The primary purpose of the study is to explore the connections between residents' perception of safety and their crime experience, and the existence of gates and fences in multi-family housing communities in urban areas. For cultivating discussions regarding the connections between gated community territory, safety, and crime experience, this study classifies apartment communities according to the conditions of their gating and fencing: gated communities, perceived gated communities, and non-gated communities. It investigates residents' perceptions of safety and their opinions and managers' opinions on gated territory and safety. The major findings from the surveys are: Residents felt safer in gated communities than in non-gated communities. Residents' perceptions of safety in perceived gated communities were similar to those in gated communities. These results reflected the territoriality issue for improving residents' perceived safety in apartment communities. Residents' perceptions of safety in architectural spaces showed that residents' fear of crime in public and semi-public spaces must first be addressed in order to ease residents' fear of crime in an apartment territory. The reality of crime in apartment communities differed from residents' perceptions of safety. Gated community residents reported a higher crime rate than nongated community residents. In addition to gates and fences that define apartment territory, such elements as patrol services, bright lighting, direct emergency buttons, and visual access to the local police were indicated as the important factors for improving residents' perceived safety. Some architectural factors and demographic factors exhibited statistical correlations with residents' perceptions of safety. Those were types of communities, dwelling floor level, educational attainment, family size, and annual income. For predicting residents' perceptions of safety in their apartment territory, multiple regression models were obtained and residents' neighborhood attachment was also considered in the multiple regression models. The apartment community managers emphasized direct maintenance issues and residents' social contact with neighbors for improving residents' perceived safety. In conclusion, design and managerial suggestions for safer communities were proposed. For creating safer multi-family housing communities, territoriality and related architectural conditions and managerial considerations and residents' participations are emphasized. The concept of community programming for safer multi-family housing communities is suggested.