NONDELINQUENT 12- TO 15-YEAR-OLD BOYS FROM PUBLIC SCHOOLS IN BALTIMORE, PHOENIX, AND INDIANAPOLIS WERE COMPARED TO ALL AVAILABLE DELINQUENTS IN THE JUVENILE JUSTICE SYSTEM AT THE TIME OF THE SURVEY. LD IS ASSESSED WITH RELATION TO DEFINITIONAL AND OPERATIONAL PRINCIPLES. THE BOYS WERE GIVEN DIAGNOSTIC TESTS TO DETERMINE IF THEY HAD LD, AND OBSERVATIONS WERE MADE ALONG WITH THESE STANDARDIZED TESTS TO MAKE A VALID ASSESSMENT. AN EXTENSIVE SYSTEM OF QUALITY CONTROL WAS ESTABLISHED TO ENSURE ACCURACY AND VALIDITY OF THE TESTING AND OBSERVATION SESSIONS AND OF THE DECISIONMAKING PROCESS. AN ALGORITHM WAS ESTABLISHED TO INTERPRET TEST SCORES SO A DETERMINATION OF LD COULD BE MADE. THIS ALGORITHM ALLOWED FOR UNIFORM DECISIONS BETWEEN ALL GROUPS TESTED. WHEN THE DATA WERE ASSESSED, SOME CORRELATIONS WERE NOTICED BETWEEN CERTAIN GROUPS OF TEST SCORES. AFTER ALL DISCREPANCIES WERE TAKEN INTO ACCOUNT, A NEW FINAL DECISION RULE ALGORITHM WAS SET UP. THE STATISTICS AND TEST RESULTS SHOWED THAT WHATEVER FACTORS ARE AT WORK IN THESE COMMUNITIES DIVIDING THE BOYS INTO ADJUDICATED AND NONADJUDICATED POPULATIONS ARE ALSO ASSOCIATED WITH THE PHENOMENON OF LD. THIS, HOWEVER, DOES NOT SAY THAT ALL LD BOYS WILL BE DELINQUENT. THERE WERE MORE LD BOYS WHO WERE NOT DELINQUENT THAN THOSE WHO WERE. RECOMMENDATIONS ARE MADE TO REEVALUATE THE TEST DATA IN ORDER TO ELIMINATE SOME OF THE DISCREPANCIES AND POSSIBLY REDUCE THE NUMBER OF BOYS WHO WERE CLASSIFIED WITH LD. TABULAR DATA AND OBSERVATION CHARTS ARE INCLUDED, ALONG WITH REFERENCES. (BAC)