THE JOLT PROGRAM INVOLVED VISITS BY JUVENILE DELINQUENTS TO THE STATE PRISON OF SOUTHERN MICHIGAN; VISITS OCCURRED TWICE WEEKLY AND LASTED ABOUT 2 1/2 HOURS. PARTICIPATING YOUNGSTERS HAD TO BE MALE, HAVE HAD AN ARREST OR PETITION FOR AN OFFENSE THAT WOULD BE CRIMINAL IF COMMITTED BY AN ADULT, AND BE ACCOMPANIED TO THE PRISON BY A PARENT OR LEGAL GUARDIAN. PRISON VISITS INCLUDED A TOUR OF THE PRISON AND AN INTENSIVE CONFRONTATION SESSION WITH SELECTED PRISONERS. THIS EVALUATION INVOLVED ANALYSIS OF 227 JUVENILES; A RANDOM ASSIGNMENT PROCEDURE WAS EMPLOYED TO ELIMINATE SYSTEMATIC BIAS. ALTHOUGH PRE-JOLT DATA WERE COLLECTED ON ALL 227 CASES, FOLLOWUP DATA FOR A 3-MONTH PERIOD AND A 6-MONTH PERIOD WERE RECORDED ON COHORTS OF 169 AND 84 CASES, RESPECTIVELY. ALL DATA WERE EXTRACTED FROM FILES MAINTAINED BY THE REFERRING LOCAL JUVENILE COURT. POST-JOLT ANALYSES FOCUSED ON THE FOLLOWUP OFFENSES AND DETENTION EXPERIENCES OF MEMBERS OF THE EXPERIMENTAL AND CONTROL GROUPS AT 3 AND 6 MONTHS FOLLOWING JOLT. ANALYSIS REVEALED NO SIGNIFICANT DIFFERENCES BETWEEN THE GROUPS AT EITHER INTERVAL, AND THOSE DIFFERENCES WHICH WERE DISCOVERED TENDED TO REFLECT THE INFLUENCE OF LIVING IN AN INTACT FAMILY SITUATION AND INVOLVEMENT IN EDUCATION AND EMPLOYMENT ROLES, RATHER THAN PARTICIPATION IN THE JOLT PROGRAM. THE PREPONDERANCE OF EVIDENCE SUPPORTS THE CONCLUSION THAT JOLT IS NOT AN EFFECTIVE CRIMINAL DETERRENT. REFERENCES AND AN APPENDIX ARE INCLUDED IN THE EVALUATION. (LWM)