In the August 25, 2008 issue of Chemical & Engineering News, there's an interview with Carol Henry (behind a paywall). Henry is a consultant who used to be vice president for industry performance programs at the American Chemistry Council (ACC). In the course of the interview, Henry laid out a set of standards for doing research that she thinks all scientists should adopt. (Indeed, these are the standards that guided Henry in managing research programs for the California Environmental Protection Agency, the U.S. Department of Energy, the American Petroleum Institute, and ACC.)
Here are Carol Henry's research standards:
- Data should be owned by the scientist, not the sponsor of the research.
- Results should be released and published without having to wait for sponsor approval.
- Inappropriate sponsor interference should be eliminated.
- Open access to data should be required -- not just for federally funded research, but also for industry and other sponsored research.
- Disclose funding sources.
- Establish conflict of interest and bias policies.
If these practices were uniformly adopted, do you think they would improve the credibility of industry-funded research? Would it take more than just adherence to these standards?
Do you think research in academic settings that is not industry-funded conforms to these standards? How about research done by government scientists?
Are there any items on the list that you would drop? Any items you would add? Any you would clarify? (Myself, I'd like some more detail on what counts as appropriate sponsor interference in item 3.)
Finally, in terms of getting scientists on board:
"I think the scientific community agrees with these standards, but I don't think that the community has come together to say we need to do these things universally," Henry says. She does not advocate requiring such standards through formal regulations, however.
In the absence of formal regulations, will industry-funded research really abide by these standards? What kind of response could the scientific community mount to industry-funded scientists who decided not to abide by the standards?
That's a good baseline for an ethical code in industry sponsored research.
What is needed, though, is an actual audit/oversight process.
You see this problem a *lot* in IS research projects. Unlike some industry-sponsored research in other fields, where you have to be careful of conflict of direct economic interest, in IS you're usually embedded pretty deep in an organization, so it's not just the external economic interest you have to worry about, it's internal-to-the-organization political interests.
Ethical guidelines are really important, but if you really want trustworthy results, you can't just have scientists nod their heads and sign on a dotted line that says they're going to follow the guidelines, you need to have some sort of audit mechanism documented and in place. You don't need to audit every study, but you need to keep the process honest.
If security research tells you anything, it's that preventative measures take you only so far! :)
When we did consulting for the US Army Corps of Engineers, our contract stated (as I recall) that our data was ours to use for academic purposes as soon as the preliminary report was accepted.
A colleague took a contract with a confidentiality clause. He found some things the company didn't want to know. They said thank you vary much, here is your money, and keep your mouth shut. That was a first and last time for him.
Always interesting to see C&EN discussing ethics. This is the trade paper of the ACS, of course, which took our membership dues in order to lobby against the interest of its member research scientists by promoting PRISM and various lies against open publishing.
Is this supposed to apply to industry researchers as well, or do we not rate as "all scientists" and "the scientific community"? If the former, 1 and 2 are non-starters.
Incidentally, is it really the case everything Carol Henry managed at all those institutions met all those criteria, including open access data and freedom to publish at will? Not that I have any information otherwise, but I'd be very surprised.
Do you think research in academic settings that is not industry-funded conforms to these standards?
Well, that depends on who you consider "the scientist", doesn't it? It sure doesn't apply to grad students or postdocs in their relationship with the PI.
The C&EN article sure makes it sound like Carol Henry thinks these standards apply to all scientists -- not just those funded by industry money but also those employed directly by industry.
I, too, would be surprised if these standards were anything but a struggle to apply in some of the projects Henry oversaw. But her assertions that all scientists, in their scientific-heart-of-hearts, are down with them certainly raises some interesting questions about the extent to which scientific research done in industrial settings and funded by industry money might be different, and differently regarded by the rest of the scientific community, if these standards were the way scientists operated.