Wanting to expand on the research I started for my Timeline lab, I sought disease related data to analyze and visualize with Tableau. After being disappointed with the minimal datasets my initial search turned up, I was ecstatic to discover WHO’s Mortality Database. So ecstatic that I neglected to heed the opening warning, “These files do not constitute a user-friendly data collection which the average user can download and access.“
I also refused to be deterred when the dataset exceeded Excel’s limit of 1,048,576 rows, as I was able to open and work with it directly in Tableau. Although the data was overall very clean, it did require quite a bit of work transposing country codes into more useful country names (geographic locations) and cause codes into diseases. Upon completion, I then created groupings of larger, more encompassing causes of death categories. Using WHO’s broader subcategories as guides, I was able to sort over 100 individual causes into 18 major groups. This Excel documents my grouping decisions.
Despite the massive size of the WHO dataset, I quickly realized its limitations. WHO relies on countries to voluntarily submit data resulting in over one third of the world’s population being discounted. The below (very incomplete) map displays the countries included in the dataset.
The limited global reach of WHO’s data also made it impracticable to pursue my initially planned investigation of infectious diseases. Realizing the challenges of getting too granular with my research, I pivoted to a bigger picture approach – ultimately displayed in three Tableau dashboards.
The first dashboard, ‘WHO Mortality Data‘, introduces the data and acknowledges its limitations.
The second dashboard, ‘Global Mortality’, takes a closer look at causes and ages of deaths
Finally, the third dashboard, ‘How will I Die?’, is a snapshot of leading causes of death per age range.
Although I was unable to tell the story I was hoping to express with this dataset, there is still much more that could be done with it. Tying in additional data on countries’ populations (also available from WHO) would help to put these numbers into perspective. Even though causes of deaths are difficult to analyze (due to the varied reporting techniques) more time could be spent reflecting on specific age groups. Further exploration of the data would also allow for further exploration into Tableau’s capabilities.