This is an oft discussed problem, but this article from American Prospect does a particularly good job overviewing the issue of industry biasing scientific studies. The problem is that when ones financing is so often tied to a particular result, even ignoring overt twisting of data, there is the problem of unconscious bias. Ideally peer review and the importance of repeating experiments would overcome this. Sadly many experiments simply aren't repeated, due to funding issues. And peer review can frequently fail, depending upon how it is done.
It seems to me that this is one area where more government involvement is actually beneficial. (Yes I am a Republican with occasional libertarian tendencies) I say that not to have government bureaucrats controlling some other aspect of life, but simply because the funding has to be separate from the scientists in some fashion. I suppose this could be done were all the companies in each industry to get together and pay into a fund that does the research independent of what any particular company is doing. In effect that is what the insurance industry manages to do with industry. However I'm very skeptical that industry would do this without the threat of government action.
While medical research is the obvious target in this controversy, I should add that it touches upon all areas of science. Genetic engineering is one obvious example. Indeed there have been numerous controversies there the past few years. But I think it could apply equally to research on everything from plasma physics to semiconductors. When potentially millions, if not billions of dollars are at stake, counting on human objectivity simply isn't wise. Further, we have rather large national laboratories that already do a lot of research on topics related to industry. So the structures are in place, not to mention existing organizations that assign research grants. What is simply needed is some reformation of grants given directly by industry.
On a related note is the movement to take the Open Source concept, so popular in computer software and apply it to science and technology. Science Digest has an interesting article on this topic. There the problem is less bias, although I do think it would resolve a lot of bias. The problem is that money follows the diseases people with money have. After all if you are an industry you have to recoup the millions of research dollars you invest to find cures. The downside to this is that billions are spent on weight loss, hair growth, and the other diseases of the western world whereas tropical diseases that probably could have treatments or vaccines are neglected.
The problem with Open Source, besides often killing companies with products that face Open Source competition, is that the money still has to come from somewhere. Something that Open Source proponents often neglect. If, however, there could be some way of reconciling Open Source with Industry then perhaps we could have a win-win situation. Make all government funded research inherently Open Source in a fashion that expands what is Open Source. Whether this could be done in a fashion that would please everyone is rather doubtful, in my opinion. Further, it would do nothing about industry research. The only way I could see it working (in the US anyway) is if we had industry contribute to a general fund that funded research. But I simply can't see why Industry would want to pay for something without having a strong chance of recouping its costs. Something that Open Source doesn't seem able to do.
Perhaps this is an experiment that will initially take off in Europe or smaller nations. (Especially as the technology for research comes down in cost)
This is something that is still quite fresh for me (having left academia six months ago). I come out of that with a different take than you, though similar in many ways. I think the NIH mega-grants are doled out in a very political fashion. It is extraordinarily difficult for young researchers to get a piece of the action. Although, the trend in collaborative grants is rectifying this and is very positive.
I think the bulk of “Industrial” grants are not of the sort that compromise good research. I admit that there is a ton of pressure to produce results expeditiously (which, is antithetical to academic sensibilities), but in most cases they are relatively small investments (from the corporate perspective) that they want to leverage in their business (this is true for statistics, engineering, chemistry, etc.). The exception are those research projects whose pass or fails results in billions of dollars, and that is very rare.
As for “open source” research: the model works extraordinarily well when the cost of research is low or zero (i.e., that is except for time – which is what grad students are for). Things like animal or human studies or cases where materials or equipment are exorbitant are just too difficult to execute on “charity”. As you mention, it all hinges on sustaining industry. I just can’t see how you could.
There's definitely politics, but then there is politics in everything. Of course coming to realize the amount of politics even in science can be very disillusioning. On the other hand that's life.
Regarding the impact of industrial bias - the problem is that the areas where it is so important the have unbiased information is precisely where it is often most biased. Frankly if physics or engineering research is a little biased, the worst case scenario is misspent money. Bad, but not the end of the world. When medical research is biased, people die or worse.
I agree with the open source issue, however if scientists are working for the government and there are government grants, then I think open source might work, although I still have problems with it. (Frankly I have problems with open source in general - I think that outside of a few applications like apache or gimp, it tends to dry up research and technology rather than produce it) On the other hand I can see this being a great way to ensure that research aids third world countries.
Like you though I have strong doubts it would work. I think the problem with open source technology are the limits in using it. (i.e. your code becomes open source) If you did that with say pharmaceutical research, it may just have the consequence of no one being able to make money on the research and therefore no one investing in making actual drugs using the research...
I've closed comments in order to avoid spam since I don't check this older blog as much anymore.
and Theology
Number of unique visitors:
Blogged by Clark Goble