Structural and social psychological determinants of prisonization

Journal of Criminal Justice, Vol. 27, No. 5, pp. 427–441, 1999 Copyright © 1999 Elsevier Science Ltd Printed in the USA. All rights reserved 0047-2352/99 $–see front matter

PII S0047-2352(99)00014-8

STRUCTURAL AND SOCIAL PSYCHOLOGICAL DETERMINANTS OF PRISONIZATION

Brent A. Paterline Department of Criminal Justice North Georgia College and State University Dahlonega, Georgia 30597

David M. Petersen Department of Sociology Georgia State University Atlanta, Georgia 30303-3083

ABSTRACT This research examined the structural and social psychological determinants of prisonization among a sample of 239 male inmates in a maximum security prison. The study examined the process of prisonization more fully than has been done in the past by developing a theoretical model that integrates measures of importation, deprivation, and inmate self-conceptions. To date, the self-concepts inmates bring with them and sustain in prison have been a neglected aspect in the study of prisonization. The results of this study demonstrated that while deprivation model variables were shown to be the better predictors of prisonization, certain importation model variables were not without significant impact and the two models do explain more variance together than either one separately. Measures of self-conception were of limited value in predicting prisonization. © 1999 Elsevier Science Ltd. All rights reserved.

INTRODUCTION The study of inmate subcultures began with the pioneering work of Clemmer, who coined the term prisonization to refer to the adoption of the folkways, mores, customs, and general culture of the inmate subculture (Clemmer, 1940, p. 270). Clemmer’s research later incited one of the more stimulating debates in criminological

literature between the deprivation and importation models as predictors of prisonization. The deprivation model emphasizes the importance of the pressures and problems caused by the experience of incarceration in creating an inmate subculture. The importation model, on the other hand, emphasizes the effects that preprison socialization and experience can have on the inmate social system. Rather than employing ei-

427

428

B. A. PATERLINE and D. M. PETERSEN

ther model as an explanation of prisonization, there seems to have emerged in more recent years a consensus among researchers, which promotes the integration of both theories into one model. The integration of the importation and deprivation models has generated a number of studies, which have increased our understanding of prisonization. These studies, however, have been limited in several ways. First, previous research has implemented only a small number of select variables to test the importation and deprivation models. Second, researchers have failed to incorporate other models of prisonization, such as inmate self-conception. Finally, measures of importation and deprivation have not been used in a multivariate context to examine the relative importance of each variable in predicting prisonization. The purpose of this study was to advance penological research by addressing the above issues which are, as such, logical extensions of previous research. This investigation began where a number of studies had left off, by examining the process of prisonization more fully than has been done in the past. In order accomplish this, the importation and deprivation models have been expanded by incorporating a larger and more inclusive set of independent variables as predictors of prisonization. To test the importation model the research not only utilized several variables measuring inmate criminal history, but also employed several independent variables assessing inmates social status prior to imprisonment, such as race, marital status, and employment. The deprivation model was evaluated by the use of multiple independent variables, including attitudes toward violence and changes in life satisfaction, two new measures created for the purpose of this research. In recent years, penologists have also concluded that current models of prisonization need to be expanded to encompass a more interactionist approach to prisonization (Thomas, Petersen, and Zingraff, 1978; Wright, 1991; Zamble and Porporino, 1988). There are other factors, beyond the preprison lives of the inmates, which may be important predictors of prisonization, even when resistance to institu-

tional life is quite high. One such factor is selfconcept, which would permit researchers to better examine the interaction between inmates and their social environment. This research advanced previous research by utilizing four new measures of inmate self-conception as predictors of prisonization. Finally, and perhaps most importantly, this research utilized multivariate procedures to examine the relative importance of each set of variables in predicting prisonization. A major contribution of this research was to determine the importance of each independent variable in predicting prisonization through the use of a multivariate regression analysis. By using a procedure, which integrates both personal and structural variables in a multivariate context, this research was expected to lead to an improved explanation of the prisonization process within inmate populations.

THEORETICAL MODELS The Clemmer (1940) investigation of prison life in a maximum security prison helped to increase social scientists’ awareness of the latent inmate community that exists side by side with the formal organization of the prison. An important contribution made by Clemmer was the concept of prisonization, which he defined as “the taking on, in greater or lesser degree, the folkways, mores, customs, and general culture of the penitentiary” (Clemmer, 1940, p. 270). Clemmer believed that all inmates suffer certain influences he called “universal factors of prisonization,” which prepared and often shocked new inmates into readiness to enter the prisonization process. Clemmer identified several universal factors of prisonization, such as the inmate’s acceptance of an inferior role, learning to adopt to the regulations and structure of the prison, and learning to become passive about one’s own needs—many of which were automatically taken care of by the institution. In the mid-1950s Sykes expanded on Clemmer’s work in his analysis of an inmate social system in a maximum security prison. Accord-

Determinants of Prisonization

ing to Sykes (1958) all prisoner subcultures (countercultures) exhibit a common pervasive inmate value system, regardless of the location and characteristics of the institution. This value system takes the form of an explicit inmate code, which is used as a guide for behavior in inmates’ relations with fellow prisoners and guards. The inmate code, therefore, summarizes the behavioral expectations of the inmates’ social system. The deprivation model is the theoretical position that argues that the conditions within prisons account for the formation of prison countercultures. Deprivation theory argues that prisonization is an adaptive process employed by inmates to cope with the social and physical deprivations of imprisonment (Sykes and Messinger, 1960; Thomas and Petersen, 1977; Tittle, 1972). Faced with similar deprivations, inmates begin to try to solve their problems collectively. Once such a collective response occurs, an inmate society begins to form, “a society that includes a network of positions, which reflect various types and levels of commitment to subcultural norms as well as adaptive reactions to the problems of confinement . . .” (Thomas and Petersen, 1977, p. 49). The creation of such a subcultural system is seen as an effective means of resolving many of the problems of prison life. Proponents of the deprivation model argued that the subculture into which inmates are assimilated is a reflection of the pains of imprisonment that are caused by the structure of the prison organization. This means that an understanding of inmates’ attitudes, values, and behavior can be gained through an examination of the influences that are indigenous to the prison setting (Thomas and Cage, 1977). In support of the deprivation model, research has shown that prisonization is related to: the number of times one has been in prison (Gruninger, 1975); powerlessness or alienation (Guenther, 1978; Hyman, 1977; Neal, Snyder, and Balogh, 1974; Smith and Hepburn, 1979; Thomas, 1975; Thomas and Poole, 1975; Thomas and Zingraff, 1976; Tittle and Tittle, 1964); interaction with fellow inmates (Morris and Morris, 1963; Wheeler, 1961); and orientation toward staff (Gruninger, 1975; Schwartz, 1971). The deprivation model has also been beneficial

429

in explaining prison homosexual behavior, and types of prison leadership (Akers, 1977). Criticisms of the deprivation model developed into what is known as the importation model of prisonization. The basic position advocated by those who support this model is relatively simple. Preprison experiences, particularly those involving the adoption of criminal values, and personal characteristics of the inmates affect the degree of assimilation into the inmate subculture (Irwin, 1970; Irwin and Cressey, 1962). If the deprivations of confinement were the sole determinants of the extent to which inmates become assimilated into the inmate subculture, then, given the common problems of adjustment, every inmate would become highly prisonized. This has not been supported by prison research (Thomas and Petersen, 1977), and although the inmate subculture has been established prior to the time a given inmate enters prison, variations in the receptivity to the subculture cannot be accounted for solely by the structural conditions of confinement. One’s adaptation and receptivity to the inmate system is shaped by his/her socialization prior to confinement. The importation model has found support in research linking the adoption of the inmate subculture to general social demographic factors such as age; race; educational attainment; and preprison socioeconomic and employment status (Alpert, 1979; Jensen and Jones, 1976; Kennedy, 1970; Schwartz, 1971; Thomas, 1973, 1977b; Wright, 1989); criminal history, such as prior convictions, number of arrests, offense type, age at first arrest or conviction, and amount of previous criminal experience (Alpert, 1979; Cline, 1968; Kennedy, 1970; Schwartz, 1971; Thomas, 1973; Wellford, 1967; Zingraff, 1980); identification with criminal values and attitudes toward the legal system (Thomas and Poole, 1975; Zingraff, 1980); identification with broad social, political, and religious ideologies (Irwin, 1980; Jacobs, 1976); and contact with the extraprison world (Thomas, 1973; Tittle and Tittle, 1964). Despite the oppositional nature of the deprivation and importation models, researchers have recognized the necessity of integrating these two positions into a more comprehensive

430

B. A. PATERLINE and D. M. PETERSEN

model (Thomas and Petersen, 1977; Thomas, Petersen, and Zingraff, 1978; Zingraff, 1980). A single-minded subscription to only one of the theories is rarely seen in the literature today (Kalinich, Stojkovic, and Klofas, 1988). Researchers have suggested that both deprivation and importation variables explain prison adjustment and have attempted to combine the models into a single theoretical perspective (Leger and Barnes, 1986; Thomas and Petersen, 1977; Thomas, Petersen, and Zingraff, 1978). Several researchers have suggested that a complete model of prisonization should include measures of inmate extraprison relationships, self-attitudes, and the self-identities that inmates maintain in the institution (Faine, 1973; Leger, 1981; Wright, 1991; Zamble and Porporino, 1988). Zamble and Porporino (1988), for example, believed that the great majority of theorists have ignored how individuals with particular personality characteristics or selfconceptions react to the conditions and situations of prison life. Future models should more fully examine the relationship between person and environment by placing more emphasis on inmate self-conceptions. Gecas (1986) contended that when examining any socialization processes, social scientists have developed models, which lack an adequate theory of motivation. Failing to examine selfmotivations, socialization becomes either too deterministic or too unpredictable in terms of outcomes. Gecas suggested that the self-motivations can be found within the self-concept. By attributing motivational properties to the self-concept, the self can be viewed as an agent in its own creation. Possessing a self-concept means that individuals are motivated to maintain or enhance it, to conceive of it as efficacious and consequential, and to experience it as meaningful or real (Gecas, 1982; Gecas and Schwalbe, 1983). Three self-motivations or dimensions of self can be distinguished: self-evaluation, self-efficacy, and identity salience. The self-evaluation or self-esteem motive refers to the motivation to view oneself favorably and to act in such a way as to maintain a favorable view of oneself. This occurs through various self-presentation strategies and/or through reconstructing the environment to make it re-

flect a more favorable view of oneself (Greenwald, 1980; Rosenberg, 1979). These adaptation strategies are even more evident within coercive organizations, such as prisons, asylums, and work settings. Inmates who enter prisons have been exposed to the depersonalizing and stigmatizing effects of the legal system. These experiences, combined with the additional societal degradation they experience from serving time and the coercive structure of the institution itself, constitute a “massive assault” on the self-esteem of those imprisoned (Sykes, 1958). The self-efficacy motive refers to the degree to which one perceives oneself as a causal agent in the environment (Gecas, 1986). As a dimension of selfconcept, self-efficacy refers to how individuals conceptualize themselves as active persons who have control over their world. Previous research has shown that these motives have behavioral consequences. Those individuals who believe that they maintain control of their destiny are more likely to take steps to improve environmental conditions and be more resistant to attempts to influence them (Bandura, 1977, 1982; DeCharms, 1968; Gecas and Schwalbe, 1983). These individuals will, therefore, resist socialization and/or prisonization attempts that undermine their sense of authenticity. A third important self-motive is identity salience and addresses the question of whether the various identities constituting the self-concept are meaningful and “real” to the individual (Gecas, 1986). According to these theorists the self is organized into various identities (Burke and Reitzes, 1991; Callero, 1985; Stryker, 1968, 1980). These identities are “parts” of the self and are internalized positions that exist insofar as a person participates in roles and/or structured relationships. Persons may have many identities, limited only by the number of structured relationships and/or roles in which they participate (Stryker, 1968, 1980). Identity theorists proposed that identities are arranged in a hierarchy of salience, and that the higher an identity is in the hierarchy, the greater the probability that the identity will be invoked in a variety of interactions. Importance of an identity can be defined as the degree to which a person’s relationships to specific sets of others

Determinants of Prisonization

depends on his/her being a particular type of person. The more important an identity is to a person, therefore, the more committed he/she will be to that identity (Burke and Reitzes, 1991; Stryker, 1968, 1980). Commitment to identities is a major source of motivation for individuals to act in accordance with the values and norms associated with these identities. Socialization experiences that undermine the values and norms associated with salient identities may fail. In the case of prisonization, inmates who place a high emphasis on many “valued” or “respectable” social identities may be less likely to become fully integrated into the inmate subculture.

431

not included in the scale. Any item-to-scale correlation that was not equal to or greater than .50 was deleted from the final scale. Prisonization. This research utilized the concept of prisonization as its dependent variable. Prisonization is generally described as the process of accepting the normative structure of the inmate social system. The measure of prisonization was provided by a seven-item Likerttype scale. Responses for each scale item were scored on a five-point scale (strongly agree, agree, undecided, disagree, strongly disagree). The higher the scale score on this measure, the higher the level of prisonization. The scale had an overall mean of 21.96 and a standard deviation of 3.99.

RESEARCH METHODOLOGY Sampling and Data Collection The data for this investigation were collected through personal interviews administered to a sample of sentenced male inmates in a maximum security prison operated by the Federal Bureau of Prisons. The institution housed 2,200 male felons, and 440 were randomly selected from this population. Ninety of the inmates were deleted from the sample before the study began because either they did not speak English or were not present in the institution at the time of the study. During the interview process, 115 respondents would not cooperate or could not be scheduled for interviews. A total of 239 inmates were successfully interviewed (67.5 percent of the sample and 10.8 percent of the total inmate population). Operation of the Variables This research utilized twenty independent variables as predictors of inmate prisonization. The manner in which all the measures for this research were operationalized is described below. (A copy of the scales used in this study is available from the senior author.) Item selections were accomplished for all scales through the use of a principle components factor analysis. Any item that did not have a factor loading of at least .40 on the initial unrotated factor, was

Measures of importation. Several variables were used to assess inmate criminal history and were measured by a one-item response from the inmate. Age at first conviction and number of arrests were both interval variables used as indicators of involvement in criminal subcultures before incarceration. The mean age of first conviction is 23.52, with a standard deviation of 7.81. The mean number of arrests is 4.59, with a standard deviation of 2.81. It is expected that the younger one’s age at his/her first conviction and the greater number of arrests one has on his/ her criminal record, the greater the degree of prisonization. Current offense is the type of felony conviction for which an inmate is presently serving time. This is a dichotomous variable in which offense type is classified into either violent or nonviolent. Ten percent of the inmates were, at the time of this research, serving time for violent offenses. It was expected that inmates who were serving time for violent offenses would be more prisonized, because one of the major values of the inmate subculture is toughness and the use of physical force. Prior incarceration is also a measure of involvement in a criminal subculture and was measured on an interval scale by noting the number of times inmates indicated that they had been previously incarcerated. Nineteen percent of the inmates had no record of prior incarceration, 32 percent had one prior incarceration, and

432

B. A. PATERLINE and D. M. PETERSEN

49 percent had two or more prior incarcerations. Following the logic of the importation model, it was expected that the more times one had been in correctional institutions, the greater his/her adoption of criminal values, and thus the more accepting he/she would be to the inmate code. Social roles and statuses are a central link between the individual and the larger social structure. In prison, inmates bring with them certain valued statuses that may “anchor” them to the extraprison world. These socially valued statuses may be used to determine one’s commitment to legitimate society. This research utilized five variables to determine several preprison social identities: marital status, race, educational attainment, employment status at time of arrest, and number of months employed during the two years prior to incarceration. It was expected that those inmates with more highly valued social identities were less likely to become assimilated into the inmate subculture because of their previous commitments to legitimate society. Marital status was measured by a dichotomous variable by noting whether or not an inmate was married. In the present sample, 46 percent of the inmates reported being married. Race was measured by a dichotomous variable indicating whether an inmate reported himself as White or Black. Fifty-seven percent of the inmates reported being White. Educational attainment was measured by noting the number of grades inmates had completed prior to being incarcerated. Forty-eight percent of the inmates reported having less than a high school education, 29 percent reported having a high school diploma, and 23 percent reported having at least some college. Preprison income was measured on an ordinal scale by noting the normal monthly income for inmates prior to their arrests. Fifty-six percent of the inmates reported making more than $600 a month prior to being arrested. Preprison employment status was measured by two variables: whether or not an inmate was employed at the time of arrest and the number of months employed during the two years prior to imprisonment. Sixty-seven percent of the inmates were employed at the time of their arrests. Eighteen percent of the inmates were not employed

during the two years prior to their confinements and 40 percent of the inmates were employed the entire two years prior to their confinements. Measures of deprivation. Prisons are similar to other types of formal organizations in that they are characterized by a system of rules, a rigid hierarchy of authority, a reliance on coercive power, and a low degree of individual autonomy. Prison officials often use the coercive structure to attain and maintain a desired level of social control which, in turn, often generates strong feelings of alienation among the inmate population (Sykes, 1958; Thomas and Poole, 1975; Thomas and Zingraff, 1976). Perhaps the most important dimension of alienation for correctional research is powerlessness, which can be defined as a feeling of helplessness and subordination to power, which is vested in others (Neal, Snyder, and Balogh, 1974). If proponents of the deprivation model are correct, one would expect to find high rates of prisonization among those inmates who have heightened feelings of contextual alienation or powerlessness. A measure of contextual alienation comparable to that reported earlier by Thomas and Zingraff (1976) was employed in this study. The measure consisted of four Likert-type attitudinal items that were derived from a larger pool of initial items. The higher the scale score on this measure, the greater the feeling of powerlessness. The mean of this variable was 16.29, with a standard deviation of 3.13. A major consequence of confinement is the tendency for inmates to develop an oppositional attitude toward both the prison institution they have been forced to become a part of, and the staff of the institution. When such oppositional attitudes are present it is unlikely that inmates will be supportive of the formal goals and policies of the institution. It is expected that the most oppositional inmates will report relatively high levels of prisonization. A six-item attitudinal scale was developed to measure opposition to the prison organization. The higher the score on this measure the more profound was one’s opposition to the prison organization. The mean for this measure was 17.13, with a standard deviation of 4.31. This research measured life satisfaction through seven dimensions of an inmate’s life:

Determinants of Prisonization

family relationships, friendships, social involvements, religion, physical health, economic security, and personal character. Responses to each dimension were scored on a three-point scale of importance (very satisfied, satisfied, not satisfied). Measures of preprison and present life satisfaction were created by summing the satisfaction responses across the seven dimensions. Scores from present life satisfaction were subtracted from the scores of preprison life satisfaction to create an index measuring a change in life satisfaction. The mean for this index was 1.45, with a standard deviation of 3.16. Low scores reflect a negative change or decrease in life satisfaction. It was expected that inmates with a negative change in life satisfaction would attempt to find consolation through integrating into the inmate subculture. A common psychological consequence of institutional oppression and coercion is aggression. Among prisoners, this aggression may be redirected into violence in which one prisoner attacks another, and may take the form of murder, assault, or rape. One would expect that those inmates who fail to be aggressive and ruthless would be less prisonized, because the inmate social system places a high emphasis on toughness and “being a man.” Attitude toward violence was measured by a six-item Likert-type attitudinal scale. Items were derived from a larger initial pool of items and any item-to-scale correlation that was not equal to or greater than .50 was deleted from the final scale. Responses for each scale item were scored on a five-point scale (strongly agree, agree, undecided, disagree, strongly disagree). The higher the scale score on this measure, the more favorable one’s attitude toward the use of violence. The mean of this variable was 20.18, with a standard deviation of 3.72. It was expected that those inmates with a positive attitude toward violence would have a higher degree of prisonization. Prisonization is typically high among inmates who maintain a low expectation of their postrelease life chances (Thomas, Petersen, and Zingraff, 1978). The future expectations of inmates were measured by two variables: postrelease expectations and future life satisfaction.

433

Both measures indicate inmate perceptions as to whether or not imprisonment is viewed as so destructive that reintegration into family, social, and occupational roles is unlikely. Postrelease expectations was measured by a four-item Likert-type scale derived from a larger pool of initial items. The variable had a mean of 14.74, with a standard deviation of 2.24. The higher the scale score on this variable, the higher are inmate postrelease expectations. It was expected that those inmates with positive postrelease expectations would have a lower degree of prisonization. Postrelease expectations, however, may be too narrow a measure to assess the multidimensional character of future expectations. This research posed questions concerning the future expectations of seven dimensions of life satisfaction (physical health, economic security, family relationships, friendships, personal character, social involvements, and religion). Responses to each dimension were scored on a three-point scale of importance (very satisfied, satisfied, not satisfied). A total future life satisfaction index was created by summing the response products across the seven dimensions. All item-to-scale correlations were .50 or greater. The mean for this measure was 12.68, with a standard deviation of 2.47. High scores reflect a positive expectation toward one’s future life satisfaction. It was expected that those prisoners with low future life satisfaction expectations would be more integrated into the inmate social system.

Measures of self-conception. Sociological approaches to prisonization have suffered from an inadequate and underdeveloped conception of the inmate. The main reason is that researchers have failed to address inmate self-conceptions when examining assimilation into the inmate subculture. This research organized self-concept into three important dimensions: (1) self-evaluation, (2) identity salience, and (3) self-efficacy (Faine, 1973; Gecas, 1982; Gecas and Schwalbe, 1983). These dimensions of self are important to prisonization theory primarily because they provide a fuller and more adequate conception of the acting subject in the prisonization process.

(1) Prisonization (2) Race (3) Marital status (4) Education (5) Preprison employment status (6) Months employed before imprisonment (7) Income (8) Number of arrests (9) Number of times in prison (10) Age at first conviction (11) Type of offense (12) Alienation (13) Oppostion to institution (14) Attitude toward violence (15) Change in life satisfaction (16) Future life satisfaction (17) Postprison expectations (18) Present identity salience (19) Past identity salience (20) Self-efficacy (21) Stigmatization

Variables

2.04

2.067

.06 .229** .214* .05 2.09 .091 2.244** 2.107 .066 2.001 .084 .07 .091 .193 .061 .257

.139*

2.27** .037 .216** .112 2.229** 2.035 .36** .591** .529** .084 .167* 2.352** .197* 2.03 2.446** 2.311** .082 .093 2.003 2.148* .063 2.172* 2.026 2.033 2.055 .005 2.018 2.05 2.066 2.07 2.012 .029

1 .05

(3)

1 2.011 .147

(2)

1 .003 2.021 2.14*

(1)

.078 .11 2.111 2.109 .015 2.083 2.149* 2.003 2.023 .03 2.09 .08 .179* .139 .018 .187*

2.059

1

(4)

2.69** .053 .259* .127 2.206** 2.173* .026 .154* .188* .092 .04 2.126 .151 .188* 2.157 2.07

1

(5)

1 2.021 2.3** 2.214** 2.276** .107 2.072 2.248** 2.182* 2.081 2.08 .179 2.224** 2.23** .24* .136

(6)

(7)

1 .178* .003 .119 2.132 2.008 .057 .127 2.033 .004 .008 .154 .101 2.109 .139*

ZERO-ORDER CORRELATIONS AMONG VARIABLES (N 5 239)

TABLE 1

1 .462** 2.353** 2.08 2.012 .111 .144* 2.013 .088 2.087 .117 .082 2.135 2.02

(8)

1 2.424** 2.047 2.035 2.002 .097 .019 .097 .067 .051 .124 .047 2.01

(9)

1 2.014 .014 2.169** 2.194** 2.043 2.04 .128 .101 2.159 .139 .106

(10)

(Continued)

.08 2.009 2.031 2.053 2.02 2.036 2.078 2.002 2.044 2.11

(11)

434 B. A. PATERLINE and D. M. PETERSEN

*p , .05; **p , .01.

(1) Prisonization (2) Race (3) Marital status (4) Education (5) Preprison employment status (6) Months employed before imprisonment (7) Income (8) Number of arrests (9) Number of times in prison (10) Age at first conviction (11) Type of offense (12) Alienation (13) Oppostion to institution (14) Attitude toward violence (15) Change in life satisfaction (16) Future life satisfaction (17) Postprison expectations (18) Present identity salience (19) Past identity salience (20) Self-efficacy (21) Stigmatization

Variables

1 .509** .187* .049 .018 2.348** 2.249** 2.069 .529** .334**

(12)

1 .347** 2.164* 2.174* .398** 2.249 2.069 .529** .334**

(13)

1 .168* .062 2.134 .226** .103 2.302** .18*

(14)

1 .039 2.264** .299** .13 2.13 2.162

(15)

(CONTINUED)

TABLE 1

1 2.309** .036 2.002 2.088 2.029

(16)

1 .278** 2.03 .391** .398**

(17)

1 .519** 2.241** 2.064

(18)

1 2.009 .131

(19)

1 .288**

(20)

1

(21)

Determinants of Prisonization

435

436

B. A. PATERLINE and D. M. PETERSEN

The first important dimension of self-concept is self-evaluation. One would hypothesize the more stigmatized inmates perceive themselves to be, the more likely they will find comfort in the inmate subculture. In this research, perception of stigmatization was measured by a four-item Likert-type attitudinal scale. The higher the scale score was on this variable, the higher perception of stigmatization. The measure had a mean of 12.59 and a standard deviation of 2.18. A second important dimension of self-concept is identity salience. According to identity theorists the self is organized into various self-identities based on roles or relationships with others (Burke and Reitzes 1991; Callero, 1985; Stryker, 1968, 1980). The concept of identity salience was measured in this research by determining the importance of five socially valued identities: occupation, family, friendships, social involvements, and religion. Responses to each identity were scored on a three-point scale (very important, important, not very important). A measure of total present identity salience was created by summing the products of the importance responses across each of the five different identities. The present identity salience index had a mean of score 8.89 and a standard deviation of 2.05. High scores reflect a high importance given to many different social identities. It was expected that inmates who placed a high importance on many social identities, would be less likely to become socialized into the inmate subculture. A measure of total preprison identity salience was also created by summing the products of the importance responses across five different past social identities: occupation, family, friendships, social involvements, and religion. Responses to each identity were scored on a three-point scale (very important, important, not very important). The preprison identity index had a mean of score of 9.14 and a standard deviation of 2.20. High scores for the index reflect a high importance given to many different past social identities. It was expected that inmates who placed a high importance on many past identities would be less likely to become socialized into the inmate subculture. A third dimension of self-concept is self-efficacy. As a dimension of self-concept, self-efficacy

refers to how individuals conceptualize themselves as active persons who have control over their world. Previous research has shown that these beliefs have behavioral consequences. Those individuals who believe that they maintain control of their destinies are more likely to take steps to improve the environmental condition and be more resistant to attempts to influence them (Bandura, 1977, 1982; DeCharms, 1968; Gecas and Schwalbe, 1983). One would therefore expect that persons with a high selfefficacy are less likely to be socialized into the inmate subculture. Self-efficacy was measured by a six-item Likert-type scale. Responses for each scale item were scored on a five-point scale (strongly agree, agree, undecided, disagree, strongly disagree). The higher the scale score on this variable, the higher self-efficacy was rated. The scale had a mean of 18.70 and a standard deviation of 3.20.

RESULTS The zero-order correlation results for all the variables used in the study are presented in Table 1. In general, the independent variables measuring deprivation appear to be the strongest predictors of prisonization. Inmates were more likely to be prisonized, the greater their opposition to the prison institution and staff (r 5 .591), the more positive their attitudes toward the use of violence (r 5 .529), the greater their feelings of alienation (r 5 .360), and the lower their postrelease expectations (r 5 2.352). Inmates who had a higher future life satisfaction expectation, also had greater degrees of prisonization (r 5 .167). This relationship was opposite of the expected direction. One’s change of life satisfaction (from past to present) was not significant in predicting prisonization (r 5 .084). Of the importation variables, two measures of criminal history were found to be significant predictors of prisonization. The more times one had been arrested (r 5 .216) and the younger one was at first conviction (r 5 2.229), the greater the degree of prisonization. Type of offense (r 5 2.035) and the number of times one was placed

437

Determinants of Prisonization

in an correctional institution (r 5 .112) were not significant in predicting prisonization. There were also several measures of preprison employment history that were significant predictors of prisonization. The greater the number of months an inmate was employed before entering prison, the lower the degree of prisonization (r 5 2.270). Those inmates who were employed at the time of their arrests (r 5 .139) had a greater degree of prisonization. This relationship was significant, but was opposite of the expected direction. Educational attainment was also found to be a significant predictor of prisonization. Those inmates with higher education attainment had lower degrees of prisonization (r 5 2.140). Other importation variables measuring social status were not found to be significant predictors of prisonization, such as race (r 5 .003), marital status (r 5 2.021), and preprison income (r 5 2.037). The correlation matrix in Table 1 shows that of the measures of self-conception, self-efficacy is the most important predictor of prisonization. Those who had greater feelings of self-efficacy had a lower degree of prisonization (r 5 2.446). As predictors of prisonization, perception of stigmatization and present identity salience were opposite of the expected direction. Those inmates who had higher feelings of stigmatization had lower degrees of prisonization (r 5 2.311) and those inmates who placed greater importance on several valued social identities had higher degrees of prisonization (r 5 .197). Past identity salience was not a significant predictor of prisonization (r 5 2.030). In order to fully examine the influence of the independent variables on inmate prisonization, a multivariate regression analysis was performed. For the regression analysis, the full range of scores on the dependent variable of inmate code adherence were used, dichotomized variables were treated as dummy variables, and ordinal measures were treated as interval. Only standardized regression coefficients were used in order to assess the relative importance of each independent variable. A stepwise multivariate regression analysis was used to determine the importance of each set of variables as predictors of inmate prisonization.

Three regression models are presented in Table 2. In the first regression model prisonization was regressed upon ten measures of importation. The total amount of variance explained by the importation model was 13 percent (r 2 5 .130). The results of the first regression analysis show that number of months employed before one is arrested is the strongest and most significant predictor of prisonization. The greater the number of months one worked before arrest, the lower the his/her degree of prisonization (b 5 2.271). Age of first conviction and educational attainment were also significant in predicting prisonization. The older one was at his/her first conviction (b 5 2.175) and greater one’s educational attainment (b 5 2.125), the lower his/ her degree of prisonization. The results show that if an inmate was not employed at the time of his/her arrest the greater is his/her adoption of the inmate code (b 5 2.122). Those inmates who had numerous

TABLE 2 REGRESSION COEFFICIENTS FOR SUBSCRIPTION TO INMATE CODE (STANDARDIZED)

Equation with Subscription to Inmate Code Independent Variables Race Marital status Educational attainment Preprison employment status Months employed before imprisonment Preprison income Number of arrests Number of times in prison Age at first conviction Type of offense Alienation Opposition to institution Attitude toward violence Change in life satisfaction Future life satisfaction Postprison expectations Present identity salience Past identity salience Self efficacy Stigmatization R2 R2 n *p , .05; **p , .01.

(1)

(2)

(3)

2.019 .057 2.006 .027 2.125* 2.108

.083 .022 2.104

2.122

2.121

2.108

2.271* .055 .116 2.076 2.175* 2.027

2.142* 2.050 .064 .026 2.034 2.049 .057 .37** .363** 2.076 .011 2.127*

.130

.532 .402

2.14* 2.051 .043 .049 2.039 2.051 .02 .348** .352** 2.081 .016 2.098 .058 2.12* 2.082 2.037 .551 .019

438

B. A. PATERLINE and D. M. PETERSEN

arrests were also more prisonized than those with low arrest rates (b 5 .116). These two relationships were in the expected direction but were not significant at the .05 level. Several measures of importation were not important predictors of inmate code adoption. Type of offense (b 5 2.027), preprison income level (b 5 .055), inmate marital status (b 5 2.006), race (b 5 2.019), and number of times in a correctional institution (b 5 2.076) were not related to inmate prisonization. The second regression model presented in the second column of Table 2 incorporates six measures of deprivation with the eleven measures of importation. This importation/deprivation model is better in explaining prisonization than the importation model alone. When the deprivation model variables were entered into the equation, the total amount of variance explained by the model increased to 53 percent (r2 5 .532); a change in r2 of 40 percent. The results show that the deprivation model variables account for more of the variance in prisonization than the importation model variables. The deprivation model variables account for 40 percent of the explained variance in the dependent variable, while the importation variables alone only account for 13 percent of the explained variance in the dependent variable. Among the deprivation model variables, the two strongest and most significant predictors of prisonization were inmate attitudes toward the institution and inmate attitudes toward the use of violence. Those inmates who had greater oppositional attitudes toward the institution had higher degrees of prisonization (b 5 .370) and inmates who had supportive attitudes toward the use of violence also had higher degree of prisonizations (b 5 .363). Those inmates who had lower postrelease expectations also had higher rates of prisonization (b 5 2.127). There were several measures of deprivation that were not important predictors of prisonization. Change in life satisfaction (b 5 2.076), future life satisfaction (b 5 .011) and contextual alienation (b 5 .057) were not significantly related to the degree of inmate prisonization. The third regression model incorporated four measures of self-conception into the importation/deprivation model. This model is presented

in the third column of Table 2. The total amount variance explained by the model was 55 percent (r 2 5 .551). A change in r 2 of only 2 percent. Past identity salience was the only measure of self-conception that was a significant predictor of prisonization (b 5 2.120). Inmates who placed greater importance on several past social identities had lower degrees of prisonization. Present identity salience (b 5 .058), feelings of stigmatization (b 5 2.037), and feelings of self-efficacy (b 5 2.082) were not important predictors of prisonization when entered into the regression model.

DISCUSSION AND CONCLUSION The findings of this investigation coincide with some previous research, which concludes that the merger of the importation and deprivation models explains more of the variation in prisonization than either analyzed separately (Leger and Barnes, 1986; Thomas and Petersen, 1977; Thomas, Petersen, and Zingraff, 1978; Zingraff, 1980). Measures of deprivation in the current study were also more important predictors of the degree of prisonization than were measures of importation. This finding was consistent with other researchers findings (Akers, Hayner, and Gruninger, 1977; Thomas, 1977a, 1977b; Thomas, Petersen, and Zingraff, 1978; Zingraff, 1980). These findings seem congruent with past prisonization studies, though there are two major differences. First, the deprivation variable, alienation, was not an important determinant of prisonization. This finding contradicts several past studies, which found contextual alienation or powerlessness to be one of the more important variables in predicting prisonization (Guenther, 1978; Hyman, 1977; Neal, Snyder, and Balogh, 1974; Thomas, 1975; Thomas and Poole, 1975; Thomas and Zingraff, 1976; Tittle and Tittle, 1964). Second, this research found there is a strong positive relationship between holding an approving attitude toward violence and adoption of the inmate code. Inmate attitudes toward violence have never been used in past research as a predictor of prisonization. In fact, attitudes toward violence and opposition toward the institu-

Determinants of Prisonization

tion were the most important variables in predicting prisonization among this sample of inmates. Literature on correctional institutions acknowledges that an inmate’s self-conception is an important factor in understanding one’s adoption of the inmate code (Faine, 1973; Wright, 1991; Zamble and Porporino, 1988). Subsequent research, however, has not substantiated this theoretical connection in the prisonization process. One of the purposes of this research was to assess the importance of inmate self-conceptions in the prisonization process. This was attempted by implementing four measures of self-conception: present identity salience, preprison identity salience, self-efficacy, and perception of stigmatization. The measures of self-conception used in this research, however, did not significantly contribute to an understanding of prisonization. One reason measures of self-concept were found to be unimportant may be that the measures of self-conception utilized in this research were based upon extraprison identities and standards. Prisons, however, are different from other socialization experiences in that inmates may totally lose or repress their extraprison sense of self. Like bootcamps, mental hospitals, and other types of “total institutions,” prisons are designed to resocialize and change inmates’ sense of self. The prison experience is a depersonalizing process in which inmates are stripped of their civilian or preprison identities. They are denied outside roles and possessions, put in uniforms, shaved, given numbers, and subjected to other ceremonies of depersonalization (Goffman, 1961). A new self, adapting to the situation, is gradually reconstructed, but unlike the old self it is based on the master status trait of prisoner. Preprison roles may no longer be meaningful, rather, inmates may take new roles and identities such as the punk, merchant, outlaw, or politician, which allow them to cope and survive in the inmate society. Measures of self-conception based on extraprison identities and statuses may be invalid measures and may prove to be unimportant in predicting prisonization. Future research needs to further investigate the roles and identities inmates adopt while incarcerated and how these identities are related to the process of prisonization.

439

To further an understanding of the prisonization process, research needs to address the consequences of prisonization in terms of postrelease adjustment. If an inmate enters a prison and becomes well integrated into the inmate society, how will this socialization process affect his/her postrelease life? Future studies should examine the extent and ways in which prisonization influences postrelease adjustments and whether or not prisonization has a permanent effect on the values, attitudes, and behavior of inmates. This cannot be accomplished, however, through a simple cross-sectional research design. Research needs to be longitudinal and should examine inmates throughout their institutional stay and their postrelease life. The general contention is that the effects of prisonization are long-term and encourage inmates to become hostile to the legal system, to employ physical violence as a means of solving problems, and to accept and value interpersonal associations with those who engage in criminal activity. If this is true, one might expect those inmates who are most integrated into the inmate subculture to have the highest rates of recidivism and the most problems adjusting to the postprison world. On the other hand, prisons may also serve as “schools of crime” where inmates may receive vocational training from experts in the field. In prison, inmates can learn how to forge checks, crack safes, avoid deception by police, and generally enhance their criminal skills. If this is the case, then rates of recidivism among the most prisonized inmates may be relatively low because they would be less likely to be caught by law enforcement officers. An alternative view is that prisonization is only a short-term adjustment to the immediate problems and pressures of confinement made by individuals who almost always know that they will someday return to the outside world. In this case, the attitude and behavior changes caused by prisonization may be only temporary and may not permanently effect an inmate’s postrelease life. Inmate homosexual activity, for example, is found in virtually all correctional institutions, however, few researchers and few inmates believe such behavior is a reflection of “real” or permanent homosexuality. Rather

440

B. A. PATERLINE and D. M. PETERSEN

most view it as a normal adaptation to an abnormal situation, and most inmates return to their previous heterosexual lifestyles after release. Whether the effects of prisonization are permanent or temporary cannot be proven without a sufficient longitudinal study. Much of the research based on the concept of prisonization has mistakenly taken a narrow view of the factors that influence what happens inside the prison. Prison organizations are greatly influenced by a variety of external influences including the political context in which they operate and the criminal subcultural patterns of the outside world such as gang membership and drug use. This has led several researchers to challenge the old conception of the prison social system. Marquart and Roebuck (1985), for example, found that the norms against snitching and ratting do not always apply. Rameriz (1984) concluded that staff and inmates are not necessarily opposed on many important issues and Stastny and Tyrnauer (1982, p. 35) see prisons as “detotalizing” power bases in which control is shared between the inmates, the warden, the guards, the courts, and a number of other external groups, which monitor prison life. Such studies may suggest that the inmate subculture is changing. Future research needs to address the fact that prisons may no longer be characterized by one dominate inmate social system. This may be due to the fact that the character of correctional populations has changed substantially. Current inmate populations consist of a more diverse group of offenders, serving longer sentences, for a wider range of offenses. There are greater concentrations of mentally disabled offenders, drug and alcohol abusers, young violent offenders, and racial and ethnic gangs. Modern prisons may therefore be characterized by a number of small segmented, mutually exclusive gangs or groups, which may be held together by the contraband economy, but not by other values. The current inmate populations are increasingly divided along gang lines, most of which are based on street gang origins and are racial or ethnic in nature. These gangs frequently engage in power struggles for control of the institution, fighting among themselves and attacking prison officials and correctional offic-

ers. This new type of fragmented inmate society may be more violent and less stable than those of the past. This presents a problem for future researchers who desire to find or conduct research on a common pervasive inmate subculture. If this is the case, prisonization may be becoming too general and too crude as a construct.

REFERENCES Akers, R. L. (1977). Type of leadership in prison: A structural approach to testing the functional and importation models. Sociological Quarterly 18(3):378–83. Akers, R. L., Hayner, N. S., and Gruninger, W. (1977). Prisonization in five countries: Type of prison and inmate characteristics. Criminology 14(4):527–54. Alpert, G. P. (1979). Patterns of change in prisonization: A longitudinal analysis. Criminal Justice and Behavior 6:159–74. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review 84:191–215. Bandura, A. (1982). Self-efficacy mechanisms in human agency. American Psychologists 37:122–47. Burke, P., and Reitzes, D. (1991). An identity theory approach to commitment. Social Psychology Quarterly 54:239–51. Callero, P. L. (1985). Role-identity salience. Social Psychology Quarterly 48:203–15. Clemmer, D. (1940). The prison community. New York: Holt, Rinehart and Winston. Cline, H. F. (1968). The determinants of normative patterns in correctional institutions. In Scandinavian studies in criminology, ed. N. Christie. Oslo: Scandinavian University Books. DeCharms, R. (1968). Personal causation. New York: Academic Press. Faine, J. R. (1973). A self-consistency approach to prisonization. Sociological Quarterly 14(4):576–88. Gecas, V. (1982). The self-concept. Annual Review of Sociology 8:103–12. Gecas, V. (1986). The motivational significance of self-concept for socialization theory. Advances in Group Processes 3:131–56. Gecas, V., and Schwalbe, M. L. (1983). Beyond the looking-glass self: Social structure and efficacy-based selfesteem. Social Psychology Quarterly 46(2):77–88. Goffman, E. (1961). Asylums: Essays on the social situation of mental patients and other inmates Chicago, IL: Aldin. Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision of personal history. American Psychologist 35:603–18. Gruninger, W. (1975). Criminal maturity, prison roles, and normative alienation. Free Inquiry 3(1):38–63. Guenther, A. L. (1978). Alienation, inmate roles, and release ideology in a penitentiary setting Ann Arbor, MI: University Microfilms International. Hyman, J. M. (1977). Alienation and prisonization. Criminology 15(2):263–65. Irwin, J. (1970). The felon. Englewood Cliffs, NJ: PrenticeHall. Irwin, J. (1980). Prisons in turmoil. Boston, MA: Little, Brown & Company.

Determinants of Prisonization

Irwin, J., and Cressey, D. (1962). Thieves, convicts, and the inmate culture. Social Problems 10:142–55. Jacobs, J. B. (1976). Stratification and conflict among prison inmates. Journal of Criminal Law and Criminology 66:467–82. Jensen, G. F., and Jones, D. (1976). Perspectives on inmate culture: A study of women in prison. Social Forces 57:57–78. Kalinich, D. B., Stojkovic, S., and Klofas, J. (1988). Toward a political-community theory of prison organization. Journal of Criminal Justice 16(3):217–30. Kennedy, W. C. (1970). Prisonization and self-conception: A study of a medium security prison. Ph.D. diss., University of California, Los Angeles. Leger, R. G. (1981). Labeling and its consequences in a closed social system. British Journal of Criminology 21:109–22. Leger, R. G., and Barnes, H. G. (1986). Black attitudes in prison: A sociological analysis. Journal of Criminal Justice 14:105–22. Marquart, J. W., and Roebuck, J. B. (1985). Prison guards and snitches: Deviance within a total institution. British Journal of Criminology 3:217–33. Morris, T., and Morris, P. (1963). Pentonville, a study of an English prison London: Routledge and Kegan Paul. Neal, A. G., Snyder, E., and Balogh, J. K. (1974). Changing alienations as consequences of imprisonment. Paper presented at the Annual Meeting of Social Problems. Montreal, Canada. Rameriz, J. (1984). Prisonization, staff, and inmates: Is it really about us versus them? Criminal Justice and Behavior 11:423–60. Rosenberg, M. (1979). Conceiving the self. New York: Basic Books. Schwartz, B. (1971). Pre-institutional versus situational influence in a correctional community. Journal of Criminal Law, Criminology, and Police Science 63:532–43. Smith, C. F., and Hepburn, J. R. (1979). Alienation in prison organizations: A comparative analysis. Criminology 17:251–62. Stastny, C., and Tyrnauer, G. (1982). Who rules the joint: The changing political culture of maximum security prisons in America. Lexington, MA: Lexington Books. Stryker, S. (1968). Identity salience and role performance. Journal of Marriage and the Family 30:558–64. Stryker, S. (1980). Symbolic interactionism: A social structural version. Menlo Park, CA: Benjamin/Cummings Publishing Company. Sykes, G. M. (1958). The society of captives Princeton. New York: Princeton University Press. Sykes, G. M., and Messinger, S. (1960). The inmate social system. In Theoretical studies in social organization of

441

the prison, ed. R. Cloward. New York: Social Science Research Council. Thomas, C. W. (1973). Prisonization or resocialization. Journal of Research in Crime and Delinquency 10:13–22. Thomas, C. W. (1975). Theoretical perspectives on alienation in the prison society. Pacific Sociological Review 8(4):483–99. Thomas, C. W. (1977a). Theoretical perspectives on prisonization: A comparison of the importation and deprivation models. Journal of Criminal Law and Criminology 68:135–45. Thomas, C. W. (1977b). Prisonization and its consequences: An examination of socialization in a coercive setting. Sociological Focus 10:53–63. Thomas, C. W., and Cage, R. J. (1977). Correlates of prison drug use: An evaluation of two conceptual models. Criminology 15:193–210. Thomas, C. W., and Petersen, D. M. (1977). Prison organization and inmate subcultures. Indianapolis, IN: BobbsMerrill. Thomas, C. W., and Poole, E. D. (1975). The consequence of incompatible goal structures in correctional settings. International Journal of Criminology and Penology 3:27–42. Thomas, C. W., and Zingraff, M. T. (1976). Organizational structure as a determinant of prisonization: An analysis of the consequences of alienation. Pacific Sociological Review 19(1):98–117. Thomas, C. W., Petersen, D. M., and Zingraff, R. M. (1978). Structural and social psychological correlates of prisonization. Criminology 16(3):383–93. Tittle, C. R. (1972). Institutional living and self-esteem. Social Problems 20:65–77. Tittle, C. R., and Tittle, D. P. (1964). Social organization of prisoners: An empirical test. Social Forces 43:216–21. Wellford, C. (1967). Factors associated with adoption of the inmate code. Journal of Criminal Law, Criminology, and Police Science 58:197–203. Wheeler, S. (1961). Socialization in correctional communities. American Sociological Review 26:697–712. Wright, K. N. (1989). Race and economic marginality in explaining prison adjustment. Journal of Research in Crime and Delinquency 26:67–89. Wright, K. N. (1991). A study of individual environment, and interactive effects in explaining adjustment to prison. Justice Quarterly 8:217–42. Zamble, E., and Porporino, F. J. (1988). Coping, behavior, and adaptation in prison inmates New York: SpringerVerlag. Zingraff, M. T. (1980). Inmate assimilation: A comparison of male and female delinquents. Criminal Justice and Behavior 7:275–92.