Auto Theft in Vancouver:
Time, Transit, and Zoning
Initial Steps
As with most projects, the data that was procured required adaptation in order for it to be consistent with the other sets of data.
Clipping to form the Study Area
The first step that was required to be performed was clipping the different data files to the study area. The mask that was used for this process was the city boundary file obtained from the City of Vancouver. This file, however, had to be converted from line to polygon before it could be used as a mask. Once this was completed, all remaining layers were clipped to this extent. An additional step was required for the Zoning Districts layer, as the zoning districts extend out beyond the shore where cars could not be present. Using the DMTI water layer, the symmetrical difference tool was used to confine the Zoning District to the land.
Projection
To ensure that all layers were displayed accurately relative to each other, the data layers were projected. Due to the location of Vancouver, the NAD 1983 UTM Zone 10N projection was employed. By using a projection intended for the given study area, error caused by distance from the datum is reduced.
Crime Data and Address Locators
In terms of time spent fixing data, the crime data obtained from the City of Vancouver required discreet values for prime locations. The reason for this was that, in order to protect the privacy of individuals, the actual addresses of the crimes were suppressed such that the spatial resolution was decreased. This meant that auto thefts were reported to the nearest 100 block following the of using two captial Xs to mask the last two digits of the address. An example is 18XX East 22nd Avenue. To overcome the challenge of providing discrete locations for points that ambiguously exist in an area, the points were provided with the address that placed them closest to the middle of the block.
This process was complicated by the address locator, as it required valid addresses for matching. The City of Vancouver's road network file was not adequate for address matching as it did not contain all of the necessary elements. As a result of this, DMTI's road network was used instead. This combination of a table from the City of Vancouver and a road network from DMTI meant that all addresses had to be adjusted to reflect the pattern and convention that the DMTI file used. Once this task was completed, only one theft was spatially unaccounted for, as it was located on an access road which was not present in the DMTI file. This theft, which occurred at Access road and East 1st Avenue, was dropped from the study. It is important to note at this point that a potential source of error for this study is that different address locators may place the location of given thefts at different locations based on how the address locator functions.
The above map, Bus Stops and Auto Theft in Vancouver, acted as the base map for two of the three analyzed variables within this study. This map includes all of the layers except for the zoning districts from the City of Vancouver.
_____________________________________________