Billionaires: Wealth in Numbers (Process Book)
Team: Best Of Four
-
Elizabeth Jeon
- ID: 490460
- e.jeon@wustl.edu
-
Sam Nordhus
- ID: 485897
- snordhus@wustl.edu
-
Oen McKinley
- ID: 509427
- m.oen@wustl.edu
-
Eric Tabuchi
- ID: 501415
- e.tabuchi@wustl.edu
Original Visual Designs
When it came to initially tackling the problem of trying to visualize a dataset of billionaires, our team thought about the need to see how they stack up with each other and possibly with the common person.
When it came to simple visuals, we thought about a basic "Hall of Fame" that would rank the top X billionaires with basic information alongisde their name as it would fit the rich person aesthetic.
Following this initial train of thought, we came up with the idea of a scale where users could drag a billionaire from the top X wealthiest billionaires and see how much wealthier they are than a selected person from a given input country.
This would showcase the disparity of wealth between the richest and average of society. This could then be juxtaposed by going from the individual to a collective by having a map visual showing off where in particular billionaires resided.
If given time and ambition, this map would be capable of "zooming" into displaying a new map view of the specific country in question such that the markers displayed become more visibly apparent to the user.
Other than these ideas, our thoughts also turned towards the basic visualizations learned in class, ranging from a standard bar or pie chart as well as a heat map. For something like the bar chart and heat map, there would be
added functionality to specifically sort / filter through to see certain billionaires under given conditions in addition to seeing the range of wealth under those conditions.
Overview and Motivation
Our motivation is to understand the factors that contribute to billionaire wealth as well as their demographics to see if it
can shed light on economic trends. We also want to know if the distribution of their wealth will align with a broader interest in understanding
economic inequality and its implications in the broader context of social sciences. Our academic interests revolve around economics, sociology, and data science.
We aim to combine elements from these fields to visualize the wealth of billionaires in hopes of seeing how the wealthy rank amongst themselves and the common person.
Related Work
What inspired us was the narrative design approach to visualizations. In our particular case, we felt it would be effective storytelling by easing
the user from simple to more complicated visuals. Starting with pie charts showcases basic distributions of billionaires, leading it down to a bar chart and
map that will unearth more specific and intricate details not immediately noticeable. As a result, we were inspired by easy-to-understand visuals to achieve this
narrative goal.
Questions
We are trying to answer how billionaire wealth is distributed amongst humanity. We are interested in knowing if there are specific industries and trends noticeable in
some of the wealthiest people on the planet. We are curious to see how bilionaires in the same geographical area stack up compared to others. For example, are billionaires
from nearby European countries similar in how they became billionaires or do they reflect the industry typically seen by that country. These
questions will evolve throughout the project as we uncover interesting observations, causing us to ask why so many billionaires are grouped in such a way.
Data
The datasets were collected from Kaggle and the World Population Review which provide information on billionaire statistics and country income respectively.
The data was filtered specifically to fit the needs of the visualization requirements. For something like a pie chart, more features will filtered out and passed
into the visual to display. To fit the needs of the geographical visuals, additional data was added to the billionaire dataset in the form of adding the longitude
latitude values corresponding to where they generally live. Additional cleaning of the dataset was done to make sure countries of different geographical datasets had
matching names / values to ensure ease of implementation.
Exploratory Data Analysis
The simplest visuals (regardless of whether or not the visual was simple to implement) was the first we initially used to look at surface level information.
The pie charts provided a means of showing how billionaires were distributed in very simple ways which fed into how the bar chart would be designed
to sort for more finer specificities behind what made up the slices in the pie charts above it. The map chart was another initial design to answer a simple question
about billionaires (i.e. where they tend to be located). We have gained insights into how we can utilize the initial findings into improving what currently
exists while also influencing the design of future visuals.
Design Evolution
Initially, our map chart was made using Leaflet which provided a good starting ground for visualizing billionaires
by geographical location. Our design might evolve to utilize a different means of displaying location data in a way
that allows for more customization options in how those billionaires are plotted on the globe. We also considered the
use of a heat map instead of a chloropleth map to visualize the population density of billionaires, but it was justified
that the use of two different map visuals would align well with each other and thus provide more clarity overall.
Implementation
The intent of the interactive visualizations is to have the user hover and discover for themselves aspects about billionaires.
When it comes to the bar chart, the intent is to have the custom color coding and sorting methods to reveal immediate trends based
on who the wealthiest people were based on the parameters set.
Evaluation
What we have learned from the data by using the visualizations is that billionaires are on the older side. Considering the size of countries like
the United States and China, it is not suprising that those two had the most billionaires overall. Our visualizations are effective in showcasing
trends or groupings within billionaires based on where they live and how they became billionaires. Improvement could be made into the narrative storytelling
in showcasing how potentially specific trends align up with a desired story we would like to tell.
Milestone 1
Our first milestone involved laying the groundwork for our visualizations. We created the visuals we felt were sufficient to tell a narrative about billionaires and simply
stacked them on top of each other to easily check each other's progress on them. Everything other than the visuals were made temporary before finalizing the styling and eventual final layout.
Milestone 2
With Milestone 2, we have created a single "screen" that displays one visual at a time. Pressing on the tab bar buttons at the top switches to one of the many visuals along
with an appropriate description provided detailing what the visual represents. That way, things become more compact and less overwhelming for the user.
Design Evolution
When it came to the initial milestone 1 results, there seemed to be a sufficient number of visuals available to convey the message and information we wanted to tell about billionaires.
At the same time, given the number of visuals on-screen at once, it did seem a bit confusing and overwhelming for a user to encounter. Given the need to also scroll vertically, there might be
crucial information that the user may not see at the time of interacting with the specific visual. As such, it was decided to, when deciding to stylize the webpage, update how the visuals are
displayed. Instead of multiple visuals on-screen at once, only one visual will be "active" at a given time with a series of buttons that switch the display over to the other visuals. For added clarity,
switching a visual will also update a description that describes what the visual is and what features exist for it (including units). That way, users can decide from themselves how they would want to experience
the visualizations.
User Testing
For user testing, we wanted to know how easy it would be for the user to accomplish simple tasks. We compiled a series of tasks involving all visualizations, starting from the basic pie
chart to eventually using the complex Leaflet map. Here is the list of tasks we came up with for testing purposes:
- Can you figure out what percentage of billionaires are male?
- Who is the highest ranking billionaire in the United States?
- What billionaire has the highest rank in the Technology industry?
- Who are the top 5 billionaires in St. Louis?
- Which countries have the most billionaires?
User Tesiting Results
Percentage of Male Billionaires:
- 87.2%
- 87.2%
- 87.2%
- 87.2%
- 87.2%
Highest Ranking Billionaire in the United States:
- Elon Musk
- Elon Musk
- Elon Musk
- Elon Musk
- Elon Musk
Billionaire with the Highest Rank in the Technology Industry:
- Jeff Bezos
- Jeff Bezos
- Jeff Bezos
- Jeff Bezos
- Jeff Bezos
Top 5 Billionaires in St. Louis:
- Paul, David, Jim, Rodger, Robert
- Paul, David, Jim, Rodger, Robert
- Paul, David, Jim, Rodger, Robert
- Paul, David, Jim, Rodger, Robert
- Paul, David, Jim, Rodger, Robert
Countries with the Most Billionaires:
- United States
- India, China, U.S; after showing which visual to use: U.S, China, darkest..
- U.S, China (darker countries)
- U.S, China, Russia
- U.S, China
Summary of General Feedback:
- Clear filter and reset button for bar chart
- Better chart titles on buttons
- Better indication of which visualization is which for buttons
- Use a more contrasting color for dots on leaflet map
- Show legend for the world map (legend cuts off)
Edits we made
- Made a clear button for bar chart
- Changed dot colors to black for map
- Edited maps so legends are not cut off
- Fixed pie chart hover function bugs
- Fixed descriptions for each map
Evaluation
Through our visualization, we learned a lot of insights. For example, we noticed that countries that generate less wealth have less billionaires.
One tester was surprised to see how many billionaires were in the U.S. It was also interesting to see that billionaires are more likely to be Male, Older, and Self-made.
Technology is generally the wealthiest sector Noticeable gaps in wealth between billionaires, especially in certain industries Elon Musk dominates Automotive industry.
I think the combination of all our four visualizations provides an great overview of a socioeconomic analysis of billionaires. Our visualization could be improved further
by potentially collecting more data about other socioeconomic factors like education levels per country, gini indicies (income equality), happiness indicies, and ease of business indicies.