Use People-First Language and Labels

Learning Objectives

After completing this unit, you’ll be able to:

• Generate people-first language in your work.
  
• Discuss how titles, text, and labels affect meaning.
  

The Impact of Labels

It’s important for data communicators and researchers to remember that data reflects the lives and experiences of real people. If your data is about people, make it extremely clear who they are.

In visualizations, labels matter. Language is living, breathing, and ever-changing. It’s only logical that certain labels that were used in the past are no longer acceptable—and may in fact, be offensive. In your data analyses, for example, the best approach is to use full labels such as Black people, not Blacks. 

The language in the legend below is noninclusive. On the More Black and More Poverty scales, Poverty refers to an experience, not a static description, and More Black refers to skin color, not people. More inclusive language might be “Larger proportion of people experiencing poverty” and “Larger Black population.” In fact, this legend was later changed from More Black to Larger Black Population to put emphasis on people, not skin color. 

Use People-First Language

The labels you use can play a role in perpetuating racial stereotypes and other forms of oppression. In your visualizations, you can address this by naming these forces of oppression, how they operate, and their historical context through titles, annotations, labels, and notes rather than burying them in surrounding text.

Examining the words you choose for titles, text, and labels reveals a great deal about how the world views a group of people. In crafting these labels, data storytellers can resist many forms of oppression and strive to make language “people first.” To do this, you first have to understand the labels you currently use.

Emphasizing the person rather than the quality (people with disabilities rather than disabled people, or Black people rather than Blacks) humanizes the information instead of cataloging people into a statistic. There are many nuances to data gathering that can be ignored or lost when the label offers a static description. 

For example, a study referred to incarcerated people as “inmates” when measuring the rate of mental health diagnoses they receive. While inmates may seem neutral and objective, it dehumanizes people by labeling them by their crimes and punishments. Inmates in this context also ignores the role racism and discrimination play in how likely incarcerated people are to receive a mental illness diagnosis. 

A more accurate reflection of the findings of this particular study could be that people of color are less likely to get a mental health diagnosis or even that white people get more mental health diagnoses. This shifts the focus from what people of color lack to the unfair advantages of the dominant group and racial disparities in the jail system, and references people rather than inmates.

Researchers and data analysts must talk with communities they’re studying, and with the consumers of the research, to understand and identify their preferred terminology to gain more ownership of the data, research, and policy recommendations.

Be Mindful of Evolving Language

Surveys or datasets that use outdated or less preferred terminology can present a tricky situation in terms of language. After all, language is fluid. Terms and phrases used to describe people and communities are constantly evolving. 

For example, the term Latinx is being used instead of Hispanic/Latino. Sometimes terms take time to gain a foothold in usage. Or there might be differences in preferred terms stemming from the intersectionality of politics, age, and race or ethnicity. You may be understandably reluctant to change a word or phrase used in an original survey that’s part of a final report, graph, or dashboard. But it’s necessary when communicating data with an empathetic and inclusive perspective.

Order Your Data with Intention

Often, little thought is given to how estimates in tables or bars in graphs are ordered other than how they appear in the raw data. Similarly, the impact of this ordering is overlooked. This is another area where you can make your work inclusive. There is a historical legacy from which this sort of ordering and implied racial hierarchy derives. Many of the largest demographic surveys conducted in the United States order race starting with white and Black as the first two options.

Who is shown in the first row in a table or the first bar in a graph can affect how readers perceive the relationship or hierarchy between groups. How you order your data may unintentionally reflect who you view as the default group against which others are compared, or who is the intended audience for your visualizations. Always starting with white, men, or straight can make these groups appear as the most important groups. 

When deciding how to order racial and ethnic groups, consider:

• If your study focuses on a particular community, present that group first.
  
• The final order should reflect the point you are trying to make.
  
• If there is a quantitative relationship that can guide how the groups are ordered (sorted alphabetically, by population size, sample size, or effect of the results), opt for that.
  

Resources

• Guide: Do No Harm Guide: Applying Equity Awareness in Data Visualization
• Website: Diversity, Equity, and Inclusion in Data Visualization: General Recommendations
• Website: Race and Ethnicity in the US Census