POLICE STATISTICS FOR VIOLATIONS BY MINORS (90,470 CASES IN 1977) DO NOT REFLECT THE ACTUAL NUMBER OF OFFENDERS EITHER OVERALL OR WITHIN SUBGROUPS (E.G., FOREIGN WORKERS) BECAUSE MULTIPLE CRIMES BY THE SAME INDIVIDUAL ARE RECORDED AS SEPARATELY. THE ACTUAL NUMBER OF YOUTHFUL OFFENDERS IS THUS LOWER THAN THE OFFICIAL FIGURE (BY 13 PERCENT IN NORTH RHINE-WESTPHALIA). DELINQUENCY FREQUENCY FIGURES TEND TO BE LOW BECAUSE CHILD DELINQUENTS ARE ONLY REGISTERED IF THEY ARE OVER 6 YEARS OLD, WHILE THE REFERENCE GROUP COVERS ALL MINORS BETWEEN AGES 0 AND 13. FURTHERMORE, CRIME RATES FOR THE 6-TO-13-YEAR-OLD AGE GROUP ARE AVERAGED AND DO NOT SHOW THAT RATES ARE LOWER FOR YOUNGER CHILDREN THAN FOR OLDER ONES. IN DESCRIBING CRIME DEVELOPMENTS, POLICE STATISTICS STATE ABSOLUTE PERCENTAGE INCREASES WITHOUT EXPLAINING THE RELATIONSHIP OF CRIME RATES IN THE REFERENCE YEAR OR PERIOD TO THE OVERALL CRIME PICTURE. ANOTHER SOURCE OF STATISTICAL DISTORTION IS THE VARIATION IN POPULATION SIZE AND COMPOSITION. COMPARISONS OF CRIME RATES AMONG GERMAN JUVENILES TO CRIME RATES AMONG CHILDREN OF FOREIGN WORKERS FAIL TO CONSIDER DROPS IN GERMAN BIRTH RATES AND INCREASES IN FOREIGN WORKER BIRTH RATES. THUS, THE AVERAGE RATES AND ABSOLUTE OFFENDER FIGURES DO NOT REVEAL THE ACTUAL HIGHER CRIME FREQUENCY FOR GERMAN YOUTHS AND LOWER FREQUENCY FOR IMMIGRANT CHILDREN. SIMILAR TYPES OF DISTORTION OCCUR FOR SEPARATE AGE GROUPS. INACCURACIES COULD BE AVOIDED BY DIVIDING MINORS INTO SMALLER AGE GROUPS. REGIONAL DIFFERENCES IN CRIME RATES WHICH CANNOT BE ATTRIBUTED TO METHODOLOGICAL VARIATION REQUIRE FURTHER INVESTIGATION. NOTES ARE SUPPLIED. --IN GERMAN.