Bolt Beranek and Newman Inc.
To appear in
K. Sheingold & M. Tucker, (Eds.), Restructuring for Learning with
The Role of Computer Technology in Restructuring Schools1
Bolt Beranek and Newman Inc.
Computer technology and electronic networks have been slowly infusing the schools (see Becker, 1986). This equipment is unlikely to end up in closets or even sit idle most of the time, because of the widespread and growing use of such technology in both business and the home. Hence there is a kind of "authenticity" (Brown, Collins, & Duguid, 1989) to using this equipment in the eyes of students and teachers; students want to use the technology because it represents the future. In a society where most work is becoming computer-based, "school work" cannot forever resist the change.
When a technology becomes widespread, whether it is the book, the automobile, or television, it has ramifications throughout society, including education. For example, the invention of the printing press and the book had profound effects on education (Boorstin, 1983; Eisenstein, 1979). It made the ideas of universal literacy and public schooling possible, and led to a deemphasis in teaching the art of memory. The automobile and bus led to the consolidation of rural schools and the dispersion of people to the suburbs, and in turn to the split between urban and suburban education, and busing to achieve racial integration. Television and video technology is even now having profound effects on education, such as the decline of print culture and the rise of a visual culture, low tolerance for boredom, and the loss of innocence for children (Postman, 1982). Similarly the computer and the electronic network are likely to have profound impacts on education, and it behooves us to consider these as we think about the issue of restructuring schools.
There are two views of education that have been at war for centuries: the didactic or information-transmission view, and the constructivist view (Brown, Collins, & Duguid, 1989; Cohen, l988a). The didactic view is the prevailing view among the general public. It holds that teachers should be masters of particular knowledge domains and that their job is to transmit their expertise about these domains to students by lectures and recitations. Students should memorize the facts and concepts of the domain, and practice the skills of the domain until they have mastered them, and they should be able to demonstrate their mastery on appropriate tests. The opposing constructivist view, which is characteristic of Dewey, Vygotsky, and Montessori, holds that teachers should be facilitators, who help students construct their own understandings and capabilities in carrying out challenging tasks. This view puts the emphasis on the activity of the student rather than on that of the teacher. Despite its predominance in the leading education schools (Cohen, 1988a), the constructivist view has made little headway in penetrating public education in America, or more generally in the world at large. But the trends I describe below may change that.
There are three different uses of technology in classrooms: (l) as tools for carrying out tasks, such as word processors, spreadsheets, programming languages, and electronic network systems, (2) as integrated learning systems, such as WICAT has developed, which include a set of curriculum exercises that students work on individually and which keep records of student progress for guiding the student and reporting to the teacher,2 and (3) as simulations and games, such as "Rocky's Boots" or "Where in the World is Carmen San Diego," where students engage in computer-based activities designed to be motivating and educational. The argument in this paper is that the tool uses of technology are most likely to be the way computers are widely used in classrooms, and that integrated learning systems and simulations (though important for educational purposes) will only penetrate schools to the degree that tool uses provide a rationale for buying computers. So the trends discussed below assume the tool uses of computers, though they apply to other uses as well.
It is obviously difficult to anticipate all the effects of computer technologies, and it may well be that I will overlook some of the most important of these. But researchers have begun to observe how these new technologies are impacting the schools, so we can at least make some informed speculations. There are at least eight major trends that can be identified from the literature and from observations in schools where computers are being used by teachers.
1. A shift from whole class to small group instruction
Where teachers use computers, normally one or two students are assigned to each computer. Teachers do not find it feasible to maintain all the students in lockstep, and so they move to an individualized instruction model of teaching (Schofield & Verban, 1988). This shows up in Gearhart, Herman, Baker, Novak, and Whittier's (1990) data on Apple Classroom of Tomorrow (ACOT) classrooms as a dramatic decrease in teacher-led activities (from over 70% of the time when computers are not in use to less than 10% when computers are in use) and a corresponding increase in independent or cooperative activities. This means teachers begin to talk to individual students, and develop an idea of what their understanding and their confusions are. Usually teachers have an inflated idea of what their students understand. So watching individual students struggle with problems may give teachers a better understanding of their students. It also means that students are more likely to go at their own pace and often in their own direction (Scardamalia, Bereiter, McLean, Swallow, & Woodruff, 1989), which for teachers can create problems of control.
2. A shift from lecture and recitation to coaching
As part of the shift from whole class to individualized instruction, there is a shift from didactic approaches to a constructivist approach. Schofield and Verban (1988) document this shift in terms of language, where there is a shift from second person constructions ("You should do this") to first person constructions ("Let's try this"). Gearhart, et al. (1990) document this shift in ACOT classrooms from teacher-directed activities (approximately 70% of the time off the computer to less than 10% on the computer) to activities facilitated by the teachers (from about 20% to 50%). The introduction of a third party, the computer, into the situation encourages the teacher to play the role of a coach in many of the same ways that a piano encourages the teacher to play the role of a coach in a piano lesson. Much of the learning is meant to take place between the student and the computer, and this puts the teacher into the role of observer and guide to make sure those interactions are beneficial to the student's learning.
3. A shift from working with better students to working with weaker students
In whole class instruction, teachers carry on a dialogue with their better students (Schofield & Verban, 1988). This is because it is the better students who raise their hands to offer ideas. Teachers do not like to call on weaker students, because they do not `want to embarrass them in front of the class. In a classroom where students are working on computers, the teacher is naturally drawn to students who need help, which is generally the weaker students. Schofield and Verban (1988) documents that in one classroom where there were individual computers, two of the weaker students received four to five times as much attention from the teacher as the more advanced students. We see this same shift in the classrooms we have observed in New York and Cambridge. However, as Schofield (personal communication) points out there may be a tendency for the teacher to overlook students who need help, but do not ask for it, because the teacher is usually very busy in these classrooms.
4. A shift toward more engaged students
In settings where computers have been put at the disposal of students in some long-term activity or project, researchers have reported dramatic increases in students' engagement (Brown & Campione, in press; Carver, 1990; Scardamalia, et al., 1989; Schofield & Verban, 1988). For example, Carver finds that students who were so bored with their classes that they would sleep through them, are eagerly engaged in a project to construct a HyperCard museum exhibit about their city. Similarly, Schofield and Verban report that students compare how far along they are in their geometry curriculum and even fight over who gets to use the computer during the time between classes. Dwyer, Ringstaff, and Sandholtz (1990) cite several examples in ACOT classrooms where teachers were encouraged to do more activities on computers, because students were so highly engaged during such activities. It may be that the reported increases in engagement are due to the novelty of the computer, but it is unlikely that this accounts for the entire increase. To the degree that the computer supports long-term effort rather than short exercises (a shift that Gearhart et al. (1990) find in computer-based language arts, but not in mathematics), there is suggestive evidence from these studies that students become invested in the activities they carry out on computers.
5. A shift from assessment based on test performance to assessment based on products, progress, and effort
Assessment in most classes is based on students' performance on tests given after different sections of the curriculum are completed. The introduction of computer technology and the shift to individualized instruction (see above) moves assessment away from the classroom test, which seems inappropriate to teachers under the circumstances. Schofield and Verban (1988) report that the geometry teacher they studied moved toward assessing students based on the effort and progress they made: in that case the system would not let them go on until they had solved each problem. Where the teacher sets up a project-based curriculum, then evaluation of students tends to be based on the products that emerge from the student's efforts. But for the present this creates problems for many teachers, because they do not know how to objectively assess such products. This problem has been solved for writing assessment in terms of wholistic and primary trait scoring methods, and clearly some such scheme is needed for project-based work (Frederiksen & Collins, 1989; Wiggins, 1989; Wolfe, 1989).
6. A shift from a competitive to a cooperative social structure
In the normal classroom, students are working individually and competing against each other for grades, except where students drop out of the competition because of social pressures or repeated failure. Brown and Campione (in press), Newman (1990), and Scardamalia, et al. (1989) find a shift toward a more cooperative social structure in their classrooms, where a network provides a common database for students. Scardamalia, et al. report how students comment on each other's notes, telling what they find interesting and what they cannot understand. Dwyer, et al. (1990) report striking increases in cooperative behavior in ACOT classrooms as reported from the teachers' journals they collected. Gearhart, et al. (1990) found an increase in cooperative behavior in mathematics classrooms (from 10% of the time without computers to about 40% with computers), but essentially no cooperative behavior in language arts classrooms in either case. Even Harel (1990), who had fourth graders working independently to produce a Logo program to teach fractions to third graders, found students sharing ideas and expertise on how to accomplish certain things in Logo. However, Schofield and Verban (1988) found an increase in competition in the geometry classroom they studied, and it may well be that integrated learning systems generally encourage students to compete to get through the material faster. One study in Israel (Hativa, 1989) suggests that this depends on how easy the program makes it for students to compare their progress.
7. A shift from students all learning the same things to learning different things
An underlying assumption of the educational system is that every student must learn certain basic knowledge and skills. This assumption leads to failing students who haven't mastered parts of the curriculum, and directing student's efforts to their weaknesses rather than their strengths (Drucker, 1989). The electronic network and shared database foster a different view of knowledge, where expertise is spread among different participants and brought together in a common space (Pea, in press). The National Geographic Kids Network (Foster & Julyan, 1988) is an embodiment of this idea of distributed knowledge, where students all over the country collect scientific data and exchange ideas with each other and working scientists. Because of the trend toward individualized education, there is likely to be a secondary trend toward breaking the lockstep of everyone learning the same thing in the same way at the same time. This trend can be seen in the classrooms described by Dwyer, et al. (1990), where students worked on different parts of complex projects, such as a model of their city; in the classroom described by Carver (1990), where students studied different aspects of their city to develop a museum exhibit; in the classrooms described by Scardamalia, et al. (1989), where students conducted research on different social studies and science topics; and in the school described by Newman (1990), where students collected different data on the weather. So the lockstep approach in schools, where everyone had to master all of the same knowledge and skills, is likely to change with the advent of computer technology.
8. A shift from the primacy of verbal thinking to the integration of visual and verbal thinking
As Postman (1982) has argued, the invention of the book transformed society from concrete, situated thinking to abstract, logical thinking. The visual media, i.e., television, cinema, and computers, have begun to develop a new kind of visual thinking, and a number of educators (e.g., Bransford, Sherwood, Kinzer, & Hasselbring, 1987; Wilson, 1986) have begun to explore how to use visual media to enhance learning. The computer and electronic network potentially provide instant access to the world's accumulated knowledge, in both verbal and visual forms. This development may slowly undermine the primacy of the book, the lecture, and their accoutrements, such as the multiple-choice test and the recitation class.
These effects of technology are subversive to some of society's most deeply held beliefs and assumptions about education. In particular they make tenuous the view that the teachers' job is to impart their expertise to students, and that the role of assessment is to determine whether the students have acquired the imparted expertise. So, inadvertently, technology seems to be coming down on the side of the constructivists, who have been trying to change the prevailing societal view of education, unsuccessfully to date.
Resistances to Technology
Cohen (1988b) and Cuban (1986) have argued persuasively that computer technology is likely to have little effect on the schools. They argue that to the degree technology is flexible, it will be bent to fit existing practice, and to the degree it can not be bent to fit existing practice, it will not be used. People interested in restructuring schools need to understand the resistances to change, some of which are particular to technology, and some of which are general, in order to identify the key leverage points on changing a well-entrenched system.
Any restructuring of schools can only take place over an extended period of time. The effects of the printing press were still being felt hundreds of years after its invention in the development of public education. So I will take a long-term view of how restructuring schools might take place and where a sustained effort is worthwhile.
Over the long term, important current issues, such as the costs of computer technology, its unreliability, and teachers' unfamiliarity with its use, become non-issues. The costs continue to fall, and as computers become more integral to everything we do, this trend can only accelerate. It is a fundamental trend in economics that in relative terms the cost of goods decreases and the cost of labor increases (Drucker, 1986), so that compared to teachers' salaries, computers will appear incredibly cheap in the next century.
The problem of teachers' familiarity with computers will also decrease as people come to rely on computers for writing, calculating, and communicating. This can already be seen to be happening: It is easier to type into a word processor than to write by hand.3 It is easier to do your taxes on a computer than to do them by hand. And it is easier to send electronic mail than to post a letter. These uses will become commonplace among college students, secretaries, and bookkeepers, so there is every reason to believe they will become commonplace among teachers. The problems of dealing with computers, such as getting them fixed, will become minor when they are used much of the time.
But of course the resolution of those kinds of problems does not mean that computers will be used in schools. Television is pervasive in society, and will probably never be widely used in schools. So why should computers come to be widely used when television is not? My argument is that the computer's most pervasive uses, which are related to work, are becoming necessary to accomplish school goals. Schools are in the business of teaching students how to read and write and calculate and think. As the computer comes to be an essential tool for doing these things in society at large, its use by students for doing these things is inevitable. We do not teach people how to drive cars by having them ride bicycles, nor will we teach people how to work by having them use paper and pencil, arithmetic procedures, and library card catalogues, when work has become computer-based.
There is a related argument that computers make the teacher's job more difficult, just as do television and film strips on the one hand, and the new science curricula of the 1960's on the other. The latter required teachers to put in extra time gathering materials together and saddled them with a difficult management problem of coordinating a class of students working independently on experiments or discussing the meaning of what they had done. It is true that computers make management more difficult when there are only a few computers in the classroom. The teacher has to figure out what other students will do when they are not working on computers, or has to allow a few students to miss a lesson while they work on the computer. But again, these are only problems in the transition to a society where most work involves computers. If students have ready access to a computer at all times, such as with a portable computer that can be connected to a network in different places, then these management problems go away. Students will do much of their work on computers instead of working with text books or worksheets. The management problem, then, is likely to be similar to that teachers currently face when children are working individually or in small groups. To the degree the tasks students are doing with computers are more engaging than those they currently carry out with textbooks and worksheets, it will make the teacher's job easier.
Another argument against the widespread use of computers is that teachers are not willing to give up their control and authority over students to the computers. There are two aspects of this argument. One aspect is that teachers want to be masters of everything that comes up in their classrooms, and because computers contain more information than teachers can possibly master, they will lose authority. The other aspect is that teachers like to hold the attention of students, and if students are off working on their own, then the teacher has lost their attention, as well as control over what they are doing. The first issue is currently exacerbated by the fact that teachers do not know a lot about computers, which as I argued earlier is a problem in the transition to a more computer-literate society. But there is a residual problem of giving students access to more knowledge than a teacher ever can master, together with the second aspect of students going off on their own. Both aspects of the problem of control can only be overcome by a changed view of the teacher's role to that of a facilitator of students' self-learning, rather than as a dispenser of information. Such a change in belief will not come easily, and will only come about slowly with the introduction of computers into school, as I discuss below.
Dwyer, et al. (1990) report a a difficulty that many of their teachers feel when they allow students to work on computers in ACOT classrooms. They seem to feel guilty that they are not teaching the students and they feel nervous about all the talking and sharing of information among the students. These feelings alternate with very positive feelings that the students are highly engaged and actively learning. So ACOT teachers in the initial phases tend to vacillate between enthusiasm for having students do a lot of their work on computers, and pulling back to use their old teaching methods in order to keep the class under control. Dwyer, et al. argue that it is important for teachers as they work through the transition to a more constructivist view of teaching to have the support of other teachers who have worked through or are working through the same transition.
Some people argue that teachers are not capable of using computers effectively. For example, in science labs they usually have students follow a fixed procedure (unlike scientific experiments), so that students know at each step what is supposed to happen. The argument is that when teachers use computers they will also follow a rigid format, since this procedural approach stems from a desire to make sure all students succeed. In fact, the computer-based integrated learning systems, such as WICAT's math curriculum, partially incorporate such an approach. This argument is surely correct to the degree that computers can be fit by teachers into their normal way of doing things. But the tools and simulations in computers are not content free. They make it possible for students to take over part of their own learning. To the degree computers support students' autonomous learning, and it is the goal of most educational software designers to do so, the particular pedagogical approach of teachers will be less decisive in determining how students learn.
A general view in organization theory is that American schools form a loosely coupled system (Weick, 1976) and while they readily adopt changes at the periphery of the system (e.g., model schools, computer labs), it is very difficult to make pervasive changes at the core of the system. While this may not be the reason constructivist teaching methods have failed to penetrate the schools (Cohen, 1988a), it surely will slow down any change that is introduced. But, if computers are widely perceived as necessary for school work, it will not stop their general adoption. In the next section I outline a set of principles designed to speed up adoption of any beneficial innovation.
Counterposed to the view that schools are a loosely coupled system, is the view that American schools have developed a system of institutions including the graded school, multiple-choice testing, curriculum and materials, teacher education, and lecture and recitation methods that are interlocking and self-sustaining. If you perturb any one of those parts of the system, the other parts will pressure the system to return to its original state. All of these institutions derive from and support a didactic model of education. Cuban (1986) makes an argument of this kind in terms of what he calls "situationally constrained choice," which incorporates (l) school and classroom structures, and (2) the culture of teaching, including the beliefs of teachers. These work together in his view to restrict what teachers can do in adopting different innovations.
On this argument, if you try to introduce computers for students to do their work, then it will be sustained only to the degree it fits this prevailing institutional structure. Since computers undermine the lecture and recitation methods of teaching, and promote the student as self-learner, they do not fit this institutional structure, and will squeezed out by it. Integrated learning systems, such as WICAT, have dealt with this problem by preparing curricular materials that fit easily into the current system. The materials mimic the kinds of test items in prevailing practice and so they produce gains on the tests that the current system embodies. They may have some early success in penetrating schools, because they have tried to fit into the current system. But my argument is that it is the tool-based uses of computers in society that will ultimately sustain their penetration of schools. The interlocking system described can certainly slow down the process, but it cannot prevent it, because the nature of education must inevitably adapt to the nature of work in society.4
Finally, there is a major resistance to the infusion of technology into the schools from the underlying belief structure in the society about the nature of education (Cohen, l988a, l988b). This didactic view of education holds that teachers must be experts in their field and that their job is to transmit their knowledge and skills directly to students. On this view learning involves memorizing essential facts and concepts, and performing procedures until they are automatic. The practices we cited above, such as the lecture and recitation methods of teaching, and testing for acquisition of facts, concepts, and procedures, are manifestations of this underlying societal belief about the nature of education. The constructivist view, that education should attempt to create environments where students can construct their own understandings and skills, is held only by a small minority of educators, and has no chance of affecting practice until the underlying societal belief changes. On this view, technology will only be used to reinforce existing practices, such as drill and practice and multiple-choice testing.
I believe this argument is essentially correct and important for technologists to understand. But even if technology is allowed into the schools under the guise of reinforcing existing practice, once there it will take on a life of its own. It is important to stress that many of the tool uses of computers (e.g., word processing, mathematical computation, graphing of data) are quite compatible with current practice. Teachers will not object to students typing their essays, or even in the long run to their using computers to solve mathematical problems. Once teachers let computers in the door, then the kinds of effects described in the first section of the paper will occur and teaching practices will change. And just as a change in practices with respect to racial integration led eventually to a change in racial attitudes,5 so a change in practices will slowly lead to a change in the educational beliefs of the society.
However, the arguments I have made so far only suggest that a change to a more constructivist education is likely to occur over the long run. A more salient question is whether there is anything that can be done to speed up the change. The next section proposes a structural change in school systems that would speed adoption of any change that improves educational practice, whether involving computers or not. The final section addresses the issues of how technology can most effectively be deployed to foster educational reform.
Principles for the Design of A Self-Improving School System
A major problem is that the present structure of schooling militates against change. Students are assigned to schools and are required to go to them. If they are bad schools they will continue to exist: there is no way for them to fail. The only thing a school system can do to fix a bad school is to send in a new principal, and usually she is prevented from making many changes due to constraints of the situation.
Another problem is that it is difficult to start new schools successfully. The problem isn't that parents or teachers are prevented from starting schools, but that the incentive is for parents to keep their children in free public schools rather than paying for them in private schools. So the only schools that are started (other than those funded by foundations) are schools for wealthy parents. This is not where our major educational problems lie: they lie particularly among poor and minority populations.
What we need to encourage innovation is a system that fosters creation of new schools and allows failing schools to die, particularly in our large urban areas where the problems of American schooling are concentrated. Such a system would stimulate existing schools to do everything possible to insure their survival. We need incentives and constraints that operate to make sure that the most difficult students and problems are dealt with, and that natural selection operates on the basis of the quality of the schooling and not on some extraneous basis, such as the race of the school principal, the quality of the athletic program, or the endowment of the school with facilities or technologies. A new system especially needs to avoid the current problem of creating schools that serve as dumping grounds for the educationally disadvantaged.
In order to facilitate innovations in schools, I would like to propose the following design principles. They are an attempt to synthesize the essential elements of various proposals that have been made for a redesigned school system (Chubb & Moe, 1990; Reigeluth, 1987; Tucker, 1989).
1. A mechanism whereby a group of parents and teachers in a school district can start a school.
The idea is that if parents and teachers in a school district want to start a school and they have a minimum of, say, 25 to 50 students, they should receive funds from the district at least equal to the current cost per pupil in the district. They also should receive space in a current building proportional to the number of students, from one classroom to an entire building.6 Since there will also be costs associated with starting a school (money for books, technology, etc.), these should be provided by a special fund on a per pupil basis. This fund should also provide resources for expansion of schools to take on more pupils. In addition, the school district should provide staff for encouraging successful schools, either within or outside the district, to set up branches in the district.
2. A mechanism whereby schools are closed.
If an existing school loses enrollment below a certain minimum (say 20 pupils), then it should be closed, and its students forced to choose another school within the system (see below).
3. A national agency should provide information on each school to parents and children.
To make effective choices, parents and children need to be provided information relevant to the educational policy and success of the schools, such as the kind of information available in national guides to colleges (Reigeluth, 1987). This kind of information is best collected by a national agency, to avoid dishonesty by local officials. The kinds of information the agency might provide include information about dropout rates, test scores of students in the school, college entrance and graduation rates of graduates from the school, random samples of opinions of former students and their parents, descriptions of the school's operation and facilities by neutral observers, occupation profiles of former students, etc. Ideally the test scores provided would be based on a "systemically valid" testing system (see Frederiksen & Collins, 1989). Information should be provided to all parents and children who will be making a school-choice decision in the near future with respect to all the schools they might consider. Where a school is new, only a statement of intent is possible, unless it is a branch of an existing school or coalition of schools.
4. Students above some age level should be provided alternatives to further schooling.
If students wish to drop out of school above some age level, for example, 12 years of age (Sizer, 1984), then they should be allowed certain options. One option might be to leave school, if they can find full-time employment with a legitimate business enterprise. Another option might be full-time participation in a licensed program, such as a music camp or boy scouts. Most important, there should be a national alternative service program, such as VISTA, that will accept any student over the legal age. But students who take one of these options before age 18 should be encouraged once a year to enroll in a school of their choice to continue their education. As Drucker (1968) argues, we should be encouraging continuing education, where people receive education throughout their lives, rather than extended education, where they are kept out of the workforce through a longer and longer adolescence.
5. Schools should be allowed to select the students they prefer, but there should be incentives to choose hard-to-place students.
If the proposed system is successful, different schools will specialize in the kind of education they offer. This means that their educational policies will probably be more successful for certain kinds of students than for others. If the system restricts schools' ability to select their students, it will restrict their ability to specialize. This would undercut a major goal of the plan. That raises the problem that schools may all want to accept certain kinds of students and reject others. To offset this tendency, greater financial resources should follow the hard-to-place students. In fact, the resources need to be enough greater to offset the systematic preferences of schools, which suggests some kind of market mechanism. Both Reigeluth (1987) and Tucker (1989) have suggested such a mechanism.
These principles are designed to produce a system where there will be both individual schools and coalitions of schools with specialized goals. There might be technology-based schools, art schools, Montessori schools, essential schools (Sizer, 1984), college preparatory schools, special schools for handicapped children, vocational schools, schools for girls, schools of design and engineering, schools for gifted in particular fields, back-to-basic schools, schools for particular minorities, bilingual schools, and even comprehensive schools that avoid specialization.
This goes against a philosophy of having every kind of student in every school in order to foster overall integration of society. I would argue that specialized schools should be restricted from discrimination in the same ways that colleges and businesses are restricted.7 But to the degree schools want to cater to students with particular interests or abilities, they may develop techniques that are particularly effective. The economic argument for the benefits of specialization applies equally well to schools as to business and labor. The moral argument against specialization loses force, given the inevitable disparity between urban and suburban school systems and the widespread tracking in the comprehensive schools.
One might argue that most parents and students will pick schools on the basis of proximity, or athletic ability, or better facilities, even if you provide them with information to make choices on the basis of educational quality. It is certainly true that most people will make their choices partly on such bases. But most people make choices by considering multiple factors, so that educational values are likely to be a factor to some degree in their decisions. The effect of proximity can be diminished by having multiple schools in each building, so that choices are made among equidistant schools. The effect of athletics could be diminished if we eliminated interschool athletic competition (as opposed to intraschool competition) in favor of Little Leagues or professional sports programs. The effects of facilities will be diminished if we equalize the distribution of resources on a per pupil basis as proposed in the first principle. To the degree school effectiveness is weighed at all in people's choices, it will bring a gradual improvement in the quality of schools. The more it is weighed, the faster the improvement.
Such a plan does not assume that parents know what is best for their children. There will undoubtedly be schools that emphasize drill and practice rather than thinking and that teach creation science rather than evolution, and these will appeal to many parents. But such problems are pervasive in the current system; over 80% of elementary school teachers think the phases of the moon are caused by shadows from the earth, and that the seasons are caused by changes in distance of the earth from the sun. The proposal does not solve these problems, but it would make it easier for people like Marva Collins (the black woman in Chicago who started an academically-oriented elementary school) to start schools. I would argue that most parents would want their children to go to such schools if they were available.
Another argument against the plan is that rich parents will subsidize the schools they send their children to by various means and this will undermine the mechanisms for establishing educational equity and for placing less desirable students. If parents want to subsidize the schools, that is in fact all to the good: It will give schools more resources to improve education. Whatever parents contribute is not likely to unbalance the funding of education more than the current system of suburbs with high per pupil expenditures and cities with low per pupil expenditures. However, if equality in educational opportunity is in society's interests, as I believe it is, then there is a rationale for offsetting parent subsidies with higher per pupil expenditures for schools that do not receive such subsidies. In principle, a market mechanism for placing less desirable students would automatically act to offset such subsidies, since the prospect of subsidies would enhance the value of students from wealthy families. So a market mechanism might be the least controversial way to offset parent subsidies.
One of the arguments that might be made against such a proposal is that it will produce a system like the college system in America, and colleges are not noted for their willingness to innovate. In fact, the most tradition-bound colleges, such as Harvard, are the most prestigious and therefore their practices serve as models for other colleges. This pattern inhibits the introduction of new practices, and serves to maintain the didactic approach to education that pervades the traditional colleges.
In organization-theory terms (Scott, 1987), the problem of educational improvement derives in part because it is difficult for consumers to tell a better product from an inferior product, unlike with restaurants and medical treatments. So in choosing colleges people rely mainly on prestige, and since prestigious colleges obtain the best students and most famous professors, they appear to be better on paper than their educational practices warrant. This effect tends to undermine the drive for self-improvement of any such plan in education.
However, I think it can be argued that in fact colleges have been much more innovative than the public schools in America and form the strongest part of our educational system. Certainly from the point of view of infusion of technology and flexibility with curriculum, colleges have been much more innovative. For example, there is more pressure on students in colleges to do their work on computers, and it seems likely that within ten years every college student in America will have their own personal computer. And when new disciplines emerge, such as psychology or computer science, they are much more readily adapted into the college curriculum than the public school curriculum. The continual birth and death of colleges encourages all colleges to seek their own market niches and to create programs that parents and children will find valuable. It is particularly among the less prestigious colleges, which serve the non-elites, that experimentation and improvement through natural selection occurs. In public schooling it is with the non-elites that our major problems lie, so that innovation is likely to occur where it is most needed under the proposed plan.
If a diversity of schools arises, and people are given the information necessary to make informed decisions, then the thesis of this paper is that the system will evolve toward better schools. The more effective schools will thrive and multiply, the less effective schools will die out. Existing schools and their personnel will do everything they can to enhance their chances for survival. There might evolve a preponderance of certain types of schools (e.g., essential schools), but that would only happen if they fulfilled the educational goals of a majority of parents and children. However, it is important to recognize that such a plan will not solve many of our schools' problems: it will only make it easier for change to occur in a very resistant system.
The Uses of Technology to Foster Educational Reform
The arguments in this paper have several implications as to what course of action school reformers and technologists should take to foster change in schools to make them compatible with the way society is changing. In the next century, an educated person will need to be able to learn and think in a computational environment. Most schools do not teach students these abilities now, and so a major change ought to be made in the way schools function.
One implication is that the first step is to put computers with powerful tool applications into the schools in as large numbers as possible. Many people might object to this step, particularly in light of the Apple Classroom of Tomorrow (ACOT) efforts, which have had at best marginal success to date (Baker, Herman, & Gearhart, 1989). They would argue that it is better to spend resources developing good educational software, teacher training, or computer coordinators, in order to make sure the technology that goes into the schools is used effectively. The trouble with that argument is that it presupposes that good educational software or teacher training or computer coordinators will lead to more effective use. In a few cases that is true, but on a wide scale it is likely to fail. I would argue that if you have computers that are easy to understand and that are powerful for doing school work, then people will eventually figure out how to use them. Using computers effectively in schools is difficult because of all the resistances described above, and so most things you spend resources on to improve usage will not work. We should not expect efforts such as ACOT to succeed immediately. But society at large is making the transition to computers, and the massive educational effort to make the transition is reaching both students and teachers; Simon (1987) refers to this as "education by immersion." So my argument is to put powerful, easy-to-use computers into place, so that society's retooling of itself will have something to work with in the schools.
Let me also add that the most powerful educational uses of computers in the future may not be their tool uses. Rather, the uses of computers for simulation, reflection, and video may be even more powerful educationally. But it is the tool uses that are becoming necessary to do work, and their usefulness to students and teachers will become readily apparent to everyone. The other uses of computers will come into play once computers have found their way into extensive use by schools:
Computers as simulated environments: Computers allow students to carry out tasks they cannot normally carry out in school, from running a business or city to troubleshooting a faulty circuit. The possibility of doing tasks that are difficult or impossible to do in school is one of the major values of computers for educational purposes (Collins, 1990; Papert, 1980).
Computers as reflective environments: Another powerful use of computers is for students to compare their own performance to other people's performances on the same task (Bransford, Franks, Vye & Sherwood, 1989). For example, in teacher education there might be a video segment of expert and novice teachers teaching some subject matter to students, with critiques on each lesson by experts from different viewpoints and explanations by the teachers of what they were trying to accomplish. Then a student teacher could compare a video of their own teaching to those videos of other teachers (Collins & Brown, 1988; Lampert & Ball, 1990).
Computers as video environments: There are vast video libraries of information that have accumulated over the last 100 years, and the output will multiply with the commercialization of the video camera. Video is a concrete medium, and people remember visual information more easily that verbal information (Bower, 1972). Having access to visual materials and explanations may well extend people's ability to learn, particularly those who have difficulty learning from books and lectures (Bransford, et al., 1987; Wilson, 1986).
In summary, because the nature of work is changing to incorporate computers in many aspects, the nature of school work will make a parallel change. This means that computers will come to be seen as necessary tools for students and teachers in their school work. But there are other powerful potential uses of computers for educational purposes. These uses will develop more slowly, but are likely to occur as computers become commonplace in schools and homes. All these uses of computers tend to be subversive to the prevailing didactic view of education in society. Using computers entails active learning, and this change in practice will eventually foster a change in society's beliefs to a more constructivist view of education.
Anderson, J.A., Boyle, C.F., & Reiser, B.J. (1985). Intelligent tutoring systems. Science 228, 456-468.
Baker, E.L., Herman, J.L., & Gearhart, M. (1989). The ACOT report card: Effects on complex performances and attitude. Paper presented at the annual meeting of the American Educational Research Association, San Francisco.
Becker, H.J. (1986, June). Instructional uses of school computers: Reports form the 1985 national survey. Newsletter of the Center for Social Organization of Schools. Baltimore, MD: Johns Hopkins University.
Boorstin, D.J. (1983). The discoverers. New York: Random House.
Bower, G. (1972). Mental imagery and associative learning. In L.W. Gregg (Ed.), Cognition in learning and memory. New York: Wiley.
Bransford, J.D., Sherwood, RD., Kinzer, C.K., & Hasselbring, T.S. (1987). Macro-contexts for learning: Initial findings and issues. Applied Cognitive Psychology, 1, 93-108.
Bransford, J.D., Franks, J.J., Vye, N.J., & Sherwood, R.D. (1989). New approaches to instruction: Because wisdom can't be told. In S. Vosniadou & A. Ortony (Eds.), Similarity and analogical reasoning. New York: Cambridge University Press.
Brown, A., & Campione, J. (in press). Fostering a community of learners. Human Development.
Brown, J.S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32-42.
Carver, S.M. (1990). Integrating interactive technologies into classrooms: The Discover Rochester project. Paper presented at the annual meeting of the American Educational Research Association, Boston.
Chubb, J.E., & Moe, T.M. (1990). Politics, markets, and America's schools. Washington, DC: Brookings Institution.
Cohen, D.K. (l988a). Teaching practice: Plus ça change.... In P. Jackson (Ed.), Contributing to educational change: Perspectives on research and practice. Berkeley, CA: McCutchan.
Cohen, D.K. (1988b). Educational technology and school organization. In R.S. Nickerson & P. Zodhiates (Eds.), Technology and education: Looking toward 2020. Hillsdale,NJ: Erlbaum.
Collins, A. (1990). Cognitive apprenticeship and instructional technology. In B.F. Jones & L. Idol (Eds.), Dimensions of thinking and cognitive instruction. Hillsdale, NJ: Erlbaum.
Collins, A., & Brown, J.S. (1988). The computer as a tool for learning through reflection. In H. Mandl & A. Lesgold (Eds.), Learning issues for intelligent tutoring systems. New York: Springer.
Cuban, L. (1986). Teachers and machines. New York: Teachers College Press.
Drucker, P.F. (1968). The age of discontinuity. New York: Harper & Row.
Drucker, P.F. (1986,). The frontiers of management. New York: E.P. Dutton.
Drucker, P.F. (1989). The new realities. New York: Harper & Row.
Dwyer, D.C., Ringstaff, C., & Sandholtz, J. (1990). The evolution of teachers' instructional beliefs and practices in high-access-to-technology classrooms. Paper presented at the annual meeting of the American Educational Research Association, Boston.
Eisenstein, E.L. (1979). The printing press as an agent of change. New York: Cambridge University Press.
Foster, J. & Julyan, C. (1988). The National Geographic Kids Network. Science and Children, 25(8). Washington, DC: NSTA.
Frederiksen, J.R., & Collins, A. (1989). A systems approach to educational testing. Educational Researcher, 18(9), 27-32.
Gearhart, M., Herman, J.L., Baker, E.L., Novak, J.R, & Whittaker, A.K. (1990). A new mirror for the classroom: Using technology to assess the effects of technology on instruction. Paper presented at Apple Classroom of Tomorrow Symposium, Cupertino, CA.
Harel, I. (1990). Children as software designers: A constructionist approach for learning mathematics. The Journal of Mathematical Behavior, 9(l), 3-93.
Hativa, N. (1989). Competition induced by traditional CAI: Motivational, sociological, and instructional-design issues. Paper presented at the annual meeting of the Amercian Educational Research Association, San Francisco.
Lampert, M., & Ball, D. (1990). Using hypermedia technology to support a new pedagogy of teacher education. National Center for Research on Teacher Education Issues paper, East Lansing, MI: Michigan State University.
Newman, D. (1990). Opportunities for research on the organizational impact of school computers. Educational Researcher, 19(3), 8-13.
Papert, S. (1980). Mindstorms. New York: Basic Books.
Pea, R.D. (in press). Distributed intelligence and education. In D. Perkins, M. West, J. Schwartz, & M. Wiske (Eds.), Teaching for understanding in an age of technology.
Postman, N. (1982). The disappearance of childhood. New York: Delacorte.
Reigeluth, C.M. (1987). The search for meaningful reform: A third-wave educational system. Journal of Instructional Development, 10(4), 3-14.
Scardamalia, M., & Bereiter, C. (in press). Higher levels of agency for children in the zone of proximal development: A challenge for the design of new knowledge media. Journal of the Learning Sciences.
Scardamalia, M., Bereiter, C., McLean, R.S., Swallow, J., & Woodruff, E. (1989). Computer-supported intentional learning environments. Journal of Educational Computing Research, 5(1), 51-68.
Schofield, J.W., & Verban, D. (1988). Computer usage in teaching mathematics: Issues which need answers. In D. Grouws & T. Cooney (Eds.), The teaching of mathematics: A research agenda (Vol. 1). Hillsdale, NJ: Erlbaum.
Scott, W.R. (1987). Organizations: Rational. natural and open systems. (2nd ed.). Englewood Cliffs, NJ: Prentice Hall.
Simon, H.A. (1987, Spring). The steam engine and the computer: What makes technology revolutionary. Educom Bulletin, pp. 2-5.
Sizer, T.R. (1984). Horace's compromise. Boston: Houghton-MiMin.
Toffler, A. (1970). Future shock. New York: Random House.
Tucker, M.S. (1989, June 21). Creating an 'entrepreneurial' school system. Education Week, p. 36.
Wattenberg, B.J. (1974). The real America. Garden City, NY: Doubleday.
Weick, K.E. (1976). Educational organizations as loosely coupled as systems. Administrative Science Quarterly, 21,1-19.
Wiggins, G. (1989, May). A true test: Toward more authentic and equitable assessment. Phi Delta Kappan, pp. 703-713.
Wolfe, D.P. (1987, December). Opening up assessment. Educational Leadership, 24-29.