Post #4352847

@CatherineFlick@mastodon.me.uk @Floppy@mastodon.me.uk It's not so much that they don't have them, it's that there's disagreement (or, more often, lack of awareness) about what needs to be referred to them. Subjects like psychology tend to be better at it because the answer for them is basically everything. It's pretty hard to conduct any kind of experiment in psychology that doesn't require ethical review. And, in medical and medical-adjacent disciplines, ethics review is part of your liability shield: it makes it the institution's liability, not yours, if something goes wrong. There are a lot of things in the borders. I remember sitting through a fascinating talk using a dataset that was provided by the largest telco in Pakistan. A few of us started to look uncomfortable about half way in and someone put their hand up and asked about ethics-committee review. The (Pakistani) speaker said it wasn't necessary because it was done in Pakistan: they'd assumed that the review was necessary for this kind of research only if the GDPR applied. After the talk, the ethics committee got involved. Catherine probably knows this far better than me, but my experience is that it's subjects that don't normally do human-facing things that tend to be the absolute worst at realising what all the safeguards are when they do research in this space. If most of your papers are about how some non-biological system works, you might never have been told about ethics committee approval before you suddenly decide you need a user study.