Project Checkpoint 1 - Example

Note

This is a minimal example representing “passing” or ‘C’ level work, to give you a baseline for how to proceed. ‘B’ or ‘A’ level work will require a more thoughtful and thorough description of the data and cleaning, deeper or more complex research questions, and/or more polished visualization sketches. Some notes about areas for improvement are included in the callouts below.

Introduction

My dataset is the Olympics data from TidyTuesday. This dataset shows all the Olympic athletes ever, and gives information about their sport, the country they are from, whether they won a medal, etc.

The data comes from a Kaggle dataset created by user RGriffin, who scraped the data from www.sports-reference.com in May 2018.

Here are a few rows of the dataset:

tuesdata <- tidytuesdayR::tt_load('2024-08-06')

tuesdata$olympics |> 
  dplyr::sample_n(6)

id	name	sex	age	height	weight	team	noc	games	year	season	city	sport	event	medal
3229	Gyula Alvics	M	28	186	91	Hungary	HUN	1988 Summer	1988	Summer	Seoul	Boxing	Boxing Men’s Heavyweight	NA
31516	Alfred Frederik “Fred” Eefting	M	21	183	94	Netherlands	NED	1980 Summer	1980	Summer	Moskva	Swimming	Swimming Men’s 200 metres Backstroke	NA
35269	Solenne Nadge Figus (-Saint Marie)	F	25	178	59	France	FRA	2004 Summer	2004	Summer	Athina	Swimming	Swimming Women’s 200 metres Freestyle	Bronze
36759	Robert Frank	M	34	NA	NA	Switzerland	SUI	1936 Summer	1936	Summer	Berlin	Art Competitions	Art Competitions Mixed Sculpturing, Reliefs	NA
80591	Aiko Miyamura	F	24	173	60	Japan-1	JPN	1996 Summer	1996	Summer	Atlanta	Badminton	Badminton Women’s Doubles	NA
20202	Chen Hsiu-Hsiung	M	32	169	70	Chinese Taipei	TPE	1968 Summer	1968	Summer	Mexico City	Sailing	Sailing Mixed Three Person Keelboat	NA

Note

Good:

• This description explains who created the data, from where, and when. It provides links to the sources so the reader can find them.

• It somewhat explains what information is in the dataset; i.e., information about athletes, as opposed to scheduling information or other details.

• Showing a random small snippet of the dataset is typically useful, to get a feel for what it looks like.

Bad:

• This is lacking in the “why” and “how” of the data creation. It is important to know what motivated the creation and sharing of a dataset, as well as the specific process that the creator used to assemble it.

• This description is unclear about the observational units, i.e., what each row represents. (It is not true that each row is a unique athlete!)

• This description could give much more detail on the observed variables present, such as how they were measured and what a typical value looks like.

Data Cleaning

This data was cleaned by using janitor to reformat column names.

The user RGriffin who scraped the data also checked for misentered data in the columns Name, Gender, Height, and Weight.

Note

Good:

• We did not just look at the Tidy Tuesday cleaning and call it a day! We followed the path of the data creation to figure out what other cleaning and wrangling took place.

Bad:

• We were not specific enough about the cleaning done by RGriffin. What did they alter in those columns and why? What other anomalies did they look for?

Explorations

RQ 1: Olympic success by country

Which countries win the most Gold, Silver, and Bronze medals?

RQ 2: Sex over time

Are there a higher percentage of female sports over time?

Note

Good:

• These two research questions are well-defined and answerable with the dataset.

Bad:

• These are not particularly deep. “Which countries win the most?” is not going to provide a new insight beyond what most people already know, and I think we can expect from the outset that female sports have increased over time.

RQ 3: Olympic success corrected by population

Which countries win disproportionately more medals compared to their population size?

Additional Data: Populations of each country

RQ 4: Correlation between female wins and education

Do countries with better support for education for women also tend to see more success in female sports categories?

Additional Data: Education trends in each country

Note

Good:

• Question 4 digs deeper into the relationships between a country’s culture and government; and outcomes at the Olympics. This is a good question.

Bad:

• Question 3 is a bit less interesting - you might uncover 1 or 2 interesting countries that win disproportionately, but you aren’t telling a rich story.

Visualizations

Note

Good:

• Plots are appropriate to data type and address RQs

Bad:

• Not a lot of thought is put into these as far as good plot design, annotation and storytelling, etc.