Source: Mapping Inequality: Redlining in New Deal America (v3)
…is to actively incorporate equity awareness when working with data and generating data products (e.g. visualizations). Not all visuals are overtly racist or discriminatory, but we as data practitioners must recognize and work against misusing or misrepresenting data in ways that can harm communities and perpetuate systemic discrimination.
Urban Institute’s Do No Harm Guide, by Jonathan Schwabish and Alice Feng (2021)
guidelines on how to apply a DEI lens to not just words, colors, icons, etc., but also the process of crafting communication products
includes checklists & tool kits that focus on the often hidden / subtle ways that data analysts and communicators fail to incorporate equitable awareness in the data they use and they products they create
Scheme by Andy Kirk from his book Data Visualisation: A Handbook for Data Driven Design, recreated by Cédric Scherer
Discuss how you perceive, interpret, and comprehend this map with your learning partner(s). What thoughts or emotions does it invoke? Why do you think you feel those emotions?
04:00
- Arrows reminiscent of battle map
- Red = color of danger
- LARGE, direct arrows (larger than most countries)
- Title language portrays criminal activity
- Doesn’t show where migrants actually come from
Discuss how you might redesign this map to make us feel less
“like we are being overrun by huge numbers of anonymous enemies, coming at us en masse from all corners of the globe to disrupt our orderly lives”
-Vermulen, De Korte, Van Houtum
04:00
Put people first
“if your data [are] about people, make it extremely clear who they are or were”
-Jacob Harris in Connecting with the Dots
Each square represents US service member who died in Iraq or Afghanistan. Source: Faces of Death (New York Times); (Note: this is an archived page and a bit glitchy)
Put people first
Use personal connections to help readers and users connect with the material
Pair data-driven charts with personal stories, or help readers understand the “far vs. near” (i.e. overall metrics + smaller ranges / groups of data)
This map focuses on one of the deadliest days in Baghdad, with 114 separate episodes of violence resulting in the deaths of 160 Iraqi citizens and police officers | Source: A Deadly Day In Baghdad (New York Times)
Put people first
Use personal connections to help readers and users connect with the material
Use a mix of quantitative and qualitative approaches to tell a story
Focusing on just numbers without context can overlook important aspects of a story, such as the “why” and “how”
Mapping the theme of EV charging convenience, which was most tied to specific places. Above, transcript snippets around positive associations with convenience. | Source: Attitudes and Experiences with Electric Vehicles, by Diana Lavery
Put people first
Use personal connections to help readers and users connect with the material
Use a mix of quantitative and qualitative approaches to tell a story
Create a platform for engagement
Interactivity allows users to find themselves in the data or discover stories.
Explore reports by region and officer. | Source: Citizens Police Data Project, by Invisible Institute
Put people first
Use personal connections to help readers and users connect with the material
Use a mix of quantitative and qualitative approaches to tell a story
Create a platform for engagement
Consider how your framing of an issue can create a biased emotional response
Crime maps are designed to show the frequency and location of crime incidents.
Trulia crime map (San Francisco) | Source: When the Designer Shows Up In the Design, by Lena V. Groeger (ProPublica)
Put people first
Use personal connections to help readers and users connect with the material
Use a mix of quantitative and qualitative approaches to tell a story
Create a platform for engagement
Consider how your framing of an issue can create a biased emotional response
“Rather than looking at where crimes are committed, we looked at where prisoners live, and the maps that resulted showed the urban costs of incarceration and suggested how those dollars might be better spent on investing in communities.”
Source: Million Dollar Blocks, by Columbia’s Center for Spatial Research
Put people first
Use personal connections to help readers and users connect with the material
Use a mix of quantitative and qualitative approaches to tell a story
Create a platform for engagement
Consider how your framing of an issue can create a biased emotional response
Recognize the needs of your audience
Make sure visualizations are accessible so that folks with disabilities can view and use content online. Similarly, avoid overly technical or jargon-heavy language that make information inaccessible to a broad audience.
Image sources: NCEAS, Design102, call focus
Some chart types (e.g. bar, line, pie) abstract content by collapsing all people represented into one shape. Instead, try using individual points or even icons (i.e. anthropomorphize your data graphics).
Source: Sabah Ibrahim (@reina_sabah), on X (formerly known as Twitter)
Screenshot from ProPublica’s graphic that allows you to visualize unemployment data aggregated at any three of the following intersections: race, gender, age class, education level, income level:
The US Census Bureau’s annual American Community Survey collects social, economic, housing, and demographic data which help to inform how trillions of dollars in federal funds are distributed each year.
While the official U.S. poverty rate in 2019 was 12.3%, the plot below, by Schwabish & Feng (2021) (figure 08) shows the large variation in estimated poverty rates for each of the 139 racial groups from the American Community Survey. The overall poverty rate for each major racial groupings are included as well.
Aggregation is sometimes used to combine multiple (or many) groups when samples sizes are too small to display them individually. Consider adding a message / reference to small or missing groups to be inclusive:
Source: Screenshot from a UCSB-administered survey (Feb 2024)
“Collecting inclusive data and building inclusive tools and visualizations can make the experience better for all users.”
Some alternatives:
Another race (or gender, etc.)
Additional groups
All other self-descriptions
People identifying as other or multiple races
Identity not listed
Identity not listed in this survey
04:00
02:00
02:00
02:00
03:00
03:00