A new study by Robert Tai of the University of Virginia compares students who studied broad science topics to those who studied fewer subjects but to greater depth. He found that those who studies few topics to depth performed better in college science classes. This raises some interesting questions for science education.

One thing there seems to be broad agreement on is that we need better science education. Not everyone agrees how to accomplish that goal – whether through better textbooks, better teachers, more multimedia approaches, or more hands-on learning. There is a surfeit of opinions but a scarcity of outcome data. There is some but not enough to squelch the diversity of opinions based largely on ideology and experience.

I have a personal and professional interest in this question. I am a science teacher at the post-graduate (i.e. medical school) and professional level. I also popularize science through my podcasts, blogs, and lectures. And I am also the father of two daughters (currently aged 9 and 6) and am actively involved in their science education, as well as monitoring their public school science education.  What this means is that I have strong opinions about science education, but also wish them to be better informed by actual evidence. So I always take an interest when these types of studies come out.

This particular variable – depth vs breadth – is one I have thought about before. I have always been of two minds on the topic. On the one hand I feel that breadth of scientific knowledge is important. Science as a discipline should all hang together, a property known as consilience. This is because science is exploring the same natural world, so discoveries in one area should at least be compatible with discoveries in another.

Further, there is insight to be gained by looking across multiple disciplines within science. It enables one to see patterns of knowledge, information, systems, and the ways in which nature works. Taking an interdisciplinary approach to problems is very popular (if rarely practiced) because of the recognition that knowledge from one area may solve a problem from another.

On the other hand there are also deep insights to be gained from studying one area to great depth. Depth of knowledge reveals patterns and insights not accessible to the dabbler. There are also general lessons that can be learned from deeply studying almost any topic. These lessons can then be broadly applied.

For example many children have a fondness for dinosaurs (as do my daughters, but I have also fostered an equal fondness for birds). By studying dinosaurs or birds in detail they have learned about categorization, attention to detail, and evolutionary principles of nestled hierarchies, diversity and disparity.  They also have learned about how to learn, how to find and access desired information, and the joy of knowledge.

Each approach, therefore, has its strengths and weaknesses. Breadth without depth leads to a superficial understanding of how science works, and such knowledge may not stick. Depth in only a narrow field without breadth gives a parochial view of science, and may also give one a false sense of expertise in all areas of knowledge.

Therefore my position has been that the best approach is to strike a balance  – learn about all areas of science to a basic level but also delve as deeply as you can go in one or a few areas. It is probably not a coincidence that this is the approach taken in medical education. Medical students learn a little bit of everything, but then specialize in one area, and may in fact subspecialize in a very narrow area. Academics often reduce their focus to a single topic and try to learn everything there is to know about that one very narrow field. In the applied science of medicine, at least, the balance approach has prevailed.

I also find that most career scientists have a keen interest in all of science outside their specialty, and become well-informed lay persons in their spare time.

This new study does not really change my opinion on this topic, although it may give me more of an appreciation for the value of in-depth science study at the high school level (as opposed to waiting until college). It makes sense for the reasons I stated above. Learning a little bit of chemistry, biology, and physics in high school may not give sufficient knowledge for it to really stick. And without sufficient depth, real scientific principles may not come across. For example, at the basic level of knowledge you don’t really get an appreciation for the controversies in science and how they are resolved. Only when a topic in given in depth do you get to the limits of knowledge where science really happens.

It’s also possible that the reason students who took advanced courses in one science did better later on is because teachers capable of teaching the more advanced courses were better science teachers overall. Also, this study is retrospective so there can obviously be self-selection in those students who took advanced courses (although the study tried to control for this). So while this study is provocative, and I do think it reflects a real effect, it is far from being the final word on this question. There are still many variables to work out.

What will need to be done is for school systems to implement changes to their curriculum and then measure the outcome of those changes.  Again, this will introduce many variables, but it would be a good real-world test to guide future science curricula.

Perhaps the best approach at this time would be to require students to take basic-level courses in the major sciences and then an advanced course in the topic of their choice. Further, the advanced course can require a thesis or other project that would require each student to study a specific topic to even greater depth. This balanced approach is most likely to get the best of both worlds, and also reflects what happens in the real world.

An alternative approach would be to have basic level courses include in-depth study in one subject area – just as an example of the field. So the combined approach can be taken within one course.

But I will point out that I do not think the answer to our science education problems lies solely in any such tweaking of the curriculum. There is no substitute for quality science teachers, and so far that seems to be the dominant variable in the equation. While I think we should continue to monitor and tweak the curriculum, it is not going to be a quick fix. Investing in and demanding higher standards for science teachers is probably going to be the most important measure for improving science education.