Crowdsourcing Neighbourhood Delineations in Wellington, New Zealand

Abstract

Neighbourhoods are important in our everyday lives, but physical definitions of neighbourhoods are often ambiguous. Official representations of neighbourhood boundaries used to present geographic outcomes poorly reflect individuals’ perceptions of their neighbourhoods. Existing methods of collecting neighbourhood delineations commonly consist of small sample sizes or stratified surveys on residents of individual neighbourhoods. By reducing effort and potentially increasing response rates, a crowdsourcing approach may be effective in collecting neighbourhood delineations across an entire city.

This thesis presents results from a web-based application used to crowdsource neighbourhood delineations from residents of Wellington City, Aotearoa-New Zealand. Over eight hundred responses were analysed to investigate how personal characteristics impact neighbourhood boundaries, determine areas of shared neighbourhood geographies based on overlapping demarcations, and examine how participants neighbourhood delineations compare to official representations of neighbourhoods. Case studies of a range of geographic features are provided to explore how they impact neighbourhood delineations.

This thesis found transport choices significantly impact perceived neighbourhood area; neighbourhood areas differ markedly in terms of consensus about their boundaries; and there are both similarities and discrepancies between official and perceived neighbourhood boundaries. A variety of geographic features were found to function as different perceptual elements in informing neighbourhood delineations. Crowdsourcing was a practical method to collect neighbourhood perceptions with possible implications for official neighbourhood boundaries. Finally, recommendations for future research aiming to crowdsource neighbourhood delineations were made with a combination of quantitative and qualitative methods being of high value.