American Association for the Advancement of Science Fellows Forum 2008
Keynote Address to Newly Inducted Fellows of the AAAS
President H. Kim Bottomly
February 16, 2008

"Unchanging Science Education in a Changing World"

Thank you for inviting me here today. I am pleased to be among the first to greet the new fellows of the AAAS. I extend a special greeting to my colleagues from Yale. As many of you know, this is my first year in many decades out of my laboratory. I want to talk to you today not about the many things I learned during those decades, but about what I have come to realize in the six months since I left my lab.

The problem I want to discuss today is a two-part one. First, too few students are interested in science. There is a rapidly escalating global need for a science and technology labor force. Second, should we as institutions of higher education be doing something different?

I am an immunobiologist and was a faculty member at Yale for 27 years. While my primary focus was research, I also taught undergraduates. One of the frustrations for me was that at Yale, like at many universities, students come with the aptitude and interest in science but are not inspired to continue with this interest. If they do remain in science at Yale, it is often primarily to go to medical school. That used to bother me, and it really shouldn’t have. I will return to that point later. Now that I am a college president, I am focused on this issue from a different point of view. Several articles have been written that suggest that liberal arts colleges provide a special environment for generating interest in science. And I now can see why that is so. But what is important is how can we transfer this good effect to research universities. The transferal is important, because no matter how good a job small liberal arts colleges do, it is the large research universities that educate the vast majority of science majors. If we want to put significantly more science majors into the pipeline, it is at the large research universities that we must do it.

Let’s begin by asking: Why should we be concerned about the number of science majors? Lately, many have predicted a coming shortage of science and engineering workers at all levels, a shortage that could cause us as a nation to lose our competitive edge. Is this true? If so, how does this comport with the fact that many write that the future for science graduate students in many fields seems bleak, with a reported shortage of academic jobs and intense competition for research grants making this career option unappealing.

Today I want to make a case that we do need to inspire more students to be scientists. To accomplish this will require changes in how universities approach science teaching, how universities provide research opportunities, and how you as inspiring role models should be involved in this process.

Let’s begin by addressing the first question: do we really need an increased science and engineering workforce? In the U.S., the number of science and engineering jobs has been growing almost 5 percent per year, compared to just over 1 percent per year for the rest of the labor force. The Bureau of Labor Statistics has projected that those occupations will continue to increase at three times the rate of all occupations, and they predicted a 47 percent increase in the number of science and engineering jobs by 2010. While the number of these jobs has been growing at almost 5 percent per year, the total number of science and engineering degrees earned only grew at a rate of 1.5 percent. So this much is very clear: our production of science degrees is not keeping up with the increase in science jobs.

There also are other factors that point to the possibility of a coming national problem in filling all of these jobs. Retirements from the scientific labor force will be significant over next decade or so. I say “or so” only because I know that we baby boomers have not acquired grace with our age – we won’t go out easily. But we will leave, and in ever-increasing numbers. Already, a large percentage of the science and engineering degree holders are 50 or older; 44 percent of Ph.D.s is that old. A labor force with such substantial numbers in their 50s is one destined to shrink very soon, because the data show that one-half of bachelor’s degree holders leave their jobs by age 62 and that one-half of Ph.D. holders leave by age 66. If you had asked me five years ago, I would have said that the only way I would ever leave my lab was when they carried me out feet first ― and yet here I am, even ahead of the curve. So it is clear that the baby boomers in the workforce will inevitably retire in the next 20 years whether we want to or not.

Even though the rate of degree production has been below the growth in the labor force, this hasn’t become a problem yet because of the large number of talented, foreign-born science and engineering graduates who migrate to the U.S. The proportion of foreign-born science students and workers has been rising steadily. By 2003, the proportion was 25 percent for bachelor’s degree holders and 40 percent for those with Ph.D.s. Already, one-half of our postdocs are foreign born.

Furthermore, many have pointed to a strong possibility that fewer of the foreign-born will come to the U.S. in the future due to escalating global competition in these fields. These are trends, not actual problems as yet. Even though foreign born students earned more than one third of all U.S. science doctorates in 2005, most of those recipients still plan to stay in the United States after graduation. But it does point to the magnitude of the problem if global competition did start cutting into those that we have come to count on remaining here. And this is not just true of the industrial labor force — it will affect our labs as well: foreign born workers accounted for 55 percent of science and engineering postdocs in the U.S. in 2005. Potential shortages at the highest level, doctoral and postdoctoral, raises the question of whether we can maintain our leadership in science when we don’t have the world’s talent at our fingertips.

I should pause to note at this point that I am only focusing on the problems for us. It goes without saying that the expansion of science and engineering expertise globally is a good thing for the rest of the world and will benefit many. I will refer you to the work of Wellesley College faculty member Robert Paarlberg who recently published a book about the rest of the world, entitled Starved for Science. It is also the case that more universities overseas will offer new opportunities for U.S. students, and also a better pool of applicants for jobs. But for the U.S., it will be a problem.

Having said this, I should add that some economists have argued that because the workforce argument is based on labor projections and not hard data, we don’t really know that it will happen, especially because, they believe, even if there is a science worker labor shortage, the resultant demand will create a larger science worker supply. Their argument is that it will be a self-correcting process; we don’t need to worry. Yet, like always, these economists assume a perfect market. Demand won’t necessarily create supply in an imperfect or flawed market ― that is, if there are significant barriers to the creation of that supply. And there are. I’m going to talk about one of those barriers today ― there are others as well, as I will note in passing.

Let me first, as an aside, add a broader view here. We need to educate more scientists. Yes, we need to preserve our talent pool to replace retiring faculty and potentially diminishing foreign-born academic pools. But we also need to educate scientists for a broader set of career options. We need to put more people who truly understand science into the non-academic science labor force. Why? Because we need lobbyists, we need lawyers, we need politicians, we need teachers, we need writers, and we need people in all walks of life who truly understand what science is all about. Science suffers greatly from the public’s lack of understanding about what we do and how we do it and what we need to do it with. This means we need more science-educated citizens in all walks of life.

Where are the barriers?
You have all read about our “leaky pipeline,” which always gets blamed for this problem. It leaks in many places, but I want to focus on the undergraduate years today, both because I think there are very significant losses at that point and also because I think we really can do something about it.

Whatever happens in high school, there still is considerable interest in science among entering college freshmen. At competitive schools, 40 percent of those entering express an interest in a science major. At that point they intend to major in science. Yet we lose about one-half of those students to non-science disciplines, so only about 20 percent or less continue with their plans and receive science degrees. And that is not because we bust out those who can’t cut it. Many of those we lose are extremely bright and motivated students who go on to excel in other areas

This is not a new phenomenon and has been consistent for the past two decades. Not only has this not changed, but our way of educating scientists hasn’t changed much during this time period either.

Why do entering freshmen lose interest? Are some schools doing it better? It has been reported that liberal arts education enhances the likelihood that students will graduate with a science degree. I looked at the data comparing selective liberal arts colleges and universities, either Ivies or top research non-Ivies, for their ability to attract students intending to major in science and engineering and for the percent of students graduating with science and engineering degrees.

What I found was that universities attract more students intending to major in science but colleges have a better yield. Liberal arts colleges yield about 75 percent of intending science majors as ultimate science degree holders. Ivy League universities yield about 40 percent and selected non-Ivy League universities 50 percent.

Why do liberal arts colleges do it better? It has been reported from a number of studies that the following are important: small class size, low student faculty ratio, research opportunities with faculty (not graduate students). Yet most science-interested students apply to universities thinking that the opportunities to do science are greater there (active, well known scientists, state of the art equipment, many more laboratories, etc.), and they leave science there in greater numbers. Part of the problem, I think, is that we still teach science like a pyramid ― laying the foundation carefully and slowly and in great detail and making sure that the foundational work is done before we reveal the point.

Most students come into college vaguely intending to major in something they know about. From high school they know about broad major areas only, so they start out thinking about majoring in science, history, math, English, art and a very few others. Then they arrive at college and discover the vast array of possibilities. Classics? Who knew you could actually major in that? Sociology? What is that, anyway? Students change their intentions ― that is to be expected. But while other fields both lose and gain students during this period (and their total percentages do not decline much from freshman to senior year), the sciences mostly lose without gaining. The true science devotees, like me, and like most of us here, don’t drop out. But we do lose those bright students who aren’t absolutely dedicated to science coming in. Other fields gain adherents, but we gain very few. Why do we lose many of those bright students who are on the cusp? One easy answer might be found in our introductory courses. I looked at lower-division courses at good places in a number of fields. Here’s an example of what I found, their exact course topics to be covered. In Introduction to Biology, these are the listed topics:

Circadian Control of Neurogenesis
Biological macromolecules
Cell membrane structure and function
Eukaryotic gene regulation

Compare that to these topics from lower-division courses in history, classics and sociology:

Women, Sex, and War in Athens
Crime and Deviant Behavior
Pirates in the Roman Empire
Human sexuality and intimate relations in modern society
The Nazis and the Final Solution
Greek men: drinking, poetry and sexuality
Urban manhood before the Civil War

Now let’s do a mental experiment. Imagine yourself to be a bright 18-year-old who likes science but is not yet particularly committed to it. You need to choose a course that sounds interesting. Which might you go for: Eukaryotic gene regulation or Women, Sex and War in Athens? Many, many choose the latter. What can we do about this?

Your first reaction may be the same as my first reaction when I thought about this. Nothing, we can do nothing because our courses are so very different ― we need to focus on what students may consider the minutiae because that is the nature of our field. We need to build that pyramid. But I am not so sure of that. The course that has women, sex and war as a topic does focus on minutiae and methodology: detailed historical dating methods, subtleties of Greek translation, geopolitical descriptions. But it starts with a big picture first, not with the minutiae. It is an Introduction to Classics course, but even titles itself differently: Uncovering the Ancient World. Scholars in the classics know that they will have trouble attracting majors, and they do something about it. If they structured their courses the way introduction to science courses are structured, the course would be called Introduction to Classics and the topics would be: Greek syntax, Greek and Persian administrative structure, and the complexities of calendric systems. This would not attract any majors. Yet they do attract majors, and they cover all of that.

We need to develop more good science courses for lower division students, courses that make it clear that science is an enormously creative enterprise. We know that, but undergraduates often don’t, and surveys show that their reasons for changing often involve the perception that science is just about amassing facts ― other disciplines are the creative ones.

In 1991 Wellesley College initiated a study with Sloan Foundation funding to determine what influences students to chose or avoid studying science as undergraduates. This longitudinal study and sample of students from 10 different entry years turned up some interesting facts.

First they found that very few students are actually recruited into science after they get to college. Thus it is about retention.

Second, interviews of students who left science reported that they prefer subjects with multiple interpretations rather than precise answers. We know that success in our field depends on an individual’s ability to both seek precise answers and tackle multiple interpretations. The challenge, then, is to demonstrate to students early that science is not just a series of formulaic exercises with right answers, but instead a creative series of trials and interpretations. This is rarely apparent to students at the introductory level.

Third, the students interviewed also revealed that the standard competitive environment inhibited a full educational experience and drove students away from science. Instead science majors insisted that collaborating with other students was an important part of their educational experience and that teamwork was an important skill that was valued.

These findings led Wellesley to revamp its science curriculum, reducing the size of the first year courses, providing dynamic laboratory sections led by faculty, emphasizing teamwork and continuing mentorship past graduation.

So what would be a good science lecture course? I will leave that to you. But Harvard’s big “Science is Sexy” party on Thursday was a good beginning.

Considerable work has been done by the National Science Foundation (NSF), the Howard Hughes Medical Institute and others to develop new methods for undergraduate science teaching. These methods involve the absence of one-way lectures, using teams of students to solve problems, peer-led team learning and other innovations. The beginnings are out there for you to find.

But the practical reality is that there is no incentive for faculty members to teach creatively. It takes time. In many places, there are strong disincentives. Departments control course development and teaching assignments. Tenure deliberations don’t consider teaching excellence. Therefore the only way the new methods are employed is if individual faculty members are motivated to do it. This is why I am speaking to you today about this. You are the leaders in your field and you can lead the way.

But we need something more that will ignite undergraduates’ passion about science, something that will actively keep them there. One crucial factor, many studies have shown, is very early undergraduate involvement in research. This is more difficult than restructuring undergraduate courses. This factor, I should note, is only effective when undergraduates have an active part in the research ― washing test tubes doesn’t do it. How do you keep a research project going aggressively while actively involving undergraduates who have little skill, and who are transient? It may be enormously effective, but it is difficult to achieve. It is easier to do at liberal arts colleges because there are only undergraduates, and there are different research expectations. Universities and scientists need to partner to provide good research experiences.

One good example is the Dartmouth Women in Science Program. I bring this up because I believe it would work for all students in science and illustrates the importance of a partnership between the institution and the faculty member in science education. They began with a specific invitation to incoming students to participate in the program. The offered paid first-year internships which involved hands-on lab experiences working one-on-on with a faculty member. This first-hand experience with scientific inquiry culminated in a poster symposium at the conclusion of the internship. They also had a peer mentoring program, an industrial mentoring program and a bimonthly newsletter. The program began with 45 research internships and quickly grew to 100 internships. In just seven years of the program, the proportion of women declaring majors in science and engineering increased from 12 to 24 percent. Do note this critical point: Dartmouth College provided support for the program, making it possible and easier for the faculty to participate. Universities need to provide funds and to think about rewards (such as counting undergraduate research activity as teaching, etc.)

In 2006 the NSF published the results of its large study on the effect of involving undergraduates in research. NSF had conducted a broad-based, nationwide evaluative study of its support for undergraduate research. The study involved four different surveys and about 15,000 subjects, but the findings were uniformly positive and remarkably robust and consistent. There were few appreciable differences in effects among graduates of different types of schools or even between those who began their undergraduate education at a two-year college and those who began at a four-year school. A good research experience appeared to transcend the place in which it was had.

The survey findings supported widely held beliefs about the positive effects of undergraduate research. Such research experiences, it was found: increased the likelihood of obtaining a Ph.D., had a strong positive effect on students’ understanding of the research process, increased students’ confidence, greatly increased students’ awareness of academic and career options in science and increased their interest in pursuing science studies and careers. Students who were actively involved in the culture of research — attending conferences, mentoring other students, authoring journal papers and so on — were the most likely to experience positive outcomes.

Now here is a problem to be overcome. NSF interviews with those mentors directing the undergraduate research determined that faculty derived a great deal of personal satisfaction from working with undergraduates in a research setting, but less than 40 percent agreed with the statement, “Mentoring undergraduates is viewed favorably in my department’s tenure/promotion review process.”

I am telling you all this today because most of you are at universities. Most of you teach science. Most of you have students in your laboratory. Most of you recognize that we are not doing a good job of selling science. Most of you probably agree that what I am talking about is true. But ― ah yes, there is that “but.” But designing an exciting introductory course takes time. You do not have much time and you have other pressing priorities. And taking undergraduates into your laboratory is also time-consuming and difficult. Undergraduates take more time than they give. Though there may be some personal rewards for accomplishing this, there are few institutional rewards. Besides, even if you wanted to do something innovative, most of you don’t know how to do this. You teach as you were taught. And so it goes.

It is clear to me that we must now do something. We now know that some aspect of liberal arts education can be translated to the university setting successfully.

I am hoping that we can find a few people at each university who are willing and able to do this. But there is something all of you can set as a goal: you can take the time to be inspirational. Now that I am at Wellesley College, I have observed the power of teaching. I have been talking to many very successful alumnae: all have stories; all had special professors who influenced them ― again the power of teaching. I can look back at my life and remember those defining moments that set me on the road to becoming a successful scientist. In high school, there was that advanced biology teacher who let us inject hormones into chicks and study changes in secondary sex characteristics. Looking back, I realize that the linking of sex and science for 17-year-olds was brilliant. In college, there was that professor who took the time to allow each member of his undergraduate class to conduct their own experimental study. Elaborate equipment isn’t always necessary. My experiment involved analysis of the sensory mechanism of crabs. I collected the crabs from Puget Sound and carried out my experiments in my community dormitory sink. This experience was at the University of Washington, not a small liberal arts college. I know that most of you can recall those inspirational figures in your own educational past. The role of “inspiration” in science persistence hasn’t been studied and I think it should be. It is a major factor ― I am sure.

As new fellows of the AAAS you are eminently qualified to inspire. You are leaders in your field, your research is exciting. You will make contributions through your research. But by virtue of your position, you have also been given the responsibility to inspire the next generation of scientists. I am asking you to take that responsibility seriously. I hope you rethink your approach to teaching undergraduates. I hope you find space and time in your lab for some undergraduates. I hope that you inspire them to be as passionate about science as you are. Take this part of your job seriously. If each AAAS fellow here today inspired just one additional bright undergraduate a year to stick with science, think of the aggregate effect on the discipline.

Seeing you all here makes me realize that the scientific enterprise is in very good shape today and with your help it will be in good shape tomorrow as well. I congratulate you on becoming a fellow.