Data science ethics: Misrepresentation

Lecture 11

Dr. Mine Çetinkaya-Rundel

Duke University
STA 199 - Fall 2022

10/4/22

Warm up

While you wait for class to begin…

Clone your ae-08 repo.
Any questions from prepare materials? Go to slido.com / #sta199. You can also upvote others’ questions.

Announcements

Slight change of plans for this week – dedicating entire week to data science ethics
Make sure to catch up with reading and (required) videos by Thursday
Highly recommend watching thew optional videos as well – if not by Thursday, over the break

Misrepresentation

Stand your ground

In 2005, the Florida legislature passed the controversial “Stand Your Ground” law that broadened the situations in which citizens can use lethal force to protect themselves against perceived threats. Advocates believed that the new law would ultimately reduce crime; opponents feared an increase in the use of lethal force.

What does the visualization, published by Reuters on February 16, 2014, say about the number of firearm murders in Florida over the years?

03:00

COVID cases

In May 2020, the state of Georgia published the following visualization. What does this plot say about the number of COVID cases in the most impacted counties?

What changes need to be made to this graph to fix it and make it less misleading?

03:00

Another one on COVID cases

What is misleading about the following two visualizations (from here) that are presented together to show the trajectory of COVID?

How it started (Jan 3, 2022)

How it’s going (Feb 10, 2022)

Redistributive taxes

What is misleading about the following visualization (shared here) about Americans’ views on redistributive taxes?

Application exercise

`ae-08`

Go to the course GitHub org and find your ae-08 (repo name will be suffixed with your GitHub name).
Clone the repo in your container, open the Quarto document in the repo, and follow along and complete the exercises.
Render, commit, and push your edits by the AE deadline – 3 days from today.

Recap of AE

Not all visualizations are equally informative. But, perhaps more importantly, some might be misleading.
Always be cautious of a visualization, particularly one with an agenda, at the first glance. Look closely to check if it might be put together to mislead, instead of clarify.