ALTERNATIVE CAUSAL MODELS FOR THE PREDICTION OF CRIME RATES, PRISON ADMISSIONS, AND PRISON RELEASES ARE PRESENTED IN THIS DISSERTATION. THE STATIC MODEL USED STANDARDIZED REGRESSION COEFFICIENTS GENERATED FROM DATA FOR THE YEAR 1970, AND THE DYNAMIC MODEL USED DATA REFLECTING CHANGES IN THE RELEVANT VARIABLES FROM 1960 TO 1970. VARIABLES USED IN THE STATIC MODEL ALLOWED AN EFFICIENT PREDICTION OF THE CRIME RATE. THE PRIMARY DETERMINANT FOR THIS MODEL WAS PERCENT URBAN. AGE DISTRIBUTION AND PER CAPITA INCOME WERE ALSO SIGNIFICANT PREDICTORS. PERCENT NEGRO WAS FOUND TO HAVE NO SIGNIFICANT EFFECT ON THE CRIME RATE. THE STRONGEST PREDICTOR OF PRISON ADMISSIONS WAS, HOWEVER, PERCENT NEGRO, FOLLOWED BY EDUCATIONAL LEVEL. STRUCTURAL CHARACTERISTICS WERE FOUND TO INFLUENCE PRISON RELEASES, WITH PRISON ADMISSIONS BEING THE PRIMARY DETERMINANT. POLICY CHANGES MADE DURING THE SIXTIES REGARDING PRISON ADMISSIONS REDUCED THE EFFECTIVENESS OF THE DYNAMIC MODEL TO PREDICT CHANGES IN ADMISSIONS. CHANGES IN PRISON RELEASES, HOWEVER, WERE PREDICTED FROM CHANGES IN ADMISSIONS INDICATING SOME MANIPULATION ON THE PART OF OFFICIALS TO KEEP THE SYSTEM IN BALANCE. CHANGES IN STRUCUTRAL ATTRIBUTES WERE FOUND TO ACCOUNT FOR CHANGES IN THE CRIME RATE. THE DYNAMIC MODEL ILLUSTRATED THAT THE CRIMINAL JUSTICE SYSTEM WAS GENERATING MORE CRIME THAN COULD BE PROCESSED, WHICH RESULTED IN FEWER PEOPLE GOING TO PRISON AND GREATER USE OF ALTERNATIVES SUCH AS PROBATION. THE FINDINGS INDICATE THAT CRIME IS A PRODUCT OF SOCIETAL DEVELOPMENT AS IMPLIED BY MANY THEORIES AND THAT THERE IS NO NEED TO SEPARATE CRIME AND IMPRISONMENT FROM OTHER PRODUCTS OF THE SOCIAL STRUCTURE. REFERENCES AND TABLES ARE PROVIDED. (AUTHOR ABSTRACT MODIFIED--TWK)