Perhaps you are designing a clinical trial or creating demographics for developing a survey. The cardinal rule whispered throughout your enterprise--garbage in, garbage out reminds you to ask better questions.
Questions that will most likely be interpreted the same by respondents or as close to "same" as we can muster are carefully developed. You have reviewed research methodology to understand the constructs of how we ask questions and what types of questions are the best design for gathering rigorous and actionable data.
You have even updated your tools to include robust racial demographics. But why? Are you looking for biologic differences or the impact of social determinants? I never rely on race but map to social constructs, genetic variants (identified within disease states), and geographic trends.
"Any two individuals within any so-called race may be as different from each other as they are from any individual in another so-called race."--Race-The Power of an Illusion
For context, the US Census describes the following race categories:
If you work with data you might already have an awareness that biologic differentiation would relate to geography--not race specifically. We roll up a lot of information into a single check box.
If we want to know about lung capacity, drug metabolism, or diet, why don't we just ask the questions rather than think a check box can tell us anything about the populations we study?
The racialized society we live in has been under construction for three centuries. How can we unmake race unless we first confront its enormity as a historical and social reality, and its emptiness as biology?--Race-The Power of an Illusion
You may have read the recently published article, Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites. I wanted to introduce the article with a bit of the context above.
The present work examines beliefs associated with racial bias in
An archived podcast, from the Hidden Brain, Remembering Anarcha, Lucy, and Betsey: The Mothers of Modern Gynecology highlight the perceptions studied in the PNAS article cited above. Shankar Vedantam describes how black patients continue to receive less pain medication for broken bones and cancer.
When designing surveys do not take short-cuts. It doesn't make sense to use the "proxy" of race for the questions we aren't asking--dig deeper in defining demography--let science and perhaps even history be your guide.
Well, doctors tell me they're using race as a shortcut. It's a crude but convenient proxy for more important factors, like muscle mass, enzyme level, genetic traits, they just don't have time to look for. But race is a bad proxy. In many cases, race adds no relevant information at all. It's just a distraction.
How to avoid asking irrelevant questions regarding race when designing demographic studies
When designing surveys I suggest embedding information such as World Region of Birth of Foreign Born, Ancestry, or other data available from the 2010 Census, American Community Survey, or American Housing Survey.