The concept of density has been an appealing notion used by many scholars, since it is, at first glance, objective, quantitative and neutral (Churchman, 1999). Starting from the grand urban theories, social and spatial changes in environment have been related to density notions. Churchman’s prominent study on the concept reveals that both density as a phenomena and its measurements are complex varying among fields and scholars. For neighborhood quality (NQ) studies, which focus on the measurement of quality in the smallest socio-spatial unit of urbanism, density has been a crucial component with its various dimensions. However, due to the complexity of the concept its integration into these studies is again various. Therefore, the comparison of the impact of density on the quality of neighborhoods cannot be clearly made within cross-contextual studies. Density has not been fully understood with its relation to NQ. The main purpose of this study is to analyze interrelationships of density and NQ with its different dimensions. Considering neighborhood as both a social and spatial unit, this study includes social domain as well by integrating the notion of social interaction in this relationship. The conceptual framework of this study is based on understanding the role of different densities and social interaction on overall perceived NQ. In measuring overall perceived NQ, this study uses the conceptual model of perceived NQ of Connerly and Marans (1988), considering it the most comprehensive model in literature. Their model uses only objective residential density as a measurement of perceived NQ. This study argues that perceived NQ cannot be affected by only one type of density. In order to broaden an understanding of the effect of density on perceived NQ, it is necessary to measure density with its wider definition and various meanings. For this purpose, Churchman’s conceptual study on density is integrated with the NQ model of Connerly and Marans. This will help understand how different densities affect overall perceived NQ. Density and social interaction are used as independent variables. Perceived NQ, the dependent variable, is measured as overall and with its global indicators as satisfaction and attachment. Density concept is analyzed in a two-level approach: objective and perceived density. Four urban residential neighborhoods with different objective densities were purposefully selected as cases. To decrease the effect of other variables, respondent’s socio-economic status and general homogeneity of the neighbors were used as control variables. All the four cases were residential neighborhoods known to be off-campus student housing areas located in the same region and all were on the university’s bus route. Different objective densities grouped under residential and public space density were measured by the use of the neighborhood plans in an interval scale. Low-density\high-density matrix was developed for both residential and public space densities. Perceived densities, social interaction and perceived overall NQ were measured by a self-administered survey taken from a convenience sample of 30 respondents from each neighborhood. Perceived density was measured in an ordinal scale with perceived residential and public space density. Social interaction was measured as actual and perceived social interaction within neighborhood. Actual social interaction was measured with number of visits within neighborhood, number of neighbors known by name and considered as friends. Perceived social interaction was measured in an ordinal scale with satisfaction of social interaction in the neighborhood. Overall perceived NQ was measured with neighborhood satisfaction and attachment in an ordinal scale. The findings showed that density and perceived NQ had overall a positive linear relationship. Rather than objective residential density, objective public space density had a more significant effect on the overall perceived NQ. Especially public space densities measured for shared outdoor spaces (shared gardens and courtyards) and social amenities (community hall, pool, sports area, playground) were considered effective in the increase of perceived NQ. Objective residential density and social interaction demonstrated a combined affect in overall perceived NQ. Perceived densities varied related to the actual social interaction level of the respondent. Overall, neighborhood satisfaction rates were more parallel to overall perceived NQ rates than attachment rates as given by respondents. This was interpreted as the low attachment experiences of students to their housing environments, which showed satisfaction as a more reliable predictor in measuring NQ for similar contexts. The significance of this study is related to two important aspects. First, the results will help researchers develop an understanding in the effect of different densities on perceived NQ. Second, this study with its case selection is one of the few off-campus student housing studies, which defines a peculiar type of user group that needs attention in human-environment research.