1

I'm editing a quantitative paper in sociology based on a Statistics Canada survey. The survey uses the term "visible minority" in a yes/no question asking if participants identify as a member of a visible minority group. This term is used across many Statistics Canada publications, but "visible minority" is rejected by many scholars today and for that reason I don't want to use it in the paper.

In the statistical demographic chart in the paper I propose to use the word "Race" to designate the demographic category (along with other categories like age and gender) and "Racialized" and "Nonracialized" as the two possible subcategories. I would then add a footnote to explain that these terms diverge from the terms used by Statistics Canada, giving the reason for the changes. Does this seem like a good solution?

Eggy
  • 1,445
  • 9
  • 16
  • 2
    Race is probably impossible to define unless you want to post the genetic information of every subject. What does it mean to be "Black"? I show a tiny fraction of Northeast Asian heritage, impossible to track. Am I Asian? I also might have a tiny fraction of "Native American" (actually Canadian) heritage. What am I anyway. All such questions are fraught and probably misleading unless you are tracking discrimination. And even then. – Buffy Aug 08 '21 at 18:42
  • 1
    Start here: https://www.gocomics.com/boondocks/1999/04/28 and read a few comics forward. Look for pages featuring Jazmine. There are a few other intervening. – Buffy Aug 08 '21 at 18:45
  • 4
    I mean, it’s undoubtedly true that race is a social construct and there’s no “objective” question you can ask here, but there’s also an issue of asking questions in a way that users of the survey will understand. So this depends a lot on your audience. The answer for a census of all citizens is going to be different from a survey of history grad students. So what’s your audience? – Noah Snyder Aug 08 '21 at 19:29
  • Please do not post questions about the content of your research. – Anonymous Physicist Aug 08 '21 at 20:09
  • 1
    Anonymous Physicist, can you explain why you said this? It's not a question about my research. I'm an academic editor trying to find the right terminology. – Eggy Aug 09 '21 at 03:28
  • 1
    @Buffy as a geneticist, I can say that the term "race" has no connection to genetics. Peoples genetics do vary from geographic location to location, but those variations don't generally correspond to what society regards as races. Race is defined by society and the individual in some combination that is not always agreed on by both (see your comic link). Self-defined race is a perfectly good word to use in a sociology study in general (although in this case I'd agree with Dan Romik below that since the participants havn't self-defined their race, its probably not good in this case. – Ian Sudbery Aug 09 '21 at 08:43
  • ""Racialized" and "Nonracialized" as the two possible subcategories" at least seems very wrong to me. – Bryan Krause Aug 09 '21 at 18:39
  • Bryan Krause, why exactly does this seem wrong to you? – Eggy Aug 09 '21 at 19:35
  • Isn't ethnicity a more neutral term still full of sociological implications? I am a chemist, by the way... – Alchimista Aug 10 '21 at 15:30

2 Answers2

10

I’m not a sociologist, so take my opinion with a grain of salt. But it seems to me that the correct term to use in the writeup is exactly the same term that was used when the data was collected. That is, if the question on the survey form people who participated in the study answered was about “race”, then that is how you need to report the data. If it was about “racialization”, that is how you should report it, etc.

If you don’t use the identical language to what was used in the data source, you risk causing confusion, distorting the meaning of the data, and opening yourself up to accusations of making the change in the reporting because of some personal or political agenda, and/or accusations of being a sloppy researcher.

Now, if you yourself disapprove of the terms that were used in the data source, or don’t want to be seen as endorsing them, you can use a device like “(sic)” to emphasize that the term originates in the data you are using.

Edit to address rephrased question: using a footnote instead of “(sic)” is also okay, but in my opinion the footnote should explain what the actual terms used in the survey are and not just say that your terms diverge from the original ones. The main point is to be transparent with your paper’s readers so they have the information they need to accurately interpret the data you are giving them, and aren’t led to suspect you of any weird shenanigans. If you cite data from a survey, you have a scholarly responsibility to describe it accurately even if you disapprove of some of the labels used.

Dan Romik
  • 189,176
  • 42
  • 427
  • 636
  • It isn't clear to me that this decision about terminology is taken late in the process rather than in the design stage, nor that it was gathered by another. The same question arises during design. – Buffy Aug 08 '21 at 20:10
  • @Buffy I answered the question that was asked, which was specifically about the writeup. Seems pretty clear to me. – Dan Romik Aug 08 '21 at 20:43
  • I have reworded my question to clear up this confusion. Dan Romik did understand my intended meaning. The question is about the write-up, not the survey design. – Eggy Aug 09 '21 at 04:21
  • @Eggy thanks for the clarification. I edited my answer to address your reworded question. – Dan Romik Aug 09 '21 at 05:23
  • Dan Romik, you made an excellent point and in fact you put your finger exactly on the problem. The study is based on data from Statistics Canada, which uses the term "visible minorities" across its publications. That term is rejected by many scholars for good reason and Statistics Canada has come under fire for it. The question in the Statistics Canada survey inquired about people's status as a member of a "visible minority" group. But I cannot use that term in the write-up as it would get flack from journal reviewers. Perhaps I should use "Race" as the category and "Racialized" and "Nonracial – Eggy Aug 09 '21 at 03:37
  • Dan Romik, thank you, I will include a thorough explanation of the original and modified terms and the controversy surrounding "visible minorities." In fact I believe Statistics Canada is in the process of revising its terminology, having recognized the multiple problems with this term. – Eggy Aug 09 '21 at 15:51
1

If you don't mind wordiness, try "Racial/ethnic self-identification". If you want to be extremely cautious, you could also try "Cultural self-identification", though that risks creating a certain amount of confusion. The 'self-' qualifier isn't strictly necessary, but does help assuage any worries that people are being objectified as something other than how they see themselves.

In either case you can use standard racial and ethnic nomenclature: White (Caucasian descent), Black (African descent), Hispanic, various Asian labels, etc. I'd avoid using 'non-white' unless you are constrained by the data. 'Non-white' is a tone-deaf relic of colonialism, implying as it does that the only important or relevant distinction is whether one is white.

P.s. And yes, incidentally, there is no such thing as race from any objective scientific perspective. The human genome is extremely restricted: a typical species of bird has two to four times the genetic diversity of the human species, and we don't see any need to divide those up into races.

Ted Wrigley
  • 465
  • 3
  • 6
  • The question on the original survey reads something like "Are you a member of a visible minority group?" and the two response choices are "Yes" and "No." So it's not a question of labeling the individual groups, but of categorizing the entire demographic (e.g., "Race") and the two possible answers to the question. Whether race exists is unrelated to this question, but the dilemma is that governments have many reasons for needing to identify races, like distribution of federal funding, and thus language is needed to identify racial and ethnic groups. – Eggy Aug 09 '21 at 19:43