Occasional Paper II: What Works: Leadership— Challenges for the Future

Some Economics of Effective Science Education

Stephen R. Lewis, Jr.
President--Carleton College

Arguments for Consideration:

  • We have a serious national need to improve the scientific literacy of our work force and citizenry and the capacity of our scientific and technological manpower base.
  • Science and mathematics education at the undergraduate level is the critical point at which to attack the national problem.
  • We know what works in producing in producing effective education.
  • Economic studies consistently show that the payoff to a college education comes from earning the degree, not from taking courses toward partial fulfillment.
  • Effective undergraduate science and mathematics education may be no more costly per degree produced than traditional approaches which result in higher attrition.
  • A "market" solution-based on individual student, family, and institutional decisions will not yield a result in programs adequate to meet the national needs.
  • Making an adequate investment in effective undergraduate science and mathematics education would be a cost-effective way for governments and major national foundations to join in a new partnership with colleges and universities in serving the national interest in areas of science, technology, and education.

Broadening the discussion, redefining the problem, and understanding the whole process are critically important in finding effective solutions to problems.

With a diminishing lead in fields of science and technology, a low and declining rate of scientific literacy in the work force, and a decline in the number of citizens involved in baccalaureate and Ph.D. study in science, mathematics, engineering, and technology (SMET), our country needs to find a more effective means of addressing and supporting education in sciences and mathematics at all levels.

I argue here for new ways to approach the issue of financing programs in undergraduate science and mathematics from the perspective of achieving both national and institutional goals. In my argument, I address the need for a national partnership focused on the economics and management for effective SMET, and the need for institutions to understand what can and must be done to prepare for the next century.

I write out of convictions based on experience: more than 30 years as an operating policy economist in national governments and 25 years as an administrator in two science-active colleges. The perspective of policy analysis is an important one; I persist in thinking that economics is a useful, though widely misunderstood discipline. In reality, science all human activity is constrained, there can be no definition of the "best" without regard to resource constraints. And, there can be no definition of an "efficient" solution to a problem without an understanding of the objectives or the goals of those seeking the solution.

At the level of state or national policy making, one frequently hears complaints by economists that "efficient" solutions to problems were not followed due to "political" constraints; or, complaints from politicians that economists donât understand political realities and values.

Decisions and choices about policy are best made when there are clear understandings of the full range of constraints and of the ultimate objectives. Something which appears to be a constraint need not be one if the problem is understood more fully. Indeed, much of the current "continuous improvement" literature in the corporate world, as well as the literature of policy analysis, suggests that broadening the discussion, redefining the problem, and understanding the whole process are critically important in finding effective solutions to problems.

Analysis of National Objectives

Let me discuss some economics in the context of a national partnership.

Three national objectives seem to have wide acceptance:

  • improving the quality of science understanding of those majoring in science or math;
  • improving the quality of preparation of those who take science and math as part of their general education;
  • increasing the quantity of undergraduate majors in the sciences and mathematics, especially among persons of color and women.

National, we want to achieve these objectives at the lowest possible cost, recalling that "efficiency" can only be defined in relation to the objectives of those seeking the solution.

The national objectives, and the efficiency with which we meet them, have to be seen from several viewpoints including that of students, faculty members, and the individual institutions where education takes place. The perspectives of individuals and organizations beyond the campus are also critical: those who provide the direct financial resources for studentsâ educations (principally families and state governments), the national scientific community and the private corporations and foundations who support sciences, those who employ science- and math-literal workers, and the Federal government (Both as fiscal agenda and on behalf of all citizens).

This means a more iterative process between faculty and budget administrators; more variables to be considered simultaneously by individual faculty, by departments, and by divisions; more concern with the ultimate objectives than with the individual courses; more rewards and incentives based on the real objectives of the educational process: promoting learning by our students.

The undergraduate educational experience is the critical link in achieving our national objectives.

The reason is simple. Undergraduate institutions produce those who go on to teach in K-12; they produce the individuals who will be the political leaders, the local school board members, the Federal, state, and local officials who will make policy on science and on issues where scientific literacy is critical.

Undergraduate science programs, whether in two-year colleges, liberal arts colleges, comprehensive or research-intensive universities, are part of the "intellectual machine tool" or "human capital/goods-producing" sectors of the American economy. Quality education in the sciences and math at the K-12, undergraduate, and graduate levels as well as the quality and size of the national scientific research and development establishment are a function of--that is, they depend upon--the quality and quantity of science education at the undergraduate level.

To the extent that we are concerned with the full participation of U.S. citizens in science-based activities, the focus on undergraduates is need to ensure that U.S. citizens have adequate baccalaureate training to compete effectively with foreign students for admission to the best graduate programs.

The issue o f improving K-12 math-science education is equally important, and the science-active colleges have much to contribute in two respects. First, they can and have taken (in the case of Maine's Math & Science Alliance) the lead in developing consortia of schools and colleges to provide in-service training and curricular development for K-12 education. Second, as suggested by some Pennsylvania data, they are preferred providers of teachers to the K-12 public school system.

From a national perspective, therefore, the effective functioning of the science and mathematics education programs at he undergraduate level is of fundamental importance. This is simple, straightforward production economics.

Some Further Economics

Since a research-rich, discover-based, lean, lively curriculum developed in a community of learners works at the undergraduate level, the next questions for the policy analyst--both at the institutional and the national level--are: What's the cost? Is it cost effective?

Here it is critical to distinguish between several units of analysis: cost per student enrolled; cost per course or lab taught; and cost per baccalaureate degree produced. It is important--at the institutional as well as at the national level--to include all costs, including appropriate shares of libraries, computer centers, and physical plant--items normally not allocated to departments, much less to courses, in typical college accounting or budgeting systems.

Studies of the economics of higher education have shown that the payoff from undergraduate education comes with the achievement of a degree, not with partial completion of a program (although some data suggest this is not as strong a finding for women). What's important for an institution, a program, or the nation, the, is not the intake, or the enrollment level, but the output. Yet, comparative cost studies generally examine the educational costs per student enrolled at an institution. Given the systematic differences in costs between lover-level and upper-level undergraduate instruction, and the different attrition rates at different types of institutions, costs per enrolled student will not accurately reflect real differences in the costs to produce baccalaureate degrees. Such studies will systematically bias the results against colleges with high graduation rates.

For example, independent colleges sometimes show higher costs per enrolled student, though some recent studies (e.g., for New York State) indicate no substantial differences in costs per student enrolled between public and independent colleges of arts and science. Since they have higher completion rates, independent colleges (and public institutions with apparently richer resource levels)are probably more efficient and less costly when the measure is the cost of producing baccalaureates.

While we need some further research to confirm, contradict, or modify this mixture of established results and educated guesses, the arithmetic of increased retention and low marginal costs of added students at the advanced levels strongly suggests that changed approaches, even with consequent added costs at the introductory levels, are likely to be worthwhile.

Let me give you an example of what this might mean in practice:

  • Traditional undergraduate science and math programs treat first-year courses as weeding-out courses--they act "as a filter, not a pump." this often means low class sizes at upper levels and consequent high cost from low student/faculty ratios. If the productivity of faculty, space, instrumentation is to be raised, one way to do it is to reduce the attrition rate from lower-level to upper-level courses--one of the objectives of the curriculum and approach at the science-active colleges. Ed Buchwald, longtime Chair of Carletonâs exceptionally effective geology program , is fond of saying, "geology majors are made, not born" The department treats its introductory course as a proselytizing course for majors and as a result has well-filled upper-level courses. Per graduate, the costs are modest, because the "conversion rate" from elementary, to intermediate, to advanced courses is so high.

Such broader considerations about the economics of undergraduate science and mathematics education must be taken into account at the national level as staff at funding agencies and governmental offices develop policies and programs focused on clear, commonly-agreed upon national objectives for science and mathematics education.

That is critical; equally critical is to broaden the scope of discussion at the institutional level. Faculty, administrators, and trustees at colleges and universities must find new ways of talking with one another and of redefining problems sot hat high-quality results can come with low-cost methods. This means a more iterative process between faculty and budget administrators; more variables to be considered simultaneously by individual faculty, by departments, and by divisions; more concern with the ultimate objectives than with individual courses; more rewards and incentives based on the real objectives of the educational process; promoting learning by our students.

Institutional Perspectives

All institutions of higher education, event he most heavily endowed, are under extreme and growing financial pressure and are dealing with issues of finding the most cost-effective ways of providing education. While a quarter century ago relatively few institutions engaged in long-range planning and budgeting, it has now become the norm for successful institutions.

With institutions, we weigh competing claims of different academic disciplines, the support services (computers, libraries, communications, technicians) needed directly by academic programs, the support for other student needs (advising, health care, housing, career advice), and the needs of general overhead services (accounting, building maintenance, insurance, personnel services). We look for educational payoff in making those budget choices. And, we project costs and revenues ahead under many alternative economic scenarios.

Average costs of education vary among different areas. Typically, lower-level courses are less expensive than upper-level courses. And, typically, the laboratory sciences are more expensive than the social sciences and humanities due to the laboratory space and equipment requirements as well as the need for smaller section sizes in laboratories.

From the institutional point of view, the impact of diverting relatively large amounts of general funds to support specific, and already expensive, parts of the curriculum is greatly eased by external funds which come on a matching basis. in this decade's climate of budget austerity and competition over the division of an institution's educational resources the importance of such external validation, as well as external partial funding, cannot be emphasized too strongly.

To illustrate this point, within science and mathematics faculties, if external and internal funding for support of stimulating and imaginative education (including collaborative research with students) becomes more available--with recognition and reward at both the local and national level--it might not be surprising to find faculty shifting from funded research to funded curriculum development. If we want to capture the energy and the imagination of top scientists for the teaching enterprise, we have to ensure it's as exciting to be in the lab with undergraduates as it is with post-docs.

To further illustrate, upon building or remodeling facilities, institutions testify to considerable increases in their pool of applicants. The initial Project Kaleidoscope studies estimated that for the 400 or so science-active undergraduate institutions, the facilities needs over the next decade (new and renovated) could be met by an added $1.5 billion per year, while instrumentation costs might be in the range of $50 million annually. Obviously, such added resources would be fully effective only if they add resources to the lab-rich discovery-oriented curricular approaches that have been shown to be effective. But, if the resources are available, then we can expect higher output of science and math baccalaureates, augmenting the supply of K-12 teachers, degree-level labor force entrants, and Ph.D. students.

Programs to provide direct support for undergraduate student participation in research with faculty fare critical in convincing the most talented science and math undergraduates to purse graduate education. Student-faculty research groups nurture communities of learning that are demonstrably effective introducing good science education generally. The continued funding of the direct costs of these student-focused programs is an essential component of a national policy on science education. It is important to emphasize that general institutional funds are not a legitimate source of financing for summer programs that benefit individual students, so programs must cover full direct costs and not be on a matching basis.

It is worth noting that NSF grants to institutions for curriculum reform, facilities, and instrumentation, as well as grants for undergraduate research participation and experiences, cover nothing more than direct costs. Indeed, in the majority of cases, they require matching funds from the recipient institutions as well. Thus, the full amount of Federal dollars supporting these programs are subject to peer review and to Federal guidelines, both ex ante and ex post. In an era of increased attention to accountability, these are extremely important attributes of any Federal program, and they are both legitimate concerns and welcome components of a renewed partnership.

Creative use of resources to change the way students learn science cost-effectively must be a joint effort involving faculty, administrative staff, and governing boards. For starters, most traditional budgeting systems in the liberal arts colleges provide neither the accounting information nor the flexibility of decision-making to permit academic departments to know, much less to reallocate, the various costs associated with a department's activities. We need to generate the information so that faculty can consider options of trading off alternative uses or resources--space (including improvements), instrumentation, support and technical staff, faculty time, consumables, library resources, computer time, access charges, and even travel funds--as they consider how best to do science.

This will not be an easy process. The information is not readily available on a cost-accounting basis; lines of budgetary authority generally dictate that the total resources actually employed in instruction are controlled by several different administrative units. Building and raw space cannot costlessly be transformed into personnel, animals, or network access fees.

As math-science capabilities declined among high school graduates, the normal process of competition has led undergraduate institutions to lower admission requirements--an example of how market behavior defeats national objectives.

Yet, the insights of the continuous improvement movement suggest that we to focus on the entire process, not on the individual pieces. This means breaking down the administrative barriers that constrain budgets; this means involving many different people in the discussion of how best to deal with individual courses, course sequences, or major programs.

If we expect faculty to be creative, we collectively have to allow them the requisite flexibility. Similarly, if faculty expect resources to be made available, they must consider the full range of constraints within which institutions operate. Just as politicians and economists have to understand each other and participate jointly in the making of good public policy, so, too, faculty and administrators have to understand each other and participate jointly in thinking through the best, most efficient, most effective ways of helping our students learn to do science. This collaboration is even more essential now than it was in the past because of the rapid and continuous obsolescence of techniques and instrumentation, the continued explosion of knowledge, and the demonstrable concern of all of our publics with the costs of the higher education enterprise.

Institutional Considerations

How can we make institutional decision-making about budgets and financing work better? Let me suggest some areas we need to consider:

  • Among the challenges of efficient and effective science/math education are those of dealing with new and complex instrumentation and with the continuing specialization within disciplines. One theme of what works in the lab-rich, discovery-based approach is that of reducing "coverage" in introductory and intermediate courses and increasing the degree of real understanding of the scientific process. At the upper level, there is a danger of producing large numbers of specialized, low-enrollment courses that are expensive in faculty time and equipment. One of the many criticisms of higher education is the "Proliferation" of course specialties, and the related increase in costs per students. How do we arrive at the "lean, lab-rich curriculum" (in PKAL rhetoric) that is productive both from the perspective of economics and from student learning outcomes?

Faculty and their administrative colleagues need to explore options that will meet the objective of providing the latest instrumentation to keep faculty at the edge of research, and for students to use as they learn to "do science" in their discipline. One possibility to provide new, more specialized instrumentation when distances are small will have to be sharing between institutions.

  • Just as the possibility of sharing specialized equipment needs to be considered, so, increasingly does the sharing of faculty in particular specialized fields. Outside the sciences, for example, Carleton and St. Olaf--both in the same town--each employ an historian with a geographic specialty who teaches half-time at each college. Without this arrangement, each of us would either have excess staff--or would not be able to offer courses--in this area.

As important as sharing across institutional lines is the possibility of sharing within an institution. A key course in the Computer Science sequence at Carleton, for example, is taught by a physicist. Such sharing within an institution needs to be encouraged.

There are other economic implications for the increasing specialization of the disciplines. When faculty insist on teaching only their own specialties, especially when there is "ownership" of specialized courses or topics by individual faculty, it is very hard to keep costs--especially those of specialized courses--down. This is where faculty must take a broader perspective.

For example, in department where:

  • there are traditions of switch-hitting, so that courses by faculty on leave are taught by other regular faculty;
  • faculty take turns and assume responsibility for introductory courses in effort to meet the entire needs f the department and academic program of the institution;
  • departmental faculty feel a sense of ownership of the whole curriculum, and not just of his or her individual courses or specialties; and
  • departmental faculty as a whole design and implement advanced courses sot that there is synergy between student learning and faculty research
  • here one can bring average costs down by using all resources, including faculty time, more efficiently.

Budgeting processes and the incentive and reward systems have to be developed that favor the faculty and the departments who take the broader view.

  • At the institutional level, the key to success in enhancing the effectiveness of learning for students is a faculty that is continuously looking for the best, most efficient, and most effective ways to teach, and to encourage student learning. This means continuous learning for faculty members as well and requires institutional support appropriate to the needs of individual faculty at each career stage. Both my own experience and my observation of younger and older colleagues over the past thirty years make it clear that the agenda for an individual's professional growth is likely to change dramatically over time.

Colleges and universities, and their deans and presidents, provosts and department chairs need to have the flexibility to respond to different needs to different faculty at different stages in their careers, sometimes campus leaders have to take the initiative in such matters, and not be merely responsible. To some extent, individual needs can be accommodated by modest funds or time, to be used at the discretion of the faculty member.

More important, however, are funds that can be used as inducements, rewards, or as "enabling" grants to direct faculty development toward continuous improvement of the learning experience for students.

The dean-end, and I believe fundamentally demeaning, discussion of teaching versus research and the highly publicized disillusionment with teaching quality and effort at some universities, has clouded the issue of continuous learning and faculty development. "Research" that is part of the undergraduate learning process constitutes a particularly effective form of learning and of teaching. "Research" that rebuilds or restores of expands the intellectual capital of the faculty member is absolutely essential if each succeeding generation of students is to have the benefit of the latest in instrumentation, knowledge, and technique. "Research" can take many forms; tenure committees, deans, and presidents must take a broad view of what constitutes research if a lively, purposeful, student-oriented, and scientifically current faculty is to be built and sustained.

All of these--securing the sophisticated instrumentation, getting faculty to think beyond their own specialties and focus on the institutional objective of the most effective, most efficient, and most productive approach to teaching and learning--are considerations that must be a joint effort of faculties and senior administrators, with support from governing boards.

Perspectives of Those Who Pay

But partnerships need to address issues beyond the financing of individual programs. In any mixed economy, the questions of what is left to private decisions and what is affected by government incentives and government finances is critical. Economists like to examine where markets work, and where they fail to work, to achieve national objectives. The failure of the educational system to achieve the desired level of scientific literacy for the work force and of science capacity for research and development is, in an important respect, a "market failure": individual decisions by students and families, institutions and their faculties, and private companies have brought us to a widely perceived scientific deficit. The reasons are not surprising.

Education and training are long-term investments, and the payoff can be quite uncertain. When curricular choices becomes possible in middle and secondary schools, the information on payoffs is at best rudimentary. At the undergraduate level, the payoff to science and mathematics is not seen as great; traditional methods of teaching introductory courses are intended to select the few who can easily go on. The incentives to pursue math and science simply are not there. Therefore, the intrinsic interest needs to be stimulated or students won't make the effort.

To some extent institutions and accrediting bodies have overridden individual preferences by insisting on requirements in mathematics and science. but as math/science capabilities have declined among high school graduates, the normal process of competition has led undergraduate institutions to lower admission requirements--an example of how market behavior defeats national objectives.

There obviously have been exceptions to the expected institutional responses, and those exceptions have been concentrated in the science-active institutions. Their success has been due to the commitments of their faculties, administrations, and governing boards to allocate resources for staff, space, and instrumentation, to create curricula that excite and entice students and faculty to work together, and to aggressively recruit students interested in science and mathematics. (One way to lower average costs, if there is any element of excess capacity, is to increase the number of students in courses!)

External resources have helped institutions lean against the prevailing wind. The private foundations active in science have been of very substantial importance to many science-active undergraduate institutions. During several periods in the past 30 years, particularly in the 1960s and to a smaller extent in the past several years, NSF programs have provided external assistance as well. Without the external resources provided through competitive processes, even the majority of science-active colleges could not have accomplished what they have in terms of science productivity.

Case Study: Reengineering Science Education

Action Needed

I believe it is relatively easy to design programs--externally funded and focuses within institutions--that are focused, efficient, effective, and of modest cost that will address a series of serious national problems.

We can make a significant impact on the nation's capacity by a relatively modest investment in a key link in the production process for scientific training: the undergraduate level.

There will continue to be under-investment in math-science education at K-12 and at undergraduate levels unless incentives are provided both to increase that investment and to make it more productive. This is the case for Federal action, and for an active collaboration of the Federal Government, undergraduate institutions, research institutions (a prime "consumer" of baccalaureates from science-active colleges), and the national foundations.

A larger program of peer- reviewed, matching grants for facilities, instrumentation, and curriculum development is needed from the Federal government. Major resources would continue to come from the major private foundations who support science and mathematics. Both external sources would join with the larger pool of core fund coming directly from the institutions and their constituents. All these resources could be managed according to the principles outlined above.

We know a great deal about what kinds of approaches are effective--and efficient-- at the undergraduate level. While they often require higher levels of investment in facilities and instrumentation that traditional approaches, they are also more productive. We should move ahead with a national partnership of the Federal government, the major private foundations, and the science-active institutions to address the problem and implement the solution.

Undertaking such a joint effort is perhaps the most essential and challenging of all the tasks that face us.

If we expect faculty to be creative, we collectively have to allow them the requisite flexibility. Similarly, if faculty expect resources to be made available, they must consider the full range of constraints within which institutions operate.