Collecting and using race and ethnicity information in linguistic studies

Abstract

Despite being regarded by some as the most humanistic of the social sciences, linguistics has been criticized for its undertheorized application of the notions of race and ethnicity. This white paper is written for practicing linguists. We provide definitions of these terms and develop attendant issues that contribute to their complexity, such as the multiplicity and fluidity of racial identification. A survey of methods texts reveals that limited attention is given to race and ethnicity in training researchers. To address this need, the bulk of the paper uses a “challenges and recommendations” format to work through common design concerns and suggest better practices. We consider issues pertinent to collecting information about self-identification in a range of study types, from quantitative, experimental, computational or intuitional approaches to qualitative and mixed methods designs. We consider the advantages and disadvantages of eliciting demographic data using multiple-choice, free-response and interview formats, and offer recommendations drawing on best practices from within linguistics and its sister fields. Ethical concerns are raised, including using locally constructed labels, respecting communities, analyst positionality, recognizing the potential for harm. Throughout, brief examples are provided where possible to speak concretely to linguists’ concerns. The final section presents a detailed case study of the decision-making process for a multi-phase research project in which ethnic identification was explicitly investigated. Our goal is to provide researchers with tools to reflect on their own study design, reflect on their own responsibility to participants and communities, and design study prompts that allow more nuanced representation of race or ethnicity.