I'm going through the training that Purdue requires before I submit any research protocols to the Institutional Review Board. It's good to care about your faculty doing ethical research, but I confess it is taking FOREVER.
So I'm reading the part on the problems with peer review, and I come across this chestnut:
Gender bias may occur in reviewing.
Okay, so far so good, I think. Maybe this is an enlightened group of folks writing this who are aware of the research that says that reviewers are biased against manuscripts authored by people with identifiably female names. I continue to read:
Some studies show that female authors were accepted more by female reviewers than by male reviewers.
Well, crap. I am now appalled that they framed it as though it was the bias of the women reviewers towards women authors, and not the bias of many men AND women reviewers against women authors!
So I was going to turn this into a big long chewy blog post, referring to all kinds of fab literature on gender bias and publication and so forth. But then I found that Zuska had done it for me. Thanks, Zuska! :-) Instead it will be a mere appetizer to whet your appetite for Zuska's post instead.
My opinion, however, remains that this particular statement in the online training is stupid.
- Log in to post comments
was this supposed to be the "balance" statement and the other stuff was there too? Or. Was. This. It?????
Some studies show that female authors were accepted more by female reviewers than by male reviewers.
That statement makes it entirely unclear whether:
(1) Female reviewers accept female authors at disproportionately high rates,
(2) Male reviewers reject female authors at disproportionately high rates,
(3) A little of both, or
(4) Only some studies show this but not others.
You're certainly right -- it is an excessively stupid statement.
The president of my grad student organization maintains (and I largely agree) that ethics training is essentially useless. For the people who ARE ethical it's redundant, and for the people who AREN'T it just teaches them how to get around the rules. I don't know about the people who might be sitting on the fence. Is there such a thing?
Sally, you've made two errors in your position.
The first is that your personal idea of what constitutes ethical is dead certain to be at variance with everyone else's personal ideas in some areas. The worst arguments are not betweeen objectively 'right' and 'wrong' ethical actions but between two sets of people with incompatible beliefs as to what constitutes 'right' and 'wrong' ethical actions.
The second is that everyone knows what the 'official' ethical rules are (which you've implicitly assumed are identical with your own, which they probably are not).
Ethical training sucks. It always comes off corny, idiotic and contrived. But lack of ethical training sucks even more. Without the training you have no 'common standard' that everyone has been taught and can use as an objective yardstick. Instead you have N different subjective standards where N is identical to the number of people.
Personally, I think all training on ethics/bias should start with participants doing a few IAT tests - the site is here:
https://implicit.harvard.edu/implicit/
It's salutary to have your underlying biases revealed to you, and it puts in context the need to be aware of your biases when making decisions.
I think U Michigan's handbook on practices to adopt in hiring has a really good explanation of how biases arise and what you can do to mitigate them and hire diverse and excellent faculty. A pdf version is at
http://www.umich.edu/~advproj/handbook.pdf
On the subject of biases, FSP has a great blog post on A-Zism and author credit rankings at http://science-professor.blogspot.com/
The problem I have with most ethical training is that it's really easy to be dismissive and think "well that is a problem for some people, but I certainly don't/won't do that". I think Rachel's suggestion to make people aware of their own biases is a good one.
Thanks for the comments, and especially the links. Harvard's implicit bias test I did a few years ago - surprise surprise, I'm biased in favour of women, just like the ethics training said I would be. ;-) I'm going to make another post with respect to some useful links on bias, in case folks are interested.
I'm interested to see that you have a
My implicit preference showed a mild bias against women.
This is despite being one, finishing top of my class, winning scholarships and working in a professional field with a lot of professional women.
My parents raised me on the slogan "Girls can do anything", I got lego sets, carpentry sets and an (largely unused) microscope for presents, and they expected me to have a career.
Still, social conditioning must play a part!
Ethical training sucks. It always comes off corny, idiotic and contrived. But lack of ethical training sucks even more. Without the training you have no 'common standard' that everyone has been taught and can use as an objective yardstick. Instead you have N different subjective standards where N is identical to the number of people.
Exactly. Plus, even aside from increasing people's understanding of a common standard, training makes it easier to punish people for violating it (without the "oops, I'm so sorry, I didn't realize my behavior was wrong" excuse).
This is the same logic as the usual "don't plagiarize" instructions given in classes. They are typically too short to be fully instructive for someone completely ignorant of the concept of plagiarism (except as a hint to go learn more). The point isn't to teach students about the precise boundaries of plagiarism. The point is to make sure they can't use the excuse that they were never warned.
The implict preferences test reported that I (by a slight margin) tend to associate females with science and males with liberal arts. Given that I've historically studied and worked in fields heavily dominated by males (physics, math, electronics and computers), that is an interesting result.
I suspect, however, that the fact I am strongly Asperger's has a confounding effect on the test. I may not 'interact' with it the same way as an ordinary person might.
Yup, it's them wimmin an all ther biases. *eyeroll*
Did you complain to whoever is responsible for that stuff? If not, do so.