Assignment 2: Deceptive Visualization
As visualizations increasingly play a role in how the general public consumes news and information, it is important to realize how a visualization design may influence what a viewer concludes and remembers about the data. In this assignment, you will identify a dataset of interest and design two static visualizations to communicate the data. One design will be an earnest representation of the data, whereas the other will be a deceptive visualization that aims to mislead the viewer. For this second image, the design challenge is to mislead without using obvious distortions or omissions.
Your task is to design two static (single image) visualizations of your chosen dataset. One visualization should aim to effectively and earnestly communicate insights about the data, whereas the other visualization design should aim to mislead the viewer into drawing the wrong conclusions. You will also provide a short write-up (no more than 4 paragraphs) describing your design rationale for both visualizations.
For this assignment, we will consider an earnest visualization to be one where:
- The visualization is clear and interpretable by the general population
- The visual encodings are appropriate and effective for the intended task
- Data transformations are clearly and transparently communicated
- The underlying data source (and any potential bias) is clearly communicated
On the other hand, a deceptive visualization may exhibit the following properties:
- The visual representation is intentionally inappropriate or misleading
- Titles are skewed to intentionally influence the viewer's perception
- The data has been transformed or filtered in an intentionally misleading way
- The existence or source of bias in the underlying data is unclear
For the earnest visualization, your goal is to be as clear and transparent as possible to help viewers answer your intended question. For the deceptive visualization, your goal is to trick the viewer (including the course staff!) into believing that the visualization is legitimate and earnest. It should not be immediately obvious which visualization is trying to be deceptive. Subtle ineffective choices in the design should require close and careful reading to be identified.
For both visualization designs, start by choosing a question you would like to answer. Design your visualization to answer that question either correctly (for the earnest visualization) or incorrectly (for the deceptive visualization). You may choose to address a different question with each visualization. Be sure to document the question as part of the visualization design (e.g., title, subtitle, or caption) and in your assignment write-up.
Your write-up should contain the following information:
- Your name and UW netid
- The specific question each visualization aims to answer
- A description of your design rationale and important considerations for each visualization
Recommended Data Sources
To get up and running quickly with this assignment, we recommend using one of the following provided datasets or sources, but you are free to use any dataset of your choice. You must use the same dataset for both visualizations, but you may transform the data differently or choose to address a different question for each design. These datasets are intentionally chosen to cover politically charged topics for the simple reason that these are typically the types of data where deceptive visualizations may proliferate.
The World Bank Data, 1960-2018
The World Bank has tracked global human development by indicators such as climate change, economy, education, environment, gender equality, health, and science and technology since 1960. You can browse the data by indicators or by countries. Click on an indicator category or country to download the CSV file.
Greenhouse Gas Emissions, 1990-2019
The Organization for Economic Co-operation and Development (OECD) has compiled data for the emissions of all participating countries broken out by the pollutant (e.g., carbon monoxide, methane, etc.) and by different sources (e.g., energy, agriculture, etc.). While the linked interface can be somewhat hard to navigate, you are free to browse alternate themes from the panel on the left (such as education or health). You can download the data by selecting "Export" at the top of the chosen table.
Data: Greenhouse Gas Emissions
DEA Pain Pills Database
The Washington Post has published a significant portion of a database maintained by the Drug Enforcement Administration (DEA) that tracks every opioid from their manufacturer, through to distributors, and into pharmacies in towns and cities across the United States. Note that this is a very large dataset with many different facets, so you may want to focus on a particular area or set of attributes of interest.
Important Note: In order to download and use this data you may need to enter your email in order to access the Washington Post Data Access page.
Here are some other possible sources to consider. You are also free to use data from a source different from those included here. If you have any questions on whether a dataset is appropriate, please ask the course staff ASAP!
- data.seattle.gov - City of Seattle Open Data
- data.wa.gov - State of Washington Open Data
- data.gov - U.S. Government Open Datasets
- U.S. Census Bureau - Census Datasets
- IPUMS.org - Integrated Census & Survey Data from around the World
- Federal Elections Commission - Campaign Finance & Expenditures
- Federal Aviation Administration - FAA Data & Research
- NOAA Daily Weather - NOAA Daily Global Historical Climatology Network Data
- yelp.com/dataset - Yelp Open Dataset
- fivethirtyeight.com - Data and Code behind the Stories and Interactives
- Buzzfeed News - Open-source data from BuzzFeed's newsroom
- Kaggle Datasets - Datasets for Kaggle contests
- List of datasets useful for course projects - curated by Mike Freeman
The assignment score is out of a maximum of 15 points. We will determine scores by judging the soundness of your visualization designs, the duplicity of your deceptive visualization, and the quality of the write-up. Here are examples of aspects that may lead to point deductions:
- Obvious identification of the earnest and deceptive visualizations.
- Ineffective visual encodings for your stated goal.
- Missing indication of the main analysis question.
- Missing or incomplete design rationale in write-up.
We will reward entries that go above and beyond the assignment requirements to produce effective (and deceptive) graphics. Examples may include outstanding visual design, effective annotations and other narrative devices, exceptional creativity, or deceptive designs that require the write-up in order to properly identify the misleading design components.
This is an individual assignment. You may not work in groups.
Your completed visualization images and write-up are due Wednesday 10/20, 11:59pm PT on Canvas. Submissions will be reviewed as part of a subsequent peer review assignment (due: Monday 10/25), so try to avoid a late submission; assignments submitted late may not be included as part of the peer review and thus not receive peer feedback.
You must submit your assignment using Canvas. Please upload two image files (PNG or JPG) of your visualization design, named using the pattern "uwnetid_a2_earnest.png" (for your earnest chart) and "uwnetid_a2_deceptive.png" (for your deceptive chart), replacing "uwnetid" with your UW network login; this is the same as your @uw email address, not a numeric id number. Please follow the file naming convention exactly, use the correct file extension for your images (either .png or .jpg), and be sure your images are sized for a reasonable viewing experience. Viewers should not have to zoom or scroll in order to view your submission!
Please use the correct file names for your submission; typos that require manual correction by the course staff may result in point deductions. Do not worry about resubmissions: Canvas will automatically add suffixes to files if you upload multiple versions. This is fine, you don't need to worry about the suffix. Remember, the visualization itself should not give away which design is earnest and which is deceptive; the file names will be randomized by the course staff prior to peer review.
In addition, submit your write-up as a plain text file, named using the pattern "uwnetid_a2.txt", with content that follows the instructions above.
The design of this assignment is largely based on the experience and recommendations of Niklas Elmqvist at the University of Maryland, College Park, and a similar assignment from Arvind Satyanarayan from the Massachusetts Institute of Technology.