Identity: Please Check One

Editor’s Note: This article previously appeared in a different format as part of The Atlantic’s Notes section, retired in 2021.

As a kid, I was great at standardized tests. I aced the fill-in-the-blanks, the vocab words, the trains passing at 25 miles per hour. I filled in each bubble completely, in No. 2 pencil, with no stray marks or smudges, and finished with time to check my answers—time I proudly didn’t use. But there was always one test question I couldn’t answer so confidently: the race/ethnicity question on the personal info page.

My mom is Filipina and came to the U.S. when she was 7. My dad is white and solidly American in heritage, with a couple hundred years between him and the British Isles. Before it became standard practice to let people “check all that apply” on the race question, I grew up choosing the Asian/Pacific Islander box but felt to varying degrees that I had lied by leaving my father out of it. When race and ethnicity were listed separately, I wondered uneasily if perhaps I was really Hispanic/Latino—given my grandparents’ Spanish names and the fact that my elementary-school classmates tended to think I was Mexican—but I never checked that particular box.

The one time I actually got the question wrong was in sixth grade, when, for the first time, I was given a test that listed Asian and Pacific Islander as separate categories. This posed a problem: I had grown used to checking the Asian/Pacific Islander box and even came to appreciate the broadness of its generalization. I’d always thought I was Asian, but, technically speaking, the Philippines are islands in the Pacific, right?

I asked my teacher, who seemed baffled, probably less by the geography than by the fact that I didn’t seem to know my own race. Eventually, we went with Pacific Islander. That evening, my mom, being Asian, was taken aback. “We’re not Pacific Islanders!” she said. “That’s like Hawaii. Or Fiji. Did you not know?”

I shrugged. To be great at standardized tests is to learn how to filter out the subtleties. If no answer seems to fit, or if too many do, you pick the one that seems closest. You remember that what you think is true matters less than the logic of the test.

But standardized tests are not particularly great at people. There are far too many of us to fit into just a handful of boxes. And, as Sowmiya Ashok wrote for us recently, that’s something census-takers are increasingly finding out:

Something unusual has been taking­­­­­­ place with the United States Census: A minor category that has existed for more than 100 years is elbowing its way forward. “Some Other Race,” a category that first entered the form as simply “Other” in 1910, was the third-largest category after “White” and “Black” in 2010, alarming officials, who are concerned that if nothing is done ahead of the 2020 census, this non-categorizable category of people could become the second-largest racial group in the United States.

Ashok goes on to discuss the history of the census form and the considerations that officials take into account when shaping questions, including the different methods of asking about race that have been tried. Here’s one response from a reader:

As someone who comes from parents of two racially different backgrounds—both of which are typically represented within the stereotypical “who are you” questionnaire—it is interesting, yet comforting, to see the increase in discussion about the “other” category. Though I rarely struggle to define my racial background (unless, on occasion, I am limited to selecting only one option), knowing that consensus and questionnaire makers are having discussions about how to create more viable options that pertain to a multitude of under-represented individuals is pleasing. Having these conversations is necessary to achieve an accurate representation of all individuals.

This next reader’s story takes me back to that moment of confusion in sixth grade:

This reminds me of when I was in high school and my grade was taking some standardized test. When we were filling out the personal info part, a Latino kid asked a teacher what he should put for race because the only options were black, white, Alaskan native, Asian, Native, and no Latino/Hispanic option for ethnicity. After much thinking the teachers just told all Latinos to fill in white. Obviously, many were weirded out because they weren’t white.

This reader comes to the question from a different perspective:

As an old white guy I’ve checked “other” for years. I refuse to be put in a category, that can be discriminated against.

Of course, as another reader points out, this data isn’t intended to be discriminatory:

The utility of these classifications is to track the effects of discrimination, social mobility, educational policies and their effects and other important information needed to know what’s going on in American society—historically steeped in racism and racial designations that were created to keep everyone “in their place.” France refuses to classify people in this way and has no real data on how many of its citizens are alienated by racism.

Happily, American society is evolving, racism is increasingly seen for the monstrous nonsense it is, people are intermarrying, and many more people are deciding not to fit themselves into the old slots. But this is creating a tracking/statistical problem, which will need to be worked out.

So how do we work it out? To me, these reader comments highlight several different tensions in how we think about putting people into boxes. As an individual, there’s the desire to be accurately, precisely described, and there’s the desire for privacy. There’s the desire to feel you belong to a group, that a category exists to welcome you. But there’s also the reluctance to be boxed in, as if that category were all that defines you. On a broader social level, there’s the need to track demographic patterns—with the hope that one day, that data won’t reveal patterns of inequity. There’s the need to ensure that the data isn’t misused, that the tracking system doesn’t become discriminatory. And all of those concerns come into play each time we check—or design—a box.

What are the benefits of putting people in boxes, and what are the downsides? We’d like to hear your thoughts, particularly if you’ve got a personal connection—heritage that resists categorization, or expertise as a census-taker. Please send us a note at