When programs "don't work" with everyone: planning for differences among correctional clients.
The need to accommodate important differences among clients is not a new idea. In the 1960s and 1970s, programs within the California Youth Authority and elsewhere used personality-based classification models to "match" clients to interventions. Evaluators of such applications of "differential treatment" found they enhanced program effectiveness. More recently, "differential treatment" has become subsumed under the Responsivity Principle,(4) which is receiving growing attention in corrections. The responsivity principle suggests that clients should be assigned to those interventions that are most able to meet their needs and styles of learning.(5) The California programs classified delinquents according to personality types, but personality is not the only individual characteristic that should be considered in program assignments; intellectual functioning, maturity, stages of change, sex and ethnicity also are considered to be relevant to client responsivity.
Effective application of the responsivity principle requires sound methods of assessment of responsivity traits such as those listed above. However, such assessments are viewed as secondary to risk assessment systems. They are used as "internal classifications" - the classification system applied by correctional facilities once offenders are in their care - recognizing that even when groups are separated according to risk, they are still highly heterogeneous.
The personality-based or psychological classification systems have the longest history of systematic practice in corrections. Examples of such systems include: a) Client Management Classification (CMC)(6), b) I-level,(7) c) Megargee MMPI-based Prison Typology(8) and d) Quay Behavioral Categories.
The following proposes a personality-based model for adult male offenders. This model represents a simplified version of personality-based classification models. It suggests five offender types and puts forward treatment implications for each. The types are common to each of the personality systems listed above. Although all but the CMC and the Quay behavioral typology enumerate more than five types, recent research on adult, male, prison inmates suggests that it is possible to simplify them and still make meaningful distinctions among offenders. This requires collapsing types when using the systems that typically recommend nine or more types. While this has some support in the corrections literature, these suggestions are still being tested in a number of studies currently under way at the University of Cincinnati.
The purpose of proposing these five types and their treatment recommendations is to: a) aid correctional practitioners in making meaningful distinctions among clients in their daily interactions, b) assist in the process of matching offenders to meaningful treatment programs and c) offer a responsivity-based classification methodology that is not overly complex. In today's correctional environment, a classification system overburdened with complex assessment methods and numerous types with complicated distinctions is not likely to be a realistic option for most agencies.
This article also makes treatment recommendations for each type. Readers are cautioned, however, that differential treatment/responsivity is not widely practiced or researched in correctional practice. We draw our treatment recommendations from: a) the few studies known to have used differential treatment to its full potential, b) recommendations offered by the designers of the classification systems and c) inferences drawn from the type descriptions. This article draws heavily from results of programs such as the Community Treatment Project(9) and other differential models as well as studies that have noted differential adjustments to correctional settings. Thus, we are discussing a classification and treatment system that is grounded in the research literature, but one that clearly will need further testing.
Differential Treatment and Personality-based Classification
There are six key assumptions underlying differential treatment: a) correctional populations can be classified into relatively homogeneous subgroups, b) each subgroup poses different treatment or supervision needs, c) classification helps correctional personnel identify treatment and supervision strategies most appropriate to each group, d) failure to differentiate can result in some offenders being assigned to inappropriate supervision and treatment strategies, e) prospects for successful intervention are increased by the assignation of offenders to appropriate treatment and f) failure to examine evaluation results according to differences among offenders can mask the treatment effect (i.e., successes are cancelled by failures).(10)
The type descriptions that follow should not be viewed as a substitute screen for mental illness, sex offending, substance abuse or violence. Moreover, while the classification systems are described as psychological, they are not intended to be diagnostic of mental health disorders. A "type" simply describes personality, interaction styles and crime dynamics. If we refer to a type as "neurotic," for example, we are describing an anxiety-prone individual, but the description is not intended to be a clinical diagnosis of neurosis.
Personality Types and Treatment Implications
The first personality type is the committed criminal. Individuals classified into this type fit social stereotypes of criminals. They are ones for whom most correctional policies are designed. Closer observation, however, reveals that this type of offender comprises no more than 45 percent of a federal prison population, depending upon type of institution and type of classification instrument.
As the label denotes, this offender is not uncomfortable with the label of "criminal." He espouses anti-social values and exhibits a greater tendency to lie than other offenders. This offender does not suffer from self-esteem problems, is self-centered and an exploiter of others, and has an extensive prior record. He tends to have more criminal peers than other offenders, and much of his antisocial behavior is learned and rewarded by peers. This offender can be aggressive and tends to demonstrate many of Stanton Samenow's "thinking errors" - blaming victims, exploiting others' trust and expecting to be successful with minimal effort.
Although the asocial tendencies of this group have caused some to label them "untreatable," a number of appropriate treatment strategies have emerged in recent years. Common to these newer interventions is an attempt to shift offender values and thinking patterns that support or even cause criminal behavior. Other recommended goals of treatment include altering situations that reward anti-social behavior, teaching respect for personal boundaries and keeping responsibility for behavior on the offender. Furthermore, offenders classified into this type may be more amenable to confrontational modes of therapy than others.
The second personality type is the neurotic acting-out offender. If we focus on the neurotic acting-out dimensions of this offender, we note an offender whose behavior is a defense against anxiety. While these offenders verbalize a need for change, they actually have little tolerance for unraveling or talking about problems. Anxiety in these offenders may result in substance abuse for medication of feelings and may instill a strong need for others to see them as calm and in control. This offender also uses grandiosity and super-adequacy as a cover-up. In prison settings, staff may be tested for whether they can be trusted. Furthermore, since this offender's anxiety is linked to his criminality, careless confrontation can escalate acting-out behaviors, rather than curb them. This offender also demonstrates numerous disciplinary infractions and there is potential for violent and explosive behavior.
In earlier studies, we found many of the offenders classified as Able or Delta, two types on the Megargee-based prison typology to be in this group. Thus, while these inmates had 4,9 or spike 4 MMPI profiles, often associated with Antisocial Personality Disorder (psychopathy), their MMPI classifications were found to correlate with anxiety scores on other assessments (e.g., the Jesness Inventory and I-level interview method). This is not entirely inconsistent with the research on psychopathy, which sometimes refers to secondary neurotic qualities, or secondary psychopathy, as individuals who behave in a psychopathic manner, but demonstrate defended, highly anxious qualities as well (primary psychopaths are characterized as demonstrating no anxiety, even in situations that typically would evoke great anxiety).
The primary focus of treatment for this offender has been the reduction of use of destructive defense mechanisms. In addition, we recognize that emotionality and anxiety must be addressed in ways that acknowledge this individual's disdain for discussion of personal problems. At the same time, communication should be open and direct to avoid the offender's fearful misinterpretations. Programs such as Anger Control,(11) Aggression Replacement Training(12) and Cognitive Skills(13) seem especially applicable.
Practitioners also may benefit from the notion of "counseling on the run," where insights are given in passing, as carefully planted seeds for further thought, rather than in long, drawn-out discussions. While confrontation, or paying too much attention to "what's wrong," is not likely to prove helpful, it may be helpful for counselors to model healthy expressions of feelings, in ways that "normalize" anger, fear, frustration and vulnerability.
Unlike neurotic acting-out types, the neurotic anxious offender is more likely to manage anxiety through self-control and introspection. Defenses tend toward intellectualization and superficial insights into problems and fears. Counselors view neurotic anxious offenders as more likely to become "stuck" in dysfunctional patterns of behavior, due primarily to fear and anxiety. Within this group, instances of substance abuse usually are derived from the use of medication for these insecurities. Offenders classified as neurotic anxious also may have problems trusting others, and these issues may cause problems with others in incarceration settings. As inmates, they require programmatic support. They often experience high stress in prison settings and their frustration can lead to aggression.
While many of the treatment recommendations for the neurotic acting-out offenders apply to this type as well, there are some specific directions to consider. For example, this may be a group of offenders which can benefit from insight-oriented therapy, because they are more willing to explore feelings and discuss problems. Thus, court-ordered therapies (if possible) and family therapy seem more amenable to this group of offenders than for others. For similar reasons, support groups make some sense. In addition, cognitive therapies designed to address self-defeating cognitive patterns and irrational fears also are appropriate.
Situational offenders can comprise a substantial proportion of the inmate population, especially in minimum security settings. On the surface of these type descriptions, we observe rather benign behaviors and personality patterns. For example, situational inmates tend to have a positive self-concept although they are not comfortable with the criminal label or with their crimes. They endorse pro-social values, evidence stable relationships with others, and have minimal or no prior records. Their criminal behavior occurs during periods of adversity and is situation-specific rather than indicative of a life style.
Further examination of this offender, however, reveals adjustment difficulties in prison and clear treatment implications. For example, stress and fear have been noted in maximum security settings and these inmates often receive many official citations for insubordination, which may be attributable to limited prison experience. Since this offender's criminal behavior typically results from some adversity, treatment should focus on problem-solving and coping skills. Failure to recognize this inmate is one of the tragedies of our current correctional policies. In including situational offenders in interventions that address criminal thinking patterns and values, for example, we may be inadvertently teaching such values, worsening the prognosis of an individual who, in fact, has a very good chance of succeeding.
The fifth personality type is the dependent offender. We do not have a clear picture of this type, because immature types failed to converge in a clear manner in the construct validity tests conducted in the prison classification study noted earlier. Our suggestions, therefore, emerge from a limited array of tests. Given this caveat, we offer a picture of this offender as one who displays poor emotional control in prison settings and has a tendency to use drugs and alcohol while incarcerated. This inmate may become institutionalized and incur numerous adjustment difficulties. He or she also may exhibit high fear and stress in incarceration settings and become vulnerable to more aggressive inmates. Most research pertaining to this type of offender suggests that they benefit from a highly structured environment where rules and expectations are clear.
Assessment and Classification Strategy
How do correctional clients receive one of the classifications listed above? The systems vary in terms of their assessment procedures: the Jesness Inventory is a 155-item paper and pencil test administered to correctional clients; CMC is a semi-structured interview; the I-level interview is a semi-structured interview and a clinical rating process which, due to its length, is seldom used anymore; the Megargee MMPI-based Typology requires administration of the Minnesota Multiphaisic Personality Inventory (scoring rules for transferring MMPI profiles into the Megargee typology have been changed to fit the MMPI2 and will be published in the near future); and the Quay Adult Internal Management System (AIMS) is a behavioral checklist completed by staff.
The literature often faults the complicated nature of psychological classification processes which in turn generates a host of reliability, validity and resource problems. Having used all of the above systems except CMC, however, we have learned that difficult is not necessarily better. The more efficient systems were just as adequate in terms of their reliability and validity as systems that required more expenditure of staff resources.
Readers familiar with psychological classification in corrections will recognize the type descriptions offered above. Scholars also note theoretical(14) and empirical similarities across different personality-based typologies. Treatment recommendations, however, are largely hypothetical and will require further research. This is primarily due to the fact that correctional practice has not begun to adequately use the responsivity principle.
Research currently is being conducted at the University of Cincinnati to further our understanding of these types and to expand upon their treatment recommendations. These studies use the psychological typologies to examine the relationship between these types and recidivism; differentiate among child molesters; and understand offenders' amenability to correctional programming.
In addition, we are conducting a large, statewide evaluation of cognitive skills and cognitive restructuring programs which, among other objectives, seeks to determine whether the programs are more successful with some offenders than with others. Greater attention to differential effectiveness also promises better program evaluation results, because such evaluations no longer "mask" the treatment effect. When we ask for whom did this program work, we are likely to find that it helped some more than others or that it "worked" with some and not others. When we ask, instead, whether it worked with everyone, as we do in most evaluations, we are likely to find disappointing outcomes, because we have no way of knowing whether the results of our successes are being cancelled out by our failures.
As noted earlier, only CMC and the Quay AIMS list five types; the other systems vary from nine to 13 types. In these cases, types are collapsed as follows:
Jesness Inventory: Cultural Conformist (CFC) and Manipulator (MP)
Megargee MMPI-based Typology: Charlie and Foxtrot
Quay AIMS: Asocial Aggressive and Manipulator
I-level Interview: Cultural Identifiers
CMC: Limit Setting
Jesness Inventory: Neurotic Anxious (NX)
Megargee MMPI-based Typology: Baker, George, Jupiter
Quay AIMS: Neurotic Anxious
I-level Interview: Neurotic Anxious (NX)
CMC: Casework Control
Jesness Inventory: Situation (SE) and Adaptive (CI)
Megargee MMPI-based Typology: Easy and Item
Quay AIMS: Situational
I-level Interview: Situation (SE)
Dependent (requires further testing):
Jesness Inventory: Asocial Passive (AP) and Immature Conformist (CFM)
Megargee MMPI-based Typology: How, Baker (perhaps) (requires further testing)
Quay AIMS: Immature Dependent
I-level Interview: Immature Conformist
CMC: Environmental Structure
Neurotic Acting Out:
Jesness Inventory: Neurotic Acting-Out (NA)
Megargee MMPI-based Typology: Able and Delta
Quay AIMS: none
I-level Interview: Neurotic Acting-Out (NA)
1 Palmer, 1992
2 Gendreau and Ross, 1987; Palmer, 1992; Van Voorhis, 1987
3 Carbonnel, Megargee, and Moorehead, 1984; Megargee and Bohn, 1979; Van Voorhis, 1994
4 Andrews, Bonta and Hoge, 1990
5 Andrews and Bonta, 1994
6 Learner, Arling and Baird, 1986
7 Jesness, 1996: Warren, 1966
8 Megargee and Bohn, 1979
9 Palmer, 1974; Palmer, forthcoming, Warren, 1966; Warren, 1983
10 Andrews, Bonta, and Hoge, 1990; Gendreau and Ross, 1987; Palmer, 1992; Van Voorhis, 1987, 1994
11 Novoco, 1975
12 Goldstein and Keller, 1987
13 Ross & Fabiano, 1985
14 Warren, 1971
Andrews, D. and J. Bonta. 1994. The psychology of criminal conduct. Cincinnati: Anderson Press.
Andrews, D., J. Bonta and R. Hoge. 1990. Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17(1): 19-52.
Carbonnel, J., E. Megargee and K. Moorhead. 1984. Predicting prison adjustment with structured personality inventories. Journal of Consulting and Clinical Psychology. 52: 280-294.
Gendreau, P. and R. Ross. 1987. Revivification of rehabilitation: Evidence from the 1980s. Justice Quarterly, 4(3): 349409.
Goldstein, A. and B. Glick. 1987. Aggression replacement training. Champaign, III.: Research Press.
Hare, R. 1970. Psychopathy: Theory and research. New York: Wiley.
Jesness, C. 1971. The preson typology study: An experiment with differential treatment in an institution. Journal of Research in Crime and Delinquency, 8:38-52.
Jesness, C., W. DeRisi, P. McCormick and R. Wedge. 1972. The youth center research project: Final report. Sacramento, Calif.: California Youth Authority and American Justice Institute.
Jesness, C. 1996. The Jesness inventory manual. North Tonowanda, N.Y.: Multi-Health Systems.
Learner, K, G. Arling and S. Baird. 1986, Client management classification strategies lot case supervision. Crime and Delinquency. 32(3):254-271.
Levinson, R. 1988. Development in the classification process. Criminal Justice and Behavior, 15(1):24-38.
Martinson, R. 1974. What works: Questions and answers about prison reform. The Public Interest. 35:22-54.
Megargee, E. and M. Bohn. 1979. Classifying criminal offenders: A new system based on the MMPI. Beverly Hills, Calif.: Sage Publications.
Novaco, R. 1975. Anger control: The development and evaluation of an experimental treatment. Lexington, Mass.: Lexington Books.
Palmer, T. 1974. The youth authority's community treatment project. Federal Probation, 38(1): 3-14.
Palmer, T. 1992. The re-emergence of correctional intervention. Newbury Park, Calif.: Sage.
Palmer, T. (forthcoming). Individualized intervention with young multiple offenders: The California community treatment project. Hampton, Conn.: Garland Press.
Quay, H. 1984. Managing adult inmates: Classification for housing and program assignments. College Park, Md.: American Correctional Association.
Quay, H. and L. Parsons. 1972. The differential behavioral classification of the juvenile offender. Washington, D.C.: Department of Justice.
Ross, R. and E. Fabiano. 1985. Time to think: A cognitive model of delinquency prevention and offender rehabilitation. Johnson City, Tenn.: Institute of Social Science and Arts.
Samenow, S. 1984. Inside the criminal mind. New York: Times Books.
Van Voorhis, P. 1987. Correctional effectiveness: The high cost of ignoring success. Federal Probation, 51 (1): 56-62.
Van Voorhis, P. 1994. Psychological classification of the adult male, Prison Inmate, Albany, N.Y.: SUNY Press.
Warren, M. 1971. Classification of offenders as an aid to efficient management and effective treatment. Journal of Criminal Law, Criminology and Police Science, 62: 239-268.
Warren, M. 1983. Application of interpersonal maturity theory to offender populations. In W. Laufer and J. Day, eds., Personality theory, moral development and criminal behavior. Lexington, Mass.: Lexington Books.
Warren, M. and the staff of the Community Treatment Project. 1966. Interpersonal maturity level classification: Diagnosis and treatment of low, middle, and high maturity delinquents. Sacramento, Calif.: California Youth Authority.
Patricia Voorhis is a professor in the Division of Criminal Justice at the University of Cincinnati. Kimberly Spencer is a research assistant in the Division of Criminal Justice at the University of Cincinnati.