Wednesday, May 22, 2013

A Concept-Driven Graduate Education

Like most graduate departments, we are constantly evaluating and modifying our graduate program but rarely do we step back and look at the big picture. What are we trying to accomplish?

Graduate education, like undergraduate education, has been subjected to serious study over the past few decades. Guttlerner and Van Vactor (2013) have just published a brief review in Cell that summarizes the goals and how they are trying to achieve them at Havard.

The authors begin by pointing out the importance of a concept-driven curriculum.
The modernization of science education requires a shift from a content-driven curriculum to an interdisciplinary, concept-driven curriculum (Association of American Medical Colleges and Howard Hughes Medical Institute, 2009; American Association for the Advancement of Science, 2009; National Research Council, 2003, 2009). Such a curriculum organizes information around unifying concepts and frees educators from the insurmountable task of presenting the complete breadth of an ever-expanding scientific knowledge base (D’Avanzo, 2008). Concept-driven education is increasingly seen as fundamental for contemporary research scientists and physicians (Association of American Medical Colleges and Howard Hughes Medical Institute, 2009).
This may seem obvious but I bet there are very few graduate programs that emphasize core concepts. The tendency is to mount graduate courses that explore the latest results in whatever field the lecturer is working on. These may be interesting courses for graduate students but if they lack understanding in basic core concepts then they won't be prepared to critically analyze the papers they're studying.

We saw an example of this when the ENCODE results were published. Many students are well-positioned to understand the techniques behind the ENCODE study but if they don't understand molecular evolution they will not be able to interpret the results. The goal of graduate education is not just to produce excellent technicians.

Nobody says that changing the way we teach graduate student is going to be easy. Gutlerner and Van Vactor recognize the difficulties ...
There are several barriers to innovation of interdisciplinary, concept-driven curricula that also develop skills in analytical thinking, experimental design, and technological fluency. Developing a cohesive graduate curriculum requires that diverse teams of faculty collaborate to reach consensus on the core conceptual learning goals for the curriculum and how to best meet these goals. Furthermore, particularly for large programs and schools, there is a challenge as to how these concepts will be aligned among multiple courses taught by different faculty members, what specific content will be used to illustrate these concepts in each course, and what content will be left for self-education. Graduate education must teach students to think independently, learn how to best access existing information, and acquire new knowledge on their own. Therefore, the curriculum must help develop these skills of the autodidact, while also identifying significant conceptual gaps in students’ backgrounds. It is a challenging task indeed to create an integrated curriculum that addresses prior misconceptions and gaps in knowledge and develops skills in experimental design, critical paper reading, and technical fluency and, at the same time, fosters the development of creative, independent, critical thinkers.
The program at Harvard has four core courses: molecular biology, cell biology, genetics, and protein biochemistry. Apparently the lecturers are developing a student-centered learning approach in those core courses.

Those courses seem quite broad and it would be nice to see some specific examples of graduate level concepts that are covered in those courses. The challenge, as the authors note, is to arrive at a consensus on what constitutes a core concept for graduate students. What did they decide at Harvard?

I thought I'd look at the course content on the BBS (Biological and Biomedical Sciences) website.

BCMP 200 Molecular Biology looks like the stuff we teach in our third year molecular biology course. Knowledge of that material is pretty much a requirement to get into our graduate program. There doesn't seem to be an emphasis on concepts. Genetics 201 looks like a typical introductory genetics course. Our students take it in their second year.

Cell Biology 201: Molecular Biology of the Cell looks like a typical survey course with no particular emphasis on core concepts. BCMP 201: Proteins: Structure, Function, and Catalysis seems to cover the same ground that we cover in our undergraduate courses.

Our experience with graduate courses at American universities is that they are taught at a much lower lever than we expect in Canadian graduate school courses. Most of them appear to be remedial courses designed to bring students up the the level that they should have attained before being accepted into a Ph.D. program.

One thing surprised me. The core courses are not required courses in the graduate program! Doesn't that defeat the whole purpose of core courses that are designed to address gaps in understanding? I'd have to see a more detailed syllabus to see how the concept-driven curriculum is being implemented at Harvard by my suspicion is that their view of "concepts" is far different than mine. I thing they are emphasizing core knowledge, not concepts.

One of the core concepts that I would teach in that program is a modern understanding of molecular evolution based on population genetics and Neutral Theory. What other core concepts should be taught in graduate school?

Gutlerner, J.L. and Van Vactor, D. (2013) Catalyzing Curriculum Evolution in Graduate Science Education. Cell 153:731-736. [Full Text + PDF]


  1. My impression is that there is a very big difference between what graduate education consists of in for example, math and physics on one side, and biology on the other. In those disciplines, you take a lot of courses as a graduate student, and they are real, hardcore course that you have to learn from. While in biology, they are few and often not very challenging. The cynical view is that this is because math and physics are disciplines in which typical undergraduate education is hopelessly insufficient for you to be able to do research, simply because of how advanced the cutting edge is compared to what can be reasonably taught in four years (you need to have covered a laundry list of very advanced math areas in order to even begin to fully understand what's going on theoretical physics). While in biology, it is possible to do a lot of work in a narrow area without knowing much about anything outside of it, and certainly, without seeing "the big picture" at all. That graduate students tend to be seen as primarily cheap labor rather than scholars who should be able to justify the "Ph" letters in their "PhD" only makes the situation worse. Basically, the impetus is to get you at the bench ASAP, while nobody really cares about intellectual development

  2. To switch to a "core concepts" focus it seems obvious that universities will need to de-emphasize the importance of publications and authorship in faculty hires because that's why there is a rush to "get you to the bench ASAP."

  3. What are we trying to accomplish?

    Oh, that's easy: to get as high quality cheap labor force as possible.

  4. critical paper reading, and technical click here and, at the same time, fosters the development of creative, independent, critical thinkers.