Warning: I’m feeling cranky today.
It’s great to see scientists getting excited about doing interdisciplinary work – for example, in science studies.* It’s embarrassing, sometimes, to see how bad they are at it.
You’ve probably seen, like I have, requests to participate in various on-line surveys – often coming via social media, although email blasts are a thing too. Just as a few recent examples, I’ve been invited to express my views on citation practices, interpretation of authorship order, produce-shopping preferences, travel intensity, and more. (I’m not counting just-for-fun Twitter or blog polls – heck, I do those – I’m talking about surveys that are identified as being for research.) I’ve gotten in the habit of clicking on these invitations to do a little research of my own. My research question: how many of these “research” surveys have research-ethics board (REB) approvals?
You can probably guess the answer from this post’s title. It’s maybe ¼ at most.
If this shocks you, then you know more about human-subjects research than (apparently) ¾ of the people who are trying to do it. Well, no, that’s not quite right. You know more about human-subjects research than ¾ of the people who are trying to do it by broadcasting links to online surveys.
Just in case you don’t know this (and you mightn’t, if you don’t do this kind of research: in most countries research involving human subjects (including surveys, not just Phase III vaccination trials!) must be approved by your institution’s research ethics board.** If you do the work without that approval, most journals will refuse to publish it, granting agencies will refuse to fund it (or may retract previous funding), and various other unpleasant things may happen to you. The REB will consider things like your plan to secure informed consent, your curation of personal data with respect to ethics and laws around privacy, and more. They’ll also assess the quality of your survey design – and this is not a small consideration, give that it’s hard to imagine someone being oblivious enough to do human-subjects survey research without REB approval, but also competent enough to design human-subjects survey research correctly.
Actually, that last point is a key one, and worth putting another way. If you’re considering doing research that involves surveys (whether online or not), you need to realize that survey construction, deployment, and interpretation are hard. As scientists, most of us don’t know much about that. If you don’t realize that your survey needs REB approval, well, that’s pretty good evidence that when it comes to survey methods, you don’t know what you’re doing. So find a collaborator who does! We all have friends and colleagues in the social sciences who have the relevant expertise. Be interdisciplinary by involving experts from multiple disciplines – don’t be interdisciplinary by dabbling in things you clearly don’t understand.
Yup, I was right. I am cranky today.
© Stephen Heard December 21, 2020
Image: Under the magnifying glass, CC0 via Pixabay.com
*^”Science studies” is the discipline that considers how scientists do their work and how science is done – not philosophy of science, but the culture of science and the systems we use to plan, conduct, and disseminate research. As you can probably tell from reading Scientist Sees Squirrel, if I were starting over and needed to pick a different field, science studies would be high on my list.
**^There are some complications to this, but no, you don’t get a free pass if you claim you’re just doing the work in your spare time. And no, it doesn’t matter whether your institution is a university, an NGO, a government lab, or something else.