Straight Talk About STEM Education

What's the freshman experience like at Harvey Mudd?
Goroff: All of our students come here expecting to do something in a STEM field, and our core program aims to prepare them to address the big problems in the world. In addition, anything we can do to erase the myth of a lone genius as the exemplar of what it takes to succeed in science and show them that science is done in connected communities rather than by someone holed away in an office really helps with the retention problem.

Pfirman: For many liberal arts students, their only exposure is through a science requirement. Unfortunately, these classes have often dumbed down the science. At Barnard, there are first-year seminars that had traditionally been offered by the English and humanities departments, and we decided to offer one in the sciences that sneaks in the science along with the rest of the subject matter. After the class, the students reported they felt a greater connection to science and saw a greater role for science in their lives.

How hard is it to revise an intro course, and what are the limiting factors?
Summers: Part of it is knowing what to do. NSF [National Science Foundation] and HHMI [Howard Hughes Medical Institute] have funded pilot projects that are, in effect, experiments designed to modify the curriculum, for example, to make biology more quantitative. What's needed now are the data that show the impact of these programs.

Pfirman: Faculty really want to be effective. It's not just what content is appropriate but also how to deliver it and how to assess it, so that you can do a better job next time. … What would really help is a way to find out what's out there that has worked well. I was just looking for a lab module to go with a unit on the ocean's role in the carbon cycle. It's a fairly basic concept. But I couldn't find anything on the Web.

What would you recommend for a faculty member who wants to do something but doesn't know where to look?
Summers: When we decided that we wanted to try to retain more students in chemistry, we looked at bridge programs--a summer program for incoming freshmen--including how to study in groups. Hal White at the University of Delaware has been doing good things with small groups. But these things cost money, and you need teaching assistants. So we had to go to the administration to get money to test this experiment. But now that we've done it and put the results on the Web, I would hope that other faculty could take the data and show it to the dean and say, "This is what worked there, and I think we can do the same thing here."

Goroff: I'd encourage faculty to get involved with national groups, too. It doesn't make sense to reinvent the wheel, much less the flat tire. Project Kaleidoscope [pkal.org] is a network of young faculty who are sharing ideas. Project Next [maa.org] is another such effort.

UMBC is one of several institutions that are trying to increase retention rates among minority students and narrow the achievement gap. How is that working?
Summers: Our president, Freeman Hrabowski, began with the premise that there are a large number of high-achieving minority students, in particular African Americans, who are just not being retained once they enter college to study science. He goes after them in high school. He brings them to campus with their families, and we start talking about earning a Ph.D. degree.

Over the past 20 years, we've graduated more than 500 students, a large percentage of whom have gone on to get graduate degrees. … And we've tried to quantify the experiment with controls. We bring about 200 students to campus and make about 100 offers--and these are full scholarships--knowing that only about 50 will actually join the program. And before we make an offer, we make parents sign a waiver that says even if you turn us down, we want to be able to track them. So now we have a database. It turns out that if they go elsewhere--and these kids get offers from Ivy League schools and everywhere else--they are half as likely to graduate with a science or math degree and more than five times less likely to go to graduate school.

CREDITS (LEFT TO RIGHT): KEVIN MAPP/HARVEY MUDD COLLEGE

Flying high. (Top) Harvey Mudd students test carbon fiber poles and the output of LED lights in a project for the Federal Aviation Administration. (Left) Meyerhoff graduate fellows take a break from the lab for an outing to Harpers Ferry in West Virginia. (Right) Barnard's Maggie Chan collects and characterizes sediment samples from the Hudson River bed near Manhattan. CREDITS: MICHAEL F. SUMMERS; M. STUTE

A lot of them have taken AP classes and could place out of chemistry. But they take the freshman courses, anyway. And the key is doing well in those first courses. Because if they are successful, then the research faculty are eager to get them involved in their labs.

What if they're not successful?
Summers: There are counselors in their first semester who keep track of every pop quiz and test that students take. If they have problems, they are immediately paired up with other classmates who are doing well in a particular subject. And we pay seniors to be mentors. We'll also call the parents, and there's a parents' association that provides money and time. This is a group effort.

People have accused the program of cherry picking by only taking the most able students. What's your response?
Summers: We are taking students who are most likely to succeed based on high school performance. But we are only taking a small fraction of that pool--we get 2000 nominations a year and about 800 applications [for approximately 50 slots]. So in a way, we are cherry-picking. But we don't apologize for it. And we only take a small percentage. People should be cherry-picking in every state.

Are environmental programs a good way to attract students into STEM fields?
Pfirman: I see the environment as the liberal science field of this century. It brings together all the disciplines, and you have a problem-solving focus that makes them naturally work together. It also gives students a link to their own lives.

"I see the environment as the liberal science field of this century. It brings together all the disciplines . . . and gives students a link to their own lives." --Stephanie Pfirman, Barnard College

CREDIT: BRUCE GILBERT

Goroff: Nature doesn't give us the great problems of the world labeled as physics or chemistry. So we want students to have that broad background. But having an in-depth experience is useful, too.

Even if students aren't thinking about careers, however, we need to be paying more attention to them, because the opportunity costs to go into a science career are much different than for someone in China or India. We shouldn't be surprised if students are thinking about law school or business school if we haven't paid attention to finding ways of making a science career fulfilling and rewarding--and I'm not just talking about money--without first having to spend 10 years as a graduate student and a postdoc.

How career-oriented are UMBC students, and does that push them into or out of STEM fields?
Summers: We have some students who have known since kindergarten that they want to be a medical doctor, and that's fine. But in general, I think we need to be careful about pushing students into careers at too early an age. I know from talking to some of my graduate students from other countries that they are tested and channeled at a very early age into either science or nonscience tracks. I think it's good for students to have the opportunity to get excited by a course or a professor and have a chance to go into science.

How well are institutions doing in tracking student achievement and what happens to them after graduation?
Summers: I don't know if there's any real tracking of our general population once they leave. We certainly track those in our Meyerhoff Scholars Program. We know where they are in grad school and their postdocs and when they apply for faculty jobs, because we want to know if we are being successful.

I don't think we take the same scientific approach to our teaching responsibilities as we do to our research. In research, we are evaluated based on our performance. If we write an NSF or NIH [National Institutes of Health] grant, we have certain objectives, and our peers look at those metrics when they judge our proposals, and later on they ask if we have been successful. Well, a major part of our job is to teach and educate. I don't think it should be mandated from the top. But I think we should approach teaching as we approach our science and think about what's important. If we're not retaining students in our own areas, then we're not doing our jobs.

"If we’re not retaining the students that we’re training, then we’re in trouble." --Michael Summers, University of Maryland, Baltimore County

CREDIT: TIM FORD/UMBC

We're also trying to understand better what students are learning in their classes. We're working with the Consortium on High Achievement and Success to understand why, for example, minority students might come in with high SATs but then tend to underperform, especially in large introductory courses. One interesting approach is to ask students to come up with a flow chart or a timeline to depict what they are studying, or to reflect on what they have learned. We're doing it in an environmental measurements class, in which students take samples from the Hudson River and analyze them, and it helps me to better evaluate their work.

What can universities do to make teaching a more attractive option for their STEM students?
Goroff: I run into STEM students all the time at the best institutions who are passionate about their subjects and who would love to make a career out of teaching. But they have trouble figuring out what a career like that would look like. So I think we need to take some responsibility and think about how these labor markets work and in particular the type of communities that exist to support that goal. In many countries, being a teacher is a very honorable and community-based activity, and teachers get together all the time to improve their craft. But that happens very little in the United States, and that's too bad. There are some programs, like Math for America or Teach for America, where you can see that progression and see how to make it a career.

Is teaching seen as a successful outcome of the Meyerhoff program, or is it more oriented toward research careers?
Summers: I think that there is some tension there. We're trying to get our students to feel passionate about science, which involves doing experiments. We brag about our students who go to graduate school at Harvard. So if a student goes into the lab and gets excited about doing research, will they have that opportunity if they decide to go into teaching? Probably not. They will probably have huge teaching loads, and it's not a glamorous or well-paid profession.

The clinic program at Harvey Mudd is a good example of providing research opportunities for undergraduates. How does it work?
Goroff: It's a signature program that has been going for more than 40 years, and it allows students to partner with industry or a federal lab or venture capitalists--people who have real problems that they need to find an answer to. The importance of having a client like that really makes a big difference. … We've even turned the program into a global activity, using videoconferencing and e-mail. It not only gets them ready to work in ways that employers and graduate schools appreciate, it also allows students who may not have the highest grade point average to excel as part of a team.

How do you make sure that they don't become just another pair of hands?
Goroff: We give away the intellectual property rights, and we're not trying to make money on this. We spend a lot of time working with sponsors to pick problems that will work well for undergraduates, and we explain that sometimes finding out the original idea didn't work can be a tremendously valuable experience--and a tremendous help to the company, too.

There is continuing pressure on universities to eliminate programs aimed at special populations on the grounds that they discriminate against the majority population. Is that having an impact on the programs that you run?
Summers: At UMBC, we have had to worry about that a little bit. When the Meyerhoff program began, it was specifically for black males, and then women, and then minorities in general. After the Banneker ruling [in 1994] shut down minority science programs at the University of Maryland, we decided to open up the program to all students who have an interest in diversity issues in the sciences. … If a white student qualifies, they have to decide if they want to do all the outreach and activities that are part of the Meyerhoff program--such as serving as tutors to inner-city minority students in the K-12 system and other monthly activities--or if they would rather take a general merit scholarship from the university. The program is now about 12% Caucasian and 12% to 15% Asian.

What about at Barnard?
Pfirman: We are a women's college, so opening it up to all our students just means more women. But I'm also working on a project to advance women faculty, and one thing we've discovered is that programs to help women are also helpful for men. So we've opened up the programs to both men and women, although we tend to get more women.

How do you extend a program that's existed for decades at one institution to the hundreds or even thousands of schools that might benefit from it?
Summers: We're trying to do that with a program funded by NIH and HHMI to increase retention rates among underrepresented minorities. We've had three meetings, hosted and attended by institutions that we felt were primed for changed. To participate, each institution first had to do a quantitative assessment of how women and minorities are doing on their campuses. The information wasn't easy to get, and many said they were surprised at how poorly their institutions were doing with respect to the achievement of minorities and women once they saw the numbers. In January, there will be another meeting for the institutions to present the outcomes of things that they have put in place. …

We call this "taking the show on the road." If you go to a typical scientific meeting, there might be 15,000 at the meeting. But only about 30 would attend a session on diversity. And they are usually the same people. After spending so much money for so many years trying to increase the number of minority scientists and seeing the number increase from maybe 1% to 2% in the sciences, federal agencies know they have to do things differently. … They have been able to build campfires at some institutions, and the question now is how do you fan the flames.

"We [need] to erase the myth of a lone genius . . . and show [students] that science is done in connected communities." --Daniel Goroff, Harvey Mudd College

CREDIT: KEVIN MAPP/HARVEY MUDD COLLEGE

Why isn't there sufficient national leadership now, and what can people do to make it happen?
Goroff: I think it's always fashionable in Washington to concentrate on the dollar signs and to think that if we just spend more money, the leaders will appear and the programs will appear. But if you look at what happened to NIH after the doubling of its budget--which was a wonderful thing that is absolutely worthy of our support--what you see is a great deal of this winner-take-all kind of funding schemes and the same people getting more grants. I'd rather see us sending money to people who want to implement some of these good ideas at their institution.

Summers: When we interview people for a position, some postdocs tell us that their adviser has warned them not to get too involved with undergraduates because they will end up spending too much time on teaching, and that could ruin their career. Faculty at most research institutions are evaluated on the basis of papers published and presentations at meetings and service on review panels. There's a teaching component, but it's not weighted the same. And there's no component at all that addresses mentoring students in your lab.

What one or two things would you like to see happen in the next 5 to 10 years to improve undergraduate STEM education, and how would you measure it?
Pfirman: I think it's critical within each of our institutions that we have academic leaders on campus who understand the value of science and math and what resources can make a difference. If more science faculty are willing to step up to those leadership positions, that could make a big difference. We also need to think of our institutions as learning communities and get everybody involved in the learning process.

Goroff: We've mentioned community many times. And I think that being part of something bigger than oneself is very appealing. If we can dispel the myth of the lone genius working on their own and emphasize the social capital, that would be a step in the right direction. A second contribution would be more data about learning and mentoring.

Summers: In science, sometimes making the big discovery means being in the right place in the right time. But when it comes to our own disciplines and our students, we're all in the right place at the right time to do something. When a minority student makes a C, rather than thinking, "That's good, they've done all right," we could bring them in and ask them, "Why didn't you make a B? How can you do better?" If we're not retaining the students that we're training, then we're in trouble.