A deep-dive into data surrounding the Olympic Games!

This was completed by Oliver Quarm (NetID: oquarm2, Group 10) for IS445’s Final Project

Introduction

For the final project I analysed extensive data for the Olympic Games, spanning over a century! My goal was to obtain the most interesting data among the datasets I looked at, and present them in the most interesting way possible! To complete my mission, I used Altair and Vegalite.

My notebook can be found by clicking the button below (it should be noted the notebook doesn’t seem to fully render on GitHub – it stops rendering just before the main visualization is suppose to appear):

The Notebook!

Primary Dataset/Visualization

My primary dataset was a massive CSV of individual entries of Olympic athletes, including their age, height, weight, but more importantly their nation, event, and any medals won. The dataset is titled ‘120 years of Olympic history: athletes and results’. The dataset is from Kaggle, posted by the user @rgriffin, accessible here: https://www.kaggle.com/datasets/heesoo37/120-years-of-olympic-history-athletes-and-results. This user gathered the data from an organization called sports-reference, whose Olympic data has now branched off as Olympedia. It’s a dataset spanning 120 years of Olympic Games statistics revolving around athlete & medal information.

Below is my main visualization (it’s interactive!) for this dataset:

Contextual Dataset/Visualization

For the contextual dataset, I looked at more high-level, wholistic data for each Olympic Games host nation. I made 2 visualizations: 1 was the spending amount for various host nations, and the second was the number of events various host nations had for their Olympic Games. I sourced this dataset from here. The 2 visualizations are displayed below:

What’s happening in the charts?

Primary Visualizations:

I have a main dashboard, which was created using mark_rect() from Altair. It shows for each country that’s participated in the Olympic Games, the number of medals they’ve won overall at each Olympic Games city location that’s ever happened. The more medals a country has won in a specific city, the dark the shade of the box!

I’ve made it interactive such that:

I think this visualization dashboard is really visually appealing. I think it’s also useful for spotting trends. I found it interesting for example that when a given nation performs in a city of their own, they have a higher medal count than if they were competing in a different city. Look at GBR’s performance in the London games, Australia’s performance in the Sydney games!

Contextual Visualizations:

For the contextual dataset, I made 2 bar graphs. One was illustrating how many events various host cities put on. For each country there is 1 total bar, but within the bar can be many shades of blue. The number of shades is equal to the number of different times that country has hosted the Olympic Games. The USA holds the record for hosting it 5 times, and so naturally has held the most events in total ever. For a single instance of an Olympic Games, China actually holds the record for the most events offered at 302.

The second bar chart shows cost various countries have endured when hosting the games. Like the first chart, for each country there is 1 total bar, but within the bar can be many shades of blue. The number of shades is equal to the number of different times that country has hosted the Olympic Games. Interestingly, even though the USA has hosted the most amount of Games, it hasn’t spent the most money! Russia has for its 2014 Sochi Winter Games – a staggering USD$21.89 billion!

References: