IT WAS HYPOTHESIZED THAT DIRECT EXPERIENCE WITH CRIME (BEING A VICTIM) WOULD BE MORE INFLUENTIAL THAN INDIRECT EXPERIENCE (HEARING ABOUT CRIME FROM OTHER PEOPLE OR THROUGH THE MEDIA) IN SHAPING PEOPLE'S CRIME-RELATED ATTITUDES AND BEHAVIORS AT THE PERSONAL, GENERAL (NEIGHBORHOOD), AND SOCIETAL LEVELS. SURVEY FINDINGS INDICATE THAT THE HYPOTHESIS IS INCORRECT. PERSONAL RESPONSE TO CRIME IS INFLUENCED PRIMARILY BY DIRECT EXPERIENCES, SECONDARILY BY INDIRECT EXPERIENCES. RESPONDENTS' CRIME PREVENTION BEHAVIOR AND PERCEPTIONS OF PERSONAL VULNERABILITY TO CRIME ARE UNRELATED TO THEIR ESTIMATES OF THE CRIME RATE, TO SUPPORT FOR CRIME-RELATED POLICIES, OR TO THEIR EVALUATIONS OF THE LOCAL GOVERNMENT. GENERAL-LEVEL RESPONSES TO CRIME ARE DETERMINED BY INDIRECT EXPERIENCES AND ARE UNINFLUENCED BY DIRECT EXPERIENCES. SOCIETAL-LEVEL RESPONSES ARE DIRECTLY INFLUENCED BY NEITHER TYPE OF EXPERIENCE. THE INFLUENCE OF A CRIME-RELATED EXPERIENCE ON ATTITUDES IS MEDIATED BY THE EXTENT TO WHICH THE EXPERIENCE IS INFORMATIVE, WHEREAS THE INFLUENCE OF AN EXPERIENCE ON BEHAVIOR IS MEDIATED BY THE INFORMATIVENESS, MEMORABILITY, AND AFFECTIVITY OF THE EXPERIENCE. THE STUDY FINDINGS PROVIDE A LIMITED DESCRIPTION OF THE PROCESS THROUGH WHICH EXPERIENCES ARE CONVERTED INTO INFERENCES ABOUT THE CRIME SITUATION. SUPPORTING DATA, STUDY INSTRUMENTS, AND A LIST OF REFERENCES ARE PROVIDED. (AUTHOR ABSTRACT MODIFIED--LKM)