J-PAL’s use of inclusive language to communicate research results

 Different people may have preferred terms for their own specific identity, so when possible and appropriate, ask people how they identify or try to find out more information if it would not be appropriate or feasible to ask directly. Deferring to an individual’s self-identification confers autonomy and respect, helps avoid the pitfalls of assumptions, and bolsters precision. For example, “Latinx” is a common gender-neutral term for people of Latin American nationality or ancestry, but many in the United States don’t use the term to identify themselves. 

People-first language emphasizes an individual’s agency, autonomy, and personhood, rather than a stigmatized issue or barrier they may face. In instances where you are not able to learn how someone personally identifies (as is often the case when writing about study results), people-first language is a best practice. For example, housing advocates in the United States often use the phrase “person experiencing homelessness” rather than “homeless person.” This practice may also be relevant at the community level, such as opting for “neighborhood with a high rate of poverty” over “high-poverty neighborhood.” Following this practice recognizes that individuals and communities are not solely defined by a single (often time-limited) characteristic.

This is a simple way to avoid generalizing about groups of people, such as participants in a study. Generalizations about or euphemisms for a particular group of people lack nuance and precision, may lead to feelings of erasure, and can reduce the accuracy of results reporting. If, for example, the community we are describing is primarily Black, it may be more appropriate and accurate to specify that, instead of using a broader description such as “people of color.”

While we want to be specific when referring to certain groups of people, we know that 1) large heterogeneous groups contain people across the gender spectrum and 2) people may not identify as the gender we perceive them as. When addressing groups—such as when presenting research findings—it may be best to avoid conventional greetings (e.g, “ladies and gentlemen”) in favor of a more inclusive “welcome, everyone.” If we do need to refer to binary genders, such as when reporting study results, the terms “women” and “men” can be more appropriate than “females” and “males,” particularly when being used as nouns.

Disaggregating effects along the basis of social identities like race or gender can illustrate that many inequalities often disproportionately impact certain people of color, women, people with disabilities, and those with other marginalized identities. If disaggregating effects is not possible due to sample size or other factors related to the study design or study population, it may be helpful to note this in an endnote. J-PAL’s research resource “A Practical Guide to Measuring Women’s and Girls’ Empowerment in Impact Evaluations” demonstrates how disaggregating effects by gender is important to uncover if a program has differential impacts that may be masked when data is not disaggregated.

 Using positive or fact-based statements, rather than normative or opinion-based ones, helps us avoid bias. Using phrases that situate human behavior, disparities, and social problems in their broader historical context can help shift any presumption of blame from individuals to structures and policies. For example, it may be more accurate to note decreases in arrests rather than decreases in criminal behavior, and more apt to refer to this behavior—if at all—as criminalized, noting the structures in place that create and enforce laws. A study on criminalization in the United States may also benefit from a brief discussion of mass incarceration and historical policies that led to the current context.

Ask yourself, “would I feel comfortable reading my document out loud to the person or community I am describing?” Part of J-PAL North America’s work around diversity, equity, and inclusion is to proactively hold ourselves accountable to the communities we aim to support. Doing this quick exercise when reviewing a document can help us catch words, phrases, or overall framing of issues that don’t align with our best practices. If we are constrained by outside factors, such as when a program’s intake form only provides two gender options, we may note these parameters explicitly.