Factors affecting housing prices in Kadiköy (Istanbul): integration of MGWR and Space Syntax analysis

Purpose The purpose of this study is to examine how street network morphology, measured by spatial configuration metrics, affects housing prices in Istanbul, beyond traditional factors such as structure and location. While numerous studies have investigated the determinants of property values, the relationship between street geometry and property prices remains to be studied. Design/methodology/approach The data set was collected through web crawling. The area where for sale advertisements were concentrated and varied was selected as the study area. In addition to building data obtained through web scraping, environmental and road data were also used. Environmental data (distance to public transportation stops, parks, coastlines and business and commercial areas) were obtained from relevant municipalities and institutions. Street network data was obtained from OpenStreetMap, and Space Syntax metrics such as connectivity, choice, integration and total depth were calculated. Multiscale geographically weighted regression was used to investigate the relationship between these variables and housing prices. Findings The results show that, in addition to common structure and location characteristics, street network characteristics also influence housing prices. Among street characteristics, “selection” is the most influential. Higher values of selection, connectivity and integration positively affect prices. Total depth showed a negative correlation. Originality/value This study integrates Space Syntax-based street measurements with advanced spatial modeling techniques and offers a new perspective. It highlights the importance of urban form in shaping housing markets and provides valuable insights for planners, urban designers and real estate analysts.