Click the Original, Code and Reconstruction tabs to read about the issues and how they were fixed.

Original


Source: Statista (2023).


Objective

The original data visualisation, a pie chart displayed on Statista, aimed to demonstrate the distribution of the global population by continent in 2022. The primary audience for this visualisation would likely be individuals interested in demographic trends, such as researchers, students, policymakers, or the general audience curious about global population patterns. Understanding global population distribution is vital for various reasons, including policy planning, tracking demographic changes, and aiding in various types of global developmental research.

However, the visualisation exhibits a few key issues that could limit its effectiveness:

  • Limitation in Data Depth: The pie chart, by design, only provides a high-level overview of the population distribution. While it is efficient at providing a quick glance at the proportions of the population on each continent, it lacks granularity. This limitation potentially hinders the ability of researchers or policymakers to draw in-depth insights and make well-informed decisions. For instance, knowledge about the variation within each continent and data about specific countries could be critical for more localized policy-making and planning.

  • Accessibility Issues: The data is not completely accessible as it is part of a premium feature that requires users to sign up for a free account to access a bar graph representation. This creates an unnecessary barrier to access the information, especially for casual users who might not want to create an account.

  • Potential Color Scheme Problems: The chosen colour scheme, while visually appealing, may not be accessible to all users. For example, individuals with certain types of colour blindness may struggle to differentiate between the different segments of the pie chart. This is especially problematic when dealing with similar shades of a colour. Moreover, the implications of cultural color associations should not be overlooked, as the perception of colors can vary widely across different cultures.

Given these issues, my redesign will focus on creating a visualisation that provides greater depth of information, ensures accessibility of data, and uses a colour scheme that is visually distinct and accessible to a wide array of users, including those with colour vision deficiencies, and also culturally neutral.

The data set to be used for the redesign is a comprehensive list of populations by country and continent from the Population Reference Bureau. The chosen visualisation technique is a grouped or stacked bar chart, which offers specific advantages for addressing the issues identified. This format not only allows for representation of both the total population per continent but also gives detailed insight into the population distribution within each continent, thereby enhancing data depth. Furthermore, the careful selection of the color scheme will ensure visual distinctiveness and consider accessibility issues related to color vision deficiencies and cultural color associations.

Reference

Code

In the reconstructed data visualisation, I employed a stacked bar chart using ggplot2, a powerful visualisation package in R, to display the global population distribution across continents, with a further breakdown into regions. The data was sourced from the Population Reference Bureau, and the graph was designed with the initial problematic visualisation's objective and audience in mind: providing accessible, comprehensive information about global population distribution to demographic researchers, students, policymakers, and curious general audience.

The three most significant issues of the original visualisation that I addressed were:

  • Lack of Data Depth: While the original pie chart provided an overview of population distribution, it failed to deliver detailed, regional breakdowns within each continent. My revised visualisation, a stacked bar chart, addresses this issue by not only displaying the total population by continent but also revealing the population distribution within each continent. This greater depth of information adds value for those seeking more nuanced demographic data.

  • Accessibility: The original data was not completely accessible as it was part of a premium feature that required users to sign up for an account. To address this, I sourced data from an open-access resource, making the information freely available to all interested parties.

  • Color Scheme: The original visualisation's color scheme may have been challenging for individuals with color vision deficiencies to interpret. In the reconstructed visualisation, I adopted a carefully selected color palette that has been checked for colorblind-friendliness. I have chosen 19 sufficiently distinguishable colors, ensuring each region has a unique color and the visualisation is accessible to as many users as possible.

# Read the CSV data
df <- read_csv("2022-World-Population-Data-Sheet.csv")

# Aggregate the population by continent and region
df_aggregated <- df %>%
  group_by(Continent, Region) %>%
  summarise(TotalPopulation = sum(`Population (millions) mid-2022`)) %>%
  arrange(Continent, desc(TotalPopulation))

# Number of unique regions
num_regions <- length(unique(df_aggregated$Region))

# Color palette with as many colors as regions
palette <- c("#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7", 
             "#999999", "#004949", "#EE7733", "#009988", "#33BBEE", "#EE3377", "#BBBBBB", 
             "#24BC14", "#FDB515", "#5AC3E6", "#B3446C", "#F6768E")

# Create stacked bar plot
newGraph <- ggplot(data = df_aggregated, aes(x = Continent, y = TotalPopulation, fill = Region)) +
  geom_bar(stat = "identity") +
  labs(x = "Continents", y = "Population (millions) mid-2022", title = "Population by Continent and Region") +
  theme_minimal() +
  scale_fill_manual(values = palette)

Data Reference

Reconstruction

By addressing these issues, the reconstructed visualisation improves upon the original, providing a more detailed, accessible, and inclusive view of the global population distribution in mid-2022. Not only does it meet the original objective of demonstrating global population trends, but it also enhances the understanding of these trends by presenting a more granular breakdown of the data. The changes made did not introduce new significant issues, ensuring the reconstructed visualisation aligns with the audience's needs and retains its integrity.