Parole Prediction Research Articles

A Reader's Companion to Against Prediction: A Reply to Ariela Gross, Yoram Margalioth, and Yoav Sapir on Economic Modeling, Selective Incapacitation, Governmentality, and Race

From parole prediction instruments and violent sexual predator scores to racial profiling on the highways, instruments to predict future dangerousness, drug-courier profiles, and IRS computer algorithms to detect tax evaders, the rise of actuarial methods in the field of crime and punishment presents a number of challenging issues at the intersection of economic theory, sociology, history, race studies, criminology, social theory, and law. The three review articles of Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age by Ariela Gross, Yoram Margalioth, and Yoav Sapir, raise these challenges in their very best light. Ranging from the heights of poststructuralist and critical race theory to the intricate details of mathematical economics and criminological analysis, the articles apply different disciplinary lenses to the analysis of the actuarial turn offered in Against Prediction and set forth both substantive and structural challenges to the book. By means of a detailed reply to the three reviews, this article provides a reader's companion to Against Prediction.

Discretionary Justice: Decision Making in a State Juvenile Parole Board

AbstractThe literature in Criminal Justice is replete with commentary on both the virtues and injustices associated with adult parole board decision‐making. There has been far less attention paid to juvenile parole boards in part because so few of them exist. Institutional release decisions in most states still remain the purview of juvenile institutional staff members. This paper examines parole Authority decision‐making in a citizen‐dominated state juvenile parole board. The paper focuses on the conduct of hearings, dispositional guidelines, criteria used by board members in decision‐making and the advocacy role of parole officers and institutional staff. The influence of deinstitutionalization and least restrictive policies on parole Authority decision making is also discussed.During recent years, the literature in criminal justice has become replete with commentary on the virtues and injustices associated with parole board decision‐making. Consequently, many questions have been raised about disparate sentences, the criteria used by parole boards in making decisions, unstructured discretion, and the pressing need for guidelines and statistically accurate parole prediction instruments.The intent of this paper is to examine parole decision‐making in a state juvenile parole board where individualized treatment and rehabilitation have been long standing goals. The paper focuses upon the criteria used by board members in decision‐making, the conduct of hearings, the guideline system used to determine a recommended length of confinement and the advocacy role of parole officers and institutional staff. The influence of deinstitutionalization and least restrictive policies on parole board decision‐making is also discussed.The author is presently completing the second year of a two‐year appointment as a member of the Parole Authority under discussion. Since December, 1983, the author has continuously observed and participated in approximately 300 individual parole hearings over thirty‐seven hearing days.

Parole Guidelines: Suggested Research Questions

Three factors have supported the development of parole guidelines: criticism of the rehabilitation model as a basis for parole decision making and the accompanying movement toward a just deserts model of punishment, the development of sophisticated statistical procedures for risk assessment, and increasing questions about disparity in criminal justice decision making. This paper raises questions related to each of these three factors that should be addressed by the research community as guidelines are implemented. With respect to offense severity and sentence served, key considerations are the rationale for sentence lengths, the degree to which discretion is being structured, inmates' perceptions of the guidelines, and the effect of guidelines on time served. Under risk assessment and parole prediction, the accuracy of the salient factors as predictors, the problem of false positives, and the possibility of unan ticipated bias must be addressed. In terms of effects on the overall system, questions are raised about the extent to which discretion can be structured, the effect on correctional staffs of changing the purpose of correction, and the impact of such a shift on the already dwindling resources for the parolee. The paper closes with a call for a coherent program of research to accompany the implementation of guidelines.

Comparison of Multivariate Techniques in the Prediction of Juvenile Postparole Outcome

Several strategies for the prediction of institutional outcome have been attempted by researchers in juvenile and parole prediction. The majority of researchers have utilized a variation of multiple regression, while a few have made use of discriminant functions and factor analysis. The present study contrasted the three approaches in terms of predictive efficiency for juvenile recidivism. A total of 579 juvenile cases were selected from the Washington state juvenile archives during the calendar year of 1971. Forty predictive variables from personal histories and behavioral records of the Office of Juvenile Rehabilitation were collated and quantified for the study. The total sample was divided randomly into two groups for purposes of double cross-validation. On each of the two sample halves, a set of predictors of juvenile outcome were generated by each of three methods, factor analysis, stepwise regression, and stepwise discriminant functions. Validation of the three methods was assessed by (1) D2 or discriminability of each predictive variable on the cross-validation sample, and (2) the accuracy of classifying each juvenile based on discriminant analyses. The results showed clearly that stepwise discriminant functions provided the more consistent selection of variables with seven of 10 variables the same for all samples. Also, discriminant functions was shown to manifest the most predictive efficiency with significantly fewer misclassifications and greater discriminability. By contrast, stepwise regression and factor analysis as strategies manifested more misclassifications and less discriminability. In addition, the two methods were found to be unreliable in the selection of predictive variables for the two validation samples. Results were discussed in terms of the shrinkage problems of regression analyses and the homogeneity of variance associated with factor analysis.

Violent Crime: Prediction and Control

Violent crime has become a major public concern, threatening many citizens' sense of well-being and involving tremendous national social and economic costs. The various professions involved in the study, rehabili tation, and control of violent offenders feel increasing public pressure to protect society from crimes of violence, primarily by incapacitating the offender. Decisions to incarcerate or release offenders are often based on in dividual predictions of future violent behavior. Recent public policy con cerning career offenders represents an attempt to protect society by evaluating collective probabilities of future violent behavior. This paper examines the growing concern over violent crime, the prediction techniques that have been used to determine the disposition of an offender, the tendency toward overprediction documented by various research studies, and recent public policies enacted to curb crimes of violence. Finally, balancing individual rights and the rights of society in light of public attitude and public policy involves several moral and ethical questions. 1. President's Commission on Law Enforcement and Administration of Justice, The Challenge of Crime in a Free Society (Washington, D.C.: U.S. Govt. Printing Office, 1967), p. 18. 2. "Losing Battle against Crime," U.S. News and World Report, Dec. 15, 1974; "Crime in America: One Nation's Journey into Fright," PTA Magazine, April 1974; "Crime Wave," Time, June 30, 1975. 3. Robert Winslow, ed., Crime in a Free Society, 2d. ed. (Encino, Calif.: Dickenson, 1973), p. 157. 4. James Q. Wilson, Thinking about Crime (New York: Basic Books, 1975), p. 7. 5. Winslow, Crime in a Free Society, p. 156. 6. Ibid., p. 48. 7. James Brooks, "The Fear of Crime in the United States," Crime and Delinquency, July 1974, pp. 242-43. 8. Winslow, Crime in a Free Society, p. 157. 9. Ernst A. Wenk, James O. Robison, and Gerald W. Smith, "Can Violence Be Predicted?" Crime and Delinquency, October 1972, pp. 393-402. 10. Wilson, Thinking about Crime, p. 21. 11. Don Gottfredson, "Assessment of Methods," in Crime and Justice, Vol. 3, Leon Rad zinowicz and Marvin Wolfgang, eds. (New York: Basic Books, 1971), pp. 362-63. 12. Michael A. Peszke, "Is Dangerousness an Issue for Physicians in Emergency Com mitments?" American Journal of Psychiatry, August 1975, p. 828. 13. Edwin H. Sutherland and Donald R. Cressey, Principles of Criminology (Philadelphia: Lippincott, 1960), p. 135. 14. Jonas R. Rappeport, ed., The Clinical Evaluation of Dangerousness of the Mentally Ill (Springfield, Ill.: Charles C Thomas, 1967), p. 27. 15. Alfred J. Kahn, "Public Policy and Delinquency Prevention," in Crime and Justice, Vol. 3, Radzinowicz and Wolfgang, eds., p. 417. 16. Sheldon Glueck and Eleanor Glueck, Predicting Delinquency and Crime (Cambridge, Mass.: Harvard University Press, 1960), pp. 144-49. 17. Gottfredson, "Assessment of Methods," p. 347. 18. Daniel Glaser, The Effectiveness of a Prison and Parole System (Indianapolis, Ind.: Bobbs- Merrill, 1969). 19. Elmer H. Johnson, Crime, Correction and Society (Homewood, Ill.: Dorsey Press, 1966). 20. Gottfredson, "Assessment of Methods," p. 360. 21. Winslow, Crime in a Free Society, p. 161. 22. John P. Conrad, Crime and Its Prevention (Berkeley, Calif.: University of California Press, 1970). 23. Daniel Glaser, "The Efficacy of Alternative Approaches to Parole Prediction," American Sociological Review, June 1955, p. 284. 24. Daniel Glaser, "A Reconsideration of Some Parole Prediction Factors," American Sociological Review, June 1955, p. 335. 25. Wenk et al., "Can Violence Be Predicted?" pp. 395-98. 26. Harry L. Kozol, Richard Boucher, and Ralph Garofalo, "The Diagnosis and Treatment of Dangerousness," Crime and Delinquency, October 1972, pp. 371-92. 27. John Monahan and Lesley Cummings, "Social Implications of the Inability to Predict Violence," Journal of Social Issues, March 1975, p. 153. 28. Ibid., p. 155. 29. Ibid. 30. Ibid., p. 153. 31. Kozol et al., "Diagnosis and Treatment of Dangerousness," p. 373. 32. Monahan and Cummings, "Social Implications of the Inability to Predict Violence," p. 374. 33. Norval Morris, "The Future of Imprisonment: Toward a Punitive Philosophy," Michigan Law Review, May 1974, p. 1162. 34. Ibid., p. 1169. 35. Ibid., pp. 1165-66. 36. Ibid., pp. 1169-70. 37. Wenk et al., "Can Violence Be Predicted?" p. 402. 38. Monahan and Cummings, "Social Implications of the Inability to Predict Violence," pp. 160-61. 39. Sutherland and Cressey, Principles of Criminology, pp. 590-91. 40. John Irwin and Lewis Yablonsky, "The New Criminal: A View of the Contemporary Offender," in Crime in America: Perspectives on Criminal and Delinquent Behavior, Bruce J. Cohen, ed. (Hasco, Ill.: F. E. Peacock, 1970), pp. 199-200. 41. Timothy D. Schellhardt, "Experimental Program Musters Legal Forces against Repeat Offenders to Boost Convictions," Wall Street Journal, Aug. 19, 1976, p. 28. 42. Gresham M. Sykes, Crime & Society (New York: Random House, 1967), p. 194. 43. Schellhardt, "Experimental Program," p. 28.

Computing Mean Cost Ratings (MCR)

The Mean Cost Rating (MCR) measures the differentiation ability of those variables or groups of variables used in construct ing parole prediction or base expectancy tables. The literature of the past two decades suggests that it represents perhaps the most satisfactoiy statistical index of predictive selectivity. Its use has been limited, however, by its complex method of computa tion. The MCR is explained and a simplified procedure for com puting it is offered. In addition, the MCR is compared with Chi square, the J-value, and the Index of Predictive Selectivity, which have also been used to determine the selective ability of factors in the construction of prediction tables.

Cluster Analysis in a Parolee Sample

This study extends earlier work concerned with the classification of individuals. The method used, called Association Analysis, has received widespread attention since its first use by Williams and Lambert.1 Ordinarily, in Association Analysis, use has been made of the statistic Chi-square (or of the Phi coefficient) to measure the associations among attributes. This report discusses the use of two different statistics (Cramér's Statistic ϕ' and the multiple correlation coefficient) for the purpose of classification according to a hierarchical subdivision process like that previously used. Although the procedure is strictly one of classification, not necessarily for purposes of prediction, the attributes by which an individual is classified are those often found associated with parole outcomes and used in the Uniform Parole Reports.2 To explore the use of this technique in parole prediction, a pooling procedure was applied to the subgroups obtained, resulting in groups of paroled prisoners, each with significantly different parole out comes.

Decision and Outcome: Studies in Parole Prediction

Decision and Outcome: Studies in Parole Prediction

The Use of Parole Prediction with Institutionalized Narcotic Addicts

This report describes the application of the parole prediction model to two cohorts of parolees with histories of narcotic addic tion and suggests an evaluation strategy for determining the relative effectiveness of different treatment approaches for dif ferent types of addicts. Narcotic users released to the supervision of the New York State Division of Parole were followed for one year on parole. From this experience, the configuration of variables most related to successful parole outcome were derived. The find ings of a complementary study of California addict parolees tended to enhance the predictive ability of the over all model, although difficulty was encountered in determining the predictive stability of the suggested variables.

New Directions for Parole Prediction Research

The author received his Ph.D. in Sociology from the University of Illinois in 1966. He was on the faculty of the University of Kentucky, Sociology Department from 1964-1967 and from 1964 to 1966 served as Director of Research for the Kentucky Department of Corrections. Presently he is on the faculty of the Department of Sociology, University of South Carolina and serves as a research consultant to the South Carolina Department of Corrections. Attempts to relate parole prediction to theories of criminality have been rare. In this article, theoretically relevant variables are suggested and data are presented which support the contention that such variables can contribute significantly to parole prediction.

Problems in Parole Prediction: A Historical Analysis

Problems in Parole Prediction: A Historical Analysis

Problems in Parole Prediction: A Historical Analysis

A review of past efforts to predict parole outcome indicates that, although there have been some recent improvements in the methodology, there has been no appreciable increase in predictive power. Two related reasons for this are evident. First, the prevalent use of prison files as a data source limits the kind of research question which can be asked. Second, the non-theoretical nature of this research has prevented a systematic accumulation of knowledge relative to this problem. The success of future efforts depends on whether parole prediction research conforms to the basic requirements of scientific investigation which include utilizing theoretical guidelines and selecting variables on the basis of their theoretical relevance instead of on their availability in prison files.

Optimism and Accuracy in the Perception of Selected Parole Prediction Items

Optimism and Accuracy in the Perception of Selected Parole Prediction Items

Comparison of Multiple Regression and Configural Analysis Techniques for Developing Base Expectancy Tables

This study compares two statistical techniques-multiple re gression and configural analysis—used in developing parole pre diction tables, according to their ability to (1) differentiate be tween offenders who violate parole and those who do not, (2) predict violators from among a new group of parolees, and (3) assist administrators and researchers. First, experience tables had to be developed and tested for prediction ability. Once their accuracy in predicting had been demonstrated, they could be used as base expectancies because they had the quality of being "expected." As such, they could be used as a yardstick to evaluate correctional programs' ability to reduce these "expected" violation rates. The two methods were applied to the same body of data and the results were compared. The data consist of Wisconsin adult male offenders paroled in 1954-57 and in 1958-59. All were followed up for two years while they were on parole. The first group was used to develop the experience tables; the second group was used to test prediction ability. The tables were compared for accuracy in predicting through use of the Dc index and for accuracy in differentiating by the J index, measures developed by H. Richard John using Daniel Glaser's data gathered for federal parolees.

A Re‐evaluation of Parole Prediction: Some Comments

A Re‐evaluation of Parole Prediction: Some Comments

Some Notes on Treatment Theory in Corrections

Previous articleNext article No AccessSome Notes on Treatment Theory in CorrectionsDon C. GibbonsDon C. Gibbons Search for more articles by this author PDFPDF PLUS Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinkedInRedditEmail SectionsMoreDetailsFiguresReferencesCited by Social Service Review Volume 36, Number 3Sep., 1962 Article DOIhttps://doi.org/10.1086/641281 Views: 1Total views on this site Citations: 2Citations are reported from Crossref Copyright 1962 The University of ChicagoPDF download Crossref reports the following articles citing this article:Don C. Gibbons Correctional Treatment and Intervention Theory: Bringing Sociology and Criminology Back in, International Journal of Offender Therapy and Comparative Criminology 30, no.33 (Jul 2016): 255–271.https://doi.org/10.1177/0306624X8603000309Richard S. Sterne A Re-evaluation of Parole Prediction, Criminology 3, no.44 (Feb 1966): 3–7.https://doi.org/10.1111/j.1745-9125.1966.tb00124.x

Some Dangers in Parole Prediction

As prediction techniques are perfected and increasingly used, certain incongruities existing between parole theory and parole prediction may have detrimental effects on parole itself. The inadequacies of substandard parole jurisdictions may become less evident if only good risks are released; challenge to pro fessional skills may be reduced; the administrative advantages inherent in prediction may supersede correctional ends; the principle of individualized treatment may be threatened. The best use to which prediction can be put is in the identification of prisoners whose parole tends to bring this measure into public disrepute.

An Experiment with Ohlin's Prediction Report

Using the Ohlin Parole Prediction Report, a follow-up study was made of 347 inmates who were paroled from the Minnesota State Reformatory for Men between 1954 and 1957. The study, conducted in March, 1962, measured success or failure on parole by a single criterion: whether parolees did or did not "violate the parole agreement." Five tables detail both the overall viola tion rate and the violation rate for each of the nine score groups at the Reformatory. The actual overall violation rate of parolees was found to be 42.5 per cent; the expected viola tion rate derived from Ohlin's detailed parole experience table was 28.2 per cent. Because the overall rate and also the indi vidual rates for the nine score groups were so much higher than those reported by Ohlin, further use of this predictive instrument at the Minnesota State Reformatory was abandoned.

Parole Prediction

Parole prediction, an estimate of the probability of violation or nonviolation of parole by an offender on the basis of ex perience tables, was put into practical operation in 1933 in Illinois. Since the early days of prediction, many refinements of the technique have been suggested and many issues raised and discussed—for instance, the weighting of predictive factors, the comparative importance of dynamic versus static factors, evalua tion of the changing attitudes of the inmates in the course of the correctional process, inconsistency of predictions based on different populations of parolees, the need for continually ad justing the experience tables, attempts to develop an index of predictive efficiency, and the continued search for the most meaningful predictive factors. The perennial central problem is the comparative merit of experience tables versus the case study in deciding on the parole of an individual offender. The con clusion that parole cannot forgo the advantages to be gained from the development and judicious use of experience tables seems to be warranted.

The Development of Social Measurement

Social measurement is the one which shows the most conspicuous development in the recent twenty years among research methods. The instruments of social measurement contains two types ; (1) scales and (2) indexes. Scales take the form of linear function of items and indexes do not. In this paper we are concerned with scales only. As we know, scales should have reliability and validity, and there are rather few obstacles about reliability. As for validity of scales, the situation differs among predicting scales and internal scales. Predicting scales, e. g. marital adjustment predicting scale, parole prediction scales, Warner's I. S. C., which have external criterion to be predicted, can use the correlation coefficient (or corresponding summarizing mesaures) between the external criterion and the scale scores, as a coefficient of validity. By internal scales I mean the ones which have not an external criterion, and are examplified in socio-economic status scales and attitude measurement scales in general. Internal scales, which have internal consistency, are considered at present time also to have validity. Therefore, scaling theories of internal scales have endevored to obtain internal consistency. The first scholar who gave attention to internal consistency and carefully constructed internal scales, was L. L. Thurstone. After his equal-appearing interval scale, there appeared Likert's scale which uses item analysis as a means of obtaining internal consistency. But item analysis could not secure complete internal consistency, as R. F. Sletto and L. Guttman indicated. Before the invention of Guttman's scale, factor analysis was considered the only means of securing internal consistency. Guttman's scale analysis was a splendid and epoch-making scaling theory, but the fact that his scalability is a special case of internal consistency, was discovered in the light of L, azarsfeld's latent structure theory. This latent structure theory, which introduced the concept of trace line and defined unidimensionality definitely by mathematical formula, offered a common theoretical foundation to various scaling techniques. But it is not yet completed as a new technique of scaling. We expect that it will be further developed in the future. Meanwhile, does internal consistency guarantee the validity of an internal scale? I cannot but say “no, ” because it is only a formal property. Still there remains the question concerning what the scale measures. Unfortunately at present, we have no way except to rely upon face-validity or validating the scale through long pragmatic processes. There is another important problem to be solved. Should all internal scales have an internal consistency? Does the situation exist where a multi-dimesional internal scale is necesitated? If this is the case, such a third scale would involve a future problem in my judgment.