This thesis investigated the links between public transport use and sociodemographic and land use factors. A dataset of address information about regular bus smartcard users was sourced. These addresses were geocoded using Geographical Information Systems, and the address points derived through this process were used to calculate the percentage of regular bus users in Census meshblock spatial areas. This percentage was then compared to a number of different factors, including deprivation levels (a measure of sociodemographic status), average distance to the nearest bus stop and bus route, and a number of variables from the New Zealand Census of Population and Dwellings, using a number of forms of analysis. The number of cardholders in different residential zonings was assessed, along with the average number of trips taken per day by cardholders. Results indicate that there was a relationship between the regular Metrocard users and deprivation, and regular Metrocard users and land use, however the statistical validity of these relationships was low. Principal component analysis and regression analysis were carried out to assess what variables best explained the proportion of bus use. It was found that the presence of International Students in an area accounted for the biggest variation in the levels of bus use, along with people who were Unemployed and of Maori or Pacific ethnicity, and people who have limited access to vehicles. However, the statistical validity of these results was again low.
| Identifer | oai:union.ndltd.org:canterbury.ac.nz/oai:ir.canterbury.ac.nz:10092/4622 |
| Date | January 2010 |
| Creators | Wright, Edward Peter Nelson |
| Publisher | University of Canterbury. Department of Geography |
| Source Sets | University of Canterbury |
| Language | English |
| Detected Language | English |
| Type | Electronic thesis or dissertation, Text |
| Rights | Copyright Edward Peter Nelson Wright, http://library.canterbury.ac.nz/thesis/etheses_copyright.shtml |
| Relation | NZCU |
Page generated in 0.0021 seconds