This paper presents the validation by Confirmatory Factor Analysis (CFA) of abbreviated versions instruments focusing on the neighborhood residential level, Perceived Residential Environment Quality Indicators (PREQIs) and Neighborhood Attachment (NA). A sample of 1488 residents in various neighborhoods of eleven Italian middle- and low-population cities (from 50,000 to 400,000 inhabitants) filled in a questionnaire including 12 scales (total N items = 158), 11 PREQ scales (covering architectural-urban planning, social-relational, functional, and contextual aspects) and 1 NA scale. The sample was randomly split-half in a calibration sample (for model development) and a validation sample (for model confirmation). Results showed good fit indexes for factorial structures including overall 19 PREQIs and 1 NA indicators, each one composed of 3 or 4 items (total N items = 66). Despite the high reduction of items, the shortened PREQIs and NA yield good or at least acceptable internal consistency, and fulfil convergent and discriminant construct validity criteria. Hence they are well suited for use in research designs focusing on multiple measures of environmental quality of residential places.
Cross-validation of abbreviated Perceived Residential Environment Quality (PREQ) and Neighbourhood Attachment (NA) Indicators
FORNARA, FERDINANDO;
2010-01-01
Abstract
This paper presents the validation by Confirmatory Factor Analysis (CFA) of abbreviated versions instruments focusing on the neighborhood residential level, Perceived Residential Environment Quality Indicators (PREQIs) and Neighborhood Attachment (NA). A sample of 1488 residents in various neighborhoods of eleven Italian middle- and low-population cities (from 50,000 to 400,000 inhabitants) filled in a questionnaire including 12 scales (total N items = 158), 11 PREQ scales (covering architectural-urban planning, social-relational, functional, and contextual aspects) and 1 NA scale. The sample was randomly split-half in a calibration sample (for model development) and a validation sample (for model confirmation). Results showed good fit indexes for factorial structures including overall 19 PREQIs and 1 NA indicators, each one composed of 3 or 4 items (total N items = 66). Despite the high reduction of items, the shortened PREQIs and NA yield good or at least acceptable internal consistency, and fulfil convergent and discriminant construct validity criteria. Hence they are well suited for use in research designs focusing on multiple measures of environmental quality of residential places.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
