In The Third Wave, Toffler identified three patterns in history. The first wave began some 10,000 years when someone planted seeds and ushered in the age of agriculture. Similarly, the Industrial Revolution had a dramatic effect. In 1900, virtually 85% of all employment was connected directly or indirectly to agriculture, but this is now less than 3% in Western nations. We have now clearly entered the third wave, although we do not know what to call it--technology age, digital age, information age, knowledge age--a period when the economy and social institutions will not be geared to muscle or machines but rather to intelligence and mind power. When the history of instructional technology is written it may well be one of the most important aspects of the third wave in a contexualist approach to history.The complete history of instructional technology has yet to be written. Instructional technology or educational computing has been overshadowed by the evolution of hardware and the many developments in programming languages. From the beginning instructional technology was a sideline in the history of computing. However, the developments in technology, especially in the last decade, make instructional technology more feasible than ever, in which case we may find people searching for the past to make a complete record. Pannabecker describes three approaches to the historical study of technology: Internalist, Externalist, and Contextualist. The internalist history is concerned with the artifact rather than on how it relates to social context. An externalist approach nearly ignores the artifact and concentrates on a broader study of social or political history. A contextualist approach shows the relationships of the internal design of specific technologies as they interact with a complex of economic, political, and cultural factors. It may be a long time before a comprehensive, contextualist history is written. As Pannabecker put it:
In contrast to historians of technology, technology educators do not do history as their primary occupation. The history of industrial arts was primarily internalist and was never as extensive in scope or depth as the history of technology. But while technology education is extremely broad in scope, the central interest of technology educators is education in and about technology, that is, how people teach, learn, and otherwise transmit technological knowledge and how people can learn to (re)construct technological artifacts and culture.The origins of instructional computing are found in the work of Sidney Pressey in the early 1900s and the later work of B. F. Skinner (who wrote The Technology of Teaching in 1968), both of whom conceptualized instructional programs on teaching machines. Pressey believed that many kinds of routine instructional tasks could be managed more efficiently by automation. Skinner believed that carefully constructed lessons with feedback, information following a response to permit a learner to evaluate the adequacy of the response, could lead to improved learning. The notions of Skinner were later migrated to computers rather than teaching machines for the obvious reason that computers could provide more control, flexibility, and feedback. Before this step, however, many kinds of programmed learning in booklet form were popular in American education in the 1960s, although they quickly disappeared. To be complete, any history of instructional technology should include a vast array of devices that preceded computerization or which ran parallel with it until the emergence of modern multimedia, and such things as the Edison Responsive Environment or "Talking Typewriter," should not be forgotten for the role they played in advancing instructional technology.The most significant breakthrough in modern computing has been miniaturization of electronic components, making it possible to reduce size and costs. The largest computer built in 1944 weighed 5 tons and contained vacuum tubes that had to be replaced daily. The transistor replaced the tube and made computers smaller and more efficient and finally the silicon chip displaced the transistor, enabling computers to become smaller, increasing their power, and reducing the amount of electricity required for operation. By packing chips with many circuits into a small space, desktop computers were possible along with a variety of other small components used in numerous products. The prevalence of fast, relatively inexpensive computer technology and software has made it possible for schools to use computers in ways that were inconceivable not so long ago.
The major impetus for the growth of desktop computing was the development of a program called VisiCalc, the first "killer app" that happened to be developed on the Apple II. Availability of the first true spreadsheet spurred on sales of Apple computers to businesses and individual users, which created an installed base of computers and stimulated the development of other software applications. In turn, this challenged other computer manufacturers to enter the market. A new cycle began with development of Lotus 1-2-3 and Excel by Microsoft.
In a broad perspective there are four periods of instructional computing: (1) instruction delivered on mainframes, (2) standalone microcomputers and peripherals, (3) networked computers, and the (4) Internet. While there are probably several ways to consider the history of instructional technology, we can use the following arbitrary divisions of instructional technology spanning four decades: (a) the work of Suppes, (b) the PLATO system, (c) the TICCIT system, (d) development of the microcomputer, and (e) the Internet. Some of the efforts were overlapping but are more easily described in separate categories. For timelines of developments in computer technology, see theComputer Society website,media history timeline, and information on early MUD history.
Most research on the early applications of CAI were conducted at three institutions, Dartmouth, Stanford, and the University of Illinois. Dartmouth developed a language, BASIC, that was originally used to make CAI. One of the first outgrowths were high school lessons in the Huntington Project at the Polytechnic Institute in Brooklyn and later at the State University of New York at Stoney Brook.
Some of the first applications of instruction on computers were developed by corporations for training using multiple-choice test banks and simple expository text. These were delivered by connections of computers, electric typewriters, and teletype terminals. The ability to develop better instructional programs depended upon development of more natural programming languages and authorware, such as an initial effort by IBM called "Coursewriter" in 1960. Patrick Suppes and his colleagues at Stanford University worked on this problem in earnest beginning in 1963. Their first emphasis was on a tutorial program for elementary students in mathematics, a drill program that operated over a terminal connected to a mainframe. Suppes designed software with feedback, lesson branching, and student record keeping. Suppes' experiments set the initial standards for computer-assisted instruction (CAI) or computer-based instruction (CBI), defined within the limits of the technology at the time. In 1968, Suppes was so enthusiastic that he believed most schools would have comprehensive computer systems to deliver instruction within a decade:
A computer can handle simultaneously a large number of students - for instance, 200 or more, and each of the 200 can be at a different point in the curriculum.At Stanford, Patrick Suppes and Richard Atkinson initiated a development project in CAI with a grant from the Carnegie Corporation, and later with support from the National Science Foundation and the U.S. Office of Education. Suppes and colleagues found that CAI reading materials produced an average increase in reading of 9 months for each 8 months of instruction (Suppes & Macken, 1978). At the college level, students using CAI scored statistically better on a final exam than a control group, and fewer students dropped out of the CAI course (Levien, 1972).Suppes imagined that individualized instruction would be available in a wide variety of subjects, that students would have individualized drill, along with a tutorial system and dialogue system. He believed that computers would be able to collect information about the individual learner and make decisions based on the response pattern so that corrective feedback could be tailored. Obviously, Suppes' predictions were hopeful and not realistic, although at the time his reasoning was logical. Today, many people believe that we are just a step away from a time when computers will play a substantial role in teaching and learning, but we may be more optimistic than realistic. Certainly the ingredients are there, but as before people will have to implement the available systems. They may not. In the 1960s and early 1970s, affordable, sophisticated systems were not generally available, schools found the idea to be of no practical value because there was no apparent need for CAI. The costs and problems associated with it for most classrooms made it impractical and not much more than an interesting oddity. The justification for using drill programs or other instruction would require superiority over ordinary instruction. At the outset, computer technology was envisioned like a "horse race" to see if students learn more, better or faster with computers than with daily teacher-led lessons.
At the same time that Suppes was working on CAI programs in math, the University of Illinois developed a computer-assisted project called PLATO (Programmed Logic for Automatic Teaching Operations), which was funded by Control Data Corporation and the National Science Foundation. The lessons for PLATO were developed by teachers.
In 1972, the Mitre Corporation of Bedford, Massachusetts and Brigham Young University implemented the Time-Shared, Interactive, Computer-Controlled, Information Television (TICCIT). TICCIT used mainframes and mini-computers with television connections. Lessons for TICCIT were developed by technical personnel and combined television, graphics, and testing.
Participants needed access to a mainframe computer through a terminal, called time sharing. A terminal was a keyboard and display screen, and time sharing was the concurrent use of a computer by more than one user, which in the days of the mainframe was often a costly practice. Mainframe time was sold by both the minute and the amount of computing resources devoted to a task, so companies that owned mainframes could make extra income.
During the 1970's, the National Science Foundation supported TICCIT(Time-Shared Interactive Computer Controlled Information Television) and PLATO (Programmed Logic for Automatic Teaching Operation), which became the first large-scale efforts to develop and test computerized instruction . The Educational Testing Service evaluated the effectiveness of PLATO and reported that the materials were not very effective. Students performed better in mathematics but in all other areas they were not superior to conventional instruction (Murphy & Appel, 1977). Achievement was significantly better than traditional instruction, but students preferred lecture classes.
The Time-Shared Interactive Computer Controlled Information Television System (TICCIT) at the University of California at Irvine was headed by Alfred Bork and supported by National Science Foundation funds, Aside from small achievement effects in mathematics and English, the TICCIT project was considered to be too costly to encourage development in CAI . The combined costs of PLATO and TICCIT amounted to $60,000,000, and extraordinary amount of money at the time (Hirschbuhl, 1980).A fault of TICCIT was inadequate feedback for students. Instructors were not involved in the project and may have communicated their dissatisfaction to students who participated. Clearly, the lack of importance of the project in the lives of students and professors undermined its acceptance, a factor today known as curriculum integration.
Headed by Donald Bitzer with funding from the National Science Foundation and Control Data Corporation, PLATO was developed to create CAI materials for college, secondary, and elementary levels (Coburn et al., 1985). The program also developed an authoring language to make it easier to develop educational programs without programming. The PLATO system was different, becoming a mass of hundreds of tutorial and drill-and-practice programs. Students accessed PLATO through mainframes. In assessing achievement gains, PLATO was equal or superior to conventional classrooms, but problems similar to TICCIT were found, mainly that in order to participate the students were isolated from other students. It was not important to their teachers, and the teachers and students viewed it as something "extra" rather than important to core activities of the classroom.
CAI was an isolated exception to instructional patterns, even in the mid 1970s. In 1972, the Educational Testing Service conducted a study to determine why CAI had not gain acceptance, and found that the costs, teachers' attitudes, lack of training personnel, extreme expectations, and limited applications framed the problem.
High costs were a major factor because of the need to use mainframe computers through school-based terminals, A long distance telephone connect was necessary to access the mainframe, which could cost several thousand dollars a year. The CAI connection was located in a specific room so that CAI was apart from classroom instruction and the traditional educational model found in schools, Specialized personnel were required to keep the process running and terminals required constant monitoring. Mainframe computers were notoriously unreliable with lots of "down time" that interfered with instructional schedules. Sometimes the students' worked would be deleted and teachers and students were reluctant to rely upon computers.
Suppes' CAI, PLATO, and TICCIT demonstrated computer-based instruction would work, although the cost was initially very discouraging, due to developmental costs, research, and time sharing. NSF support was approximately $65 million, a huge amount in 1970s' dollars. Although CAI worked, there was no real need for it in classrooms. Soon the enthusiasm waned and these "grand experiments" seemed to be dying out, which to many seemed to be the death knell for computerized instruction. But just as these projects were winding down the first commercial microcomputers were introduced to the market in 1975. Soon, a variety of stand-alone computers made computer software available without a mainframe and users could do other things that were too unwieldy or expensive with mainframes. Many schools purchased computers by Commodore,Apple,Atari, and Tandy. These computers could fit on a desktop, did not have to be connected to a mainframe, did not require special air conditioning or humidity controls, and were easy to operate. They use BASIC and had graphics capability with color and sound. At the time advocates were about to abandon hope that CAI would be possible in routine educational uses, the microcomputer made it possible. Soon the majority of software for instruction was delivered on microcomputers. The emergence of the micro renewed interest in CAI and stimulated commercial development of software.
- Apple II, the first personal computer, was marketed in 1977, followed by the Apple III in 1980, and the first machines with a mouse and graphical interface, the Lisa, in 1983. The Lisa, very expensive and ahead of its time, was not successful but the Macintosh, which included the Lisa features, was introduced in 1984.
- The Apple was originally made with borrowed parts from Atari. Atari was intent on developing games with high quality graphics and was unable to compete successfully in the personal computer market.
- The TRS-80 from Radio Shack was introduced in 1977, the Commodore PET in the 1980s, and the Osborne in 1981, the first commercially successful portable computer at 23 and a half pounds.
Elmer-Dewitt (1994) claims the Macintosh played a role in making society comfortable with computers due to the visual interface borrowed from Xerox, rather than DOS commands used on IBM computers. Beekman (1990) claims Macintosh initiated a new trend in authoring with the introduction of the HyperCard, which is similar to the concept of Hypermedia today. With increasing memory and RAM and dropping prices, competition between Apple and IBM and IBM clones provided schools with cheaper more powerful computers. The microcomputer made it possible for anyone to use software without the need for "time sharing" arrangements with mainframes through terminals, and this opened up instructional software development possibilities to many companies and individual programmers (Robinson, 1991).
- The IBM 286 AT was introduced in 1984 and, because of the faith corporations had in IBM, this computer became very popular and captured a large market share. IBM, however, continued to support its mainframes and did not follow the lead it gained with desktops. The IBM PC Junior could not compete successfully with the home/school computers (Commodore 64, Apple II, Atari 800).
While the early efforts of PLATO and TICCIT were concerned with using instructional software, which many perceived as replacing teachers, the use of microcomputers in schools quickly shifted away from instruction. The typical classroom teacher did not use a computer but many districts implemented a computer curriculum, created computer labs, and hired "computer teachers" to teach "computer literacy" skills. Most early emphasis was on programming. By the middle of the 1980s, half the states had a computer literacy requirement. BASIC programming was taught, even in elementary schools, followed later by LOGO. While elementary and secondary children were learning about computers and programming them, they rarely used computers for other school subjects. The emphasis was clearly on programming, but much instructional software became available, some of it inspired by the TICCIT and PLATO programs. Control Data Corporation acquired the PLATO software but attempted to continue to sell access via mainframes on proprietary terminals. Control Data sold off its businesses and reformed due to financial problems in the late 1980's. The availability of computers in homes and schools created a demand and a market for educational software that has been met by new companies, many of them learning lessons from the successes and mistakes of the TICCIT and PLATO experiments.
A number of companies offered instructional software on floppy disks for use in school programs. One of the largest and most successful groups was the Minnesota Educational Computer Consortium (MECC); begun in 1973, it provided numerous programs in all subject areas, and commercial companies were quick to follow. A similar effort was CONDUIT at the University of Iowa. In 1986 more than 7,000 commercially produced educational software packages were on the market (Jolicoeur & Berger, 1986). Today the number of educational programs is so large as to be virtually uncountable.
For a time the videodisc was regarded as the ultimate in instructional computing because of its superior multimedia applications and the encouraging research about student achievement. There was great anticipation before its release in 1979 as a digital device read by laser. Some research has shown the videodisc to be highly successful for instruction, but most schools found them to be too expensive to buy in large quantities. Today, of course, the internet makes it feasible to deliver similar content more cheaply, so the videodisc is dead technology. The CD-ROM is able to do what the videodisc could do, and the Internet makes the CD-ROM unnecessary for certain applications.
One of the early concerns about educational software was quality. Several states and private groups created software evaluation projects, but these died out quickly because there were too many software programs to evaluate. The original intent of software evaluation was to determine if a product was effective, or the extent to which it matched or surpassed traditional instruction as measured by achievement or gain scores on tests. In a research design students can be carefully matched with others in order that two or more groups might learn the same content under different treatments. Today, like books and movies, software evaluation is based on expert opinion by reviewers rather than research. Using an instrument or set of criteria, a professional educator or other expert reviews a piece of software and makes a subjective evaluation about its quality. (See this site for a list of online reviews). Research and expert opinion are highly different approaches to software evaluation. Obviously research takes time and costs money; expert opinion is easier to obtain. Today there has been a similar phenomenon with WWW web sites. A number of people and groups have proposed different evaluation schemes about good and bad web sites, but the task of evaluating them is impossible due to the sheer number.
Finding it difficult to fit computers into classroom routines, either because of the lack of enough computers or the resistance of teachers, computers remained relegated to labs, something that was reinforced in states where explicit certification rules were enforced for "computer" teachers. As long as schools employed "computer teachers" and kept computers in labs, the programming emphasis was predominant. For a period of time, schools were divided about what to do with computers, either to teach literacy skills (programming) or teach subject matter. However, federal funding became available forChapter 1 programs to purchase computers and software for disadvantaged children. This opened the market wider for sales of computer software and the specialization of comprehensive instructional learning systems (ILS). The ILS is similar to the PLATO and TICCIT approaches because teachers are ordinarily not involved in the instructional decisions and have little interaction. In fact, one of the oldest and most successful ILS companies is the CCC, started and still owned by Patrick Suppes. Van Dusen and Worthen (1995) reported that ILS programs have not delivered on their promised potential. While they are expensive, ILS programs may be used in the same way as PLATO and TICCIT, leaving the teacher out of the loop.
Now there are many options with technology. Computers can be used for drill, didactic instruction, simulation, and tutorials. The current emphasis has shifted from programming to "curriculum integration" of computers, with the expectation that teachers and students will use computers throughout the day for a variety of purposes. The recent push to "wire" all schools at the federal level and the promulgation of outcome standards for both teachers and pupils by the International Society for Technology in Education (ISTE) clearly reveal the new emphasis.However, until such time that teachers are either replaced totally by computers or taught to share instruction with them, technology will not fit well into most classrooms, because the system does not really permit it. Skinnerian programs, PLATO, TICCIT, the ILS, and current software and the Internet do not easily coexist with classroom teachers and traditional school organization and group instruction. There are still many classrooms where computers sit idle or where computers are still in their boxes. In some schools students engage in ILS activities in the "computer room" but work on entirely different curriculum objectives in the regular classroom. Most instructional technology has been considered to be a frill or an add on. Therefore, most research has concentrated on the use of computers in competition with classroom instruction (the "horse race").
Programmed instruction uses clearly defined content, presented to the student individually in small pieces, and then a question is asked and must be answered in order to proceed. The approach is the basic structure for software design in CAI as stimulus-response patterns (Case & Bereiter, 1984; Poppen & Poppen, 1988). Teachers, however, were not involved with the presentation and were only able to watch as their students went through the lessons. In effect, the teacher was removed from instruction in much the same way as they were in PLATO and TICCIT. A major reason for the disuse of programmed learning was that teachers often interjected themselves into the process or lessened the time students could use the materials. The ILS is usually housed in a separate lab and students are pulled out to go to take there lessons and tests, but often there is no connection between the activities in the ILS and what happens the rest of the day in regular classrooms.
Much of the software was considered to be "drill-and-practice" and attracted the criticism of many writers. Schools were also soon criticized for teaching programming, which seemed like a waste of time to many people, which was added to the criticism for using drill software (Merrill et al 1992).
The availability of so-called productivity tools--word processing, database, and spreadsheet--boosted computer options and popularity in schools. Rather than just programming or using CAI, students could learn skills that might improve their school performance and thinking abilities, and teachers could learn tools that might help in classroom management and instruction. Productivity tools became important for teachers and students to learn. These are still considered to be the most important basic skills, or the current definition of computer literacy promulgated by the International Society for Technology in Education in NETS.
Introduction of the local area network (LAN) in businesses influenced schools to connect PCs (Hayes, 1995), making it possible to conceive of computer-managed instruction. The greatest appeal of the LAN is a centralized system of software management and the ability to use e-mail and have easy access to the Internet. Only 5% of schools had LANs in 1991, but today there is a push to get LANs in all schools and to have all schools connected to the Internet. The Department of Education spearheaded the plan for wiring with the assistance of President Clinton and Vice-President Gore. Clinton proposed that every classroom be connected to the Internet by the year 2000 in his 1996 State of the Union Address.
The multimedia applications of current computers have expanded their popularity and capacity for instruction, such as the CD-ROM, Internet, and other applications that allow students to create their own products, such as PowerPoint, Hyperstudio, and graphics programs. Multimedia software appeals to all the senses and stimulates high interest, appealing to students and teachers (Mitchell-Powell, 1995). Many believe that technology can be used to revolutionize learning, a belief dating back to the early days of microcomputers and time sharing (Molnar, 1980). Despite this, successful applications of computers are limited because of the need to update regularly, straining budgets (Henry, 1997). Teachers are not trained to integrate technology into instruction, and instruction remains organized for large groups. So the increase in the number of computers does not mean that teachers are using them effectively (Wiburg, 1995; Becker, 1992).
Even when teachers are involved in the process, however, they must want to use the technology, gain the skills, and have support for it in the school. Evans-Andris (1995) reported that teaching styles have critical implications for how computers are used. A teacher's prior practices and philosophy influence integration (Miller & Olson, 1995). Dwyer (1994) reported that significant changes were found in classroom environments and teaching methods after 8 years of use. Brady (1991) reported that the most effective computer using teachers had 13 years of teaching experience and worked in schools where technology was promoted. These classrooms were regarded as constructivist in that students were said to take more responsibility for their own learning, teachers worked as mentors, and there was more collaboration. The current philosophy is that computers can be used for students to show what they know rather than being limited to presentations by the teacher. Many people seem to want a new kind of school that departs from chalk, paper and pencil. Often teachers are blamed for not accepting technology and change, but even many parents and pupils do not really want to change a system that people seem to understand. There is a very conservative view of education and parents are the most conservative.
Jonassen, Carr, & Yeuh (1998) say that databases and spreadsheets can be used to integrate content areas and that students can organize and display information with these tools in ways they cannot with paper and pencil. Authoring tools such as LOGO, HTML, and HyperStudio provide teachers and students with concrete tools for building understanding and enable constructivist learning (Barab et al, 1998). In fact, much of current thinking about the future of computer instruction is based on constructivist hopes rather than programmed instruction (Morrison & Lowther, 1997; Barr, 1990; Thornburg, 1995; Muir 1994; Adams & Bailey, 1993; Strommen, 1995; Reil 1994; Wilson & Marsh, 1995; Means & Olson, 1994; Elmer-Dewitt, 1991; Ratnesar 1998). At least one seems willing to attempt the impossible, combining programmed Instruction and constructivism (Karen Smith-Gratto).
As pointed out in the section on the efficacy of instructional technology, there is ample evidence that computers can be used to achieve educational goals with higher achievement scores, shorter time to criterion, and reduced costs. The social context of schools are more important factors determining acceptance of technology. As long as the "horse race" mentality restricts computerized instruction to a simple competition with regular classroom instruction, the focus will be on gain scores in math or reading, not on the many ways that computers can be used in classrooms to support both students and teachers. And critics will still see computers as taking money away from other programs in the school and offering nothing of value (The Computer Delusion by Oppenheimer).
A significant problem with technology in education is dependability. A looming crisis in most schools is the lack of technical personnel to keep technology working and running. According to Simitus, in the early 1900's Daimler-Benz commissioned a study to determine the world-wide demand for automobiles, which turned out to be one million; "the controlling factor was the number of households that could afford chauffeurs, since in those days, one could not drive very far without needing the services of a mechanic as well as a driver to keep 'the machine' running." Most schools do not have an adequate support staff for their computers and infrastructure sufficient. I have seen schools where teachers have given up on computers because they cannot get the school district to fix them in a timely manner. Perhaps with more advances in technology we may be able to run our computers with only a few pit stops. But for now the greatest problem for computer acceptance in "curriculum integration" is not its applicability to school-related tasks but rather the inability to get service and keep it running. Long-term funding for technology is essential for maintenance and repair, scheduled replacement, training, and support. There must be a plan to assure that technology will not become obsolete or unusable.