Data classification is power – you have the power to manipulate information to better perceive your goals whether it be ethical or appropriate to do so.

Being a journalist, my intention with making these maps was to expose the ‘truth,’ meaning there is no bias in my judgements. Since my data is not normally distributed, I would not choose standard deviation as my classification method. The data set is not uniform which means equal interval is insufficient as well. Manual breaks allow for the greatest subjectivity since I am the one choosing the intervals displayed. In this case, picking representative breaks would be challenging to not seem partial. I would choose natural breaks since this method looks for the best arrangement based on the distribution. Best arrangement is determined by the Jenkins natural breaks method, which looks for significant changes in the histogram as ideal breaks for classifying data. For a journalist, this is the preferred method because it represents the data best without over/understating any census tracts. With all this being said, in the event that I were creating maps for comparison, I would use manual breaks to best represent disparities between datasets.

If I were a real estate agent with intentions of selling homes near UBC, I would choose equal interval since it displays the neighborhoods around UBC similarly to the rest of the city. “Why buy a house in east Vancouver when there are homes similar in price in the west end? Just look at this map showing how similar they are!” would be my selling line. This method misrepresents reality because this area is actually the most expensive within the City of Vancouver. This raises ethical implications for cartographers who have to choose how data is represented. Data classification is inherently subjective since somebody has to make the choice on the preferred method, which can steer the map reader in different directions when it comes to comprehending the information. There has to be caution in manipulating data since the choice for classification methods is made with the map purpose/objective in mind.

We are measuring affordability by normalizing median dwelling cost by median household income. This measures how “affordable” a house is based on household income. It’s a better indicator than merely stating housing cost because it considers how much income a household can pull in to maintain the cost. Affordability has a different definition depending on who’s using the term. We can map 5 million dollar homes and make it look unaffordable to map viewers, but chances are, whoever owns that 5 million dollar home is able to afford it – deeming it “affordable” by this standard.

The categories we used to differentiate our normalized figures was developed by the Demographia International Housing Affordability Survey which calls the method of classification “median multiple.” Affordability is divided among 4 categories which can be trusted given that we are assessing the ratio of cost to income, not merely income. By using a proportion, the same indicator can be applied across every house.

I would say affordability is a great indicator for livability, although it is not the only factor that should be looked at. Being able to live somewhere means being able to sustain yourself, your family and your home. If you are unable to afford housing in Vancouver for example, you might not find the city so livable. But if you enjoy mountains, water, greenscapes and mild weather, you might overlook housing costs (like many people do), and decide this city is very livable. You might have to rent forever, but if that’s ideal for you given what Vancouver has to offer, then you would consider Vancouver livable.