1

Which alternative names can one use to refer to a "blind experiment" (defined by https://www.thefreedictionary.com/blinds as "2a. Performed or made without the benefit of background information that might prejudice the outcome or result: blind taste tests used in marketing studies.")?

A colleague recommends to replace the word "Blind" (for the sake of inclusiveness) by "masked", "hidden", "anonymous", or "unbiased", but I fear that the referees will suspect me of not knowing that I am actually referring to a "blind experiment", and negatively evaluate my submission because of that (I am sure to commit other newbie mistakes, but I do know what a blind experiment is).

Note: This is distinct from my other question (For inclusiveness sake, what is the consensus about which usage of "blind" should be changed?) about the consensus about which of the usage of the term "blind" should be replaced: for an academic submission (to a conference, where there is no opportunity to answer to the referees, just an acceptance or rejection of the submission by the program committee based on the reports of 2 to 3 anonymous referees), I do want to replace the term "blind" so that to avoid annoying any referee...

8
  • 3
    Does this answer your question? For inclusiveness sake, which usage of "blind" should be changed? Commented May 19, 2022 at 10:42
  • I was the one asking this other question! They are related but distinct (in my mind at least).
    – J..y B..y
    Commented May 19, 2022 at 11:37
  • Why would the usage of the term "blind experiment" annoy a referee if this is the common teminology in your field? Commented May 19, 2022 at 11:46
  • @JochenGlueck: My (non-coauthor) colleague threatened to refuse any future collaboration with me if I stuck to the use of the term "blind" (when I only stated that I did not understand her objection, but that is an unrelated issue). They are a potential referee, and I assume that other potential referees might share their opinion.
    – J..y B..y
    Commented May 19, 2022 at 11:50
  • 3
    Thanks for your response! So what term does this particular colleague use in papers instead of "blind experiment"? Commented May 19, 2022 at 11:54

3 Answers 3

5

I've had an opinion along the lines of the other commenters for a while, that substituting blinded would be far too confusing to be worthwhile. However, today I used 'double-blind' in a presentation and was informed that my institution employs people with limited vision who have personally registered their objection to the term ‘blinded’ in the context of blinded trials.

In terms of practical substitutes, NASA uses the term "dual-anonymous" for peer review. This is sufficiently entrenched that they don't seem to need to define on first use.

This paper also offers a reasonably convincing argument to me that 'blind' isn't the best term even just for tangential reasons, and starts to make the argument that blinded could be replaced for scientific reasons for being imprecise. It also mentions some practical issues with its use:

She is told that she is part of a double blind trial in which she and the doctor will be blinded to the treatment. Taking fright, she withdraws her consent and goes home, terrified that this “blinding” experiment may deprive her of what little vision she has left.

The term “blinding”—commonly used in clinical trials—is particularly inappropriate in the ophthalmological setting, not least because an outcome measure of a particular trial could indeed be blindness.

They suggest "masked".

"Masked" is used as a substitute in even quite old papers without first defining it, and it's introduced in a few university texts on the topic.

The APA Manual of Style, 11th edition, says

The equivalent term masking (or masked review or assessment) is preferred by some investigators and journals, particularly those in ophthalmology (see 5.7.1, Confidentiality During Editorial Evaluation and Peer Review and After Publication, and 19.5, Glossary of Statistical Terms).

A Springer journal refers to the process as "journal’s peer-review system is masked (i.e., double-blind). Thus, leave all identifying information off the manuscript."

In fact, a Google Scholar search for 'double masked study design' returns ~800K results (compare to the 2.3 M results from 'double blinded study design'); adding flags -opthamology -eye to remove that field yields ~500k results from broadly across the life sciences.

So I at least feel reasonably comfortable that my audience will be familiar with (or can straightforwardly learn) the term "masked" in place of "blinded".

This paper offers some fairly reasonable objections to even the double- part, mentioning the blind- part only briefly:

In some situations, it can be confused with the condition of being without sight [2, 5, 12, 20, 22, 23] Some authors prefer “masking” to “blinding,” although the meaning of either term in a clinical trial may not be readily apparent to nonnative English speakers [18, 22]. Further, some authors use the terms interchangeably [5,6,7, 10,11,12, 15, 18, 24, 25], others insist that only masking be used [17, 20, 23], and still others insist that only blinding be used [2, 5, 22]. In addition, masking is sometimes used to describe how treatments are made indistinguishable [18, 19, 25, 26], whereas blinding usually indicates which groups are unaware of treatment assignment [1,2,3,4,5,6]. Finally, searching the literature for “blinded,” “partially blind,” or “fully blind” randomized trials also identifies dozens of unwanted citations to the condition of being without sight.

The reviewer comments make it clear just how much of a minefield this topic is:

“Fully blinded” makes sense only if everybody agrees on what “fully” means. The problem is not the definition proposed by the reviewer, which is excellent. The problem is whether researchers will define “fully” the same way. The literature identifies at least 12 groups or individuals that can be blinded in a trial. The likelihood that a study will blind all of 12 groups is close to zero, meaning that nearly all studies would have to be described as “partially blinded,” a term no more informative than “blinded.”

And finally there's a really good suggestion along the lines of cbeleites' Alternative (above), in the reviewer commentary, which is both scientifically unambiguous and can sidestep this problem:

We also propose that all trials described as blinded include the details in a standard “Who Knew” table that would indicate whether each of the 6 groups most commonly blinded was or was not blinded, what information they were blinded to, how blinding was implemented, and whether it was compromised during the trial.

5

I think that your best approach is to explain in the paper that

Rather than referring to our trial as "blind," because [your reasons here], we use the term "masked."

I don't think it would be reasonable to object to mentioning the former term "blind" in the context of explaining why you are not using it. It isn't as clearly bad as some other words.

4

(single|double|triple) blind experiment as well as blinding or employing a blind* against ... are technical terms with well-established meaning.

I'd like to point out that substituting something else for a well-established technical term in itself makes a text targeted for the corresponding technical audience less inclusive, burdening readers with questions about the precise meaning of the employed term. E.g., whether that particular term is used to emphasize subtle differences to the established term.
And it is much less inclusive against non-native speakers of the paper language, who are familiar with the technical terms, but to whom your substitute term may not come as a natural choice.

I'd therefore suggest that you stick to the established technical term (and, as a side note: please give a precise description of your blinding techniques ["... were blinded by .... against ..."].)

If I'd feel an additional need to explicitly express respectfulness for blind people, I'd put a footnote that I judged the potential of verbal injury to bind people against the potential loss of inclusiveness due to non-standard terminology and decided in favor of the established term.

*As a personal side note: *


* which suggests to me that we may be talking about blinds as in curtain or shutter rather than blind as in people. Which of course also has a root in not seeing.

Along this line of reasoning, if you are forced to substitute the term, single/double curtain experiment may be a possibility: "masked", "hidden", "anonymous", or "unbiased" all are meaningful in combination with "experiment", but do not have the same meaning as blinding (one may go for "masked", but at least in my field, masking is already used for multiple other entirely different effects/approaches). This creates a danger of the resulting text being misleading.
In contrast, I believe that "curtained experiment" is an entirely new term and while being burdensome for readers, it forces readers to look up and use your definition.


As an alternative, you could give a precise description of whom you blinded how and against what effects without mentioning the term blinding. This would take some more space and words, but it is anyways far more informative than saying only blind experiment.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .