The Project Method

It is not due to the importance of the project method that I devote an entire chapter to it. Rather it provides the point of departure for several important ideas about teaching and learning. Perhaps the most important of these is the "learn by doing" ideal.

One might ask just what the "project method" consists of. Could we say that a biology class is taught by the project method, and the current "project" is to learn the information on chapter ten? This is not what we generally mean when we use the term. I think the essential feature of the project method is that the "project" is somehow different than what we think of as common learning activities. A project is something tangible, such as a model of a volcano, or something at least semi-tangible, such as an event. A project is something identifiable, something that has existence of its own and a reason for its own existence. A "learning activity", in contrast, does not have a value of its own separate from the learning. When a second grade teacher drills the class with addition flash cards, this is surely not a project, not in the usual way that we use the term. It is simply a learning activity. When a fifth grader makes a model of a medieval castle, it is a project. Of course there is a blurred boundary between projects and non-projects. One grades into the other. Writing a book report might be right on this boundary. In this chapter I will analyze the "project method" and try to show some considerations and serious limitations involved with its use.

The "project method" seems to fit naturally in some subjects. When learning woodworking it seems sensible to build projects. One might first make a simple project, perhaps a lawn ornament that can be cut out of a sheet of plywood with a jigsaw, then sanded and painted. Next one might tackle a more complicated project, perhaps a simple shelf to hang on a wall. Then one might progress to successively more complicated projects.

I am not saying that woodworking cannot be successfully taught with this general strategy. But I will make a case for "breaking up the project method". I have never taught woodworking, so perhaps I will not choose the best of examples. I certainly do not offer my observations as recommendations on how to teach woodworking. But I believe it is a good point of departure.

My general objection to the project method is that it may not fit the needs of learning. First, there is the matter of practice. Skills and concepts require a certain amount of drill, and any given project may provide too much or too little practice for any given skill or concept. For example the first project, making a lawn ornament, requires sanding smooth the rough edges left by the saw. Sanding doesn't require a great deal of skill. Perhaps five minutes of sanding gives the student the basic idea of how the sandpaper works. Yet the project might require several hours of sanding to get all the edges smooth. Is this time well spent?

Now consider another skill involved in this same project, that of driving nails with a hammer. Let's suppose this project requires a brace on the back to be fastened with two nails. Driving nails with a hammer takes a little more skill than using the sandpaper. Yet the project, as I have described it, provides very little practice. Can we say that the student, as a result of completing this project, has learned to drive nails?

The project, as I have described it, gives far too much practice in sanding and far too little practice in driving nails. There are gross disparities between the amount of practice needed to learn a skill and the amount of practice provided by the project. Perhaps I have exaggerated the problem for purposes of illustration. It could be argued that with careful selection of an appropriate project these disparities would be minor, even to the point of irrelevancy. However, I would argue that the chances of finding a perfect project are zero. Even a very appropriate project will not give exactly the right amount of drill or practice on every skill or concept.

So what can be done? Must the project method be abandoned? No, the project method is valid, but I will argue that there are times when it is beneficial to break it up. In the above example I claimed that not enough practice is given in driving nails. "Breaking up the project method" could consist of simply having the students practice driving nails on scrap lumber before driving nails in the project. This practice on scrap lumber might continue as an adjunct to several projects. Driving nails is a skill. It requires practice. If the projects do not give enough practice in driving nails, scrap lumber will.

The problem of too much practice involved in a project is perhaps not so easily handled. However there are still modifications possible to break up the project method. In the example of sanding edges on the wood the teacher might do part of each students' sanding himself on an electric sander, then leaving the remainder of the sanding to the student. This might give a more appropriate amount of practice for the skill. It might be argued that the student should do all of the project himself. This seems unconvincing to me. Nobody does every part of anything. The students do not cut down the tree to get the lumber for the project. They do not mix the varnish to finish it. They do not have to make their tools before making their project. I would argue that having the student do "everything" is a logical impossibility and pedagogically undesirable. It puts the project ahead of learning, and this is the second general objection to the project method, which I will now discuss.

Uncritical use of the project method may cause the student to view the finished project as the only goal of his efforts. The real goal is learning. The project is supposed to be a vehicle to that learning, not an end in itself. Again I will use woodworking for an example, in particular the driving of nails. Suppose Johnny and Jane are doing the woodworking projects described above. Johnny is conscientious and insists on doing every bit of work on his project himself. Jane is a little more enterprising. She keeps running to the teacher for help. By the time she is finished she has done very little nailing. The teacher had to start both nails on the project I described above, and Jane's contribution was only to apply the final few blows to finish the job. Johnny, on the other hand, bent seven nails before finally managing to drive the required two nails straight. Jane's nailing looks neater than Johnny's, because Johnny's repeated efforts left marks on the wood. Jane brags about her project, but of course she learned less than Johnny. The secondary goal of having a good project supplanted the primary goal of learning.

The teacher doing part of the sanding in one example, and Johnny driving the nails himself in another example, may seem contradictory. However my point in both cases is that considerations of learning should be paramount. Neither the appeal of a neatly done project nor the appeal of a type of self sufficiency should supplant the primary consideration of learning.

If a course consists only of projects then of course it is easy to think of the projects, rather than learning, as the goal. When the project method is broken up, when a number of pure learning activities are introduced, then the students are much more aware that learning is the real goal. If the shop teacher has students practice nailing on scrap lumber until they can demonstrate a certain level of competency, then they will focus on learning. Thus it is desirable to break up the project method.

My third objection to the project method is that it tends to be expensive in time and effort in return for the benefits gained in learning. This is very similar to the first objection, but it goes beyond. Projects contain overhead, as I discussed in Chapter Five, as well as too much practice on some parts. Of course there can be very widely differing interpretations of what is overhead and what is essential learning or what is overhead and what is beneficial extra practice. I will attempt to give more specific examples of overhead as I continue.

Perhaps the most compelling argument for the project method is that in some subjects a project is absolutely necessary because learning cannot exist in isolation. It must be applied. This fits well in woodworking. Perhaps one could learn to drive nails on scrap lumber without ever applying that skill to a project. And perhaps one could learn to use a saw by cutting scrap lumber, and one could learn to measure without making a project, and so on. But surely these isolated skills ought to be applied to something real. A student who learned all these skills, but was never given to opportunity to actually use them, would justifiably feel cheated. Further, it could be argued that perhaps the most important thing a person learns in woodworking is how to organize and apply the separate skills and bits of knowledge. A project is essential for this.

An "ideal model" of the projected by drill, discussion, assignments, tests, and whatever learning activities are needed and appropriate. Then a project is assigned which allows the students to apply those principles and methods. The project must be well chosen so that the learning is actually applied, and overhead is kept to a minimum. And, of course, the project should have intrinsic worth. This is important for motivation. As the students proceed on their projects the teacher provides guidance. This can include correction as needed, and it can also include new learning as needed. In this model the project method is certainly broken up. Learning comes first, separate from the project, and the project is kept subordinate to the learning. Further, the project would not be possible without the learning. I expect woodworking, when well taught, follows this model closely.

This "ideal model" of projects can be interpreted simply as a matter of coherence. The goal of a course is to build a structure of knowledge. Any project should contribute to that goal just as any other learning activity should. Any project should be coherent with the total course just as any learning activity should be. I remember in more than one college science course the first lab or two seemed disconnected. This can be analyzed in terms of projects and coherence. A laboratory exercise is a project of sorts, and it ought to fit the ideal model I have described. But this is not always the case, especially at the beginning of the course. The first lab will probably fall after perhaps one or two lectures. One or two lectures constitutes little more than an introduction to the course, not enough time to really get into the substance of the course. A lab exercise cannot be built on the substance of the lecture if the lecture has not yet gotten to any substance. It cannot fit the "ideal" model. Thus more than once I have felt that the first lab of a course was a "throwaway". It may have had some value, but lacked coherence with the rest of the course.

I have used woodworking only as an example to introduce my thesis that the project method should be broken up and to introduce the "ideal model". I will now use these same general ideas and apply them other subjects.

A term paper is a project, a very academic project rather than concrete, but certainly a project. Is this a good use of the project method? Can you teach history, for example, by the "project method" in which the "projects" consist of a series of term papers?

Not so long ago while doing some cleaning I came across three term papers that I had written in high school many years ago, and my reaction to these papers seems relevant here. One of them was for physics. I chose the topic "The Physics of Music". This paper, though I had not seen it in many years, was totally familiar. The ideas I developed all made sense, as if I had written it recently. I learned a great deal from doing this "project". Another term paper I found was for world history, and it was a report on the country of Austria. It was totally unfamiliar. Reading my own report was like reading an encyclopedia article written by someone else. I knew absolutely none of the information. It did not even "ring a bell". A third term paper I had written was for a psychology class on the subject of perception. My response to this was in between my responses to the other two papers. It was somewhat familiar, but I remembered little of the content of what I had written.

As a method of learning, my term paper on the country of Austria was a dismal failure. I learned very little. I simply took information from encyclopedias and other references, reworked it into my own words, and produced the paper. It could be argued that to do this, to produce a coherent paper, I had to learn. To some extent this is true of course. But I think it is very limited. I found facts, I organized them, and wrote them down. I didn't really examine the information in any critical way, other than superficially organizing facts, and I certainly didn't retain the information. My endeavor had more in common with compiling a list of phone numbers than of constructing and retaining a body of knowledge.

Viewed from the perspective of learning, my term paper in physics was very successful; my term paper on perception was somewhat successful; and my term paper on Austria was a failure. It is tempting to think that interest is the key that explains the difference. I was interested in the physics of music, and in perception, but not in Austria.

The "interest first" rationale is very popular in educational circles. This rationale has two parts. The first part is the idea that children will learn what they are interested in, and the second part is that therefore we should concentrate on first developing interest. I think both parts of this rationale have some validity. But, more importantly, the validity of each part is less important than its limitations. Yes, other things being equal, children learn more when they are interested than when they are not. But other things are never equal. Much more importantly, children learn when they are placed in a situation in which they are expected to exert their efforts to learning, when conditions are set up so that their efforts will bear fruit, and when they receive benefits from their learning, the most important benefit being satisfaction of accomplishment. And yes, teachers should try to elicit interest, but there is no dependable way to do this. My world history teacher failed miserably in eliciting in me an interest in the country of Austria. I don't fault him for that. I had other interests at the time, strong interests. I can't imagine any teacher getting me interested in learning about the country of Austria at that time.

And I think there is a case to be made against eliciting too much interest. Interest is not something that can be turned on and off like a faucet. If a charismatic teacher at that time had succeeded in interesting me in Austria, it would have pulled me away from my more long term interests. Perhaps it was better for me to stick to my long term interests. When I taught science one year I discovered my predecessor had convinced a number of students to choose a career in science. This included several of my twelfth grade chemistry students. A few months into the school year I felt that some of those students were really not cut out for a career in science. Indeed by the end of the school year they had changed their minds. I don't know just how their "interest" in science had been elicited, but I suspected it was more social than academic, that the teacher made himself into a salesman for science and some students bought indiscriminately. I can't help thinking they would have been better off not to be quite so "interested" in science and to give their commitment to more suitable fields of study.

The structure of the subject matter also is a factor in the suitability of term papers for learning. Subjects with structures of accretion are different than subjects with structures of implication. A term paper in a subject with a structure of accretion can easily be assembled like a catalog. That's just about what I did for my term paper on Austria. A subject with a structure of implication cannot be easily assembled like compiling a catalog. My physics of music paper could not have been written without understanding. From this perspective, then, it might be argued that term papers are suitable only for subjects with structures of implication. I don't think it works out this way, however. Term papers in subjects with structures of implication are heavily dependent on intelligence. They might be very productive for a few, but frustrating or even impossible for the many. Term papers in subjects with structures of accretion, on the other hand, are possible for anyone, given a certain amount of effort.. Anyone could assemble the facts about Austria, type them up, apply a cover sheet, and hand it in. Not anyone could figure out the physics of music in a meaningful and productive way. So it might be argued that term papers are only appropriate for subjects with structures of accretion. I would argue that neither type of subject is more suitable than the other for term papers, but that different considerations apply in the two cases.

Efficiency is a consideration that I think is very important to consider when assigning term papers. The situation that brought this to my attention was a term paper in a college sociology class. My topic was interracial marriage. I had some interest in the subject, and I felt I gained some knowledge from the process. But I also felt it was inefficient. I estimated I spent about forty hours on the paper. This would be an hour a day for forty days, which, as I recall, is approximately how I did it. It was assigned early in the semester and was due sometime toward the end of the semester, and of course I procrastinated for about a month before getting started. But when I did get started I worked pretty steadily and evenly. Reflecting on the experience after I was all done, and after the paper was returned to me (with what grade I cannot remember), I decided that I could have acquired all the learning I gained from all this effort from one or two lectures on the subject in class, and perhaps another hour of reading and reviewing my notes. If the teacher had made the information a part of his course, therefore, I would have invested about three hours of effort for the knowledge I gained, instead of the forty hours spent getting the information on my own. Results that are worth three hours of effort in return for an outlay of 40 hours of effort gives an efficiency rating of 7 1/2 %.

This analysis rests on purely subjective premises of course. I have no way to objectively establish that I actually spent forty hours on my paper, or that I could have learned the equivalent from three hours of lecture and review. And it might be argued that the writing provided benefits beyond the knowledge that might be verbalized. Perhaps there are intangible benefits. Perhaps the time and effort to learn one small topic in detail is worthwhile, though it would be impossible to learn every topic in such detail. Perhaps learning to fit ideas together, in at least one small topic, is something that cannot be done well in the usual lecture, discussion, and test format. It might be argued that these intangibles may come only through this labor intensive method. I am not convinced of this. But even if one concedes that there are intangible benefits only to be had in this way, I would still argue that the bulk of learning cannot be gained in this way. One term paper, one project, may be worthwhile every now and then, but the vast majority of learning must be gained in more efficient ways. We don't have the time to learn very much by slow methods, for there is a vast amount of knowledge that we want the young to acquire.

Another argument for writing term papers is that "We are all teachers of English". The method by which I arrived at an efficiency rating of 7 1/2 % assumes that no benefit is gained from the writing itself. When I wrote the report in the sociology course I was not the least bit interested in improving my skills in composition. But of course that is not always the case. There are many situations in which the learning of the subject matter is secondary to the learning of composition skills. In such a situation my method of computing efficiency of learning simply does not apply. In elementary school, for example, it is quite appropriate to consider composition skills the main purpose of writing reports. After elementary school, however, it become a debatable point whether or not every teacher should consider himself an English teacher in addition to being a teacher in his own subject. My view is that at least past the junior high level it is best to leave composition to composition teachers. As a science or a math teacher I want to teach science or math, not composition, and I want my students to be free to learn science or math. And I want them to learn in the most efficient way possible. If the English teacher cannot teach a student to write then how can I expect to teach him to write? No, I can't teach him to write, but perhaps I can succeed in teaching him a little science or math. And I certainly will not tell the English teacher that "we are all teachers of math" and therefore she should look for ways to integrate math and literature.

I related term papers to projects. When discussing woodworking projects I said the project may not fit the needs of learning in several ways. I said that a project may give too much practice in some areas and too little practice in other areas. I also said that the goal of the project can detract from the goal of learning. These disadvantages apply to some degree to term papers as projects. It can be argued that the entire topic of the term paper is given too much emphasis, and everything else is given too little emphasis. It can be argued that one gets far too much practice in typing, and one gets far too much practice in organizing of information, while one does not enough exposure to the main information of the course, the framework structure. I think these arguments have merit, but most important, I think, is the problem of inefficiency. We simply don't have time to learn much by the project method. There is too much overhead.

In discussing woodworking I mentioned the "ideal" project method in which learning comes first, the project provides an object on which the learning can be applied, and the project would not be possible without the learning. It could be argued that the term papers I described fit this scenario, but it could also be argued that they do not. In the course on race and cultural relations, in which I wrote the term paper on interracial marriage, the professor might say this model fits. He might say he presented sociological principles and methods and we applied them in the term papers. I hope there is some truth to this, but I am not sure. Suppose I had written a term paper on interracial marriage for a composition class instead of for the sociology class. Would the result be the same? If the model fits then the result would not at all be the same. The composition class would be concerned with finding, organizing, and presenting information, not with applying sociological principles. Therefore a paper on interracial marriage for a composition class would be different from a paper on interracial marriage for a sociology class. My viewpoint, admittedly subjective, is that there was very little difference. Had I written the paper for a composition class it would have come out just about the same. During the first half of the course, before the term papers were assigned, we got information from the course, not principles and methods. Again, the professor might disagree with my judgment on these matters.

In the last paragraph I mentioned "principles and methods" to be applied, but I did not mention perspective. It could be argued that the information we received in the first half of the course gave us a sociological perspective that we were to apply in writing the term papers. This may be the case, and this may even be consistent with my feeling that the paper would have been identical whether written for a composition class or a sociology class. What, then, does this say about the value of the term paper? I'm not sure. I felt that my benefit from the paper was information, and the information was gained inefficiently. The professor might argue that my benefit was learning the sociological perspective, and that this sociological perspective could only be learned by actually applying it in the term paper. Had I written the same paper for a composition class, by this argument, I might have gotten the same information, but would not have gotten the sociological perspective. And, it could further be argued, that the sociological perspective was what I took the course for. I do not pretend to have final answers to these questions.

There are many projects that are similar to term papers and which can be discussed from the perspective of the principles I have been developing. When in the seventh grade my daughter had to make a "parts of speech handbook". This consisted of about six or eight pages enclosed in a plastic binder. The first page was a title sheet. Next was a page titled "Nouns", and consisting of perhaps ten sentences with the nouns highlighted in yellow. Then followed similar pages for verbs, adjectives, and so on. My daughter felt good about this project and the teacher asked my daughter if she could keep the booklet as a good example for the students next year. I didn't complain about this project, of course, but I felt it was about 90% overhead. The booklet looked good because we happen to have a computer that can make large attractive printing and even throw in a few graphic decorations, and my daughter knew how to use it. The bulk of the "project" then consisted of computer techniques, not of grammar.

The grammar content of this project amounted to a simple review exercise of identifying parts of speech. If this grammar content were in the form of an exercise in a workbook it would have occupied my daughter for perhaps ten minutes to circle the correct answers. To write out the answers instead of circling them would have taken perhaps twenty minutes. I would estimate my daughter probably spent two hours at the computer, spread over several days. By these figures then the efficiency comes out to about 8% (figuring 10 minutes of results for 120 minutes of effort) or as much as 17% (figuring 20 minutes of results for 120 minutes of effort).

This project, in my opinion is susceptible to all the faults of projects. It gave way too much practice on using the computer, (perhaps valuable in itself, but not related to learning grammar), and of course it perverted the goal from grammar to the project.

There is an interesting sidelight to this story. We have a book of the "Calvin and Hobbes" comic strip, and my daughter and I greatly enjoyed this book. One of the running themes in this particular collection of comic strips was that Calvin was convinced that a "professional-looking clear plastic binder" was all he needed to get an "A" on an otherwise poor report. It was about this time when we bought the computer. The salesman who sold it to us pointed out that the professional printing capability of the computer would help one get good grades. When we got home my daughter told me she immediately thought of the "professional-looking clear plastic binder", as, of course, I had. So of course during work on the grammar booklet there were a few references made by both me and my daughter to the necessity of finishing it off with a "professional-looking clear plastic binder", which indeed she did.

Lab experiments in science courses can be seen as projects. So one might ask if the problems of projects and term papers discussed above apply. I think they do. I remember many lab exercises in my own experience. In one physics lab I realized that we had spend over half an hour trying to measure 30 inches above the floor on the wall. There was some rationale to this. It was important to have very accurate measurements, and it was not certain that the floor was perfectly level. It can be argued that the experience was beneficial. It can also be argued that the experience was a total waste of time. I tend toward the latter perspective.

The "learn by doing" perspective is a consideration that I think is important whether we are concerned about term papers, woodworking projects, laboratory exercises in science, or learning academic subjects. Educators are fond of saying that "we learn by doing". A project is something to do. Therefore, it seems to follow, you will learn. I think this is fallacious. We learn by doing, all right, but not necessarily what we are supposed to learn. We learn by doing when we are:

doing what we are to learn, and

doing it in a productive way.

Consider the woodworking example of learning to drive nails. The "learn by doing" perspective certainly seems to apply here. How else would learn to drive nails? There might be differing views on how much practice is needed, and what considerations would most effectively keep fingers and thumbs out of danger, but no one would argue that there is any way to learn to drive nails other than driving nails. Driving nails is a part of many woodworking projects. Therefore, it might be argued, the project method is good.

My criticism of the learn-by-doing approach is not that it is wrong, but that it is often misapplied, especially in the interests of some "project". Teachers often seem to think that by doing one thing they will learn another. I referred in a previous chapter to "superintegration", giving the example of baking cookies to bring together various topics or subjects, and presumably learn them. Of course few teachers would think the actual baking of cookies teaches arithmetic. Rather they would argue that baking cookies provides the situation in which the students will have an opportunity and motivation to learn arithmetic, just as many woodworking projects provide the opportunity and motivation to learn to drive nails.

All this makes some sense, but, as I have argued, is usually inefficient. When carried another step or so teachers may fall into the trap of thinking that doing "A" will teach "B". This is clearly fallacious.

In the woodworking class, for example, the students may be expected to learn some information as well as skills. Suppose the students are expected to learn that plywood is made in factories by gluing thin sheets together that are shaved off logs, etc. Does this fit the learn-by-doing perspective? Will driving nails help the students learn this bit of information? When it is put this way the answer is obvious. No, of course, driving any number of nails will not help students learn how plywood is made. To learn about plywood they must be given information about plywood. They must assimilate this information. They must fit it in with their existing structure of knowledge. And finally, of course, they must remember it.

In the teaching of science it is tempting to translate the "learn by doing" ideal into the idea that science is learned primarily in the laboratory. I think this is a mistake. In chemistry all the shining glassware and colored bottles make the lab appealing. But I would argue that that is not where the actual learning takes place. A laboratory exercise in chemistry is usually a demonstration of a scientific principle or fact. If the principle or fact in question is not learned and understood previous to the laboratory work, then the laboratory work degenerates into the following of a recipe. Chemical knowledge, at least until one is a research chemist, comes from a textbook, not from glassware. We learn by doing all right, but in chemistry the "doing" consists of studying the textbook, asking and answering questions, doing the problems at the end of the chapter, listening to the teacher, taking quizzes and tests, and so on.

We learn to manipulate mental concepts by manipulating mental concepts, not by driving nails or engaging in any other physical skill. One does not learn swimming by writing poetry. One does not learn algebra by blowing a horn. One does not learn to can fruit by driving nails. We do indeed "learn by doing". But we can't learn one thing by doing another. We cannot learn "A" by doing "B". Learning about how plywood is made in factories is a matter of cognition, of thinking and remembering, of building a structure of knowledge. Driving nails is a matter of motor and perceptual learning.

We learn to solve equations in algebra by solving equations in algebra, not by writing a report on Descartes, or by making a model of a space shuttle, or by any other extraneous activity. Doing one thing does not result in learning another. Driving nails will not teach one about the manufacture of plywood. Driving nails will not teach one about algebra, or history, or how to tell a humming bird from a bat.

We learn to manipulate mental concepts by manipulating mental concepts. We build structures of knowledge by fitting concepts and associations together in coherent ways. We develop fluency in cognition by practice in cognition. One learns to reduce fractions by reducing fractions, not by baking cookies.

Often, I think, we judge "learning" in rather general terms. If students are engrossed in a project then we say they are "learning". I think this can be misleading. What does it mean to be "engrossed" in something? One thing it means is that the students are not bothering us. In general, I think, it is good when students are engrossed in what they are doing, but that is not all there is to it. They can also be engrossed in a mindless card game they have played since they were pre-schoolers. They can be engrossed in daydreaming. Or they can be engrossed in a cartoon on TV. We also may say that to be engrossed in something is to be self-directed, and this certainly is something a step above simple daydreaming. We observe what we might call "directedness" in the students and feel that is a good thing. I agree. It is a good thing. But the goal of that directness is very important.

When my daughter made the parts-of-speech booklet I mentioned earlier she was self-directed. But I felt the end product was not worthwhile. It had little to do with learning grammar. When a child does a project at home just because he or she wants to, efficiency doesn't matter. As a child I made model airplanes. I was very self-directed. You can't make model airplanes without being very self-directed. My model airplane making was probably very inefficient from a learning standpoint. But that is totally irrelevant. My goal was not learning. My goal is simply the enjoyment of the model airplanes.

More importantly, any home activity like this is second in priority to school work. School time is limited. There is a tremendous amount of knowledge and skills we want the young to learn, and model airplane making is not part of it. There is only a limited amount of school time available for this learning. Therefore it should not be squandered. When doing a project at school the students may appear to be engrossed. They may appear to enjoy it. They may appear to be self-directed. None of this is bad, but none of this really means they project is justifiable. I very much enjoyed making model airplanes, and I would even call it "educational" in a very broad sense, but I would not advocate including model airplane making in the school curriculum.

In Chapter Three I discussed the "assumption of right thinking", the idea that if the answer is right then the thinking behind the answer must be right. A very similar situation can apply here. If students do a project or activity and seem to be learning then it is easy to assume that the learning is the result of the project or activity. Of course that may not be the case. The project could be only part of the cause of learning. Indeed it could be totally irrelevant, or even antithetical, to the learning. In the example of my daughter's parts-of-speech booklet the activity seemed largely irrelevant to the actual learning of grammar. In the following example I think the activity went beyond irrelevant to actually being a hindrance to learning.

In a college physics course we were supposed to process data from a lab exercise on a computer "spread sheet". The advantage of a spreadsheet is that it automatically does the needed computations. For example in one column of the spreadsheet the computer can be programmed to find the average and display it in the box at the bottom of the column. I discovered some of my classmates were using the spreadsheet all right, but not in the way it was intended. There were entering the numbers in the column, as they were supposed to do, but then they found the average on their calculator and entered that number in the box at the bottom of the column. They were doing by a hand calculator what the computer was supposed to do. Thus, instead of the computer being an aid in their work, it added to their work. They were supposed to be "learning by doing", when in fact they were doing extra work and not learning what they were supposed to. The professor, of course, considered learning to use the spreadsheet as one goal of the project. And when he looked at the finished project, consisting primarily of the computer printout, I'm sure he assumed that goal was accomplished. But it was not. The project had displaced the goal of learning to use the spreadsheet, and I think it actually hindered that goal. My classmates were focused on producing the computer printout. This was the project, the goal, and to achieve this goal they were quite willing take a shortcut. The "shortcut" in this case, using a hand calculator instead of the computer, actually displaced learning to use the computer.

It is certainly possible, perhaps even likely, that before the project was completed my fellow students figured out the spreadsheet. I don't know about that. But I do know that at least at the time I observed them the project was supplanting the learning. They were not learning by doing because they were not doing what they were supposed to be learning. It only appeared on the surface that they were.

I have emphasized that one cannot learn one thing by doing another, but this should not be considered an absolute. Doing "A" will not teach "B" if "A" and "B" are entirely separate. However "A" and "B" might overlap. Indeed there may be a great deal of overlap. A boy who grows up on a farm and learns to drive a tractor long before he is old enough to drive a car will certainly benefit when he finally gets to learn to drive a car. There is a great deal of overlap between learning to drive a tractor and learning to drive a car. Sometimes circumstances lead us to capitalize on this sort of overlap. In learning to dive off a high board, for example, one might first learn to dive off a low board, with a great benefit of safety. The overlap between learning to dive off a high board and learning to dive off a low board is not 100%, but it is very substantial, and very worthwhile to exploit. A similar analysis would apply to pilots learning in a flight simulator.

Another way to look at overlap of subjects is by the idea of "supporting structure", which I discussed in Chapter Five. Learning one thing can provide supporting structure for learning another. Whether this supporting structure is worth the expense in time and effort, however, must be considered. Efficiency must be considered. Overhead must be considered. And, of course, coherence and motivation must be considered. Learning to drive a tractor may indeed help one to learn to drive a car, but it does not follow that everyone should start on a tractor. And building a model of the Globe theater may provide supporting structure for learning about Shakespeare, but it does not necessarily follow that such a project is worthwhile. It may be too expensive in time and effort to justify its benefits.

Finally I will briefly mention some considerations in choosing projects. I believe many of the same considerations apply to choosing projects that apply to designing a course of study. I discussed these considerations in Chapters Six and Seven. Then, of course, there are special considerations that apply to choosing projects. These are the considerations that I have devoted most of this chapter to. The project should provide the right amount of practice for whatever needs practice, the project should not displace learning, and the project must be reasonably efficient.

The size of a project ought to be considered. In Chapter Six I discussed "chunk theory", and many of the factors I discussed there apply here. How big should a project be? Should one project fill a semester, or even a school year? Should one project be completed in a day, or an hour? In elementary school a great many art projects are begun and completed in a half hour or even less.

A project that is too small is not greatly valued. Indeed a "learning activity" might be viewed as a project too small to have a value in itself. Thus completing a worksheet is a learning activity, not a project, because the goal is learning only. There is no value to the worksheet once it has been completed and graded. It will not be framed and hung over the fireplace. At the other extreme, a project that is too large may be frustrating, or non-motivating because it is seen as never to be completed.

Can a group project be effective as a teaching tool. I have seen on the news now and then a story about a shop class building a big group project. In one case it was an airplane. I think it is common for high school vocational classes to build houses and then sell them. It is commonly asserted that the group nature of the project adds to the motivation. I agree that people do feel some intrinsic value in group endeavor. But I would also argue that group endeavor dilutes individual motivation. A student in a woodworking class that is building a house may feel that the project is not his, as he is doing only bits and pieces of it. In a worst case scenario he may feel that is simply providing labor in return for a grade. The phenomenon of group endeavor versus individual endeavor is an important matter, but not one to be developed here.