Free Materials

Education Would Be Better If...

CAI or Teachers? Not Either/Or — But Both!

Moursund, D.G. (October 1988). CAI or teachers? Not either/or — but both! The Computing Teacher. Eugene, OR: Internatoinal Council for Computers in Education.

At the spring 1988 annual conference of the Northwest Council for Computer Education the keynote presentation was a panel discussion by Karen Billings, Sylvia Charp, David Thornberg, Tom Snyder and myself. LeRoy Finkel was the moderator, and the central focus was the future of computers in education.

The initial part of the discussion was a brief presentation by each panel member. The various points of view were mostly upbeat and can be summarized by:

1. Computers in education are a good idea and progress is continuing.
2. Computer-as-tool is great. (This was the main point I made, and it was reinforced by other speakers.
3. Routine CAI drill and practice has proven quite useful.
4. Empowering the teacher, and focusing on how to make effective use of one computer per classroom, is a great idea.
5. Teachers are wonderful. The human-to-human interaction of teacher with student is at the core of quality education.

A variety of questions from the audience focused on the same issues. Each comment about maintaining the current central role of teachers brought cheers from the audience.

As the discussion progressed, I found myself growing more and more frustrated. Two major themes had been ignored in the initial presentations and were being ignored in the discussion. One was the issue of whether students in the future will be learning any solid computer science and computer programming. Surprisingly, no panelist made a prediction in this area and no member of the audience raised the question. But that contributed only modestly to my feeling of frustration.

The second major theme that nobody seemed willing to raise was that of computer assisted instruction as a vehicle for presenting curriculum units or entire courses. Sylvia Charp, who is a strong proponent of CAI, had focused on supplemental drill and practice in her presentation. At an opportune time, I mentioned the topic and suggested that it will gradually produce a massive change in education. Sylvia Charp cheered, several other panel members immediately jumped into attack mode, and many of my former and current graduate students blanched. I was pleased that my statement had brought increased life to the panel presentation.

As the discussion continued it became clear that many people view CAI in an either/or mode. That is, they think of CAI and our traditional educational practices that make little use of CAI as being in direct competition. Either we maintain our current system or we have CAI. They do not acknowledge the fact that we already have both in many schools.

Those who oppose CAI then go on to paint a frightening picture of children spending all day chained to a soulless, inhumane machine that assumes full responsibility for their education. Many of us are brought to the verge of tears just thinking about what a terrible thing this would be for our children.

Those who favor CAI tend to talk about increased rates of learning, teacher productivity, individualization of instruction, and an increased range of learning opportunities. The picture of children learning more, better, and faster, and achieving their full potential, is heart warming.

Surprisingly, the panel discussion never got beyond these two extremes. There was no suggestion that a compromise position might be appropriate. It seems inevitable to me that during the next two decades our school systems will gradually move toward making substantial use of CAI. However, during that time human teachers will continue to play a dominant role in the overall educational process. Computers will gradually fill roles that they do better than humans. Humans will gradually move in the direction of filling roles that they do better than machines. We will have both humans and computers deeply involved in the instruction of our children.

I enjoy discussing which aspects of instruction might gradually be relegated to computers, and which aspects are best preserved to human teachers. The human brain is a wonderful thing, and there are many things that humans do far better than computers. Perhaps the most important of these is having a deep understanding of what it is to be a human being. This includes understanding human verbal and nonverbal communications systems. The very best work of researchers in artificial intelligence has not yet begun to develop computer systems that even show signs of eventually leading to systems that have such human abilities. Thus, to the extent that teachers are making use of these human abilities, they can far outperform the very best of current CAI systems.

But much of the educational process is not based on intimate, one-on-one human interaction that requires use of these human communication abilities found in all teachers. We cannot afford an educational system in which there is one human teacher for each student. Moreover, it is essential that students learn to learn from books and other resource materials, such as computerized information retrieval systems. Routine drill and practice is an important part of education. CAI can provide rich simulations, opportunities for trial and error explorations requiring higher-order cognitive processing, greater opportunities for individualized instruction than most current classrooms provide, and so on.

It seems obvious to me that our educational system would be better if it were based on a combination of well-prepared and dedicated teachers and an abundance of high quality CAI. The cost of providing a computer for every student and a wide range of CAI materials is quite modest compared to our current educational expenditures. If we devoted five-percent of current annual school budgets to this task, it would soon be accomplished. I strongly believe that we should be working toward this objective.

Standardized Testing and Computer Assisted Instruction

Moursund, D.G. (November 1988). Standardized testing and computer-assisted instruction. The Computing Teacher. Eugene, OR: International Council for Computers in Education.

There is one sure way to get a rise out of the students in my graduate computer education courses. Just mention standardized testing and the increasing role it seems to be playing in education. Most of my students become quite agitated in thinking about this, and some become downright hostile towards the school systems where they work.

Students face a barrage of standardized tests, beginning in grade school and often continuing on into graduate school. Moreover, some teachers are now being evaluated by how well their students do on standardized tests. Increasingly, teachers themselves are being required to take standardized tests, either to obtain a teaching certificate or to maintain their teaching certificate.

The educators I work with give a variety of reasons why they are troubled by the major emphasis on standardized testing. Reasons given include that such testing is a waste of time, irrelevant to the curriculum, focuses too much on lower-order skills, and is a major force moving education in an inappropriate direction. The tests seem to be driving the curriculum—teachers are teaching to the tests and students are studying methods specifically designed to raise their test scores.

Interestingly, I pick up nearly similar feelings of disquiet and fear when my students discuss computer-assisted instruction. Much of the CAI material is rather superficial, focusing mainly on lower-order skills. Deeper aspects of the human elements of teaching remain elusive to most CAI developers. There is a distinct possibility that eventually the content of CAI-based courses will become the curriculum.

Standardized Testing

Generally I maintain a neutral stance in discussing standardized testing. I have some understanding of the processes that have been followed in developing and evaluating the test items. I know a little about validity and reliability. And, of course, I understand some of the roles that computers now play in the overall process of developing standardized tests.

In recent years computers have played an ever-increasing role in standardized testing. Two trends are evident. First, there are large databanks of possible test questions, along with item analysis and other statistical data that have been gathered through use of the test items. Thus, it is growing easier to create standardized tests or other tests with specified characteristics. Second, an increasing amount of testing is now being done online. In one type of online testing, called adaptive testing, the computer system adjusts the selection of questions to the particular person being tested, making changes based on performance during the test.

Adaptive testing has many characteristics of computer-assisted instruction. Indeed, much of the CAI that is currently available can be considered as tests, with some feedback and perhaps some remedial instruction being provided while the test is being taken.

Perhaps it is the close similarity between objective testing and routine drill and practice CAI that agitates so many of my students? In both cases, a large part of education seems to be reduced to a lower-order skills, multiple choice or short answer format. The multidimensional aspects of a good student/teacher rapport are missing, along with much of the richness of a good classroom environment. Many educators find this objectionable. They know education has many important dimensions that cannot be measured through such a testing format.

Coachability of Objective Tests

Recently I read None of the Above: Behind the Myth of Scholastic Aptitude written by David Owen. In large, it is an attack on the Educational Testing Service and their widely used test, the Scholastic Aptitude Test (S.A.T.). But at a deeper level it questions all standardized tests. It is a powerful book, and I strongly recommend it to all educators.

There are a number of important points discussed in Owen's book. One is the nature of the standardized test questions themselves, and the fact that many questions are subject to multiple interpretations. Thus, one has to have or to develop a mindset somewhat similar to those who create the questions in order to interpret the questions in a manner leading to "the correct" answer.

But a deeper problem that Owen raises is the "coachability" of standardized tests. It is possible to teach to the test or to coach students so that they will do well on a particular test. A number of companies publish books that are designed to help students improve their test taking ability, and many of these books are geared toward a particular test such as the S.A.T. Indeed, there are now a number of pieces of software designed for the same purpose. Some companies advertise the purchase price will be returned if the user doesn't make a certain specified gain in their S.A.T. test score.

Owen discusses several companies that run short courses specifically designed to help students learn to make higher scores on specified standardized tests. In these courses, students learn a wide range of tricks, almost none related to increasing their understanding of the material being tested. It turns out that because of the way standardized tests are created and the way that the test constructors think, it is possible to correctly guess answers to many questions without even reading the questions!

Earlier in this editorial I suggested that the feelings my students have about standardized testing and about CAI seem to be similar. Owen's has increased my understanding of this issue. The real world does not consist of a sequence of objective questions, where success is measured by one's ability to select the one correct answer from a short list of choices. But both standardized testing and most of the currently available CAI view the world in exactly this manner. Thus, both foster teaching to the test, teaching objective test taking skills, and rewarding students for developing a good objective test mentality.

A Confrontation?

The problem of an objective text approach to education is not easily solved. Objective testing has become institutionalized, and it is now a driving force in our educational system. Moreover, most currently available CAI seems designed to contribute to this approach to education.

I suspect that eventually there will be a major confrontation between the forces that support standardized testing, objective testing, and objective oriented CAI, and those who feel that this represents a major threat to education. Currently I side with the latter group.

References

Owen, David. None of the Above: Behind the Myth of Scholastic Aptitude. Houghton Mifflin Company, 1985.

Problem Solving.

Moursund, D.G. (Dec./Jan. 1998/99). Problem Solving. The Computing Teacher. Eugene, OR: ICCE..

Of all creatures on earth, humans are the best at creating and solving problems. One of the main goals of education is to help students become even better at these endeavors.

I have been deeply interested in problem solving for many years, and I have spent much time studying this field. During the past two years, I have presented a number of workshops on the topic of appropriate roles of human brains and computers in problem solving (Moursund, 1988). Here are some of the ideas from my workshop; you may find a number of them useful in your teaching.

1. With the exception of some people who are severely brain damaged, all people have a substantial ability to create and solve problems. The human brain is designed to be quite good at this endeavor and does it routinely.
2. While some people have more natural talent than others, all people can get better at problem posing and solving through study and practice.
3. One of the most important goals of education is to help students improve their ability to pose and solve problems. Both problem posing and problem solving are higher-order cognitive skills.
4. Teachers can play a significant role in helping students to improve their problem posing and solving skills.
5. To be good at posing and solving problems within a particular domain, one needs both considerable general knowledge and a great deal of knowledge and skills specific to the domain.
6. Most real world problems are interdisciplinary. Posing and solving such problems requires a broad range of knowledge and skills from many different disciplines.
7. There is a considerable body of knowledge and skills useful over a wide range of problem posing and solving. Transfer of such knowledge and skills from an initial learning environment to a variety of somewhat different application environments is relatively difficult for most people. However, we know how to teach in a manner that helps to increase transfer.
8. Computers can solve or make a significant contribution to solving a large number of problems. The number and scope of such problems will continue to grow quite rapidly through research, advances in computer science, and advances in computer hardware/software.
9. There is some good computer software designed to help improve students' ability to pose and solve problems, and the amount of this type of software is growing.

The list could easily be extended. The point is, we know a lot about problem posing and solving. A number of people have taken some of this knowledge and created courses that can be taught at the precollege or college level. An excellent survey of eight of these courses is given in Chance (1986). Many such courses have been implemented and are backed by substantial research. In the opinions of the course creators and their followers, these courses work.

You might ask, then, why don't all students encounter such courses as part of their regular academic programs of study? There are several possible answers, and perhaps the following three capture the spirit of the most common ones.

1. The courses are too general and do not focus on any particular discipline. They do not fit well within the domain specific nature of our current curriculum.
2. It takes considerable knowledge to teach problem posing and solving. Relatively few teachers arc prepared to teach this general topic in a wide-ranging, interdisciplinary manner that facilitates transfer of learning.
3. Problem posing and solving is so domain specific that it is best integrated into the existing disciplines, and this should be done by the teachers of these disciplines.

The latter point is particularly interesting. It is an argument that students don't need specific courses on problem posing and solving because they already receive such instruction in all their courses. All teachers teach problem posing and solving. (What teachers would be willing to admit they don't do well in this regard?)

But this approach leaves us with a difficult dilemma. We know computers arc a substantial aid to problem solving. Thus, we might expect that all teachers would teach the appropriate roles of computers as an aid to problem solving within the disciplines they teach. But this is terribly inefficient for two reasons. First, it requires that every teacher have a good understanding of the computer's role in solving the problems within their discipline. This is a worthy goal, but it will not be achieved with the majority of current educators.

Second, this approach leads to considerable duplication of effort. There are many rudiments of computer use that easily transfer from one discipline or application area to another. It is not appropriate to expect all teachers to start from scratch in teaching their students to use a computer as an aid to problem solving.

I have two conclusions. My first conclusion is that all students should learn the rudiments of using a computer quite early in their educational careers. Certainly, students could be well grounded in using a word processor, database, and presentation graphics by the time they enter middle school or junior high school. All teachers at the middle school and higher could then build upon this initial level of computer knowledge.

My second conclusion is that there is a need for a course in problem posing and problem solving that takes into consideration capabilities and limitations of computers. I believe such a course should become part of the regular curriculum for all students. Such a course would require a reasonable level of maturity (some functionality at the Piaget level of formal operations) on the part of students. It would contain material and ideas that should be practiced and used over several years of schooling. Thus, the course might best be offered at the eight or ninth grade level.

There is substantial non-computer material available for use in courses on problem posing and solving. And, of course, there is substantial computer-oriented material. Thus, there is ample material for a year-length course. But even a half-year course would make a significant contribution to the education of most students.

I'd like to see such a course become common in the middle school or junior high school curriculum. If you are teaching such a course or are aware of such a course, please send me information on it.

References

Chance, P. (1986). Thinking in the classroom: A survey of programs. New York: Teachers College Press.

Moursund, D. (1988). Computers and problem solving: A workshop for educators. Eugene, OR: ICCE.

The Computer and Problem Solving: How Theory can Support Classroom Practice

Yates, Billy C. and Moursund, Dave (Dec./Jan. 1988/89). The Computer and Problem Solving: How Theory can Support Classroom Practice. The Computing Teacher. Eugene, OR: ISTE.

Introduction

Human beings, with their powerful brains and spoken language, have tremendous ability to create and solve problems. They are particularly good at developing aids to problem solving and teaching their children to use these aids. Some powerful and widely used aids to problem solving include reading, writing, arithmetic, and computers. Since one of the major goals of education is to help students get better at problem solving, it is natural that computers are of growing importance in education.

For the first time in the history of humankind, we have a machine that can emulate some of our own thought processes and therefore solve certain problems that in the past could only be solved by people. This has led to increased emphasis on teaching students to think about thinking; the field of metacognition is flourishing.

We believe that every teacher should be concerned with the classroom applications of computers as an aid to teaching, learning, and doing problem solving. Teachers should also be attentive to the educational research on computers and problem solving. The results from such research can guide and strengthen effective classroom practices or call into doubt potentially ineffective instructional methods.

Problem Solving: Some Key Ideas

There are a number of general problem solving heuristics or strategies advocated by researchers and educators that, while not applicable across all disciplines, do seem to capture some qualities that make them useful in more than one discipline. The research literature supports the contention that all students and teachers should gain a working understanding of these ideas, since doing so will likely increase their ability to solve problems.

Many of our formal ideas about problem solving can be traced back to Dewey (1910). Some early researchers (cited in Best, 1986) believed four steps were typically used in solving a problem: preparation, incubation, illumination, and verification. More recently, the mathematician George Polya (1968) suggested a series of general problem solving steps: understand the problem, devise a plan, carry out the plan, and look back to analyze the solution.

More specific to primary and secondary education, Moursund (1988) has synthesized much of the literature and has suggested a problem solving approach that combines certain aspects of John Dewey's philosophy and Polya's model of problem solving. He defines formal problems as having four qualities: givens, guidelines, goal, and ownership. The givens of a problem are what is known about the problem at the beginning. Guidelines are the steps or rules that can be used to work toward the end state or goal. The goal is the desired end result or situation. The last component, ownership, requires that the person working to solve a problem have some personal investment in its solution. As most problems don't come to us with these qualities delineated, Moursund suggests that a key idea in problem solving is developing a clear understanding of the givens, guidelines, and goal.

Research on Human Problem Solving Characteristics

There has been substantial research on problem solving. We list here 19 statements about problem solving with implications for the educational use of computers. Each one has relatively strong support in the research literature. While you can probably find counter arguments or contrary positions to each assertion, there is enough evidence to convince many educators to consider these ideas when making educational decisions.

1. To become an expert in a particular area requires both talent and at least a dozen years of hard work (Bloom, 1985). We need to provide a rich intellectual environment, so a student gains the necessary experience with more and more complex problems within the discipline. It seems evident that in many problem-solving areas, some of the needed experience can come through the use of computer simulations, and that sometimes this is a cheaper and safer approach.
2. Problem solvers who talk about the steps they are taking to solve the problem do better than those who do not describe their efforts (Berry, 1983;Carey&Tarr, 1968). Computer labs should not be "Quiet Zones." Students should be encouraged to talk to them selves and others as they work to solve problems.
3. How we think about (or represent) a problem is a better indicator of the problem's difficulty than any quality intrinsic to the logic of the problem (Fischler & Firschein,1987; Fredericksen, 1984). Students should learn to make use of a variety of different representations for a problem, including computer representations. Database, spreadsheet, and graphing software provide rich opportunity for generating different problem representations and examining the types of problem-solving approaches that can be drawn from a given representation.
4. Problem solving skills used in groups do not necessarily transfer to individual problem solving skills (Bender, 1986).The research literature supporting cooperative learning is strong (Kohn, 1987). The human interactions that come from working in pairs or as a group during a simulation or other computer-use experience are very important. However, instructional time also needs to be devoted to individual acquisition of problem solving skills, especially if the primary goal is individual problem solving skill.
5. Even with well defined problems, people tend to frame small subgoals, and may not be able to explain why they did so (Greeno, 1976). When a problem seems complex and perhaps overwhelming, the idea of systematically breaking a big problem into smaller pieces becomes very important. The computer can be used to explore different options (subgoals). When the problem context is appropriate, the use of spreadsheets and word processor can be used to allow different options to be generated and examined in a short period of time.
6. How conscious we are of our thinking processes while solving a problem is dependent on whether we are using a familiar strategy or developing the strategy as we work (Kellogg, 1982). Careful, conscious thinking about problem solving processes helps to improve one's problem solving skills. One model of how people solve problems is that they look for patterns that seem familiar and then apply standard strategies that they have previously found useful when the patterns occur. The capacity of the computer to allow for quick and varied representations of problems (e. g., graphs) may support this type of problem solving by students.
7. We seem to have a few basic general problem solving strategies for dealing with a variety of problem situations (Simon and Simon, 1962). Without specific instruction, many students will fail to develop some widely useful strategies. If we can help students acquire just a few additional strategies, we may make a major contribution to their overall ability to attack a variety of problems. Using the computer to model an event (e. g., simulations) and then providing them with problem solving strategies to apply in that or similar situations can give them a richer repertoire of problem solving strategies.
8. Precise thinking (processing) is one of the keys to strong problem solving ability (Whimbey, 1984). Precise thinking may be represented orally, in writing, in applications software, or through one of the many computer languages available. The main point is that precise thinking is a necessary part of every academic discipline in which people have ideas and points of view they want to communicate. Practicing such careful communication improves one's ability to solve problems in these disciplines.
9. Changing our perspective on a problem often aids in arriving at a solution (Scheerer.1983). Computers are a powerful and versatile aid in dynamically representing and re-representing data. Databases, spreadsheets, or graphics software can be used to represent and manipulate data to give students practice in examining a problem from different perspectives.
10. Experts outside of their domain of expertise do no better than novices. Good problem solving ability in one area does not carry over to problems in another area (Hayes, 1981; Royer, 1979). Students need a broad and general education in order to understand and cope with the variety of problems they are apt to encounter. Many problems they encounter will relate to technology, and more specifically to computer technology. All students should acquire basic computer literacy.
11. We improve as problem solvers with experience, but we seem to have difficulty transferring that knowledge to analogous problems in other domains (Reed et al, 1974). Using databases to examine different trends in varying contexts, such as economics and politics, helps increase transfer skills. More generally, all teachers should teach for transfer. When teaching the use of computer tools, they should give ample examples and time for practice in applications to a variety of disciplines.
12. On certain types of problems, the computer has been shown to be a strong problem solving aid (Pogrow, 1985; Stemberg and Davidson, 1982). It is important that students understand the capabilities and limitations of computers as an aid to problem solving within each domain that students study, as well as in a domain free sense. This means that within each domain, the teacher is responsible for helping students to gain domain specific knowledge of roles of computers in solving the problems of that domain.
13. Groups of 2-4 individuals are better than larger groups or individuals at problem solving activities that use computers (Cox and Berger,1985). This is a statement consistent with other research about cooperative problem solving in a computer environment. It is a statement that "two heads are better than one" and it is also a suggestion that too many heads interfere with each other's work.
14. The more you know about a particular area or the more knowledge you have about a subject, the better problem solver you tend to be on problems within that subject (Best, 1986). There appears to be no substitute for knowing the subject matter domain of a problem. The research in computer-assisted learning indicates that the computer can do as well and often better than traditional methods in helping students to gain basic knowledge and skills over a broad range of subject areas.
15. No general problem solving heuristic applies equally to all disciplines (Groneretal., 1983; Anderson, 1987). The basic modes of inquiry, thought, and problem solving vary greatly across domains. We know that students will get better at this transfer if they receive specific instruction and practice in it. The use of databases, spreadsheets, word processors, and graphing software can be useful in practicing these skills.
16. A prodigious memory may not enhance problem solving skills; but rather, a set of carefully refined problem solving strategies can be a more significant influence on performance in a given domain of discourse (Simon and Simon, 1962). Information retrieval, with and without the use of computers, is an effective substitution for memorization in many cases. In addition, by modeling events using computer simulations, students can develop, test and refine strategies specific to that problem area and begin to refine problem solving strategies specific to a given domain of discourse.
17. Students who have a positive attitude toward a problem-solving task do better than those who have a negative attitude (Armstrong and Me Daniel, 1986; Ross, 1983). The computer can be a strong motivational element in the classroom by providing a rich world to explore using Logo and other programming languages, games, and software that allow for the constructive interaction between students, and other software that helps facilitate enjoyable interaction with the computer.
18. Problem solving in computer programming doesn't easily transfer to problem solving in other disciplines (Jansson et al, 1987; Ross, 1983). (Also discussed later in this paper.) The teaching of computer programming is not a panacea to improving problem solving skills. Computer programming is an important discipline in its own right, and it could well be an appropriate topic to include in the general precollege curriculum. But doing so may not contribute substantially to the overall problem solving skills of students.
19. Short term, active memory is quite limited in capacity, and is estimated to hold 7 ± 2 "chunks" (Miller, 1957). Computers add a new dimension to the idea of chunking. If we teach students what types of problems are readily solved by computers (e. g., data management tasks of sorting and searching), then we provide them with "chunks" of expertise useful for solving other problems. If we can break a large problem into component problems of the sort that a computer can solve, then we can more efficiently solve a larger problem. In a broad sense, we can consider an artificially intelligent expert system to be a chunk. Increasing availability of such massive and powerful chunks will gradually have a major impact on education.

In summary, these research findings taken as a group suggest that the computer can support and augment the human problem solver. The skillful use of computers in an instructional setting can give the student exposure to problem solving methodologies and a good environment to practice these skills.

Specific Literature and Ideas on Computers as Aids to Problem Solving

While the general literature on problem solving is quite extensive, the specific literature on roles of computers in problem solving is rather limited. This section gives the flavor of such research. It provides an indication of some of the things we are beginning to understand about computers and problem solving. However, it is clear that much more research is needed.

Taylor (1980) divides instructional uses of computers into the categories tutor, tool, tutee. In each of these categories, computers are a significant aid to problem solving.

The Computer as Tool

One study using spreadsheets examined the role of productive thinking and problem solving. Productive thinking is defined as thinking based on a good comprehension of the problem rather than on rote memorization of facts and figures. Students reported that they experienced greater learning with integrative work using the computer than with repetitive non-integrated assignments involving drill and practice (Borthick and Clark, 1986).

A study by Steinberg et al. (1986), using problems requiring the storage of large amounts of information while applying problem solving strategies, indicated that the computer was a helpful aid to problem solving.

The results of a study by Dubitsky (1986), using an electronic spreadsheet to solve algebra problems, indicated the students were able to understand the workings of the spreadsheet and devised systematic problem solving methodologies and were able to transfer these skills from problem to problem.

Each of the studies cited give support to the proposition that computerized data management systems like spreadsheets and databases should be a strong problem solving teaching tool. One study looking at databases confirms this notion. Students using a computer database can find relevant information, determine if it is sufficient to solve the problem, and sort the information in a way likely to produce a solution better than students using traditional paper and pencil methods (White, 1987).

The Computer as Tutor: Simulations

Simulations that use guided discovery seem to be the best use of a computer simulation as measured by tests of scientific thinking and critical thinking (Rivers and Vockell, 1987).

Studies involving computer simulation in chemistry classes have indicated that the science skills learned using a simulation are as effective as more traditional noncomputer methods and often take less time (Choi, 1987). In a summary article on computer lab simulations, Wells (1985/86), concluded that computer simulations of lab experiments can be as effective as other instructional methods involving labs and paper and pencil exercises.

McClurg (1985) and Yates (1988) studied the use of problem solving software to help improve spatial visualization skills of students. Both studies provide support for use of software involving spatial visualization activities.

The Computer as Tutee: Programming

A 1985 study done by Clements indicated that the Logo group did significantly better than the CAI and control groups on the ability to decide the appropriate domain in which to solve a problem and on solution processes (cited in Massialas & Papagiannis, 1987; for related work see Clements, 1988/89). Another study by Rieber (1987) examined students' abilities to problem solve after being exposed to Logo in a guided discovery educational environment. The results indicate that the Logo group performed better on the problem solving measures than the control group. However, other studies have not shown positive results (Rieber, 1987).

Rose (1983) examined the effect of teaching BASIC programming on tests of logic and problem solving. The results indicate that hypothesis testing was significantly better for the experimental group but no other logic or problem-solving effects were observed.

A study conducted by Jansson et al. (1987) tested the hypothesis that computer programming would improve performance on conditional reasoning tasks. Three separate experiments were carried out using Logo, BASIC, and Pascal. No significant results were found.

Their investigation appears to place the burden of proof on the shoulders of the advocates of computer programming as a means of developing certain thinking skills.

Summary

The computer as a tool is a strong problem solving aid. Teachers using databases can teach important information processing and problem solving skills to students. Word processors and spreadsheets are also important problem solving tools.

Simulations placed in an environment where guided discovery is fostered and metacognitive activities are encouraged show strong indication of fostering problem-solving skills.

The Logo environment is consistent with much of the literature on human problem solving, but other programming environments have little research support that indicates they foster general problem solving skills in students.

[Billy C. Yates, Teachers College, Emporia State University, Emporia, KS 66801; Dave Moursund. ICCE, University of Oregon, 1787 Agate St., Eugene, OR 97403.]

References

This includes only sources cited in the body of the paper. Readers interested in a complete list of references used in researching this paper may contact Bill Yates at the address above.

Anderson, J. R. (1987). Skill acquisition: Compilation of weak method problem solutions. Psychological Review, 94, 192-210.

Armstrong, P. & McDaniel, E. (1986). Relationships between learning styles and performance on problem-solving tasks. Psychological Reports, 59, 1135-1138.

Bloom, Floyd et. al. (1985). Mind and behavior. New York: Free man.

Borthick, A. P., & dark, R. L. (1986). The role of productive thinking in affecting student learning with microcomputers in accounting education. The Accounting Review, 67(1), 143-157.

Carey, M..Foxman, P. N.,Tarr,D.B.(1968), Verbalization, experimenter presence, and problem solving. Journal of Personality and Social Psychology, 8, 299-302.

Choi, B, & Gennaro, E. (1987). The effectiveness of using computer simulated experiments on junior high students' understanding of the volume displacement concept. Journal of Research in Science Teaching. 24(6), 539-552.

Clements, Doug (1988/89). A series of articles on Logo and problem solving, with the first article appearing in the October 1988 issue. Logo Exchange.

Cox, D. A. & Berger, C. F. (1985). The importance of group size in the use of problem-solving skills on a microcomputer. Journal of Educational Computing Research, 1, 459-468.

Dubitsky, B. (1986). Algebraic problem solving in grade six: An application of the computer spreadsheet. Doctoral dissertation, Columbia University Teachers College, DA8704291.

Fischler, M. A., Firschein, 0. (1987). Intelligence: The eye. the brain, and the computer. Menlo Park: Addison-Wesley Publishing Company.

Fredericksen, N. (1984). Implications of cognitive theory for in struction in problem solving. Review of Educational Research, 54, 363-407.

Greeno, J. G. (1976). Indefinite goals in well-structured problems. Psychological Review, 83, 479-491.

Groner, R., Groner, M. & Bischof, W.F. (Eds.). (1983). Methods of heuristics. Hillsdale, NJ: Lawrence Eribaum Associates.

Hayes.J. R. (1981) The complete problem solver. Philadelphia: The Franklin Institute Press.

Jansson, L. C., Williams, H. D., & Collens, R. J. (1987). Computer programming and logical reasoning. School Science and Mathmatics. 87(5), 371-379.

Kellogg, R. T. (1982). When can we introspect accurately about mental processes? Memory and Cognition, 10, 141-144.

Kohn, Alfie (1987). It's hard to get left out of a pair. Psychology Today, 21, (October), 52-57.

Massialas, B. G., & Papagiannis, G. J. (1987). Toward a critical review of computers in education: Implications for social studies. The Social Studies, 48( 1), 47-53.

McClurg, Pat, with Chaille, C. (1987). Computer games: environments for developing spatial cognition? Journal of Educational Computing Research, 3( 1), 95 111.

Miller, G.A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81-97.

Moursund.D. (1988). Computers and problem.solving: A workshop for educators. Eugene, OR: International Council for Computers in Education.

Pogrow, S. (1985). Helping students to become thinkers. Electronic Learning, 4, 26-29.

Polya, G. (1945). How to solve it, Princeton, NJ: Princeton University Press.

Reed, S.K., Ernst, G.W. & Bamerji, R. (1974). The role of analogy in transfer between similar problem states. Cognitive Psychology, 6, 436-450.

Rieber, L.P. (1987). Logo and its promise: A research report. Educational Technology, 27(2), 12-16.

Rivers, R.H., Vockell, E. (1987). Computer simulations to stimulate scientific problem solving. Journal of Research in Science Teaching, 24(5), 403-415.

Rose, N.S. (1983). The effects of learning computer programming on the general problem-solving abilities of fifth grade students. Doctoral dissertation. North Texas State University, DA8327056.

Ross. S.M. (1983). Increasing the meaningfulness of quantitative material by adapting context to student background. Journal of Educational Psychology, 75, 519-529.

Royer, J.M. (1979). Theories of the transfer of learning. Educational Psychologist. 14, 53-69,

Scheerer, M. (1983). Problem-solving. Scientific American, 208 (April), 118-128.

Simon, H.A. and Simon, P.A. (1962). Trial and error search in solving difficult problems: Evidence from the game of chess. Behavioral Science, 7, 425-429.

Steinberg. E.R., Baskin, A.B., & Hofer, E. (1986). Organizational/ memory tools: A technique for improving problem solving skills. Journal of Educational Computing Research. 2 (2), 169-87.

Stemberg. R.J. & Davidson, J. E. (1982). The mind of the puzzler, Psychology Today, 16 (June), 37-44.

Taylor, Robert (Editor). (1980). The computer in the school: tutor, tool, tutee. New York, Teachers College Press.

Wells, G. & Berger. C. (1985/86). Student/teacher developed spread sheet simulations: A population growth example. Journal of Computers in Mathematics and Science Teaching, 5(2), 34-40.

Whimbey, A. (1984). The key to higher-order thinking is precise processing. Educational Leadership, 42 (September), 66-70.

White, C.S. (1987). Developing information-processing skills through structured activities with a computerized file-management program. Journal of Educational Computing Research, 3(3), 355 375.

Yates, B.C. (1988). The computer as an instructional aid and problem solving tool: An experimental analysis of two instructional methods for teaching spatial skills to junior high school students. Unpublished doctoral dissertation, University of Oregon, Eugene. OR.

Teacher Productivity Tools

(Editor's Message for May 1989 TCT)

Dave Moursund

Sometimes when I sit down to write an editorial, the words seem to literally flow off my fingertips. The underlying message is clear and formulating it is easy. An hour at the keyboard, and I am nearly done.

But sometimes that is not the case. Then I start, stop, restart, and so on, over and over again. In such cases I generally try to figure out what is going wrong. Am I suffering from some sort of writer's block, am I suffering from lack of direction, or is there some other major source of difficulty?

Well, right now I am spinning my wheels. I want to write about Teacher Productivity Tools, and I certainly know a lot about this topic. But I have already written and discarded a half dozen first paragraphs. What is my problem?

I think I know the answer. I don't want to offend my readers. I am afraid that the message I want to convey is not a message that teachers want to hear.

The expression "teacher productivity tools" has come to mean software such as computerized gradebooks, test generators, presentation graphics, and other software that is used by teachers as they work at their profession. A word processor is a teacher productivity tool if it is used to write and modify handouts for students, lesson plans, letters to parents, and so on.

All of that is well and good. But why isn't computer assisted instruction a teacher productivity tool? Since one goal of education is to have students learn, it would seem that software that helps students learn more, better, and faster would be considered to be a teacher productivity tool. Or, why isn't hardware and software that eliminates a major piece of the curriculum a teacher productivity tool? The lowly hand held calculator, which can eliminate significant chunks of several years of the current math curriculum, provides a good example.

Now I see the source of my difficulty in attempting to write this editorial! Almost no teacher wants to believe that a significant portion of what he or she does can be done by a computer. Almost no teacher wants to believe that large parts of the current curriculum have become nearly irrelevant because of computers. Thus, I believe, teachers have carefully limited the meaning of teacher productivity tools and have developed a definition that tends to mask the whole issue of increasing teacher productivity.

This can be contrasted with what has occurred in business and industry. For all practical purposes, the computer industry has been driven by the productivity gains accruing to computer users. IBM has yearly sales in excess of $50 billion because over a wide range of job categories, the people who use computers effectively are more productive than those who lack computer access. This has lead to huge changes in business and industry, where productivity gains lead to increased profits, or at least to remaining competitive and staying in business.

Now let me switch gears for a moment. In some of my workshops I make use of a computer attitude scale. The workshop participants are asked to respond on a scale of 1 (strongly agree) to 5 (strongly disagree) to statements that include:

1. Computers can teach better than teachers.
2. Computers can displace teachers.

Invariably the mean response on these two statements is above 4.75, with a huge majority of responses being strongly disagree.

But then the interesting discussion begins. Are there some pieces of the curriculum that computers can teach better than some teachers? Surely the answer is yes, if we are to believe the research on computer-assisted instruction. If so, then computers can displace teachers.

It is not that we now have a computer that can do all that a human teacher can do, and so might replace teachers on a one for one basis. Rather, if suitable computer-assisted instruction facilities are available, the total number of teachers needed might decline.

A similar argument is proposed for computer applications that might lead to dropping substantial chunks from the curriculum. For example, suppose that we agree that one goal of education is that students should learn to perform at the 80% level or above on a computational test covering addition, subtraction, multiplication, and division of multi-digit decimal numbers. Right now we devote huge number of hours of instructional time and student study time to this topic. This would be greatly reduced if we merely gave students hand held calculators and a little instruction in their use.

By now many of my readers will be experiencing a certain level of hostility, discomfort, or desire to enter into the discussion. The careful reader may note that I have not advocated that computers be used to displace teachers. Indeed, I strongly support that we use computers to increase teacher productivity and use this increased productivity to improve the outcomes of our schools. There are many things that humans can do far better than computers, and schools would be much better if the teachers could devote more time to such tasks.

But there are many tax payers, school board members, and legislators who are beginning to understand that computers and other related technology can indeed increase teacher productivity. Many of them would like to translate this increased productivity into decreased costs of schools. For them, this line of reasoning suggests that computers should displace teachers.

Sooner or later teacher will have to face this challenge. It seems to me that it would be better if teachers themselves brought up the issue. Teachers clearly understand that much of what they do cannot be done by computers. They clearly understand that schools would be better if they had more time to do the types of tasks that humans can do and computers cannot. Let’s raise the issue and carry it to the tax payers, school broad members, and legislators. An aggressive approach can lead to significant improvements in our school system.

Retrospective Comment 1/19/05

Less than a year ago I read one of Peter Drucker's in which he discusses productivity. Peter Drucker is a consultant and writer in business. He published his first book the year after I was born, and he is still making significant contributions to his field. One of his areas of focus has been changes in productivity. For example, at the time of the American Revolutionary War, about 90% of the population were farmers. Now, less than 3% of the population are farmers, and they produce a large surplus for export. Roughly speaking, the productivity of farmers in the US has gone up by a factor of 50 in the last 230 years.

Drucker estimates that the productivity per worker in the industrial manufacturing sector has also increased by a factor of about 50 over that period of time. He then looks at other sectors, such as education.

Our education system is highly labor intensive. It is somewhat difficult to define what we mean by productivity. Certainly it means much more than the number of "student seat hours" produced per employee. In education, we are looking for high quality results—students getting a modern education that appropriately prepares them for adult citizenship in our society.

Research suggests that reducing class sizes increases student learning. We can certainly quantify what we mean by reducing class sizes—such as a 50% decrease in average class size But, can we quantify student learning in a comparable manner? What would it mean to make an assertion that students learned twice as much in a given amount of school time? Perhaps we might mean that the learning rate doubles, and all other things such as retention and ability to use one's learning remain constant. But, suppose that students learn 1.5 times as fast, show 10% gains in both summative evaluation and long-term retention evaluation, and score higher on a test designed to measure interest in and satisfaction with schooling?

You should be able to see that productivity in education is a "bit of a sticky wicket." While I'm at it, let me further complicate the issue. Suppose that we have a curriculum in which students learn the care, maintenance, and driving a horse-drawn wagon. Over a period of years, we gradually improve the curriculum content, instructional process, and assessment. Students perform better on the tests. We do all of this without increasing the budget. Then, we might well say that we have increased productivity. However, suppose during the same period of time gasoline-powered trucks are invented and come into widespread use, greatly decreasing the use of horse-drawn wagons. Now, would we say that we have increased the productivity of our education system, as we continue to produce "product" (graduates of our program of education) who are not qualified to work in the truck-maintenance and driving industry.

Hmmm, you might say to me. You might ask, "Are you suggesting that both our curriculum and our methods of teaching are somewhat in the horse and wagon era, while our world has moved on well beyond that? If you ask this question, then you have gotten my message!

The editorial was published in May 1989, a little more than 15 years ago. The message is still fresh.