An Honor Codes’ Second Component and Research Science
October 4, 2007
Many academic honor codes boil down to two essential statements, namely “I will not cheat and I will not tolerate those who do“. For “cheat” you may read any number of disreputable activities including plagiarism and research fraud. My alma mater had this sort of thing, I know the US military academies have this. Interestingly a random Google brings up some which include both components (Davidson College, Notre Dames, Florida State Univ (which as been in the academic cheating news lately), and some which do not (CU Boulder, Baylor); Wikipedia entry has a bunch of snippet Honor Codes. The first component, i.e. “don’t cheat” is easily comprehended and followed. The second component, the ” I will not tolerate those who do” part is the tricky one.
Discussions of academic honesty and science ethics abound on Adventures in Ethics and Science (here, here., here, and especially relevant here) and on Medical Writing, Editing and Grantsmanship (here, here, here). Click the relevant post categories on each site because they each have plenty more on the topic. Open Reading Frame taps into this with a “should be shunned” prescription (here and here). Some points made in the posts and much ensuing discussion circle around questions of how do we improve research ethics. Meaning not just how do we act personally but what needs to be done on a broader scale to improve (some might say “clean up”) science. In some ways it comes down to otherwise good actors failing to step up to the plate on the “I will not tolerate” aspect of the internal Honor Code. Recent discussions with colleagues indicating that their local institutions have the “will not tolerate” component enshrined in their ethics statements for research faculty and staff makes this even more interesting. In the cases with which I am most familiar (my alma mater and one service academy) one can be equally culpable for “tolerating” as “cheating”…in theory. In practice of course I doubt this really holds up but it is a good point to ponder.
I will state baldly that in my opinion while the vast majority of scientist subscribe to the “I will not cheat” part, it is a high number who are a bit shaky on the “I will not tolerate those that do” front. In this I include myself, however not so much in that I’ve had to make an actual hard decision as yet. Rather because I suspect it would take a very high level of proof and inescapable responsibility for me to want to launch an ethics probe. We’ll get to that. My evidence? All the people who say things like this “Well, I don’t believe the data from that lab (in that paper, grant, etc)”. Or all those of you who gossip about suspiciously good / successful postdocs in your lab or collaborating labs. You know who you are! This kind of stuff is perhaps not rampant but common enough. In most of these cases, it remains at the level of gossip or, at best, an oblique reference in a conference presentation or paper. If you say it, you must believe it. And if you believe it and don’t report it to the appropriate investigating authority, well, you are tolerating it. Are you not?
So why? No great shocker there. Because it is very rare that the case is clear cut enough or the evidence so readily available as to make it a slam dunk. And usually there is little immediate and personal “cost” involved, just because someone in your field, institution or lab is faking data doesn’t necessarily affect your career after all. Yet if you decide to blow a whistle there will definitely be a cost, potentially severe depending on whether this pisses off your superiors (PI, Senior colleagues, Dean, etc). And if a case fails to be made by the appropriate authorities? Might as well kiss your career goodbye. Nobody but nobody is going to give props to someone who appears to be a little too overeager to enforce science ethics.
I hope you didn’t think I was going to have any solutions. I’m just seeking them…
October 4, 2007 at 9:05 pm
Just recently you wrote that you didn’t see the point of Open Notebook Science. Well, here it is: I don’t have to complain about not believing data from a particular lab or postdoc or whatever, if I can look through the relevant notebook. I can look through *all* of the relevant work, *all* of the results, every experiment — and decide for myself on the basis of raw data.
(Obvious objection: people will fake their lab notes. Obvious reply: that would be more work, and very tedious work at that, than doing the science properly!)
LikeLike
October 5, 2007 at 11:04 am
My problem with Open Notebook science is that I think eventually one will end up spending all and I mean ALL of one’s time repeatedly defending the stupid stuff. Every possible kind of science with which I am familiar has judgment calls, irrelevant stupid errors and areas of uncertainty. Take the sorts of ‘discussions’ that erupt on SB(tm) every now and again as your predictor for the future of “open notebook”. and it isn’t so much that you would have to defend each and every point (again, and again, and again) of your research. but you might feel obligated in a LOT of cases.
the issue is not so much faking lab notes but more of omission. unless i am missing something here. like when an experiment is repeated 12 times but only gets talked about 5. this interacts with the above. “I got tied up in traffic/at daycare/etc and didn’t make it back to perform x, y, z so the experiment was farked”. do you enter this? how about when the antibody is sucky and it takes you forever to figure this out?
and some of this gets really field specific. what constitutes an “experiment” to a bench scientist ain’t the same as for an animal behavior person or a human clinical researcher. so you could end up with some debates that are more meta to the field being played out again and again and again, in the case of a specific open notebook lab. exhausting.
LikeLike
October 5, 2007 at 11:07 am
okay perhaps the antibody thing was not a good example, on reading it, of course you would want to document this. in the abstract anyway. but for every kind of science is there not some ‘threshold’ for what you want to record that starts somewhere above the actual labor of each and every day?
LikeLike
October 5, 2007 at 2:11 pm
That argument applies quite directly to discussions of evolution vs. creation, feminism, etc etc — and yet you don’t see PZ Myers or Zuska giving up just because they come across the same dumb arguments time after time.
This “internet discussion” phenomenon is unavoidable in any system with low barriers to input. My usual strategy applies equally well to SB(TM) comment threads and an open notebook: if my interlocuter has a valid point, I respond (if only to say thanks); if not, I simply pay no heed. Besides, I doubt that I would enable comments on an open notebook — anyone unwilling to go to the trouble of sending me email probably does not have anything useful to say. And email is as easy to ignore as it is to reply to.
Plus, why would I defend the stupid stuff? Why would I not write “dropped sample 5, whole bloody thing no good, have to repeat, hate myself more than ever” in an open notebook the same way I do in my current paper notebook? Anyone inclined to attack me over that is an obvious ass, since we all do it.
There really is no threshold for recording — there is only the (deeply non-trivial!) logistical problem of how to make an open notebook as easy to use as the paper ones we are currently familiar with. The form factor of an electronic lab notebook will probably have to vary from field to field — Cameron Neylon is developing a blog based system, Jeremiah Faith uses LaTex–>pdf, Jean-Claude Bradley uses a wiki. For myself I have Star Trek dreams of a graphics tablet with good text recognition and auto-publishing to a wiki…
I really do have nothing to hide, and am quite happy to have anyone read my lab notes. You will get no more out of me than a sheepish shrug over the screwups and mis-steps. Frankly, I think it would be a good idea if Joe Public had a better idea of how science actually progresses, and one need only read a couple of notebooks to lose forever the image of scientific information as something handed down from Heights of Genius!
LikeLike
October 5, 2007 at 6:44 pm
“This ‘internet discussion’ phenomenon is unavoidable in any system with low barriers to input.”
The whole point of peer-reviewed science is that there is a relatively high barrier to input. This means that once something has vaulted that barrier, it is reasonable to make all sorts of assumptions that save a vast amount of time and effort.
Yes, once in a while, those assumptions are unwarranted. But the savings of time and effort in the long run are well worth the risk.
“Open notebook” science is a ridiculous joke. After a few weeks have passed, it takes a substantial effort for a scientist to even figure out half the stuff in their *own* notebook.
“Open notebook” implies that you can’t trust any scientist to interpret their own data and filter out the relevant from the irrelevant. If this is true, then we may as well just give up the scientific enterprise.
The last thing I want to do is waste my time looking through some notebook reading crap like, “Restriction digest didn’t work; forgot to add enzyme”. Or looking at pictures of gels that aren’t even annotated because the scientist realized they were garbage immediately, but still stuck the picture in the notebook.
LikeLike
August 12, 2015 at 6:32 am
[…] was originally posted October 4, […]
LikeLike