ANY REASONABLE MODEL OF THE DETERRENCE PROCESS CANNOT IGNORE THE POSSIBILITY THAT SOME PERSONS ARE MORE DETERRABLE THAN OTHERS. YET CONVENTIONAL DETERRENCE RESEARCH STRATEGY, RELYING AS IT DOES ON STRICTLY LEGAL VARIABLES (CRIME RATES AND PROPERTIES OF LEGAL PUNISHMENT), DOES NOT PROVIDE THE DATA NEEDED TO EXPLORE THIS POSSIBILITY; I.E., DATA THAT AFFORD AN OPPORTUNITY TO DISTINGUISH UAMBIGUOUSLY BETWEEN DETERRED AND UNDETERRED PERSONS. TO ADDRESS THIS PROBLEM, SELF-REPORT DATA FROM A NATIONAL DRUG USE SURVEY OF 1,689 PERSONS ARE ANALYZED. DETERRED PERSONS ARE IDENTIFIED ON THE BASIS OF THEIR LACK OF INVOLVEMENT IN A DEVIANT ACTIVITY (MARIJUANA USE), WHICH IDENTIFIES THEM AS CONFORMISTS, AND THEIR PERCEPTIONS OF WHAT INFLUENCED THEM TO REFRAIN FROM THIS BEHAVIOR, WHICH IDENTIFIES WHETHER LEGAL THREATS WERE INFLUENTIAL OR NOT. THE THE ASSUMPTION IS THAT PERSONS WHO CONFORM AND PERCEIVE LEGAL THREATS AS A CAUSE OF THEIR CONFORMITY HAVE BEEN DETERRED. THE ANALYSIS SHOWS THAT PERSONS WHO WERE DETERRED FROM MARIJUANA USE BY LEGAL THREATS ARE INDISTINGUISHABLE ON THE BASIS OF AGE, SEX, RACE, MARITAL STATUS, OR OCCUPATION FROM THOSE WHO REFRAINED FROM USING MARIJUANA FOR OTHER REASONS. THIS FINDING HOLDS TRUE FOR PEOPLE WHO NEVER USED MARIJUANA (GENERAL DETERRENCE) AND FOR PEOPLE WHO STOPPED USING MARIJUANA (RESTRICTIVE DETERRENCE). REASONS FOR NOT TAKING THESE FINDINGS AS CONCLUSIVE EVIDENCE THAT THERE ARE NO SOCIAL CORRELATES OF DETERRENCE ARE DISCUSSED. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. (LKM)