The overarching objectives of this project are
- Learning how to turn data into visual insights using Tableau
- Learning how to communicate insights using the appropriate visualizations.
To achieve the above broad objectives, several datasets relating to real estate prices and housing affordability in Canada were collected, explored, and visualized to communicate relevant insights. The deliverables for this project includes
- A Tableau Workbook file containing several worksheets, dashboards, and a story section located in Submissions/Workbook,
- A 10-slide PowerPoint file located in Submissions/Presentations, and
- A Jupyter notebook containing detailed procedural steps and observations located in Submissions/Notebook.
The detailed steps undertaken to complete this project includes
- Collecting relevant Canadian-specific datasets with information about residential house prices, office building prices, average weekly earnings, inflation measures in form of Consumer Price Index (CPI) information, and housing construction starts.
- Exploring the datasets to provide initial insights and develop questions and hypotheses for further visualization
- Providing appropriate visualizations in Tableau to communicate the insights gleaned from the datasets
Option 1 project type was selected and questions relating to real estate prices and affordability in Canada were explored and visualized. Detailed procedures about the question generation, answers, and steps leading to the visualization can be found in the notebook located in Submissions/Notebook. An abridged, and distilled, subset of the questions explored is provided below.
Since trend information are better captured with line charts, a line chart showing data about residential house prices, office real estate prices, and the overall housing index was used in this visualization. To aid clarity, all data were normalized using 2005 as the reference year. The percent change since the selected reference year was then used to give the datasets a common base. Results show that real estate prices have been consistently trending upwards, with residential house prices in particular accelerating in recent years.
Both a heatmap and a diverging bar chart were used in visualizing an answer to this question. From the results, it was gleaned that, interestingly, the price differences between the selected cities have been decreasing, as prices in the cheaper cities catch up to the overall average.
A line chart was used in visualizing an answer to this question. The result showed that house prices have been growing faster than the rate of increase in salaries/wages. Hence, house prices are increasingly becoming less affordable.
Annotations using reference bands/lines were used to highlight notable events, including recessions, on the trend plot of housing construction starts. The results showed that recessions that affect the broader economy lead to reduction in housing starts. Conversely, events isolated to specific sectors, like the Dot com Bubble, are less impactful on housing starts.
Details on more explored questions together with the associated Tableau visualizations can be found in the deliverables under the Submissions folder.
Overall, the project was undertaken relatively smoothly. The main challenges were with the datasets and the time limitation. Of note, the provided weekly earnings json file was unusable, and I had to re-source the file from the Open Data repository.
With more time, it would be interesting to enrich the present dataset and get more geographic granularity into the house prices across Canada.