A Comparison Study Between Google Street View and On-Site Visiting to Audit Neighbourhood Environment’s Walkability

04 Pubblicazione in atti di convegno
RAFIE MANZELAT Reihaneh, Valeri Diego, D'Alessandro Daniela

As walking for physical activity is associated with significant health benefits, several studies empirically support the role of walkable environments in inducing or hindering of walking. Researchers proposed different methods to measure the level of walkability. Google Street View (GSV) is an instrument able to assess physical features of city streets and environments as well as Site visit and local resident’s perception.

This study aims to compare the reliability of the audit tools for use with on-site visiting (OSV) and Google Street View (GSV), to assess the levels of agreement between on-site and computer-based assessment. Three neighbourhoods of Rome/Italy with different urban context and various level of socio-economic status has selected. This study compared neighbourhood measurements coded in 2017 using Street View with neighbourhood audit data collected in 2018.

The comparison reguards 61 items associated with five neighbourhoods characteristics: Traversability, Compactness, Safety, Pedestrian Facilities, and Aesthetics. To measure concordance between fıeld audit and Street View data, Intra Class Coefficient (ICC), percentage agreement and Kappa statistic (K) were used for individual items.

The level of agreement is classified as follow: < 0 as indicating no agreement and 0–0.20 as slight, 0.21–0.40 as fair,0.41–0.60 as moderate, 0.61–0.80 as substantial, and 0.81–1 as almost perfect agreement.
The analyses, found high levels of consent (≥80% agreement) for 64.7% of the items. Measures of parking, public transport stops, and man-made aesthetic had relatively high levels of consent, while pedestrian path obstruction, barriers, and cleanliness had low levels of consent.

This exploratory study indicates that Google Street View has some opportunities such as easy accessing, saving time, money and human resource that can be useful to auditing neighbourhood environments in more than 60% of environmental features.

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma