FOLLOWING A 1970 NATIONAL SURVEY OF PUBLIC AWARENESS, ATTITUDES, AND OPINIONS CONCERNING CHILD ABUSE, DAVID G. GIL OF BRANDEIS UNIVERSITY PROPOSED THAT SUCH ABUSE WAS AS MUCH A RESULT OF SOCIETAL APPROVAL OF VIOLENCE IN CHILDREARING AS THE PATHOLOGY OF THE PARENTS. THIS STUDY ALSO TAKES NATIONAL DATA AND RUNS A REGRESSION ANALYSIS, A PATH ANALYSIS, AND OTHER STATISTICAL TESTS TO EXAMINE THE RELATIONSHIPS BETWEEN PARENTAL PSYCHOLOGY, CHILDREARING ATTITUDES, SEVERITY OF INJURY, AND DISPOSITION OF CHILD ABUSE CASES. FOUR MODELS WERE DEVELOPED: A POVERTY-VIOLENCE IN CHILDREARING MODEL WHICH COMBINES THE LITERATURE'S SOCIOCULTURAL EXPLANATIONS OF CHILD ABUSE; A PSYCHOLOGICAL MODEL, WHICH FOCUSES ON THE PATHOLOGY OF THE PARENT; THE 'REINFORCEMENT' MODEL, WHICH EXAMINES THE PARENT WHO WAS SEVERELY DISCIPLINED AS A CHILD AND NOW USES SEVERE DISCIPLINE FOR THE CATHARTIC EFFECT IT GIVES; AND A COMPREHENSIVE MODEL UTILIZING ALL THESE FACTORS. IT WAS FOUND THAT CHILD-INITIATED ABUSE RESULTED IN LESS SEVERE INJURIES. IT WAS ALSO FOUND THAT THE PERPETRATOR IN THE REINFORCEMENT MODEL WAS NOT NECESSARILY MORE 'SICK' THAN NONABUSE PERPETRATORS, SUGGESTING A SOCIALIZATION FACTOR. FOR THE OTHER MODELS DIFFERENCES WERE FOUND BUT NOT OF THE MAGNITUDE NOR SIGNIFICANCE EXPECTED. IT IS SUGGESTED THAT THE MOST SIGNIFICANT CONTRIBUTION OF THIS STUDY HAS BEEN THE DEVELOPMENT OF MATHEMATICAL TECHNIQUES FOR STUDYING LARGE SAMPLES. TO DATE MOST CHILD ABUSE DATA HAS BEEN GATHERED THROUGH SMALL SAMPLES OR SUBJECTIVE CASE STUDIES. LONG-TERM STUDIES AND LARGE SCALE STUDIES ARE CALLED ESSENTIAL TO UNDERSTANDING THE PROBLEM. APPENDIXES GIVE MATHEMATICAL DATA FROM THE STUDY AND STUDY INSTRUMENTS. A BIBLIOGRAPHY IS INCLUDED. (GLR)