Comparison of Multivariate Techniques in the Prediction of Juvenile Postparole Outcome

Abstract

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.

Get full access to this article

View all access options for this article.

References

Alumbaugh, R. V. , Davis, H. G. , and Sweney, A. B. A comparison of methods for constructing predictive instruments. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT, 1969, 29, 639-651.

Anderson, H. E. Regression, discriminant analysis, and a standard notation for basic statistics. In R. B. Cattell (Ed.) Handbook of Multivariate Experimental Psychology, 1966, 153-173.

Cattell, R. B. and Tsujioka, B. The importance of factor-trueness and validity, versus homogeneity and orthogonality, in test scores. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT, 1964, 24, 3-30.

Crigler, Marcia. Factors related to postparole outcomes of juvenile offenders. Unpublished master's thesis, Central Washington State College, 1974.

Cronbach, L. J. Essentials of psychological testing. New York: Harper and Row , 1970, 433.

Dixon, W. J. (Ed.) Biomedical computer programs . Berkeley: University of California Press, 1967.

Ganzer, V. J. and Sarason, I. Variables associated with recidivism among juvenile delinquents . Journal of Consulting and Clinical Psychology, 1973, 40, No. 1, 1-5.

Glaser, D. The effectiveness of a prison and parole system. Abridged edition. New York: Bobbs-Merrill Co., Inc., 1969.

Glaser, D. and O' Leary, V. Personal characteristics and parole outcome. U. S. Department of Health, Education and Welfare, Office of Juvenile Delinquency and Youth Development, 1966.

10.

Gottfredson, D. and Bonds, J. Systematic study of experience as an aid to decisions . Research Report No. 2, December, 1961, Research Division, Department of Corrections, State of California.

11.

Gough, H. G. , Wenk, E. A. , and Rozynko, V. V. Parole outcomes as predicted from the CPI, the MMPI and the base expectancy tables. Journal of Abnormal Psychology, 1965 , 70, 432-441.

12.

Guilford, J. P. Psychometric methods. New York: McGraw-Hill Book Co., Inc., 1954.

13.

Hase, H. D. and Goldberg, L. R. Comparative validity of different strategies of constructing inventory scales. Psychological Bulletin, 1967, 48, 231-238.

14.

Horst, P. An overview of the essentials of multivariate analysis methods. In R. B. Cattell (Ed.) Handbook of Multivariate Analysis, 1966, 129-152.

15.

Jaman, Dorothy. Behavior during the first year in prison. Report III, Background Characteristics as Predictors of Behavior and Misbehavior, March, 1972.

16.

Jitodae, T. The construction and evaluation of alternative methods for predicting statuses of parolees. Unpublished master's thesis, University of Washington, 1954 .

17.

Kirby, B. C. A parole prediction study using the discriminant function. Unpublished doctoral dissertation, University of Washington, 1953.

18.

Ohlin, L. E. and Duncan, O. D. The efficiency of prediction in criminology. American Journal of Sociology, March, 1949,54,441-451.

19.

Ohlin, L. E. Selection for parole. New York: Russel Sage, 1951.

20.

Rao, C. R. Advanced statistical methods in biometric research. New York: Wiley, 1952.

21.

Reed, Judy A. and Ballard, K., Jr. Washington state parole base expectancy study. Research Report, May, 1971, 3, No. 21, Office of Research, Division of Institutions, State of Washington.

22.

Strinden, D. N. Parole Prediction using criminological theory and manifold classification techniques. Unpublished master's thesis, University of Washington, 1959.

23.

Wenk, E. A. and Emrich, R. L. Assaultive youth: An exploratory study of the assaultive experience and assaultive potential of California youth authority wards. Journal of Research in Crime and Delinquency, 1972 , 9, 171-196.

24.

Wheeler, S. A comparison of parole prediction techniques. Unpublished master's thesis, University of Washington, 1956.