be in their first university term than would inmates low on Recruitment. Perhaps fewer inmates in their first university term had served a previous federal sentence than inmates with several terms of university behind them. For these latter inmates, a Recruitment orientation was associated with a history of previous federal incarceration and with a longer duration of involvement in the university program. Disorientation correlated moderately with "previous federal term served" (r=.39, p<.0l). Inmates high on Disorientation were more likely to have served a previous federal prison sentence than were inmates with low Disorientation scores. Disorientation is characterized by feelings of uncertainty about the demands of the university program. This finding suggested 115 that feelings of Disorientation are most strongly related to the experience of repeated incarceration in federal prisons. Separation-Alienation also correlated moderately with "previous federal term served" (r=.35, p<.02), and modestly with "mandatory supervision date" (r=-.28, p<.05). Inmates high on Separation (ie. those who selected a greater proportion of these items to indicate "current feelings" than those low on Separation) were more likely to have served a previous federal sentence and to be closer to release on mandatory supervision than were inmates with low Separation scores. Separation-Alienation is characterized by feelings of conflict over the demands of inmate and student roles, but with an emerging preference for student status. These findings suggested that feelings of Separation, say in terms of role conflict, were most strongly associated with a history of previous federal incarceration. Moreover, separation behaviours (distancing from inmate norms) associated with imminent release have been observed in prison studies. The literature on the "importation model" (see Chapter 2) and adaptive patterns of inmates behaviour in various segments of their sentences, supports this finding. As inmates near release, they tend to disengage from inmate society and from behaviours associated with the inmate code in preparation for life outside. Transition-Reframing correlated modestly with "previous federal term served" (r=.29, p<.04). Inmates high on Transition were more likely to have served a previous federal sentence than were those with low Transition scores. Transition is 116 characterized by feelings of consolidation of one's student status and an appreciation for the benefits associated with being involved in the program. This finding suggested that Transition feelings were most strongly associated with a history of previous federal incarceration. Perhaps inmates in this study who had a prior federal sentence spent a longer time in the program than inmates in their first federal sentence. A marginal correlation between Transition and "university term" (r=.25, p<.08) suggested that inmates high on Transition more often had a few terms in the university program behind them than inmates with low Transition scores. Solidarity correlated with "university term" (r=.36, p<.02) and "year of sentence" (r=.35, p<.02). Inmates high on Solidarity often had relatively more terms of university program behind them and were more likely to have been sentenced longer ago than were inmates with low Solidarity scores. Solidarity is characterized by feelings of support for and leadership in the program. These findings suggested that feelings of Solidarity are most strongly associated with duration of time spent in the university program and with the length of time already spent in prison on the current sentence. Inmates in higher university terms necessarily required more time to achieve them, unless they entered prison with university credits. A large majority of these inmates did not. Perhaps this fact accounted for the correlation with how long they have already been in prison. The model suggested that inmates' achievement of Solidarity is associated with the duration of time they spend in the program. 1 1 7 This finding offered support for the part that duration of time in the program is presumed to play in the development of student roles. In summary, several significant relationships between stage and background variables stood out. There appeared to be two stages associated with being a more senior student. The first stage, Recruitment, was "oppositional" to the program, the second, Solidarity, was "solid" within the program. Senior students with high Recruitment scores may also have served a previous federal sentence. A generally accepted view is that repeated experience of imprisonment "hardens" the inmate and socializes him more deeply into an inmate code which is oppositional to attempts by authorities to change (rehabilitate) him. Conversely, inmates with low Recruitment scores were more often at the beginning of their programs and without a prior federal sentence than those with high Recruitment scores. Senior students with high Solidarity scores indicated feelings of support for the program. This result was consistent with the stage model of role enactments and at odds with the relationship between Recruitment and student "seniority." However, based on the proportion of items used to describe present feelings, high Solidarity scores were considerably larger numerically than high Recruitment scores, and therefore indicative of a more dominant state for inmates. Therefore, senior students were not "split" into those with dominant Recruitment or Solidarity scores. The next section reports how duration of program 1 18 involvement and previous history of federal incarceration accounted for variations in inmates' stage scores. Included are the effects of background variables on response patterns. Effects of Background Variables on Stage Scores This section reports the extent to which duration of program involvement and previous history of federal incarceration predicted stage scores. The model proposed that duration of program involvement (measured as university term) is a predictor of stage orientation. The general question concerned the extent to which this and other carceral (such as previous federal incarceration) and demographic variables, when treated as independent, predict stage scores. To test these effects, several analyses of variance were performed using SPSS:X ANOVA. Table 12 reports the results of the main effects of university term on stage scores. The only significant relationship between stage and university term occurred with regard to Recruitment (p<.0l). Because there were five levels of university term, a further comparision between Recruitment and university term was performed to determine whether levels of term varied systematically. SPSS:X ONEWAY with the Tukey procedure tested for significant differences amongst term levels. The Tukey test established that mean Recruitment scores for inmates in terms 1 and 3 were significantly different. Inmates enrolled in term 3 of the university program had Recruitment 1 19 Table 12 EFFECTS OF ENROLLMENT TERM ON ROLE STAGE SCORES Stage University Term Cell Means 12 3 4 5 Sums of Squares D.F. F Prob. R2 Recruitment 1, .24 1. .26 1, .52 1. .42 1. .38 .39 4 3.95 .01 .28 Disorientation 1, .43 1, .57 1. .63 1. .54 1. ,53 .19 4 .51 .72 .07 Separation 1. .62 1. .82 1. .70 1. .82 1. .68 .22 4 .37 .82 .05 Transition 2. .53 2. .67 2. .59 2. .84 2. .76 .40 4 .60 .66 .08 Solidarity 2, .38 2, .63 2. .46 2. .75 2, .78 .84 4 1.37 .28 .18 * denotes pair of groups significantly different at the .05 level scores that were .28 points higher than their peers in term 1. Inmate-students in term 3 expressed significantly more opposition in their present feelings toward the university program and prison than did respondents in term 1 of the program, possibly because more inmates in term 3 had served previous federal sentences than those in term 1. Recruitment has been characterized by items which embody oppositional and cynical attitudes toward the university program. Table 13 shows the covariate effects of carceral and background variables on stage of role enactment. The main effects were calculated first (as shown in Table 12) without regard for the influence of the covariates. Then the covariate effects were calculated. One general observation is that the covariates raised the R Squared values substantially for each stage, by as much as a factor of three. The only significant 1 20 Table 13 CARCERAL AND DEMOGRAPHIC EFFECTS ON STAGE SCORES Stage Sums of Squares D.F. F Prob. R2 Best Predicting Covariate Prob. Recruitment .53 9 2.41 .05 .66 Previous federal term served .003 Disorientation .48 9 .57 .80 .27 Separation 1.08 9 .81 .60 .31 Transition 1.08 9 .71 .68 .31 Solidarity .84 9 .66 .73 .37 overall covariate effect was on Recruitment (p<.05). Including all the covariates increased the explained variance substantially (from .28 to .66). Only one covariate (previous federal term served) was significantly related to Recruitment (p<.0l). Based on these results, "previous federal term served" could be used to predict a Recruitment stage orientation. This result also suggests that respondents who had served more than one federal term were more likely than those who had not to express "opposition" to the university program and the student role. Recruitment, more than the other stages, embodied "oppositional" statements. A second analysis of variance was performed to test more directly the effects of "previous federal term served" on stage scores. Table 14 confirmed that "previous federal term served" 121 Table 14 EFFECTS OF FEDERAL PRISON EXPERIENCE ON STAGE SCORES Stage Previous Served 1 (No) Federal Term Cell Means 2 (Yes) Sums of Squares D.F. 1 Prob. R2 Recruitment 1.25 1.50 .47 1 19 05 .0001 .33 Disorientation 1.44 1.65 .36 1 4 39 .04 .14 Separation 1.60 1.85 • 50 3 79 .06 . 12 Transition 2.55 2.77 .40 1 2 50 . 12 .08 Solidarity 2.55 2.53 .01 1 01 .90 .01 predicted (p<.000l) high Recruitment scores. Furthermore, its effect on Disorientation stage scores was significant (p<.04). These findings suggested that inmates-students who served another federal sentence were likely to have feelings of opposition and negativity toward the university program. While not proving the view that repeated incarcerations "harden" inmates and reinforce norms and values resistant to change, these findings did show that inmates' negative feelings toward a program they otherwise value are associated with previous experience in federal prisons. 1 22 CHAPTER 7 CONCLUSIONS This chapter recapitulates the framework within which the study was designed and conducted, presents conclusions, and considers limitations. The Prison Context This study portrayed prison structure and organization as important factors in the growth of inmate social systems which largely oppose the pro-social goals of correctional programs. Some inmates were conjectured to "import" into the prison criminal values held on the outside that were not attributable to prisonization brought about by incarceration. Authoritarian run prisons and opposition from inmates present formidable obstacles to correctional programs designed to achieve positive change. Successful programs were described as multifaceted. They manage to engage or neutralize the inmate social system while fostering cognitive and social development. The SFU prison program met these criteria of success. Inmates in the SFU program appeared to develop positive social identities through their student roles. Role and cognitive development associated with the program appeared to counteract the prisonization effects of incarceration. The role theory of Sarbin and Allen (1968) provided a 1 23 framework for understanding the importance of role interactions within the prison and academic ecologies. Drawing on their principles, a model was developed to explain stages of inmate role development in the SFU program. Previous accounts of this program's success relied primarily on anecdotal reviews and psychological explanations. The interaction between inmate cognitive development and social environment had been acknowledged but not operationalized. This gap was bridged by developing and testing aspects of a five-stage model which purports to portray role development among participants in the prison university program. The Model Each conclusion pertains to one or more of the three purposes of the study. They contribute to the debate over differing views of the inmate social system (see Chapter 2 — Deprivation and Importation models) and the notion that "nothing" or "some things" work (see Chapter 1). This section recalls the three purposes and lists conclusions pertaining to each. Subsequent sections provide discussions of each conclusion. The first purpose was to explicate a theoretical model to identify student roles (and associated feeling states and expectations) occupied by inmates who participate in the prison university program. Two conclusions resulted. 1. Role theory is an appropriate framework for articulating a model of prison ecology. 1 24 2. Inmates experience five distinct and sequential stages of role development. The second purpose was to operationalize the model with expert judges and inmates. Three conclusions resulted. 1. Judges found the overall model plausible and workable. 2. Judges were able to discriminate items into stages. 3. Inmates' responses confirmed intra-stage reliability. The third purpose was to examine relationships between scores obtained from operationalizing the model and various socio-demographic and prison-related variables. Four conclusions resulted. 1. The expected associations with carceral and socio-demographic variables were not confirmed. 2. Inmates' forwarding of feelings from previous incarcerations supports the importation model. 3. A counter-intuitive finding (university term by Recruitment) is probably an artifact of previous penitentiary experience. 4. The university program does foster pro-social role development. Therefore, "some things" do work. 125 Explication of the Model It was concluded that role theory is an appropriate framework for articulating a model of prison ecology. The literature on prison organization, the inmate social system, and the university program all allude to the importance of inmate roles in analyzing prison dynamics. Principles and concepts of role theory are easily applied to prison settings. Drawing on role theory, literature on the university program and conversations with inmate-students, it was concluded that inmates experience five distinct and sequential stages of role development (Recruitment, Disorientation, Separation, Transition, and Solidarity). The five stages of role development were portrayed as consequences of the duration of inmate participation in the program. The model assumes that the program fosters inmates' achievement of positive role development. Operationalizing the Model After the model was developed, four expert judges familiar with the SFU prison program assigned a pool of 70 items into five stages of role development. Agreement among judges varied from stage to stage, but was moderately high overall, and led to several conclusions. First, it was concluded that judges found the overall model plausible and workable. It had face validity. Judges endorsed the five stages of role development on the basis of their 1 26 experience with inmate-students in the various prisons in which the university program operates. The stages made sense to them, they could work on task within that framework. Secondly, it was concluded that judges were able to reliably discriminate items into stages. The items expressing feelings toward the program and prison were written in phrases the judges said were representative of the inmates they knew. Moreover, inter-judge concordance on placing items into stages was high. Finally, with regard to explicating the model, inmates' responses confirmed intra-stage coherence. Reliability coefficients showed that inmates differentiated among five sets of items in their responses, corresponding to the stages of the model. Intra-stage were higher than inter-stage correlations (Table 8). Moderate inter-stage correlations suggested that respondents viewed many Recruitment and Disorientation, and Transition and Solidarity items, as similar. A moderate and significant relationship existed between Separation, Transition, and Solidarity items as well. These findings suggested there were three pronounced "themes" among the five stages of the model, resembling a three-stage perspective transformation process (Chapter 3) or Kelly's (1955) notion of personal constructs. However, the inter-stage correlations could well be an artifact of the small sample. Before concluding that three stages might be more representative than five, additional subjects from the four prisons in which the program operates 127 should be obtained in a follow-up study. Relationships with the Model The model suggested that stage of role development should vary by duration of involvement in the program (university term) and possibly by other carceral and background variables associated with inmates. Significant relationships existed between stage scores and several inmate-related variables (Table 11). Subsequent analyses of variance further tested relationships between these background variables and stage scores (Tables 12, 13, 14). It was concluded that the expected associations with carceral and socio-demographic variables were not confirmed. Age, educational level at start of sentence, and occupational status, severity of offence, length of sentence, warrant expiry date were not significantly associated with stage scores. It was expected that inmates closest to release might show significantly more pro-social orientations than those in mid-sentence. However, no sentence or release-related variables were significant, suggesting that the development of student roles does not parallel the changes in inmate roles over the lifespan of a sentence described by Garabedian (1963). It was concluded that inmates' "forwarding their feelings" from previous federal sentences served supports the Importation model. Inmates who had previously served another federal prison term had higher Recruitment and Disorientation scores, typified as oppositional feelings closely allied to inmate norms and 1 28 codes, than inmates without that experience. The importation model proposes that an inmate's past, present, and future experiences will shape his role expectations and values while in prison. Thus, prisonization effects appear to persist from sentence to sentence. However, inmates did not strongly suscribe to either Recruitment or Disorientation items as representative of their current feelings. The dominant stages for all inmates, including those with previous federal sentences, were Transition and Solidarity, typified as pro-social and learning-oriented. These inmates appeared to feel only a little opposition toward the university program they otherwise strongly identified with through pro-social stages of role development. One counter-intuitive finding (university term by Recruitment) was probably an artifact of previous penitentiary or other personal experience. Inmates in term 3 had significantly higher Recruitment scores than their peers in term 1. It was expected that as university term increased, so would the dominant stage score, moving from Recruitment to Solidarity. Thus, this finding is at odds with the assumptions of the model. There is no clear explanation for this relationship between university term and Recruitment. Moreover, with a sample of 33 inmates and a cross-sectional rather than longitudinal study, it would be premature to abandon the model's proposition about the relationship between university term and dominant stage score. Therefore, the interactions between cognitive and social development (the 129 program goals and desired outcomes), the time required to develop these skills (duration of program involvement), and the acquisition of role expectations and values (possibly in advance of cognitive and social skill development) need to be studied longitudinally with a larger sample. It was concluded that the university program does foster pro-social role development, and thus supports the position that "some things work" in correctional programming. The model and evidence from judges and inmates generally supported the conclusion that the SFU prison program fostered a shift from past ("how I used to feel") feelings typified by statements like "I find it hard to be a student" to present ("how I feel now") feelings typified by statements like "I'm confident of my ability to learn", and "I feel more tolerant toward other peoples' views." Respondents most strongly suscribed to Transition and Solidarity (Figure 6; Table 9). These latter two stages embodied the notion of a commitment to learning and to the program. The relative lack of present but stronger past support for Recruitment and Disorientation buttresses the conclusion that the SFU program fostered positive role development in the respondents. The model and methodology employed in this study appear to provide a basis for further study, especially in a wider variety of institutional settings. 1 30 Limitations of the Study All these conclusions are subject to certain limitations. There were two principal issues regarding limits to this study: (1) Structural — what are the stages, and (2) Functional — do inmates pass through them, and what demonstrates the passage? Concerning structure, the stages were derived using role theory as a framework. Items were based on an understanding of role theory, prison education programs, and prison dynamics. Judgements about the elements particular to each stage depended on understanding how personal, social and environmental variables in prison interact. Subsequent testing of the stage model could only occur if the stage descriptions were adequate. They appeared sufficiently adequate for the judges to discern five distinct stages into which items were assigned. Secondly, a more extensive pool of items at the item construction stage might have modified the emphasis presented in the set of items finally designated for each stage. However, items did not just "happen." Many were suggested by comments made by inmate-students. Items were created on the basis of defined concepts and propositions. Items were culled from frequent reviews of each of these elements. Did the judges equally understand the task they performed? Insofar as possible, the task was presented to each judge in the same fashion, given that the presentations took place in a variety of settings. The consensus among judges on which stages the 70 items best represented was strong but not unanimous. Moreover, judges were influenced in their decisions by the need 131 to reconcile disparities in the program according to its location in a high or lower security prison. How they balanced these possibly disparate observations should be more closely studied, perhaps leading to a more explicit operationalization on the basis of prison differences. Some ambiguity was introduced by using items that lacked total agreement on their placements into stages. The stages themselves were not precisely defined by "pure" items alone. How then could the items form a basis for testing the model with respondents in this study? The decision to employ all the items was based on reasonable inter-judge agreement. A larger pool of equally rigorous items would have increased reliabilities obtained for each stage while permitting items that lacked unanimous consensus to be discarded. However, after a number of iterations generated no significantly new items, all items in the pool were included. Moreover, judging involved subjective interpretations by judges. Just because an item did not elicit unanimous placement by the judges was not a sufficient reason to discard it. The inmates constituted another set of judges for the same 70 items. Thus, all 70 items were retained for use in the card sort. Concerning function, a limitation arises from the small and restricted sample of inmates in one prison. Generalizations based on findings with this group of respondents are limited. However, this initial test of the model concerned the power of its propositions to explain social-psychological interactions in the prison university program rather than its generalizability 132 to different prison settings. The sample of respondents made up 87 percent of the university inmate-student population at William Head and about 25 percent of all inmates in the prison. Their carceral and demographic backgrounds were quite varied and appeared to be typical of inmates in the general prison population in the Pacific Region. Moreover, this inmate population was known to have the best balance of new to continuing and senior students in this university program compared to all other prisons in the region. A final limitation concerned the "snapshot in time" or cross-sectional approach to data collection used in this study. It was argued that the prison environment exerts an unrelenting influence on the minds of inmates. The configuration of personal histories respondents brought to the data collection could conceivably have been subject to a number of mitigating factors beyond the scope of the tasks performed, but central to the motivation with which they were done. Factors such as being denied or granted parole, receiving or not receiving correspondence, or any other of the recurring incidents which affect mood in a prison, could have coloured the tasks with opposition or optimism. However, the goal of this study was not to chart fluctuations that inmate-students experience but levels or stages of role orientation to the university program that they identify with. Within these limits, the function of the stages was legitimately described in terms of dominant and associated minor but interesting orientations. Due to the cross-sectional 133 approach used in this study, passage from stage to stage was not measured. Only predictors of individual stage orientation were investigated. The final chapter provides a broader discussion of the implications of this study for future research, the correctional system, and the training of instructors. 1 34 CHAPTER 8 DISCUSSION Concerning Future Research Future studies should investigate the structure and function of the model. For example, a study to test the model with inmate-students in prisons which differ in security classification would test the generalizeability of its structure. Another study should investigate the extent to which inmates transfer their student roles into the community after release. A first step in such a study would be to identify appropriate indicators of a transfer effect. The stability of the stages, especially Solidarity, needs to be investigated. How do inmate-students who face long-term incarceration maintain a role relationship to the university program that exemplifies Solidarity? Do they ever "fall off the wagon" -- pass from Solidarity back to Recruitment and Disorientation? Moreover, competing hypotheses need to be considered. Inmate "cycles" of discouragement and despair and hope and renewal, and their impact on stage of student role development, should be investigated. The relationship of the inevitable fluctuations and inconsistencies in personal and social behaviour to longer-term social identity (such as learner, student) could be investigated by case study or 1 35 longitudinal study. Is there a "conversion" that happens prior to entering the program, or even after, that reflects something like an individual's coping mechanisms rather than a socially powered transition? There are several implications for future research which flow from this study. William Head was classified as a low-medium security prison. Most respondents felt they were more involved in positive student roles than in "oppositional" inmate roles as participants in the university program. This result concurred with the view that prison practices vary by level of security and produce correspondingly different degrees of "opposition" in inmate society. Lower security results in less opposition, higher security, more opposition. Thus, a logical extension to this study is its replication in a higher security prison. More "opposition" would be expected of inmate-students in a maximum security prison like Kent than in William Head. The extent to which greater "opposition" in high security prisons affects stage of role development needs to be tested. How successful are inmates in higher security prisons in maintaining a positive role orientation to the university program, personnel, and to other prison staff? If they maintain a positive relationship to the university program absent in other formal relationships or official accounts of their backgrounds, what attributes of the university program account for it and could a similar effect be achieved in other program areas? "Having served a previous federal prison term" emerged as 1 36 an important correlate, most notably with Recruitment. A study should be made of factors associated with repeated incarceration as an adult and juvenile to determine the rate of success and participation in the university program of those who may be most thoroughly socialized into the values of an oppositional inmate subculture. Although Recruitment and Disorientation scores were significantly higher for inmates who had served a previous federal sentence than for those who had not, the amount of "opposition" to the program expressed in these stage scores was comparatively minor. Even recidivist inmates developed dominant pro-social, learning-oriented roles typified by Transition and Solidarity. The apparent failure of "university term" to account for significant variations in "stage scores (with the exception of Recruitment) needs further study. A longitudinal study where inmates are followed from Recruitment to Solidarity should be undertaken. This type of study would provide information about how quickly new inmate-students are socialized into role expectations associated with the university program. A number of case studies conducted in this fashion would enable researchers to compare inmate-students' changing perspectives of their stage of role development and instructors' assessments of their development. Moreover, this type of approach would lead to a definition of what inmates do while occupying the various stages of role development and could be used to validate the Q-sorts as well. Studying their behaviour as well as self-perceived feeling states should help explain the functions of 137 student role stages in relation to the demands of prison existence. In addition to a longitudinal study of the university program, an experimental project should investigate role development in environments employing democratic principles espoused by the SFU program, but without the humanities curriculum as content. The purpose of this study would be to determine whether a "content-free" social environment (at least insofar as academic content is concerned) fosters stages of role development analogous to those associated with the SFU program. The underlying issue is this: What accounts for variance in the development of pro-social roles? Do the interaction effects of the university curriculum and the alternative community of the program account for more variance in the development of pro-social roles than would a non-academic social ecology run on similar principles? Conversely, a study should be conducted on stages of role development associated with the university, or other academic programs, independent of the social support engendered by a physically distinct, alternative academic community to which inmates belong. Comparison studies concerning the unique and joint contributions of academic content, socially supportive communities, and interactions between them would increase our understanding of "what works" in correctional programming and why. 1 38 Concerning Prison Administration What does this study say to prison authorities who have responsibility for planning and implementing programs in prison? If their goals are to foster positive, pro-social changes in inmates, several implications flow from this study. The university program is successful because it maintains a separate and distinct existence within prison. The relationship between the university program and prison appears to be an exception to the notion of horizontal integration in lifelong education. The contrast between the norms and role expectations associated with the program and prison fosters role conflict which' has the potential to result in the formation of pro-social values and roles. Therefore, it would be counterproductive to assimilate the academic program into the prevailing prison organization. Rather, prison authorities should tolerate and reinforce, with due regard for reasonable institutional security, the autonomy of the SFU and similar programs. Prison authorities should reduce obstacles to inmate participation in the SFU program. Moreover, they should establish other programs that embody the active social principles of the SFU program if not the focus on academic content. Although there can be no guaranteed success in correctional programming, the notion of neutralizing inmate social roles and providing opportunities to develop pro-social alternatives exemplifies the "some things work" position. Prison authorities should encourage all inmates to enrol 1 39 (if they meet the literacy requirements) in the SFU program. Being a student in this program for a time may well ameliorate the negative effects of prisonization the recidivist inmate carries from sentence to sentence. Finally, authorities need to recognize that some inmates continue on as students after their release. They use their continuing student roles to construct a bridge to the outside world. By maintaining student roles they may be able to better avoid the circumstances that led to incarceration. This transfer of student roles to a setting outside prison suggests that pro-social role development associated with the SFU program is not just an artifact of coping with prison life that disappears upon release, but indicative of lasting and substantive change in inmates. It deserves support from correctional authorities. Concerning Training of Correctional Educators Several implications arising from this study should inform the training of correctional educators. First, the model of role development highlights the importance adult educators ascribe to climate-setting in its broadest sense. The distinctiveness of the SFU community is a key factor in its success. Democratic principles are put into the crucible of experience. Instructors in that environment must be able to guide and tolerate the experimental practice of theoretical principles. The model of role development suggests no alternative to the effective practice of student roles within a 1 40 community run on democratic principles. Content-bound instructors are likely to fare less well in the social experiment of the SFU program than those who allow inmate-students to take ownership of their learning. Instructors should be made aware of the importance of the roles they play (eg. complementary, audience) in relation to the overall integrity of the SFU program by means of the social reinforcement they provide to inmates. They should be aware of the importance of modelling attitudes consistent with the pro-social goals of the program. Moreover, instructors should be well-centred individuals who are able to act as consistent models in a prison environment typified by manipulation. Finally, they should consider that the SFU program is capable of transforming the lives of inmates just as liberalizing education does for learners everywhere. 141 REFERENCES Akers, R., Gruniger, W., & Hayner, N. (1976). Prison inmate roles: inter-organizational and cross cultural comparisons. International Journal of Criminology and Penology, 365-381 . Ayers, J. D., Duguid, S., Montagne, C, & Wolowydnik, S. (1980). Effects of University of Victoria program: A post  release study. Report to the Solicitor-General of Canada. Bell, R. M. (1982, May 16). An approximation to the convocation address: Tenth anniversary of the University of  Victoria prison education programme. Presented to the Second Annual Recognition and Awards Ceremonies, William Head Institution, British Columbia. Boshier, R. W. (1983). Education inside: Motives for participation in prison education programmes. Report to the Solicitor-General of Canada. Vancouver, University of British Columbia. Boshier, R.W. & Clarke, G. S. (1983). Psycho-behavioural correlates of motives for participation in prison education programmes. Proceedings of the 2nd Annual Conference of the Canadian Association for the Study of Adult Education, 1, 1-14. Boyanowsky, E. 0. (1977). The psychology of identity change: A theoretical framework for review and analysis of the self-rble transformation process. Canadian Psychological  Review, 18(2), 115-127. Burns, H. (1969). A miniature totalitarian state: Maximum security prison. Canadian Journal of Corrections, 11(3), 153-164. Clemmer, D. (1958). The prison community. New York: Holt, Rinehart & Winston. Cloward, R. (1968). Social control in the prison. In L. E. Hazelrigg (Ed.), Prison within society: A reader in  penology (pp. 78-112). Garden City, NY: Doubleday. 142 Cohen, S. & Taylor, L. (1972). Psychological survival; The  experience of long-term imprisonment. Harmondsworth, England: Penguin. Conrad, J. P. (1983). What prospects for rehabilitation? A dissent from academic wisdom. In L. F. Travis, M. D. Schwartz, & T. R. Clear (Eds.), Corrections: An issues  approach (2nd ed.) (pp. 216-226). Cincinnati: Anderson. Cosman, W. (1981). Penitentiary education in Canada. In L. Morin (Ed.), On prison education (pp. 34-42). Ottawa: Supply & Services Canada. Cousineau, F. D. & Plecas, D. B. (1982). Justifying criminal justice policy with methodologically inadequate research. Canadian Journal of Criminology, 24, 307-321. Cressey, D. (i960). Limitations on organization of treatment in the modern prison. In R. A. Cloward, D. R. Cressey, G. H. Grosser, R. McCleery, L. E. Ohlin, G. M. Sykes, & S. L. Messinger (Eds.), Theoretical studies in  the organization of the prison (pp. 78-110). New York: Social Science Research Council. Davis, M. (1976). The prison dilemna. In W. T. McGrath (Ed.), Crime and its treatment in Canada (pp. 370-375). Toronto: MacMillan. Dubin, R. (1978). Theory building. New York: The Free Press. Duguid, S. (1979). History and moral education in correctional education. Canadian Journal of Education, 4, 81-92. Duguid, S. (1980). Post-secondary education in a prison: Theory and praxis. The Canadian Journal of Higher  Education, 10, 29-35. Duguid, S. (1981). Moral development, justice and democracy in the prison. Canadian Journal of Criminology, 23, 147-162. Duguid, S. (1981a). Prison education and criminal choice: The context of decision-making. Canadian Journal of  Criminology, 23, 421-438. 1 43 Duguid, S. (198lb). Rehabilitation through education: A Canadian model. In L. Morin (Ed.), On prison education. Ottawa: Supply and Services Canada. Duguid, S. (1983). Origins and development of university education at Matsqui institution. Canadian Journal of  Criminology, 25, 295-308. Duguid, S. (Ed.). (1985). The prisoner as student (Occasional Papers on Prison Education). Burnaby: Prison Education Program and Institute for the Humanities, Simon Fraser University. Eichman, C. (1981). Humanizing punishment: The future of corrections. Journal: John Howard Society of Quebec, 2(1), 1-7. Ekstedt, J. W. & Griffiths, C. T. (1984). Corrections in  Canada: Policy and practice. Toronto: Butterworths. Fawcet, B. (1983). [Editorial]. Prison Journal 83, 3, 4-5. Garabedian, P. (1963). Social roles and processes of socialization in the prison community. Soc ial Problems, 11, 139-152. Gendreau, P. & Ross, R. R. (1979). Effective correctional treatment: Bibliotherapy for cynics. Crime and Delinquency , 25, 463-489. Glaser, D. & Stratton, J. (1961). Measuring inmate change in prison. In D. Cressey (Ed.), The prison (pp. 381-392). New York: Holt, Rinehart. Goffman, E. (1961). Characteristics of total institutions: The inmate world. In D. Cressey (Ed.), The prison (pp. 15-67). New York: Holt, Rinehart. Goffman, E. (1961a). Characteristics of total institutions: Staff-inmate relations. In D. Cressey (Ed.), The Prison (pp. 68-106). New York: Holt, Rinehart. Gosselin, L (1982). Prisons in Canada. Montreal: Black Rose 1 44 Books. Grosser, G. H. (1968). External setting and internal relations of the prison. In L. Hazelrigg (Ed.), Prison  within society: A reader in penology (pp. 9-25). Garden City, NY: Doubleday. Hepburn, J. & Stratton, J. (1977). Total institutions and inmate self-esteem. British Journal of Criminology, 17(3), 237-250. Irwin, J. & Cressey, D. (1962). Thieves, convicts and the inmate culture. Social Problems, j_0(2), 1 42-155. Jackson, M. (1983). Prisoners of isolation: Solitary confinement in Canada. Toronto: University of Toronto Press. Kassebaum, G., Ward, D., & Wilner, D. (1971). Prison treatment  and parole survival• New York: Wiley. Kelly, G. A. (1955). The psychology of personal constructs, (2nd ed.). New York: W. W. Norton. Kerlinger, F. N. (1973). Foundations of behavioral research (2nd ed.). New York: Holt, Rinehart & Winston. Knights, W. ( 1983). Bourgeois purity. Prison Journal 83, 3_, 57-60. Law Reform Commission of Canada (1975). Working paper 11:  Imprisonment and release. Ottawa: Information Canada. Linden, R. & Perry, L. (1982). The effectiveness of prison education programs. Journal of Offender Counselling,  Services & Rehabilitation, 6(47, 43-57. Martinson, R. M. (1974). What works? Questions and answers about prison reform. The Public Interest, 35, 22-54. Mezirow, J. (1978). Perspective transformation. Adult  Education, 28(2), 100-110. 1 45 Morin, L. (1981). Inmate right to education. In L. Morin (Ed.), On Prison Education (pp. 23-33). Ottawa: Supply and Services Canada. Murton, T. 0. (1976). The dilemma of prison reform. New York: Holt, Rinehart & Winston. Ohlin, L. E. ( 1956). Sociology and the field of corrections. New York: Russell Sage. Ontario Institute for Studies in Education (1978). Review of  penitentiary education and training, Phase 1: Report to  reviewers. Ottawa: Education and Training Division, Canadian Penitentiary Service. Ouimet, R. (1969). Report of the Canadian committee on  corrections — toward unity: Criminal justice and corrections. Ottawa: Information Canada. Roebuck, J. (1963). A critique of thieves, convicts and the inmate culture. Social Problems, JM[(2), 193-200. Ross, R. R. (1980). Socio-cognitive development in the  offender: An external review of the u-vic program at matsqui~penTtentiary. Ottawa: University of Ottawa. Ross, R. R. & Fabiano, E. (1981). Time to think cognition  and crime: Link and remediation. Ottawa: Department of Criminology, University of Ottawa. Ross, R. R. & Fabiano, E. (1983). The cognitive model of  crime and delinquency prevention and rehabilitation:  Assessment Procedures. Toronto: Ministry of Correctional Services. Ross, R. R. & Fabiano, E. (1983a). The cognitive model of  crime and delinquency prevention and rehabilitation:  Intervention techniques. Toronto: Ministry of Correctional Services. Ross, R. R. & McKay, H. B. (1978). Behavioral approaches to treatment in corrections: Requiem for a panacea. Canadian  Journal of Criminology, 20, 279-295. 1 46 Sapsford, R. J. (1978). Life sentence prisoners: Psychological changes during sentences. British Journal of  Criminology, J_8(2), 128-145. Sarbin, T. R. & Allen, V. L. (1968). Role theory. In G. Lindzey & E. Aronsen (Eds.), The handbook of social  psychology (2nd ed., Vol. 1, pp.488-567). Menlo Park, California: Addison-Wesley. Sarbin, T. R. & Scheibe, K. E. (1983). A model of social identity. In T. R. Sarbin & K. E. Scheibe (Eds.), Studies in soc ial identity. New York: Praeger. Scull, A. T. (1977). Decarceration: Community treatment and  the deviant — a radical view. Englewood Cliffs, N.J.: Prentice Hall. Solicitor-General of Canada (1977). A summary and analysis of some major inquiries on corrections, 1938 to 1977. Ottawa: Canadian Government Publishing Centre. Stephenson, W. (1953). The study of behavior. Chicago: University of Chicago Press. Sykes, G. & Messinger, S. (1960). The inmate social system. In R. A. Cloward, D. R. Cressey, G. H. Grosser, R. McCleery, L. E. Ohlin, G. M. Sykes & S. L. Messinger (Eds.), Theoretical studies in the organization of the  prison (pp. 5-19). New York: Social Sciences Research Council. Sztompka, P. (1974). System and function. New York: Academic Press. Task Force on the Creation of an Integrated Corrections Service (1977). The role of federal correct ions in Canada. Ottawa: Supply and Services Canada. Thomas, C. W. (1973). Prisonization or resocialization? A study of external factors associated with the impact of imprisonment. Journal of Research in Crime and Delinquency , 10, 13-21. Thomas, C. W. & Foster, S. (1976). On the measurement of 147 social roles adaptations in the prison community. Criminal  Justice Review, j_( 1) , 11-21. Thomas, C. W. & Peterson, D. M. (1977). Prison organization  and inmate subcultures. Indianapolis, 111.: Bobbs-Merrill. Whetstone, K. (1981). How the prisoner sees education. In D. Ayers (Ed.), Proceedings of the national conference on  prison education (pp. 79-93). Victoria: University of Victoria. Zimbardo, P. G., Haney, C., Banks, W. C. & Jaffe, D. (1977). The psychology of imprisonment: Privation, power and  pathology. Stanford: Stanford University. 1 48 Appendix A: Seventy Items Listed By Stage RECRUITMENT STAGE 1. This program can't be any worse than other prison programs. 2. I don't plan to work too hard in this program. 3. Maybe I can fraud it for awhile in this program. 4. I'm not interested in any subject in particular. 5. I don't care what is taught in this program. 6. I don't expect much out of the program. 7. I'm curious to see what the program is like. 8. This program is some sort of scam. 9. This program is probably like all the others. 10. I don't care what I have to study. 11. No way is this program going to change me. 12. I only want to take courses in the subject I prefer. 13. There is only one subject that I'm really interested in. 14. I'm not going to get too involved with this program. 15. I think all instructors are pretty much the same. 16. I think all courses are pretty much the same. DISORIENTATION STAGE 17. There is more work than I expected. 18. Instructors sure expect a lot from me. 19. I'm surprised at how involved the instructors are with the students. 20. I don't know if I like this program. 21. This program is different from what I expected. 22. I find it hard to be a student. 23. I'm not sure I should continue in the program. 24. I feel like I'm too isolated here in the academic area. 25. Other students take this program more seriously than me. 26. I haven't figured out this program. 27. I don't know what to make of this program. 28. Other students take this program more seriously than I do. 29. I don't know what I'm supposed ito do as a student. 30. Being a student increases the pressure on me. 149 SEPARATION-ALIENATION STAGE 31. It's getting harder to put up with prison bullshit. 32. I feel a lot of conflict between student and inmate roles. 33. As a student, I feel too cut off from the rest of the inmate population. 34. I feel pressure not to get too involved with student roles. 35. The less I have to do with the general inmate population the better for me. 36. I'm starting to feel that I can really handle being a student. 37. I would like to have even less contact with the general inmate population. 38. I've found one subject I'd be happy to spend all my time studying. 39. I'm fed up with prison games. 40. I'm less interested in activities outside of school. TRANSITION-REFRAMING STAGE 41. I feel like this program is helping me get control over my life. 42. I feel the possibility of making changes in my life because of my student experiences. 43. I'm beginning to feel that this program has some value for me. 44. I'm finding it easier to accept the demands of being a student. 45. I feel like I'm starting to get something useful out of this program. 46. I'm beginning to feel that I've got different opportunities than I've had before. 47. I feel my student status is an asset in the prison. 48. I'm looking forward to getting something for myself out of this program. 49. I feel like my student experience will help me deal with problem more effectively. 50. I feel this program could lead me into different things. 51. I spend as much time as I can on my student interests. 52. I want to be more involved with all aspects of the program. 53. I enjoy learning about many different subjects. 54. I'd like more courses in different subjects. 55. I feel the best thing will be for me to stay involved with this program. 56. Being a student is changing some things for me. 57. I'm confident of my ability to learn. 150 SOLIDARITY STAGE 58. I feel completely involved with this program. 59. I know I can rely on the student community. 60. I see the program as a means to express myself. 61. I feel a responsibility to support the program. 62. This program is about the best thing that has happened to me in prison. 63. I have good working relationships with instructors. 64. I feel equal to any challenge which might come up in the program. 65. I feel like I have something to add to the program. 66. I can gather support from other students for group projects; or activities. 67. I feel more tolerant toward other peoples' views. 68. I would defend this program to other inmates. 69. I feel supported by the student community. 70. Other students want the same thing I do. Appendix B: Offence Severity Scale OFFENCE SEVERITY SCALE MAJOR OFFENCES 1. First, Second-degree Murder and Attempted Murder. 2. Assault causing or intended to cause serious .injury disfigu rement, or mutilation. 3. Kidnapping, forcible detention/abduction, and/or hostage-taking . 4. . Hijacking of aircraft and/or piracy of sea vessels. 5. Treason. 6. Espionage. 7. Illegal possession and/or detonation of explosives which are likely to cause death. 8. Violent terrorist activities. SERIOUS OFFENCES 1. Robbery with violence. 2. Violent sex offences (i.e., rape, attempted rape, child molestations, etc.). 3. Arson. 4. Sabotage. 5. Conspiracy to traffic or import a dangerous drug. S. Trafficking and possession for the purpose of trafficking (dangerous drugs) . 7. Trafficking in illegal firearms. 8. Manslaughter. 9. Extortion. 10. Armed Robbery or Attempted Armed Robbery. 11. Prison breach. 12. Escape custody with violence. MODERATE OFFENCES 1. Possession of dangerous drugs. 2. Trafficking, conspiracy, possession for the purpose of traf ficking (soft drugs) . 3. Forgery. 4. Fraud. 5. Bribery. 6. Forcible entry. 7. Break and Enter/Breaking Out. 8. Criminal negligence causing death or resulting in bodily harm. 9. Non-violent sex offences (i.e., gross indecency, indecent as sault , incest). 10. Robbery (excluding armed robbery and robbery with violence). 11. Escape (non-violent). 12. Theft over 200 dollars. 13. Obstruction of justice and perjury. 14. Possession of stolen property over 200 dollars. 15. Possersion of a weapon for a purpose dangerous to the public peace. 16. Assault causing bodily harm. MINOR OFFENSES 1. Possession of stolen property under 200 dollars. 2. Common assault. 3. Possession of soft drugs . 4. Theft under 200 dollars. 5. Public mischief. 6. Criminal negligence not resulting in bodily harm. 7. Possession of a restricted or prohibited weapon. 8. Possession of forged currency, passports, cheques. 9. Unlawfully-at-Large. Appendix D: Carceral And Demographic Coding Sheet CARCERAL AND DEMOGRAPHIC INFORMATION Card One  Column No. 1 - 3 7-12 13 - 18 19 - 22 23 - 28 29 - 34 35 Variable Identification number Date of birth-Date of most recent sentence Warrant expiry date Mandatory Supervision date Occupational status Code Day, month, year Day, month, year Length of sentence (in days) Actual no. Day, month, year Day, month, year -1—1 1 r I i ! Unskilled-1 Skilled»2 Clerical/sales=3 Managerial/ AriTninistrative-4 Professional/ Technical=5 36 - 37 Educational level at start of sentence Grade 4 • 4 5 = 5 6 = 6 Partial Univ. 12 » J,2 "13 - 13~ 14 - 14 _15 = L5_ Univ. 16 - 16 38 Previous federal penitentiary experience? No = 1 Yes = 2 39 Current year of enrolment in university program 40 "Severity" of most serious Minor « 1 present offence Moderate = 2 Serious = 3 Major = 4 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items