Our editorial in
February 1987
included this article on the future of interactive technology in education:
The Potential for Interactive Technology
The future for learning and using technology requires vision and facts
by Alfred Bork
Computers and associated technology (the videodisk, the compact disk, a
nd sound) can stimulate major changes in our educational system. In the last few years, schools and universities in many parts of the world have acquired large numbers of computers. It is time to assess the situation.
The main advantage of the computer as a way of learning is that it allows us to make learning interactive for all students. We can then pay attention to the needs of each student by individualizing the learning experience.
But the results of computers in learning are disappointing; studies might even show that computers are doing more harm than good. Effective use of computers in learning will not occur automatically. A reasonable chance exists that it will not occur at all unless we readjust our current directions. Just because computers and other modern technology are widely present in schools is no guarantee that this equipment will be wisely employed.
I divide the topic of computers in learning into four major areas. First, I consider several serious problems facing our edu
cational system today. Oddly, they receive little attention in discussions about the role of computers in learning. I will then cover the recent use of computers, visions of future developments in education, and proposals to attain those goals.
Current Problems in Schools
Dozens of reports over the last 25 years have discussed major problems of education. Perhaps the best known, prepared in April 1983 for the Secretary of Education, is "A Nation at Risk: The Imperative for Educational Reform". This report emphasizes the serious consequences to the country of a poor educational system, as reflected by its title. I will not summarize these reports but rather mention a few critical factors about the current state of classroom education in the United States.
Typical Classes
Most students today do not succeed at the learning task. It has been estimated that in a typical class, only about one-third of the students learn the material well. In a class with 20 or 30 st
udents -- common in education in the U.S. -- it is almost impossible to give individual attention to the learning needs of each student. The two primary learning modes provided, the textbook and the lecture, do not create active learning possibilities for most students and offer the same learning experience to all students regardless of background, learning style, learning speed, or goals in life.
Poor Results
Declining scores on national and international tests are another measure. For example, recent international mathematics testing shows that U.S. students are far from the top, at all age levels and in all areas of mathematics. Almost all the developed countries are above us in these tests. Further, illiteracy has now become a serious problem in the U.S., greater than in most developed countries. The indicators in such testing are mostly negative; the problems are increasing.
Decreasing Democracy
The decline in scores becomes even more frightening when we
consider the uneven distribution of the quality of education in the U.S. Minority, poor, and rural individuals receive poorer educations than do students from affluent areas. In most states, school districts in wealthy areas spend far more per student than poorer districts. This inequality of opportunity is more likely to be a problem in the U.S. than elsewhere, given local control of school budgets, but it is a worldwide problem. Children may be created equal, but the educational system quickly changes that. We seem to be moving back to a time when education was the privilege of only a few.
Lack of Teachers
Over the next ten years the shortage of good teachers will continue to grow. This decline may well be the dominant visible problem of education during this period. There are several causes. First, we pay teachers very little compared to professions requiring similar education. The average salary of a teacher in the U.S. in 1985, according to the National Educational Association, w
as about $25,000. Second, teaching is no longer considered a desirable profession by many people in this country. Third, we train few teachers now. Only about one-fifth of the number of math teachers graduate today as compared to ten years ago. Fourth, our schools of education are, on the average, weak. Fifth, the students currently entering teaching have low test scores on such exams as the SAT. Sixth, it is estimated that about half of the teachers currently in schools will be retiring during the next ten years, primarily because they have reached retirement age. Hence, we can expect an increasing shortage of qualified teachers in the next few years. Simple solutions will not be adequate for this problem. Educational planning, with or without computers, must pay attention to this decline in both quantity and quality of teachers.
Negative Attitudes
The attitude toward teachers and the teaching profession just mentioned is paralleled by a decline in interest in education in this count
ry. We see many unsuccessful attempts to increase taxes to pay for adequate education.
Of even more concern is the attitude of most students toward learning. Our schools seem to make people less interested in learning. Young children entering school are bright and curious, enjoying much of what they do. A few years later many of them view schools as a type of prison, convinced that the failure to learn is their own problem.
If we are to improve education, we must consider new directions. These directions must recognize that we will soon have far fewer teachers than we currently have. Modern technology, particularly the computer, provides us with the possibility of meeting these problems.
How Computers Are Used Today
To support the belief that computers can lead to an improved educational system, I discuss four examples of existing computer-based learning material: the logic course at Stanford developed by Patrick Suppes, the physics course at the University of California a
t Irvine, the Writing to Read course, and the Scientific Reasoning Series. These four projects show the role of interaction. Two of them -- the courses at Stanford and Irvine -- are more than ten years old. The examples use sizable amounts of computer material, mostly with support from print and other media.
The heart of the logic course at Stanford, and of the set theory course that may follow it, is the interactive proofs. Like other good courses of this type, the focus is on getting college students to prove increasingly difficult theorems.
The interactive theorem-provers that support the computer-based course allow students to develop their own proofs, which can then be checked by the computer. Both formal and informal proofs are possible. The detailed feedback to the student is far superior to that provided in most lecture-based logic courses. The expository material is less interactive (not surprising given the ten-year age of this course) but does reflect interesting early use of digitally
stored speech. The course has been transported to other locations. Throughout much of its history at Stanford, it has been the only logic course available to students.
At the University of California at Irvine, a physics course was organized around a set of on-line exams, with each exam uniquely generated by the computer. As with most physics courses, the exams are primarily problem-based, but considerable conceptual knowledge is also required. Students stay with an exam until they perform it almost perfectly, in the tradition of mastery learning.
The exams contain most of the learning material of the course. As soon as a student has trouble with a problem on their exam, a detailed learning aid is interactively presented to the student. Often this material is closely related to the difficulty the student is experiencing. A variety of other learning modes are available, but it is possible to learn the material from the exams alone. Students are also offered a choice of content; two different courses,
mostly with different quizzes, are available. Students can choose a conventional course, but three-quarters select the computer course. A central problem of education for young children is reading and writing.
The Writing to Read course, developed by John Henry Martin, is directed at five- and six-year-old children. It is a multimedia course, with extensive use of materials other than the computer. The course is based on a phonetic system. Children learn to write (i.e., type) the words they can say, and then they can read the words that they type. (Unlike the Initial Teaching Alphabet approach, it has been found that students quickly move to ordinary spelling.)
Writing to Read has had a full summative evaluation, conducted by the Educational Testing Service in Princeton, New Jersey. The results are favorable compared to other ways of teaching reading and writing. Even to the casual observer, the performance of the children at the end of the year is almost amazing; children are typing coherent comp
ositions. The series is available from IBM.
The Scientific Reasoning Series is a collection of ten programs, directed to students of junior high school age and up. Bright elementary students will also be able to use many of them. The aim of this series is to bring students to an understanding of the nature of science.
The tactic is to place students in interactive environments in which they will behave like scientists in developing and testing concepts and theories. These environments are student-supportive, offering aid to students who are progressing slowly. The programs are self-contained, not requiring reading material, teachers, or other programs. The flavor is conversational, with the student playing an active role in the learning process. This series is also available from IBM.
The four projects just discussed illustrate what is possible with sizable developments. They suggest the computer's potential to improve education. But we need to consider further the structure of future educat
ional systems.
Two Positive Visions
We can conceive of computer-based systems of the future that will provide a much better education for all students than our current system does. To follow are two models for future educational systems: mastery learning and the proposals of George Leonard.
Given that future educational systems should perform better than current systems, we should work toward the goal of everyone learning everything well, the goal of mastery learning. This does not imply any reduction of the level of education.
The notion of mastery learning, as promoted by Benjamin Bloom of the University of Chicago and others, has a variety of components. But these are not all possible at the same time in most current educational situations without skilled individual tutors. A student is kept with a given topic until mastery is demonstrated.
This demonstration is typically through some type of exam, with the student be-
ing required to perform almost perfectly on
the exam. A student who does not show mastery continues to study the unit, with a variety of learning material provided by the instructor. Several approaches to learning the topic may be required, perhaps reflecting different pedagogical approaches and different media. Students are thus encouraged to learn at their own rate and ability.
In a sense, mastery learning is the application of democracy to education. Not only are people created equal, but we ensure equal education for all. Experience suggests that mastery learning is possible with the use of modern technology. But mastery learning does not completely specify the structure of schools. In thinking about future educational structures, certain questions must be answered. How many teachers will we have? What are the roles for the teachers? Will schools continue to exist or will part of education move to other learning environments, such as the home or public library? How do we accomplish the nonlearning aspects of schools, such as child care? How
do we develop the necessary curriculum material? How do we implement and disseminate desirable changes? What does the daily life of a technology-based school look like?
Few people have developed visions of future schools. But without such visions, meaningful planning to improve education is difficult. Consider George Leonard's approach.
A few speculative views of schools 20 or so years from now have been developed. George Leonard has developed two of these. One is described in his book
Education and Ecstasy
(Dell Publishing, 1968) and the other in an article entitled "The Great School Reform Hoax" (
Esquire
, April 1984). Both of these schools are fully consistent with the goal of mastery learning; both make extensive use of the new interactive technologies; both individualize the learning process and make it interactive for all students. Both are practical visions of an educational system heavily dependent on modern interactive technology.
In
Education and Ecstasy
,
Leonard divides the school into two segments, one for acquiring knowledge and one for affective behavior. In the knowledge dome, students work at large three-dimensional displays. The computer has a complete management system; as a student begins to work, the experiences are directly relevant to that student's needs. Brain waves are analyzed to reflect understanding, or lack thereof, in addition to the technical details already in use today. In the socialization component of the school, children work in small groups, with teachers, using encounter-group approaches.
Later, in the
Esquire
article, the two functions of the school are combined into a single program. A greater variety of instructional modes are used. And a major new ingredient is added: Students start each day by planning their schedules for the day, usually in an interaction with the computer.
How To Attain Such Schools
This last section proposes a set of steps that could lead to a future educational system
that would be, for most students, far superior to our current one. Although the problems discussed in the first section refer partially to the U.S., the approach suggested could work all over the world, including the developing countries.
To implement the models of George Leonard, the creation of many courses is a critical step in developing a new educational system. Our current courses are inadequate; they were developed before the modern learning technologies were available, they are not interactive and individualized, and they do not aim at mastery learning.
But two factors, and perhaps others, make it difficult to move to extensive immediate curriculum development. First, it is questionable whether we have enough practical experience with computers in learning, particularly with full courses, tc jump directly to remaking all courses. The second factor is the difficulty of starting a project of this magnitude directly. Not everyone is convinced that a computer-based educational system is possi
ble and reasonable.
To overcome these problems, and to proceed rationally, I suggest a two-stage approach. We should conduct a trial period to examine the concept of technology-based courses, followed by full development.
The Experiment
The purpose of this experiment is to study the effective use of the computer in education, in a focused way, to furnish information for later development. I propose the development, use, and summative evaluation of about 20 full courses that utilize the computer and associated technology, along with older learning media.
These courses should be diverse in all senses. They should be flexible enough to reflect different approaches to learning, different subject areas, and students of different ages. Courses should use a variety of production systems in the process of development. Their structure should be modular so that students can choose from several levels. All courses should be funded at levels that might lead to higher-quality courses.
We already have a reasonable understanding of the process needed to develop high-quality material, so the focus should be on highly skilled teachers as the head of the pedagogical design team. These teachers should be given maximun freedom to make the best possible decisions to aid student learning. Further, most or all teacher training materials should be considered as an essential part of the course development itself. The process of implementing each course in the schools should be studied.
The most important part of the experiment will be the extensive evaluation of these courses with typical students, both in the formative sense, during development, and in the summative sense. Summative evaluation should be conducted with at least tens of thousands of students, in a variety of types of schools, comparing these courses in detail with the existing courses in the areas involved. This full evaluation should not be done by individuals associated with the development; competent professional evalua
tors should plan and manage the evaluation.
Based on our studies at the University of California at Irvine, I estimate that this experiment in the full use of the computer in education will cost about $200 million and will require about five years to complete. After such an experiment, we will have the information we need to remake our educational system and begin its full development.
International Ramifications
An intriguing possibility to consider is multinational development. Joint activities at the University of California at Irvine and the University of Geneva in Switzerland have demonstrated that it is possible to structure computer-based learning material so that it is relatively easy to convert to a new language. Cultural differences may be present, but most of these can be taken care of in the translation process. Some areas may not lend themselves to such development, but in many areas, such as science and mathematics, courses can only benefit from international devel
opment. This approach should be particularly important for the developing countries, given their difficulties in developing full-scale educational systems.
Several of us worked at estimating the cost of full curriculum development. Conservatively, we should expect to spend about $10 billion, perhaps over a period of about six to eight years. Another way of viewing this is that we would need to invest 1 percent of our educational costs each year in such development.
If we were to place education's cost as one of national value, the cost is about one-third of that of putting a person on the moon and is about one or two days of the military budget each year. These figures assume only one language; conversion to other languages would be a modest increase.
Education worldwide has major and increasing problems. We have the capability, using the new interactive technology, to ameliorate these problems. But this will only happen if we learn from the past to plan for the future.
photo_link (115 Kbytes)
Alfred Bork, developer of highly interactive computer-based material, is professor of computer science and director of the Educational Technology Center at the University at California at Irvine.
10 Years Ago in BYTE
