But mastering the techniques in order to investigate problems should not be sufficient, in my opinion. Above all, a Ph.D. candidate needs to demonstrate a deep understanding of the basic science that underlies our current models and theories. That's absolutely necessary if you are going to be capable of challenging those models and theories.
An editorial in a recent issue of Science caught my eye because it proposes to "reimagine" graduate degrees in STEM disciplines. The authors are Ian Banks who is director of Science Policy at the Foundation for American Innovation in San Francisco (California, USA) and Prineha Narang who is a professor in Physical Sciences and Electrical and Computer Engineering at the University of California, Los Angeles (Los Angeles, California, USA). [Reimagining STEM doctoral training].
Here's part of their proposal ...
But if one of the goals of granting an advanced degree is to produce professionals who can drive innovation—applying new ideas, methods, or technologies to create value—then academic programs must be available to support that outcome too.
The solution is not to replace existing PhD programs, but to add a STEM innovation PhD track. The selective program would require deep private-sector engagement in an accelerated 4-year program in which a student immediately joins a structured research project with clear milestones, crafted by a faculty member in consultation with a relevant industry partner. During the second and third years, students would participate in 2- or 3-month internships with companies that are aligned with their research.
I do not agree that one of the goals of a Ph.D. program is to drive technology to create value. It may be a consequence of mastering critical thinking but that's not the same thing. I do not think that the average student who obtains a Ph.D. in evolutionary biology, geomorphology, or astrophysics has to demonstrate that they are "professionals who can drive innovation-applying new ideas."
Furthermore, the idea that Ph.D. programs at major universities would be partially controlled by industrial partners is repugnant. When I was training Ph.D. students there was no way that I would have teamed up with a biotech company or a drug company and allowed them to exploit one of my students for free research.
What do you think? Should Science be publishing such editorials on the prominent first page of the journal?
9 comments :
They're obviously thinking of the "T" and "E" bits of STEM (which wouldn't include the examples you give, but might include biochem insofar as it feeds in to the pharma industry). On the whole, it sounds like creating an excuse for techbros to style themselves as "Dr."
This already exists in Canada, to some extent, in the form of the MITACs program. Essentially joint industry-academia projects with some governmental support. I've not had any of these students in my lab, but some of our department labs have hosted these students with good success.
Do I think this is a good idea? No. Collaborative University-industry PhDs were tried in the UK in the 1970s. The industrial partner saw this as a way of getting routine background research carried out, while the deeper motivation remained a proprietary secret. Do I think it was a good idea to publish this editorial? Yes. The current PhD is an inefficient ritual, focusing on what is nominally one individual's contribution to a single problem defined in advance, culminating in the writing of a tedious thesis unlike any useful technical or general interest publication, and we should be talking about this
@Paul Braterman I'm only familiar with PhD programs in biochemistry and molecular biology in North American universities.
As a general rule, the research done during graduate school is open-ended - just as you would expect at the front lines of research. The general problem is laid out but it's not true, in my experience, that the thesis is based on "a single problem defined in advance." In fact, many (most?) students end up working on something very different from what they started with.
It's not unusual for students to find their own problem to work on and convince their supervisor that it's worthwhile. Two of my own students did this and ended up with a thesis that was not what I had suggested in the beginning.
The final thesis often consists of published papers or papers that have been submitted plus a lengthy introduction showing that the student understands the context and the background of the subject. Sometimes, the project doesn't work out and the results don't lead anywhere. This happens a lot in research but it doesn't mean the student can't get a PhD as long as they can convince their committee that they understand the subject and took a reasonable approach. I've approved several PhD's that fall into that category and some of the students went on to become excellent researchers.
My own thesis didn't produce an answer to the most important question I was addressing.
Often the PhD oral exam is as much about the student's understanding as it is about technical details. I was sincere when I said that one of the important goals, in my experience, is teaching graduate students how to approach a problem (i.e. critical thinking).
I don't think my approach is widely shared these days since I see plenty of prominent research labs who aren't following it. They churn out PhD's who are clones of their supervisors and don't understand the controversies in their own field. I'm thinking about students in ENCODE labs and in labs that study epigenetics, regulation, and alternative splicing.
I can't imagine how working for a biotech or drug company is consistent with the freedom that's required of a graduate student working on basic research.
Maybe those students shouldn't get a doctor of philosophy degree but some other kind of degree showing that they know how to obey corporate bosses?
Ah yes, but that brings up the fundamental problem. No-one knows what research will become useful technically, or of general interest. Ask industry to focus on only technically useful research and they'd laugh at you. That would not be research, but development. If industry, with its focus on outcomes, does not know which research will pay off, how is anyone else supposed to know? If we see further today it is not by standing on the shoulders of giants, but standing on piles of (sometimes tedious) theses and publications. When these things are written, no-one knows which ones will be stood on, nor, in this time of multi-discipline approaches, by whom.
I like the comments above. I'm also troubled by the opening of the 2nd paragraph of the posted portion of the proposal:
"The solution..."
There probably are issues with how current PhD programs are run in various places. Mine was in statistics back in '89 so it has been a while, but implying that the proposal authors know the single solution to the many problems across STEM disciplines seems to me to be narrow-minded and more than a bit arrogant.
I see the goals of industry-driven research as generally incompatible with a well-designed PhD research project. I feel that a good PhD project should be designed to be of scientific value, whether the answer is “yes” or “no” (the "yes" may be more interesting, but the "no" should at least be useful in advancing the field). Industry has different priorities, as they should. As for the assertion that “The current PhD is an inefficient ritual, focusing on what is nominally one individual's contribution to a single problem defined in advance, culminating in the writing of a tedious thesis unlike any useful technical or general interest publication”, I agree that is the case in some instances, particularly when students chose to work in “high-power” labs where there may be 30 people, the PI is rarely available, and the student’s mentor is, in practice, a post-doc. I find that most of these PhDs had a terrible experience and felt like they were just cheap labor (they were) but I put some of the onus on the student’s choice of laboratory here. I went into a small lab and had a wonderful time- I was tasked with bringing molecular biology into a strictly protein lab, and learned a very broad swath of techniques working on an interesting problem. I actually stayed for an extra year after my committee said I could graduate just because I was on a roll. To the point of the post, sure, this editorial can be published. But it should certainly be rebutted.
Unfortunately, the situation in Germany got worse since the diploma studies were replaced by bachelor/master studies to harmonize European/international stndards. When I started as a PhD student I was treated as a full scientist capable of addressing and solving problems technically and scientifically. There were nor graduate schools which in my opinion is just a means of keeping students small and dependent. In addition, there were much less PhD students than today. Of course, we had to learn things but we did this on the job either by reading or by discussions with more experienced colleagues not by joining classes again. I feel sorry for today’s PhD students who haven’t been given the chance to experience a solid undergraduate education just because the curriculum has been cut short. On the other hand, there are enough young students who despite their education are so good that they don’t need graduate schools and can run their PhD projects without much supervision. Thus, there will always be good scientists independent of the curriculum and the organization or form of studies. The problem is people who are not really qualified are carried through the system and being promoted until this becomes obvious.
I have no problem with the kind of industry-track graduate degree they are talking about. However, it clearly would require a different degree 'name' (D.Tech. or T.D. or whatever) since, as you point out, the goals are completely different and the kinds of 'skills'/abilities that are being developed are completely different. Would it make sense to have a medical degree like M.D. be modified to have an drug company or insurance company directed 'track'? Absolutely not and it would be obvious to everyone. M.D.'s are supposed to be trained to do what is best for patients' health and having drug or insurance companies entered into the training risks changing the entire goal of the degree. It's exactly the same with a PhD. In fact using PhD for that kind of degree is an attempt to steal the legitimacy and known value of the PhD degree to market a completely different degree.
Post a Comment