In this lab, we examined the ethics of data classification and housing affordability in the Metro Vancouver area.
The Ethics of Data Classification:
There are many ways to classify quantitative data. The way the map-maker chooses to represent the data will significantly impact the visual effects of the map and will influence the conclusions that the map-reader may draw. Map 1, attached below, shows 4 different data classification models using the exact same data: Natural Breaks, Equal Interval, Manual Breaks, and Standard Deviation. The choice of method allows whoever is making the map to encourage the audience to draw specific conclusions. For example, consider the following scenarios:
If the map maker is a journalist, they may want to make their data as captivating and controversial as possible to gain maximum views, attract attention to the issue of housing affordability, and promote conversation. As such, they might choose natural breaks, which shows a sensationalized version of the map by making more areas appear within the highest price range. However, they may also choose manual breaks as it gives a more accurate representation to share this information with the public and it is the most user friendly/readable map (because the numbers are round numbers).
Now consider the map maker is a real estate agent. They might choose a classification method that makes more areas appear affordable. Depending on their clientele, they might choose one that would be the most appealing to their potential buyers price range. Likely, they would choose the manual breaks method as it is the easiest to read and would be less likely to scare potential buyers. In short, there are certainly ethical implications with the choice of data classification method you choose. When creating a GIS map, one must always be thoughtful and considerate with regards to the ethics of their choice of data classification method.
Map 1: Metro Vancouver Housing Affordability displayed through 4 different data classification models
Housing Affordability, 2011 and 2016, in Metro Vancouver:
Map 2 shows the changes in shelter cost in Metro Vancouver between the years 2011 and 2016. I used manual breaks because they are the most useful method when comparing two data sets. It allows the map maker to create legend values for each map that are the same, therefore more readable in relation to each other.
The information is based on the census data from each year. The variable used was ‘shelter cost,’ which refers to “the average monthly total of all shelter expenses paid by households that own or rent their dwelling” (Statistics Canada). ‘Shelter-cost’ data does not include band housing, dwellings on reserves, or dwellings that are part of an agricultural operation operated by a member of the household. The ‘shelter-cost-to-income ratio’ is “calculated by dividing the average monthly shelter costs by the average monthly total household income and multiplying the result by 100” (Statistics Canada) and was used as an indicator of housing affordability. Using this as the only variable may be a source of error because it does not take into account any other variables that may impact housing affordability.
Map 2: Comparing Shelter Cost in Metro Vancouver, 2011 and 2016
Accomplishment Statement:
In this lab, I learned about different ways to classify data, as well as the ethical implications attached to this kind of choice. We learned how to reclassify the number of classifications and how to choose appropriate symbology (depending on whether the data is nominal, ordinal, interval or ratio). Using ArcGIS tools learned in previous labs to analyze census data, I was able to create maps that show the change in housing affordability in Metro Vancouver in 2011 and 2016.
