DEFENSIBLE SPACE IMPLIES FOUR BASIC APPROACHES TO RESIDENTIAL SECURITY: INITIAL DEVELOPMENT OF A SECURE AREA; SUBDIVISION OF LARGE RESIDENTIAL COMPLEXES; RELOCATION OF VULNERABLE GROUPS; AND EXPANSION OF SECURITY PERSONNEL IN RESIDENTIAL COMPLEXES. THE APPROACHES ARE APPROPRIATE IN DIFFERENT SETTINGS UNDER DIFFERENT CIRCUMSTANCES. THE RELOCATION TECHNIQUE IS PARTICULARLY USEFUL FOR THE ELDERLY AND IS EXEMPLIFIED IN A PHILADELPHIA PROJECT INVOLVING THE CONVERSION OF A MIXED-USAGE APARTMENT BUILDING TO A SECURE RESIDENCE FOR THE ELDERLY. TURF RECLAMATION FOCUSES ON NEIGHBORHOOD INVOLVEMENT IN ESTABLISHING AND ENFORCING NORMS FOR BEHAVIOR ON NEIGHBORHOOD STREETS. AN APPROACH TO AREAWIDE PLANNING AGAINST CRIME HAS BEEN DEVELOPED ON THE BASIS OF A STUDY OF THE MOVEMENTS OF TEENAGE GANGS. THE STUDY FOUND THAT CLEARLY DEFINED BOUNDARIES FOR NEIGHBORHOODS TEND TO REDUCE VULNERABILITY TO CRIME. SURVEY INSTRUMENTS HAVE BEEN DESIGNED FOR USE IN MEASURING THE VULNERABILITY OF RESIDENTIAL ENVIRONMENTS TO CRIME AND IN GENERATING DATA FOR PLANNING AND EVALUATING RESIDENTIAL SECURITY PROGRAMS. SURVEYS OF 10 HOUSING PROJECTS HAVE REVEALED HIGH VICTIMIZATION RATES AND LEVELS OF FEAR. AN ANALYSIS OF CONTROLLED ENTRANCEWAYS INDICATED THE NEED TO INVOLVE TENANTS IN THE PLANNING OF SECURITY MEASURES. FIELD STUDIES SUGGEST THAT ONE-DIMENSIONAL SOLUTIONS TO RESIDENTIAL SECURITY PROBLEMS GENERALLY ARE INEFFECTIVE. IT IS NOT ENOUGH TO INCREASE LIGHTING, CONSTRUCT BETTER DOORS, OR HIRE MORE SECURITY GUARDS; COMBINATION OF SOCIAL AND PHYSICAL IMPROVEMENTS IS CALLED FOR. MULTIDIMENSIONAL SOLUTIONS SHOULD BE BASED ON DETAILED, ON-SITE RESEARCH, INTERVIEWS WITH RESIDENTS, AND STUDIES OF THE CRIME PROBLEMS UNIQUE TO EACH HOUSING PROJECT. GRAPHS AND TABLES ARE INCLUDED.