A SELECTION OF SIGNIFICANT VARIABLES FROM THE CASE HISTORIES OF 300 BOYS RELEASED FROM THE FAIRFIELD SCHOOL OF BOYS, AN OHIO JUVENILE INSTITUTION, IS DESCRIBED. HALF HAD SUCCESSFUL PAROLES, HALF HAD THEIR PAROLE REVOKED. THE SAMPLE WAS REPRESENTATIVE OF FAIRFIELD SCHOOL ADMISSIONS AS A WHOLE. MOST OF THE BOYS WERE POOR, FROM LARGE FAMILIES, WERE OF LOW AVERAGE INTELLIGENCE, AND HAD BOTH OLDER AND YOUNGER SIBLINGS. CHI SQUARE TESTS AND TSCHUPROW'S T VALUES WERE USED TO TEST THE SIGNIFICANCE OF THE ISOLATED VARIABLES. THE DEVELOPMENT OF THE TEST INSTRUMENT IS DETAILED. THE FINAL QUESTIONNAIRE WAS TESTED ON ANOTHER SAMPLE OF 150 BOYS RELEASED FROM FAIRFIELD SCHOOL THE FOLLOWING YEAR. IT HAD COMBINED ACCURACY SCORE OF 0.79 AND AN OVERALL EFFICIENCY SCORE OF 0.52. THE RESULTING INSTRUMENT IS BASED ON 3 MAJOR CONFIGURATIONS: 19 PREDICTING SUCCESS ON PAROLE, AND 13 PREDICTING FAILURE. INTERPRETATION OF THE INSTRUMENT IS PROVIDED. CAUTIONS AGAINST THE LIMITATION OF SUCH TESTING ARE INCLUDED. SUGGESTIONS FOR FURTHER RESEARCH. ARE INCLUDED. AN EXTENSIVE BIBLIOGRAPHY AND A COPY OF THE PREDICTION INSTRUMENT ARE APPENDED. (GLR)