by Carol Twigg
A Report from the Broadmoor Roundtable, Colorado Springs, Colorado, July 24-25, 1996
Educom's National Learning Infrastructure Initiative seeks to create a shared understanding of how information technology can transform the collegiate learning environment to improve quality, contain costs and increase access. We believe that high-quality instructional software is a necessary ingredient for creating more productive learning environments. We also believe that software development strategies depending on the efforts of individual faculty members will not result in a sustainable, scalable body of materials, even when such efforts are generously supported by corporations, government agencies and private foundations. Instead a robust commercial market for interactive learning materials is needed.
Higher education institutions want and need high-quality solutions to the cost, access, and quality problems they face; various commercial entities have an interest in providing those solutions. But we have not yet seen the emergence of a viable market for interactive learning materials in higher education as we are beginning to see in the K-12 setting. While many agree about the desirability of creating such a market, there is neither agreement about the impediments to its creation nor consensus as to how to accomplish it.
On July 2425, 1996, Educom convened a roundtable consisting of 20 knowledgeable representatives of universities, publishing houses, hardware and software companies, and the federal government to discuss those issues and to contribute ideas for creative solutions. Meeting at the Broadmoor Hotel in Colorado Springs, Colorado, this roundtable helped us test an initial statement of the issues and establish a sense of what was possible and what was not. While it is not possible to convey each of the many contributions made during this fruitful discussion, we have tried to capture the most pertinent. This white paper is the result.
Lawrence B. Coleman
Professor of Physics
Chair of the Academic
Senate
University of California-Davis
Carole Cotton
President
CCA Consulting
Robert L. Culver
Vice President for Finance
Northeastern University
Steven L. Epstein
Vice President, Director of Multimedia
Development
Simon & Schuster
Bernard Gifford
Founder, Chair, and Chief Instructional
Officer
Academic Systems
William H. Graves
Chief Information Officer (Interim)
University of
North Carolina-Chapel Hill
Robert C. Heterick, Jr.
President
Educom
Thomas Kalil
Senior Director
National Economic Council
James Lichtenberg
Vice President, Higher Education Division
Association
of American Publishers
Mark Luetzelschwab
Director of Technology
Lenox Softworks
Mark A. Luker
Program Director, NSFNet
National Science
Foundation
Chief Information Officer
University of Wisconsin-Madison
Patricia H. Montgomery
Senior Director, Worldwide Education
Markets
Apple Computer, Inc.
Michael M. Roberts
Vice President
Educom
Carol A. Twigg
Vice President
Educom
Hal R. Varian
Dean, Information Management and Systems
University of
California-Berkeley
Richard P. West
Vice Chancellor for Business and Finance
The California
State University
Jack M. Wilson
Dean, Undergraduate and Continuing Education
Rensselaer
Polytechnic Institute
Patricia Bartscherer
Administrative Assistant
Educom
Carolyn Jarmon
Director, Graduate Program
SUNY Empire State College
Elizabeth Lawrence
Associate Professor
SUNY Empire State College
We begin with an important assumption. In the words of Bruce Johnstone, former chancellor of the State University of New York, "American higher education must become more productive. That is, our universities and colleges-public and private, selective and open access, research universities and community colleges-must produce demonstrably more education, research, and training for the resources requested from students, parents, and taxpayers. Increasing productivity is, at least arguably, an imperative of every important sector of our economy. For higher education-whose unit costs have consistently increased at rates exceeding even its own wage and salary increases, whose costs to the student and parent now exceed the reach of all but the most affluent families, and for which a major increase in state or federal tax support remains remote-the need for higher productivity is compelling and urgent."(1)
The pressures for improving productivity stem from factors like these:
The most recent reiteration of this view is contained in the vision statement for the Western Governors University: "All western governors are feeling the press of increased demand on their state systems of postsecondary education. All recognize that the strength and well-being of both their states and the nation depend heavily on a postsecondary education system that is visibly aligned with the needs of a transforming economy and society. At the same time, the states' capacity to respond to those challenges is severely constrained by limited resources and the inflexibility and high costs of traditional educational practices and by outdated institutional and public policies. . . . Affordable, accessible higher education-that is the vision of a western virtual university."(3)
The public is not simply concerned about cost; it is also equally concerned about quality. Some institutional chief financial officers go so far as to suggest that education as an industry may be on the verge of being sued. If the college catalog is the contract between the student and the institution-a representation of the goods and services provided to the consumer in return for his or her tuition-do we face product liability and negligence issues if we fail to mention the following?
In response, the public is increasingly asking: What is the value received for the dollars we invest in higher education?
Information Technology: Part of the Problem or Part of the Solution?
We take as a starting point the supposition that information technology has the potential to be part of the solution to higher education's productivity problem (rather than being part of the problem as a contributor to increased costs.) More specifically, Educom believes that the existence of a substantial body of instructional software is a key component of that solution (and, conversely, that its absence is a key contributor to the problem).
By instructional software we mean software that instructs students in a particular subject matter such as the PLATO efforts of the early 1980s, the University of Illinois' introductory chemistry series, Jim Noblitt's foreign language multimedia materials, Academic Systems' college algebra courseware, and others.
Our focus on instructional software is not intended to discount the value of other forms of computer-mediated materials and techniques, such as computer-mediated communications, networked library and learning materials, and tools to assist learners, authors, and instructors. But such forms by themselves have limited usefulness in addressing the productivity problem. Nor are we unmindful that the context-the learning environment-in which instructional software is used has a substantial impact on the quality and cost of the learning experience. In other words, the existence of software is necessary but not sufficient to address the problem.
Furthermore, we recognize that appropriate use of instructional software can lead to higher quality, more productive-i.e., more cost effective-learning environments in other settings, such as K-12 schools, corporate training programs, military education programs, and others. But our discussion at the Broadmoor focused on higher education, both on and off campus.
As a basis for our discussion at Broadmoor, we asserted the following five propositions:
Those five propositions, as elaborated by the Broadmoor discussion, are
discussed in detail in the following section.
Proposition 1. Instructional software is essential if we are to solve the productivity problem because:
Higher education costs to the consumer are rising at two to three times the rise in the consumer price index. They have been doing so for nearly two decades in spite of significant efforts at cost cutting on the part of our colleges and universities. One apparent impact of our cost cutting efforts has been to make salaries an even larger portion of the operating budget-in the 80 percent range for many institutions. This compares to 25 to 40 percent for most of the manufacturing industries and less than 60 percent for the service industries. Contrary to popular belief, the real money in college budgets is not in the administration but in faculty positions. Significant savings can only be realized by reducing personnel costs in relation to the learning outcomes produced.(5)
Real labor costs tend to rise with economy-wide productivity gains (say 2 percent per year, on average), whereas technology-based costs tend to decline due to learning-curve effects, scale economies in production, and continued innovation. Increasing technology's share of cost will reduce overall cost growth until the rate differential reduces technology's share to the point where labor again dominates. By this time, however, total cost will be lower than it would have been without the injection of technology. If the real cost of technology were to decline at a 25 percent annual rate, after ten years the alternative scenario would cost about 12 percent less than the baseline. If the rate of decline is only 10 percent, the savings ten years out would have passed 9 percent and still be rising. Given the differential growth rates of labor and technology, one can expect positive long-term returns on investment even when returns are negligible during the first few years.(6)
During the past decade or so, a variety of technology-mediated learning environments have emerged. So far, most IT-based improvements have involved doing more with more. Some increase access (such as distance learning and networked resources) and some improve quality (such as supplemental multimedia, interactive learning applications) but few control costs. Instead, most of those applications "bolt on" to the traditional classroom structure, thereby adding to the cost of instruction.(7)
With labor-especially faculty labor-considered to be fixed, IT becomes a quality-enhancing add-on. This fits the faculty culture but suffers from at least two serious deficiencies. First, scarcity of add-on funding limits IT's rate of adoption. While colleges and universities might like to pour money into more-with-more productivity enhancement, most are not in a position to do so. More fundamentally, the more-with-more approach does not address the academy's need for cost containment. One can imagine a scenario where widespread IT add-ons produce a situation like that found in medicine, where technological breakthroughs produce a spending race that eventually threatens the system's affordability. Tight financial circumstances currently inhibit such scenarios, but even if today's constraints could be relaxed, more-with-more productivity growth would eventually encounter new financial limits.(8)
The more you replicate the traditional campus model as you design mediated programs, the more your operating costs will resemble or exceed traditional campus costs-for example, instructor-led models such as televised classes or computer conference-based courses that rely on the same student/faculty "contact" as traditional models.(9)
Controlling costs means reducing the direct, personal intervention of faculty where possible in the teaching and learning process. By lessening the need for direct faculty intervention in the learning process and by increasing the ability of students to find and use learning materials on their own, we can create more cost effective instruction. One should expect that additional students could be accommodated at lower cost with technology than with traditional teaching methods. What you may ideally want is a combination of preprepared, multimedia materials that reduce the contact time between students and teachers and some interactive time when the faculty member can concentrate on the things that she is best at doing: handling interaction with and among the learners.(10)
Example: Many in higher education believe that introductory courses that depend heavily on teaching assistants are relatively cheap to offer. But those who have examined this issue more closely have found that not to be the case. As part of the reengineering process at Rensselaer Polytechnic Institute (RPI), Jack Wilson and his colleagues did full costing of their introductory courses. It was clear that the biggest part of their instructional cost was personnel, both faculty and support staff cost. Their redesign, and subsequent savings, was based on substituting some capital in the form of hardware and software for labor in order to improve productivity. They also found that they could teach at least equivalently in one-third less time. In chemistry, the savings was in faculty; in calculus, the savings was in TAs; in physics, the savings was in both. Some courses broke even in one semester, some in two years.(11)
Example: Furthermore, while the input costs may appear to be low on the surface, the costs of high attrition to the students and to the institution are significant. As Bob Culver, vice president for finance at Northeastern University, pointed out, from a financial standpoint, the first two years at many urban institutions are brutal. Reducing the 28 percent loss between the first and second year would be "money in your pocket." If educational software can affect retention, it would be a worthwhile investment. The economic model developed by Academic Systems produces savings that are realized both in increased class size-i.e., gains in the number of students that can be served-and in higher retention and greater student achievement-for example, a 68 percent completion rate in this model compared to a 50 percent completion rate in traditional classes.
Proposition 2. Targeting areas of high student demand will have the greatest likelihood of success because:
Example: Ron Bleed, vice chancellor of information technologies at the Maricopa Community College District, decided to do an analysis of the undergraduate academic program to find out where students are concentrated. Maricopa, one of the nation's largest institutions of higher education with a student enrollment of 90,000, offers 2,000 distinct courses. Interestingly, Bleed discovered that 44 percent of the student enrollment at Maricopa is concentrated in just 25 courses. Put another way, 1 percent of Maricopa's courses generates nearly half of its total enrollment!
The 25 course titles include introductory studies in English, mathematics, psychology, sociology, economics, accounting, biology, and chemistry. Maricopa's numbers are typical of all community colleges and probably not that far off for most four-year institutions.
Bleed also discovered that the satisfactory completion rate for these courses is 64 percent. A rule-of-thumb estimate of direct instructional costs yields a figure of about $38 million per year not including the costs of facilities, overhead or student support services. This means that more than $13 million per year is spent on unsuccessful attempts at learning.(12)
Example: At RPI, Jack Wilson and his colleagues targeted the large enrollment introductory courses-physics, calculus, chemistry, biology, and several engineering courses-as their first target of opportunity for reengineering the curriculum. They found that 10 courses could make a huge difference in their institution and that "another 10 courses on top of that would change the whole way the institution was put together."
IT has strong potential to increase learning productivity in the areas of codified knowledge and algorithmic skills. The areas that can profit most from IT-based strategies are those subjects that have a high volume of students, a more-or-less standardized curriculum, those whose outcomes can be most easily delineated, and over whose content faculty are less possessive. Examples of good target subjects include remedial and basic math, general education courses, and composition courses. In those specific areas, the implication is that IT should supplement human instructors whenever possible-human intervention should be oriented mainly towards making the advantages of IT accessible to all learners. In the case of math remediation, for example, that might mean monitoring student motivation and providing support at critical junctures to ensure that a student completes the program. Wherever a significant portion of a curriculum includes noncodified, nonalgorithmic knowledge, however, faculty maintain their historic advantage.(13)
So much of the discussion about uses of instructional technology in higher education centers around what is called the "adoption" problem-in other words, how are we going to get "them" to use "it." For those institutions concerned about productivity, the "One Percent Solution" offers a strategy to focus institutional attention and to clarify institutional thinking.
It is difficult to change an entire institution, an entire culture. But if we decide to increase the learning productivity of not 2,000 courses but of a mere 25-approximately one percent of the total-we can make a substantial contribution to controlling institutional costs. Should our institutions develop a strategy for helping "the faculty" or for helping the faculty in those 25 courses who teach 44 percent of the students? Should we have a strategy for acquiring multimedia materials for all courses regardless of student enrollment or for the 25 courses? Should we be thinking about redesigning all courses by integrating technology and new pedagogical techniques or should we think about reengineering 25 courses to produce the most effective learning experiences possible for students?
And suppose we set as our goal upgrading the quality of those courses to eliminate attrition and to strengthen substantially the foundation that successful students take to their future courses. The $13 million per year currently wasted on students' not learning in our Maricopa example would provide a formidable financial base from which to launch this effort. And, of course, if Maricopa joined with other institutions who spend the same $13 million on identical instructional problems to codevelop or purchase course materials, we'd be talking real money.(14)
Participants pointed out other advantages of the "One Percent Solution." Innovation in those subject areas carries a relatively low risk. In many large institutions, those courses are not "owned" by individual faculty members in the same way as are advanced level courses. They are frequently not "owned" by any curriculum in particular but rather serve as feeders to multiple programs.
Those subject areas tend to spill backward into the high school years and
forward into areas of lifelong learning, thus widening the demand for such
products beyond the college years. They offer the possibility of affecting a
very large number of students to amortize investment.
Proposition 3. Colleges and universities cannot create the requisite software on their own because:
Strategies depending on the efforts of colleges and universities to encourage individual faculty members to develop instructional software will not result in a sustainable, scalable body of materials. Despite the potential offered by such materials, many in our community have noted the failure of most institutions-even those who have worked hard to create them-to integrate their use into their academic programs. In the past, such efforts have been generously supported by government agencies, private foundations and by various corporate entities. But the incentives offered by those grant programs are microscopic, and they are all in the wrong direction. The weakness of focusing on individual faculty members as a strategy for change can be demonstrated by assessing three approaches to stimulating new learning environments at the national level.(15)
In the 1980s, IBM spent millions funding more than 3,000 individual faculty projects in its Advanced Education Projects. Similarly, the National Science Foundation curriculum reform program has spent millions of dollars on awards to individual faculty members to improve individual courses at individual institutions. NSF grants are designed to support "innovation" rather than commercialization; the NSF will only fund you if you do something new. For example, literally hundreds of algebra innovations have been funded, but they all tend to be micro-scale. They do not scale up beyond an institution and, frequently, they do not even scale up in one institution. From the beginning, those projects are misdesigned because they are planned for a pilot of 10 to 15 students. The faculty fail to anticipate what it would mean if you have 1,000 students to deal with; they ignore the fact that a model that will work with 15 students may or may not work with a thousand.
Mark Luker, of the University of Wisconsin-Madison, also added another explanation as to why attempts to get faculty to develop software within individual universities have not been successful. What has typically happened is that the faculty are provided with only a minimum amount of preparation. They think they can develop software by using various kinds of authoring tools which are supposed to be easy to use. But the development process inevitably ends up being a full-fledged programming job. Software development is an iterative process: as projects progress, additional improvements are always recommended and upgrades are always needed which, in turn, leads to more programming. And then the grant funding runs out.
Bernie Gifford spent from 1989 to 1992 as vice president for education at Apple Computer. He was convinced that the way to build software was to subsidize faculty members, and he spent a lot of money trying to do just that. Gifford cited one of many failed experiments. He approached six well-known physics faculty members to create 30 credit hours of instruction. At the end of 18 months, he had six separate programs with six different nomenclatures, six different products that did not fit together. He had to ask Apple systems engineers to "glue" them together. Gifford's conclusion: It is very difficult to build coherent courseware by supporting individual faculty members because it requires enormous discipline to submit to a common mission. Software development requires tremendous coordination.
IBM, Apple, and the NSF meant well, but all of their programs have failed to achieve results beyond the individual classroom. A wealth of experience shows that the attempt to "add-on" innovations with external support, and without internal structural change-especially commitment of resources in the educational system's core budget-has been almost totally unsuccessful. The most surefire way to tell whether an innovation is for real or is artificial is to look at its funding. Unless an innovation is paid for directly by those who stand to benefit from it, its chances to flourish are dubious at best.(16)
Colleges and universities are organized, financed, and staffed to provide
educational services and conduct research. They are nonprofit, tax-exempt
institutions. Public institutions in particular have specific mandates for
service as a condition of their existence. Were they to reconceptualize their
mission to become producers of learning materials-i.e., to go into business-they
would need to transform their basic organizational and financial structures, and
forego their special status, in order to compete with the private sector. Such a
transformation appears both unlikely and undesirable.
Proposition 4. What is needed is "self-sustained commercialization."
We need to create a robust commercial market for college-level instructional software. We need the direct engagement of those whose business is the development, production, distribution, and marketing of educational products-specifically, the publishing and digital industries.
The proper design and distribution of public and private roles in the innovation process will be debated endlessly. What R&D specialists uniformly agree about, however, is that commercialization is the minimum measure of success. Inventions that cannot ultimately be produced, sold, bought, and adopted in practical use, without coercion, don't count as successful innovations.
Within the broad context of innovation is the nitty-gritty process of inventing and introducing new technologies. This process includes basic research, transfer of the discoveries of basic research into the invention and development of new technologies, application of technologies to provide practical solutions to real problems, and the development of efficient production and marketing systems to make the new technologies affordable and accessible.
Without the expectation of profit, private investors and entrepreneurs will
not make the investments and take the risks that creating and marketing a new
technology entails.(17)
Proposition 5. Market success depends on instructional software that is disaggregated, disintermediated, differentiated, and diffuse.
As Paul Evan Peters, of the Coalition for Networked Information, has pointed out, researchers want to use network technologies and techniques to obtain journal articles without having to acquire the associated journal titles, to obtain monograph and report chapters without having to acquire the associated monograph or report titles, and to otherwise reach into the usual packages by which research publications and communications are produced, distributed, and used so as to pluck out just the information they want at just the time they want it. Readers want to pulverize the packages that publishers, librarians, and even authors have been using to produce and distribute research information. Information products and services are being repackaged, therefore, to cater to this desire. The growth of networks and networked information seem to be having this effect on markets in general.(18)
So too do teachers and learners wish to pulverize-or disaggregate-the one-size-fits-all packages of the courses and the textbooks that have characterized learning environments for the past several hundred years. Instructional software needs to be modularized such that users have the ability to pick and choose and use as needed in diverse learning environments. Disaggregation enables "mass customization" where learning materials may be customized by the adopter.
Example: The idea behind RPI's CUPLE physics project was to decompose physics into 300 different concepts that could be collected together in various ways-or reaggregated-into a range of courses. The designers wanted to provide a seamless environment for students from grade school to graduate school. Students may be sitting in graduate school with a grade-school level understanding of a particular concept; they really need to be able to go back and look at the underlying material there. Or they might be in a freshman class and actually be prepared for a graduate-school understanding of some of the material; they should be able to go forward in that particular area.
Disaggregation means to pull it all apart into its separate parts, but it also means the software has to understand the relationship of each part to every other part. As Patt Montgomery of Apple Computer said, "You don't want to create a Frankenstein monster." Two kinds of coordination are needed: (1) technology standards enabling building and sharing-interoperability-of the component parts, and (2) some kind of editorial process to coordinate the final product.
All of our modern technologies with rapid diffusion rates (high consumer acceptance) have been personal and disintermediated. There is a significant lesson in the relatively slow diffusion rate of telephony when compared to electric power, radio, and television. The telephone at its introduction and for its first 30 or more years required the mediation of a human operator. Personal mediation is moving to the background while "user friendly" and simplified but "smart" automated mediation moves to the user interface.(19)
Instructional software needs to be capable of being used by learners to a large degree without human mediation. This is not to suggest that instructional software is best used by solitary students communicating solely with their computers. Most educators agree that learning seems to be a social activity. But that doesn't mean that students must go to an ivy-covered building with brick walls; there are other ways to construct a social learning environment. The intermediary can often be another student. The same software should have the capability to be used in a traditional classroom setting or by individuals studying at their own pace, off campus.
The incorporation of such elements as diagnostic self-assessments of beginning skill level and learning style, mastery tests to assess learning achieved, and such into instructional software contribute to disintermediation. Rather than relying entirely on faculty members and academic support personnel to provide those services, such tasks can be handled by the software. The bottom line is that in order to have an impact on productivity improvement, instructional software must be capable of "off-loading" a significant number of instructional tasks traditionally performed by personnel.
Software that is differentiated can be combined and used in different ways for more than one purpose to meet different needs (for example, in high school classes, community colleges, introductory courses in universities, corporate, government and military training settings).
Diffusion implies that the more something is used, the more value it has (such as the telephone). As Perelman puts it, "A few television sets are a novelty; one hundred million televisions define a way of life."(20) Software that is diffuse seeks to be used by lots of people to drive the price down-in other words, is scalable. For example, it is designed for higher education as an industry rather than for a particular class or institution. It costs the same to build a software module for one person as for a million people; therefore, in order to be profitable, software must be scalable.
Such software makes use of those technologies that have widespread use, like
the personal computer or the television or the Internet, rather than depending
on the creation of new or specialized technologies.
Conclusion
If we more or less accept those propositions, we recognize that we are facing a familiar chicken-egg problem: Higher education wants and needs high-quality solutions to the instructional problems we face and various commercial entities have an interest in providing those solutions. But we have not yet seen the emergence of a viable market for interactive learning materials in higher education as we are beginning to see in the K-12 environment. While many agree about the desirability of creating such a market, there is neither agreement about the impediments to its creation nor consensus as to how to accomplish it.
The following questions were posed to the group as a framework for the subsequent discussion:
I'm sure you know the story of a student who tells his professor, "Look, there's a $20.00 bill on the floor." And the economics professor says, "Nonsense, if that were really a $20.00 bill, someone would have picked it up already." -Tom Kalil, senior director to the National Economic Council
Tom Kalil's story about the student and the economics professor illustrates the nature of our discussion at the Broadmoor. Participants from the commercial sector as well as some from universities expressed the traditional point of view that demand drives supply. And the reason for the lack of supply is because the commercial sector finds it hard to see that a demand for software products exists. As Patt Montgomery said, "We can't seem to get the institutions to create the market."
Others from the universities argued that we face a supply problem having to do with the availability of collegiate instructional software-or rather the lack of it. In the words of Larry Coleman, professor at the University of California-Davis, "More faculty would adopt it if they thought it was there." For most academics, the concept of instructional software is an abstract one. It's hard for them to "demand" something they have never seen or do not understand. Good examples are needed to pique their interest and stimulate their thinking. Some suggested that this phenomenon may be true of information products and services in general. Consider the introduction of word processing. The nation's secretaries were not sitting around saying to themselves, what I really need is a microcomputer! And where was the demand for the Apple IIE in the early 1980s?
The availability of comprehensive software is a necessary precondition to
changing the learning environment in particular disciplines and to increasing
productivity in higher education in general. One could be completely committed
to changing the learning environment and making it more productive by using
information technology, but if redesign is attempted in a field with no
software, the problem becomes much more difficult. It may seem rather obvious,
but in the two examples of successful adoption described below, there was
something for the institutions and faculty to adopt.
What Do We Mean by Supply? Software as Value Add
It became apparent in our discussion that we have neither a shared understanding of what is needed nor a common vocabulary to describe it. Publishers say, "We're producing software," but what they mean is that they are including a CD-ROM that illustrates a few concepts as a supplement to a textbook. The academic world says, "There's not much out there," and by that they mean that there are few products that meet the characteristics of desirable software (disaggregated, disintermediated, diffuse and differentiated) described in Part I.
What is the difference between a textbook and software? Participants felt that many faculty would adopt instructional software if it allows them to do something the book will not or if it enables them to do something they cannot currently do-in other words, if it adds value to the textbook. The value add could be in helping students learn better or increasing productivity by diminishing the amount of time faculty spend on instruction. They would tend not to adopt a current textbook simply because it is on a CD-ROM.
Pointing out that books and video tapes are forms of instructional technology, Hal Varian, dean at the University of California-Berkeley, asked, "What is it that executable code is bringing to table?" He offered the following analysis of teaching, textbooks, and instructional software to suggest an answer:
"When I teach, I ask students questions and see if they can answer them. This is a repetitive process; my presumption is the greater the feedback, the better the teaching. When I wrote an undergraduate textbook, I got some advice from a colleague. He said, 'assemble the problems, homework exercises and exam questions that you want students to be able to complete. Then write a book that teaches them how to do this. Build in a lot of testing materials since testing helps students see what they have learned.' So I built a test bank with scoring, a work book and other evaluative mechanisms into the design of the materials and in the actual text.
"The question is, can I economize on this kind of feedback in a much more efficient way than when I do it in a class or when I do it in a textbook? The computer can store and generate a test bank consisting of an infinite number of examples and questions for students to review and answer, all different and all at a comparable level of difficulty. Information technology facilitates building group memory since a test bank is a form of group memory, and group memory leads to greater consistency. Question sets both save faculty time and serve as an incentive to better student learning."
If we extrapolate from this example, we can see that the value of
instructional software is directly related to its ability to add value to the
instructional process. Electronic page turning will inevitably be met with a
ho-hum. But the more instructional software captures and improves upon aspects
of faculty mediation, the more it will be valued.
What Do We Mean by Demand? A Changed Learning Environment.
It also became clear that we are not just talking about bolting instructional software on to traditional classroom models; rather, we are talking about a different delivery model or a change in the learning environment. As Patt Montgomery said, "This is not just a content issue. We really need to look at the whole teaching paradigm." In this sense, we are talking about an adoption problem.
The group considered the question, What is the pedagogical promise of instructional technology? Stimulated by Jack Wilson's initial explication, participants constructed the following list. Note the emphasis on words like "provides," "enables," and "allows."
Participants observed that most of those items are not inherent in instructional technology; rather they are part of pedagogy. What is the value add of instructional technology? It can accomplish those pedagogical goals in a more cost-effective and higher-quality manner. As Hal Varian said, "One-on-one might be better, but we can't afford it."
To illustrate this point, Jack Wilson described how a faculty member at RPI tried to demonstrate that he could teach an introductory calculus course using all of those positive pedagogical techniques but with no technology. After two years, he was burned out and begging for a sabbatical. In addition, there was a high institutional cost for the service of the four TA's needed to grade his worksheets. And, furthermore, his model wasn't diffused; no one adopted it.
Barriers to Adoption in the Traditional Collegiate Environment
Those who see the market development problem as a demand problem tend to
focus on "adoption" and much of the Broadmoor discussion centered around the
adoption problem. The following examination of some of the barriers to adoption
assumes that the software exists to be adopted.
The Infrastructure: "We must build the cake before we can provide icing."
Several participants noted that the technological infrastructure is a necessary precondition to support software applications. As Simon & Schuster's Steve Epstein pointed out, the penetration of CD-ROM drives has enabled publishers to begin to include CD-ROMs with textbooks. "A few years ago, CD-ROMs were not being made because they were not profitable." Richard West, vice chancellor for business and finance at The California State University (CSU), noted that "even though we have a pretty good level of penetration of network technology at many of our institutions, we really don't have it at a level to assure developers that there could be widespread adoption at the local level."
While some expressed concern about the cost of developing and supporting the
infrastructure, others noted that campuses are going to be wired with or without
new instructional models. Institutions have already made-and will continue to
make-huge investments in the infrastructure; the challenge is to use it
effectively. Many publishers feel that the absence of machines is no longer a
significant obstacle. Furthermore, as Mark Luker pointed out, we are living in a
transition time for infrastructure. Until now, institutions have felt compelled
to "provide" the infrastructure by building computer labs on campus. But
increasingly, students are buying the technology just as they buy textbooks. For
example, about half of entering freshmen at the University of Wisconsin-Madison
arrive on campus already owning computers, mostly laptops, many with CD-ROM
drives. The availability of less expensive, "network" computers promises to
accelerate this trend.
Academic Culture: Teaching vs. Research
A major obstacle to adoption cited by many participants is the prevailing academic culture, which is characterized-especially in the university-by a greater interest in research than in teaching and the predominance of the classroom paradigm. As Jack Wilson put it, "How do we change the culture at the university so that we have a willing, receptive market?"
Richard West commented, "I do think that there is a private market to draw capital in once there is a willingness for the institution to innovate. Are we really going to change the classroom paradigm, and are we creating enough incentives to do that? At the California State University, for the past three or four years, we have made no secret about the fact that we would like to do more in this area."
Some voiced the belief that faculty members are unwilling to change their pedagogical orientation; that they have too many years invested in the status quo. In that regard, tenure is seen as working against an emphasis on teaching. But others hotly contested that view, pointing out that faculty members are not rewarded for teaching. "Who entrenched those faculty?" they asked. The faculty are doing exactly what they have been asked to do: bring in external funding through research or educational grants. The system is designed to reward entrepreneurs, and faculty are responding to the prevailing reward system. As administrators, we have to focus on our responsibilities in setting up a system that asks faculty members to behave in a particular way and then complaining when they do exactly what we tell them to do.
The universities we have are the ones we wanted to have when we established
them. Now the community feels we need to have a new kind of university, but all
of our incentives are built on old models. How do we have a different kind of
university? We need to redefine rewards, to change the incentive structure for
faculty within institutions. And we need to change those incentive systems
external to the university that affect faculty behavior. The NSF, for example,
has been a major player in creating the university system as it is now. We need
to use our influence in this regard as well.
Academic Culture: Who's in Charge?
Decision-making structures on campus-or the lack of them-were cited as another obstacle to adoption. As one publishing executive put it, "If I come out with a whole new way of approaching a subject, the decision to adopt is frequently made by a faculty committee. The committee system tends to support the old way. Even if a new book would be better than the current text, it might require faculty to change the way they teach, so they won't adopt the new book even though the new book is better." Instructional software simply compounds the problem. Again, there is no incentive for change.
Bernie Gifford offered the following example. The math department at a certain community college was a target of anger because of the high failure rate in mathematics courses. The nursing department went so far as to seek out a math department at another community college to teach the nursing students. Academic Systems made a presentation to a faculty committee about adopting their mathematics software. The vote was 17 to 2 in favor of adopting the Academic Systems software. But because 2 out of 19 faculty members voted against it, the contract was denied. Academic Systems even offered to give the courseware to them for free, but they refused. No one at the community college appears to be in a position to say, "This is not working."
What Characterizes Successful Adoption?
There are, however, examples of successful adoption of instructional software, even on very traditional campuses. Participants considered two such examples-the RPI studio courses and the adoption of Academic Systems software at several CSU campuses-to ascertain from each what the critical success factors were. In both examples, four factors stand out. First, the campus administration provided the necessary infrastructure development. Second, the academic areas, remedial or introductory courses, were those in which the faculty were not heavily invested. Third, the faculty could see positive results from the new learning environments in terms of greater student success. And fourth, faculty feel the results in terms of reduced workload. In other words, greater productivity leads to less work for the faculty!
Unlike grant-based endeavors, Academic Systems, a private corporation, wants to make a profit so they work hard to produce a product and ensure its adoption. Academic Systems has initially targeted the remedial market, an easy entry point with low faculty resistance to acceptance. In every case of campus adoption, the campus president has supported the model and provided sponsorship-this is a critical success factor. But this was not a top-down decision: faculty reviewed and selected the materials. Finally, after gaining some experience with the software, faculty can point to higher student achievement and better retention. This in turn can lead the way to adoption of courses in other content areas.
At RPI, everyone-faculty and administrators alike-accepted the fact that money needed to be saved and thus a climate for adoption was established. The redesign of introductory courses had increased productivity as a goal. Prior to the introduction of the studio courses, faculty were not very involved in introductory courses. Now there isn't one research faculty member who hasn't said, "This is so much better." The physics department voted to abandon the traditional course even before the new labs were completed; the engineering faculty are also in favor of the replacement. Why? Because faculty believe that the studio courses provide a better environment for both faculty and students. Because there are fewer instructional events to be staffed, workload for those involved has been reduced. There used to be 50 to 70 events; now there are 12 to 14 events. There used to be three kinds of rooms; now the course takes place in one kind of room. Making the faculty more productive means making faculty's lives better.
The implications of those examples are clear. In both cases, the adoption decision was an institutional one. The decision was not left up to individual faculty members or to committees. There was general agreement among participants that if adoption must take place at the individual professorial level, in the words of one participant, "we're doomed."
In today's academic culture, responsibility for content rests with the faculty. But a shift is occurring in higher education where increasingly the institution is, in a sense, buying content which it can control. We are seeing tremendous growth in the number of courses taught by untenured, part-time, adjunct, or temporary instructors. In addition, the decision-making structure of many community colleges, continuing education and distance learning programs is institution-based rather than faculty-based. Those environments may provide more fertile ground for the adoption of instructional software.
Alternatives to the Traditional University Market
The discussion then turned to the following topic. If we conclude that it is going to be really hard to make progress on stimulating an initial market for instructional software in the academic sector because of all the cultural factors that we have identified, there may be a lot less resistance in other venues. As Mark Luker pointed out, we do not need to fix all the problems of higher education to create instructional software for higher education. We need to think about the demand for collegiate software as being broader than the traditional research universities, broader than the campus itself. Rather than starting with what may be the least receptive audience-university professors, it may make more sense to start with a market that is more receptive.
Bernie Gifford commented that there are literally millions of adults-including those who have graduated from college, senior citizens and others- who would like to take advantage of the knowledge base of higher education. But they are not prepared to secure that instruction by coming to campus according to a relatively rigid schedule. There is plenty of business out there, and the revenue potential is enormous.
Products created for the nontraditional market could be transferred to the
more traditional one. The curriculum overlap is much greater than we think, and
the technology overlap is almost 100 percent. As Educom president Robert C.
Heterick, Jr. said, "What we're trying to do here is find a way to get our nose
under the tent." The supply-demand problems in the right market such as the
life-long learning arena are quite different than in the traditional collegiate
environment.
In their recently published monograph, Transforming Higher Education: A Vision for Learning in the 21st Century, Michael Dolence and Donald Norris present a view of higher education as a growth market, unable to keep pace with rising demand. They project that by the year 2000 the demand for postsecondary education will increase by 20 million full-time enrollments (FTE) in the U.S. and by 100+ million worldwide.(21) This anticipated growth will come not from traditional, on-campus students but from those already in the workforce: those whose knowledge needs continual updating (think engineers, medical personnel, computer programmers) or those who need to acquire new skills (think 40,000 laid-off AT&T employees).
Echoing this point of view, the Western Governors Association cites "the press of increased demand on their state systems of postsecondary education" as the driving force behind its efforts to create a virtual university. But this demand is not coming from on-campus students. Here is one of the examples the governors use in their vision statement to illustrate demand. A CEO of a small software company discovers his programmers need proficiency in C++ programming, but the nearest classroom is three hours away. He learns that other software companies are experiencing similar training challenges. Together, they approach the Western Governors University and develop a set of expected competencies and assessment approaches for certifying C++ programmers. Using these established expectations, the virtual university launches a competitive grants process for courseware development.
The governors' preferred solution to meeting demand for higher education is
not to build new universities but to "expand the marketplace for instructional
materials, course-ware, and programs utilizing advanced technology that have
already been devised by public and private sector providers, and to foster
interstate and public-private cooperation in the development of new
instructional materials that respond to unmet needs in the region." The ideas
behind the Western Governors University offer new ways for software developers
to think about providing educational services to off-campus students, including
competing with existing institutions to sell their products directly to the
customer. Ask employers what skills and competencies they need and develop
software to meet those needs.(22)
Marketing Directly to Individuals
Another way to go around the logjam of academic culture entirely would be to sell instructional software directly to individuals, the end consumer who may not even be enrolled in a course. One could go directly to the individual and say, here is a product that you can immerse yourself in to learn economics or Spanish or C++ programming-and you can buy it for a reasonable price. Is there a big enough market to make this a commercial venture? Using English as a Second Language as an example, participants speculated about how many people at any one time are engaged in ESL programs. Perhaps a million people? One would only need a small percentage of them to produce a profitable product.
Hal Varian questioned the plausibility of this idea, citing the difference
between diet books and fat farms. People need external discipline. Classes
provide feedback mechanisms and assessment, the additional supports that make
people stick to a diet. Others countered that (a) diet books are a huge market
in themselves, and (b) many people, though not all, are self-motivated. Some
people would like to teach themselves and then take an examination for credit.
And as more alternatives for certification such as the Western Governors
University begin to emerge, students will increasingly have more
options.
Competency-based Standards
One of the objectives of the Western Governors University is to establish competency-based standards that students can meet through both traditional and nontraditional routes. Several participants pointed out that in the K-12 environment, the increase in software development has been related to the rise in competency-based learning. The NCTM standards in math-agreed-upon guidelines for math excellence-has led to greater investment and competition in the marketplace. There are fewer guidelines about what higher education needs. Is it possible, they asked, to get people to agree that "this is excellence" and then create a development model around this?
One example of widespread agreement about subject matter excellence in the higher education community is the Advanced Placement (AP) exams. The AP model incented students to acquire certification independent of their high schools; in turn, colleges grant credit for the result. The advantage of competency-based standards that have broad support is that they give us a better way of making an apples-to-apples comparison of learning outcomes regardless of the instructional methods used to achieve them. Either universities or potential new entrants can say, We can demonstrate that by using this technology we can deliver the same or better education at a lower cost. One particularly intriguing idea would be to set as a condition for qualifying for federal student loans and other forms of financial aid the requirement to offer equivalency exams for $50 in addition to offering courses. This would give students a way to test out of the course and disaggregate the outcome from the process.
The Supply Side: Ideas to Stimulate Development
The discussion then turned to a series of issues related to the supply side. The question was raised, What would it cost to get a first rate usable product? What are some ball park numbers? Bernie Gifford responded that the cost of development for courseware is about $50,000 per hour of instruction or about $3 million per course. This does not include the cost of sales and marketing. But Jack Wilson also pointed out that the cost of software development is continuous: you don't just develop the courseware and sell it for 20 years. Software development is an iterative process: as projects progress, additional improvements are always recommended and upgrades are always needed which, in turn, leads to more programming. Like a magazine subscription, the cost goes on forever to incorporate new software innovations, new operating system releases, and so on. Academic Systems currently budgets $500,000 per year per course for continuous improvement efforts wherein they bring together instructors who are using the software to decide what is needed to improve the learning product.
Carole Cotton, president of CCA Consulting, pointed out that a key problem in
the software industry is how to build a recurring revenue stream. You cannot
make enough money on new licenses and new sales because the selling cost is so
high. It takes five times the effort to gain a new relationship as it does to
keep an ongoing one. You need to move to a place where 50 percent of your
revenue is in a recurring revenue stream, such as upgrades. This is true of
textbooks also.
Is the Software Market like the Textbook Market?
Many of the participants believe that the development of instructional software-and the appropriate relationship between universities and the corporate sector-is, in many respects, like the creation of textbooks. Each side needs to focus on its core competencies. When a textbook is published, the faculty member provides the content, and the publisher produces the book. In the textbook development model, publishers pay faculty outside the university system. As Richard West commented, "I think the development of instructional software, like textbooks, is institutionally assisted but not institutionally based. Our faculty participate in this market, but we don't see that money institutionally. Putting our venture capital into this area would not be something I'd think an institution should do on any large scale."
Several publishers echoed the view that the software market is not qualitatively different from the textbook market. Jim Lichtenberg, vice president of the higher education division of the Association of American Publishers, pointed out that, to some extent, publishers are already entering the instructional software market because of the lack of reliability in the textbook market. Textbooks are not selling well; students are sharing or copying or buying used books. College publishers want to survive so they are looking for alternative products that students will buy. Publishers will continue to launch software products in the marketplace. Not all of them will be good, and we must expect some failures. The marketplace talks to itself: good products will result in sales through word of mouth just as a new text gets adopted quickly if it really works. Publishers are closer to transforming than might be apparent from outside, said Lichtenberg.
Other participants countered that there is an important difference between developing high-quality, multimedia, immersion learning courseware and producing a textbook. It is not clear that the scenario is the same. The idea of a faculty member writing a text does not translate to producing software. Apple Computer or Microsoft would not permit a single person to go off for a year and come back with some software. Unlike the textbook model, the cost of software development is front-loaded. Furthermore, the iterative costs are significant in terms of programming, management, and support. Putting words on a page is a low-tech enterprise; developing software is expensive. The "production of the product" is where the capitalists' costs lie, and this is significantly different in software compared to texts. Whose responsibility is it to pay for development and production in a new model? Are publishers willing to pick up the tab?
The Role of Government
One of the questions posed to the group from the outset was, is there a role for the federal government in stimulating a market for instructional software? Tom Kalil began this part of the discussion by commenting that if there is evidence that instructional software can reduce costs while maintaining or even increasing quality, then there will eventually be a market for it. There may be a long S-curve associated with its adoption because of the organizational issues we have discussed. But if this technology really can deliver high quality education at a lower cost per credit hour and if colleges and universities fail to adopt it, then there will be new entrants in the education market. So one approach to the question of federal intervention would be to do nothing and let the market take its course.
Even if one says, no change of policy is needed, we need to recognize that the federal government is already making some investments. So another approach would be to ask, are there some things that could be done that are analogous to the federal government's support for the ARPANet and the NSFNet which have led to a really large marketplace in this area? Are there some things we can do to help grow the market at a faster rate than it would otherwise, especially since higher education to a certain extent is insulated from market forces? Are there some things that can be done that are catalytic as opposed to the sorts of intervention that the federal government does in the agricultural sector where we have large subsidies. We are not faced with a binary choice between doing nothing and engaging in really heavy-handed government intervention.
If existing incentives are in the wrong direction, what sort of incentives could the government create to encourage the kind of change we have been discussing? If we face an adoption problem, then we should fund some strategies that focus on the demand side-in other words, to change the culture at the universities to create a more willing market. For example, it may be particularly important to do research and development in the area of third-party evaluation that would help identify the pay-offs of adopting instructional software, both from the point of view of the educator who wants to know about outcomes, and the point of view of the chief financial officer who wants to know whether this will allow institutions to do more with less. If we face a supply-side problem, we should fund the producers of materials, but perhaps in a different way than we have been doing up until now.
Kalil then went on to outline some possible approaches that the federal government might take on the supply side.
Research and Development
We seem to agree that the federal government's current approach to curriculum development tends to favor small-scale innovation and faculty development as opposed to encouraging wider scale adoption. One option is not to spend any more money in this area, but to look at what we are currently doing and decide whether we are making the wisest use of existing research and development dollars. Many participants felt strongly, for example, that NSF's practice of funding individuals for curriculum innovation needs to change. DARPA, for example, does not limit itself to funding individuals working on small projects; it goes after systemic problems and funds good projects including those from private enterprise. The government could fund Disney or Apple to develop software, for example, not just individuals. Only those who have demonstrated expertise in software development should be funded.
Participants suggested that the typical government funding model of paying people upfront may not be appropriate. We have already spent a lot of money in developing content; the problem is that we do not have much to show for it as a result. Sometimes the grant recipients simply do not finish their projects. As the system now exists, there is very little accountability: no penalty if there are no deliverables, no penalty associated with not producing. And even when a product is produced, there is no incentive to market it.
Ideas for alternative funding models were to use a prize rather than a grant model-such as having a contest and rewarding only the winning software; to fund proposals in phases with recipients receiving increments of money as completion goals are met; to require coinvestment from the private sector rather than funding universities at 100 percent; to treat grant recipients as if they were small businesses, giving many seed money but fully funding only the most promising to finish the product; and to use a two-phase approach, funding development as well as adoption.
Joint Procurement
Since this market is characterized by very large fixed costs and very low marginal costs, is there a case for some sort of joint procurement among colleges and universities where 10 to 20 large universities or state systems put in $300,000 to $400,000 each? While it is unreasonable to expect a group of colleges and universities to reach consensus on a shrink-wrapped course, it might be possible to aggregate their demand and develop an RFP for a superset of disaggregated instructional modules. And if so, could the government play a role in encouraging that sort of activity by putting in 10 to 20 percent to defray costs? The government might also have an interest in being one of the participants in the procurement consortium since it currently spends a lot of money on education and training, particularly on the defense side.
Bob Culver pointed out that before universities will undertake such a consortial activity, they need to understand what their return on investment would be. Currently, they do not understand that the opportunity costs of poor retention rates are significant and that the investment would be worthwhile if they could be reduced. Bob Heterick offered the following example to illustrate the point. Roughly how many students take remedial math in a year? Pick an average credit hour cost, and you will quickly discover that it costs our institutions about $1 billion a year to do remedial math. By jointly procuring instructional software that significantly improves retention rates, the investment could be amortized in the first two or three months of use. There is no question about the pay-off to higher education; the challenge is to develop a way to collaborate to do these things and then to adopt them.
Jack Wilson countered that such an effort requires an upfront capital investment, which is not the kind of thing institutions usually do. But publishers and software companies can do it. They can put capital forward and collaborate with universities in the development effort. Partnerships are the key, and the key to making partnerships work is for each side to understand the mutual benefits.
Brokering Workshops
If the option of joint procurement seems implausible for institutional reasons, the government also has the ability to act as an honest broker. It could convene some workshops in which subject matter experts describe to publishers and software companies the sorts of functionality they require, especially in the "One Percent" courses. The government also might be willing to defray some of the overhead costs associated with pulling together publishers, software companies and subject matter experts.
Contract for Software Development
Kalil suggested that we ought to look at whether existing federal investments in job training could be employed to help promote this market. Are there current activities involving big dollars that could be diverted to fund software development? In the State of Massachusetts, for example, the federal government is pumping funding into a center designed to serve the thousands of people who need retraining because of military downsizing. The center is being established in cooperation with Peat Marwick and several universities. The military intends to fly in people from all over the country to be retrained over a 10 to 15 year period.
An alternative approach would be for the government to become a purchaser of software or courseware to use in such programs. The military has a good track record in educating its enlistees and in retaining them because they have a culture of retraining. They are also very good at creating software to teach skills. The military could issue RFPs for software to be used in retraining, software which could later be used by colleges and universities. Efforts such as these could begin to infuse software in the marketplace and help lead a slower moving higher education community.
In Response to a Weak Government Role
Several participants urged the federal government to take a more active role than that posed by Kalil. They stressed that the government is a real driver of higher education via research grants and student loan subsidies. The expansion of higher education in 1970s, for example, can be traced to the Higher Education Act of 1972, which resulted in an explosion of students able to attend college and universities because of student loans. Higher education is an industry that is consuming family income at a record rate; the average family will soon have to use its entire paycheck to pay for a college education. These same trends drove the President to try to get a handle on health care.
The President's efforts had an impact by making people realize that health care costs were getting out of control. Even though reform attempts failed, the effort had the impact of bringing the issue to people's consciousness. This resulted in a slow-down of rates and in people exercising greater control over escalating costs. In some ways, it was better to have the government try and fail than to do nothing at all. The administration's "bully pulpit" may be the most powerful tool the government has. If we agree that there will be no market for instructional software unless higher education undergoes a significant cultural change, then a more active role may be required.
Others cautioned against a more active role, pointing out that the goal of developing software and new ways of teaching is smaller in scope then trying to reform all of higher education. The analogy with the Higher Education Act breaks down, they argued, because there was no other societal force to increase enrollment access. In the case of software development, there are several industries with resources that are ready to move once they are convinced that the demand is there. Citing Lew Perelman's argument, "Unless an innovation is paid for directly by those who stand to benefit from it, its chances to flourish are dubious at best," Jim Lichtenberg pointed out that unless the universities invest in the product themselves, government intervention will have little result.
A More Activist Role for Government: The Corporation for Public Software
In response to the desire to see the government more activist in its behavior, Tom Kalil asked, "If the administration were to do something bolder, what would it be? Suppose we had $25 to $50 million? How should we spend it?"
Participants noted that the government can effectively assist in arranging
partnerships and facilitating the commitment of resources. When the NSF "funded"
the Internet in its early days, it did so by leveraging a rather small direct
investment to stimulate large investments by universities and the private
sector. Building on this model, one idea would be to create an entity similar to
the Corporation for Public Broadcasting, the Corporation for Public Software
(CPS). Its mission would be to (1) establish agreed-upon standards for learning
outcomes, (2) contract with universities and companies to develop products with
a focus on the "One Percent" areas, and (3) establish standards to support the
software. Instead of being a physical place, the CPS would be a virtual
development center with sufficient funding to coordinate the activities outlined
above. It would have all the functionality of the joint procurement consortium.
And it would phase itself out of existence once the initial market takes
off.
Participants generally concluded that colleges and universities can do a better job at changing structures than creating additional supply. Universities by themselves cannot afford to get into the software development business. Richard West suggested that the collaborative experiences that the CSU is having with Academic Systems may represent the emergence of how a market will function. As we gain some experience, other companies will be able to identify additional needs in this area, and private capital will be invested. If we can create incentives on the institutional side, the supply will be there. Our responsibilities are on the education side of the equation.
For a long time many of us in higher education have been comfortable living in an environment where a large of number of students who entered our doors did not leave with any token of achievement. We have been satisfied, for example, with community college systems where only 9 percent of the students who come to us wanting an associates degree actually secure an associates degree and with four-year schools where less than 40 percent of the students get baccalaureate degrees within seven years. We've lived with those numbers for a very long time. And, if we're happy with those numbers, then we'll continue doing what we've been doing all along.
Because of the changes taking place in our economy, because of the continuing explosive growth in the cost of higher education, and because of other competing demands for scarce public resources, I believe that we are at one of those junctures in the history of higher education where we are either going to have to cut back sharply or commit ourselves to increasing the success of the students that come to us.
Increasingly, there are larger numbers of students showing up at our campuses who don't look like the kinds of students I first taught at the University of Rochester a little more than 20 years ago. They are more diverse in all kinds of ways. Frequently, they are first-generation college students. Many of them have been underserved by their high schools, and they show up with lots of rough edges. They have the potential to be successful if we would take it upon ourselves to help them.
For reasons that are illustrated in this paper, of all the options that are available to us, instructional technology can have the greatest impact in the first two years of college. The first two years play a major role in dictating the success of students. If they do not finish the first two years, they do not come back. But technology will not be viewed as a solution unless an institution is committed to student academic success. Once you commit yourself to increasing student success, you will conclude that the lecture-presentation model is not the optimum model for promoting it. Technology becomes one of the parameters that enables us to have a different sort of conversation about education.
If institution leaders are indifferent about student achievement, if they can live with freshmen failure rates of 60 to 70 percent, then it follows that instructional software will not be important to the institution. In other words, if you are happy with the status quo, there really is no impulse to invest in technology. But if institutions care about what happens to students, and ask, how can we improve instruction, how do we change the nature and character of the enterprise, then it follows that they will be interested in what instructional software can do to help in the learning process. The transformation of higher education requires a fundamental commitment to changing the statistics of success and failure.
1. D. Bruce Johnstone, Learning Productivity: A New Imperative for American Higher Education, Studies in Public Higher Education, State University of New York, 1992.
2. William F. Massy and Robert Zemsky, Notes for the Educom Conference on Academic Productivity, Wingspread, June 7-8, 1995.
3. Western Governor's Association, "A Vision Statement for a Western Virtual University," February 1996.
4. Lewis J. Perelman, School's Out, New York: William Morrow & Co., 1992.
5. Robert C. Heterick, Jr., Keynote Address, NLII meeting, New Orleans, LA, Jan. 1995.
6. William F. Massy and Robert Zemsky, Using Information Technology to Enhance Academic Productivity, Washington DC: Educom, 1995.
7. Carol A. Twigg, The Need for a National Learning Infrastructure, Washington DC: Educom, January 1995.
8. Massy and Zemsky, Using Information Technology to Enhance Academic Productivity.
9. Carol A. Twigg, "Is Technology a Silver Bullet?" Educom Review, Vol. 31, No. 2, March/April 1996, pp. 28-29.
10. Twigg, "Is Technology a Silver Bullet?" and The Need for a National Learning Infrastructure.
11. Jack M. Wilson, "The CUPLE Physics Studio." The Physics Teacher, Vol. 32, No. 9, December 1994, pp. 518-523.
12. Carol A. Twigg, "The One Percent Solution." Educom Review, Vol. 30, No. 6, November/December 1995, pp. 16-17.
13. Massy and Zemsky, Using Information Technology to Enhance Academic Productivity.
14. Twigg, "The One Percent Solution."
15. Twigg, The Need for a National Learning Infrastructure.
16. Perelman, School's Out.
17. Perelman, School's Out.
18. Paul Evan Peters, "From Serial Publications to Document Delivery to Knowledge Management: Our Fascinating Journey, Just Begun," The Serial Librarian, 28, 1/2 (1996) 37.
19. Heterick, NLII New Orleans Keynote Address.
20. Perelman, School's Out.
21. Michael G. Dolence and Donald M. Norris, Transforming Higher Education: A Vision for Learning in the 21st Century, Society for College and University Planning, 1995.
22. Carol A. Twigg, "It's the Student, Stupid!" Educom Review, Vol. 31, No. 3, May/June 1996, pp. 42-43.
Academic Productivity: The Case for Instructional
Software
by Carol A. Twigg
© Educom® 1996, Interuniversity Communications
Council, Inc.
Educom, Suite 600, 1112 16th Street NW, Washington, DC
20036
Phone 202-872-4200 · Fax 202-872-4318