Data Visualisation Critique

Critique 

The visualisation chosen had three issues which are perceived bias, colour issues, and deceptive methods. 

Perceived Bias 

The visualisation consist of three hierarchy and it is a combination of a pie chart and a donut circle chart. Pie chart and donut circle chart utilises the division of a specific area (or proportion) of the chart to display a section (variable) which makes it hard to assess each section proportion based on the area alone. Hence, this visualisation is not helping reader to understand the data but may actually steer reader from understanding the factual data. 

Colour Issues 

This visualisation uses various colours in the shade of red, pink, green, blue, and grey. 8.4% of the human population suffer from colour blindness and that red-green blindness is the most common colour blindness in the world. In order to create a visualisation that is inclusive, this aspect must be considered. In this visualisation, since various shades of red and green is being use it will be very hard for red-green colour blind individual to understand the chart. Therefore, this visualisation does not support inclusivity. 

Deceptive Methods 

The visualisation contains three hierarchical data information. However the third and last hierarchy was only applied for a selected section of the pie chart instead of the whole chart. For example sector "Iron and Steel" is a part of the "Energy use in Industry" which is a part of "Energy" sector. However, "Unallocated fuel combustion" is also part of the "Energy" sector but has no further breakdown within the chart. At a glance, reader will be drawn to the area which has three hierarchy and would assume that this section is the biggest and most crucial aspect of greenhouse gas emission. Although this may be true, it is still unfair to blindly "forget" the other sectors of greenhouse gas emission that is being shown in the visualisation as a whole. 

It seems like the author lead their reader to only focus on the sector which has three hierarchy. Furthermore, having some part of the visualisation containing three hierarchy while others only have two creates inconsistency within the visualisation and questions the credibility of the visualisation altogether (despite having a credible data source). 

Reconstructed Visualisation

Reconstruction of the original visualisation to a horizontal bar chart is achieved using R programming language, with x-axis showing the emission percentage while the y-axis shows the sub-sector. We have chosen to only visualise the first and second hierarchy as it has the most detail and complete information within the visualisation. The bar chart is also colour-coded based on the first hierarchy: "Agriculture, Forestry & Land Use (AFOLU)", "Energy", "Industrial Process", and "Waste"; this is identical to the inner pie chart within the original visualisation. 

I used the Colour Brewer tool to determine which colour scheme is colourblind friendly for the reconstructed visualisation and I applied it to the graph. The decision of the colour used also depends on the type of data that we are displaying which is a categorical data and that a complementary colour scheme is used to represent data like this. 

Full Project PDF