Comparative Analysis of Frequency Ratio, Analytical Hierarchy Process, and Logistic Regression Methods in Selecting the Location of Rural Dwellings: the Case of Keban Stream Basin (Elaz??)

Abstract

Rural areas in Turkey are undergoing significant transformations, but not as much as cities. Rural areas are evolving in many aspects, including land use, economic structure, and lifestyles. Rural housing is one of the most important indicators of change. The migration of rural dwellings from settlement centers causes the settlement pattern to loosen and disperse, similar to urban sprawl. Thus, existing agricultural/livestock lands and natural elements are changing and transforming. The change/transformation process that rural areas undergo as a whole is conceptualized as new rurality in the literature. The adoption of the dual lifestyle and the urban lifestyle in the countryside, which are indicators of the new rurality, has increased the demand for country residence and second homes in the countryside. Thus, more housing is being built in areas near major transportation axes, particularly near cities and in rural areas. Making suitability analyses and creating projections by employing geographical data in the selection of the locations of these newly built houses is important for the future of existing agricultural areas and rural planning. Performing site selection analyses with different methods demonstrates the advantages of one method over the others and provides more accurate analyses. In this study, logistic regression (LR), analytical hierarchy process (AHP), and frequency ratio (FO) methods were used for site selection suitability analysis. The study area, Keban Stream Basin, is located in the northwest of Elazig, within the borders of the Keban district.The basin has a surface area of approximately 187 km(2). According to the results on the maps, the basin has an average land area of 12.5 km(2), which is suitable for new rural housing construction. Although each model has its advantages, the LR and FO methods produced more suitable results than the AHP method.