Multidimensional perception of residential environment quality and neighbourhood attachment in the urban environment

This study tests a path model in which residential satisfaction, operationa lized in terms of multidimensional perceived residential environment qualit y (PREQ), predicts neighbourhood attachment considered as the final criteri on. Socio-demographic (age, sex, socio-economic level) and residential (len gth of residence both in the neighbourhood and in Rome, number of persons l iving together) variables are also included in the theoretical model as pre dictors of both PREQ and neighbourhood attachment. Using a multidimensional questionnaire for the measurement of PREQ and a unidimensional scale for t he measurement of neighbourhood attachment, the study focuses on a sample o f 497 inhabitants from 20 different neighbourhoods in the city of Rome. The multidimensional PREQ questionnaire comprises 20 different scales covering four main areas: architectural and town-planning features (six scales); so cial relations features (three scales); punctual and in-network services (s ix scales); context features (five scales). First, the path model is tested separately in each area using structural equation analysis. Then, the best predictors emerging from each area, together with all socio-demographic an d residential variables, are included in a final model. This model shows bo th the relevance of predictors from all four areas in predicting attachment , and also a hierarchy between the areas in the power of the prediction (co ntext area giving the most powerful predictors, services giving the weakest ones, architectural and town-planning, and social relations having interme diate importance). Length of residence in the neighbourhood and socio-econo mic level are the most relevant of the residential and socio-demographic va riables. Results are discussed with reference to the multicomponential natu re of the process of neighbourhood attachment. (C) 1999 Academic Press.