Organizations learn only through
individuals who learn.
Hopefully, by this point in this book you have some ideas about how you would create an organization such as the one described by Senge. The problem still remains that even if employees want to learn more, they may not be able to learn at a rate that will keep the organization ahead of the competition. Traditional educational methods certainly have little to offer to solve this dilemma.
Probably nowhere in modern society are there more opportunities for improved efficiency and effectiveness than in American education. Training methods in business are no more efficient, but their inefficiency is usually hidden by the fact that considerable time learning to use what was presented in the classroom is done in on-the-job training.
There is no point in reciting all the problems of modern education in America. The poor showing of U.S. education in comparison with the other major economies of the world is well known. It has been documented too frequently, and still nothing seems to change.1 Even though by many measures of our economy and standard of living, we outperform almost all countries, if we continue to lag behind educationally, it can only have negative effects on both.
The promise of technology in improving education at work and at school has yet to be realized. In my opinion, this is not because computers can’t increase learning rates substantially but because those who design educational software and those who use computers to teach often don’t really understand what learning is all about. So-called computer-assisted instruction (CAI) often turns out to be nothing more than a fancy page turner with beautiful graphics for the students, adding little that they couldn’t get from reading an “old-fashioned” book and less than they could get from an “old-fashioned” lecture. At least in the old-fashioned lecture, students could ask questions! Even though the word interactive is appearing more and more in educational settings and with commercial learning companies, if education is not truly interactive, it doesn’t teach much. There are still too many educators who believe that the solution to student learning lies in more hours in the school day, more days in the school year, and smaller classes. Unfortunately, this kind of thinking does not encourage considering new ways of increasing the efficiency and effectiveness of teaching methods. In fact, there has been a consistent reluctance on the part of the education community to examine teaching methods.
Siegfried Engelmann, professor at the University of Oregon, wrote War Against the Schools’ Academic Child Abuse, a book that should be of interest to anyone who cares about education.2 He defines academic child abuse as “the use of practices that cause unnecessary failure.” Engelmann cites two studies that deserve to be repeated here. Coles (1978) reviewed approximately 1,000 research studies on “learning disabilities.” Not one considered the relationship between instruction or other school factors and the learning disability! Similarly, Alessi (1988) discovered that of about 5,000 children referred by school psychologists for remedial education, not one case was considered to be the fault of the curriculum, the teaching practices, or the school administration.
Consider these findings against the following quote by Engelmann:
During the years that I’ve worked with kids and teachers, I have never seen a kid with an IQ of over 80 that could not be taught to read in a timely manner (one school year), and I’ve worked directly or indirectly (as a trainer) with thousands of them. I’ve never seen a kid that could not be taught arithmetic and language skills. During these years, however, I’ve become increasingly intolerant of reforms formulated by naïve spectators who don’t really understand what school failure is and how it can be reversed.
Cathy Watkins3 summarizes the problem by saying:
Suggestions about how to address the problems of education have included changing the content of instruction, raising educational standards, increasing the amount of instructional time, increasing pay for teachers, and a long list of other “solutions” that would change just about every structural and functional aspect of education except how children are taught.
The traditional method of teaching has changed little since about the eleventh century and usually works as follows:
Business has basically adopted this same model. The only difference is that the homework is on-the-job training (OJT). Typical of training in business settings is the training that I received when I was learning to use basic computer software. Even though we were all on the computer, the teaching model required that we all stay together in learning. This was a most inefficient method because some in the room had considerable experience with computers and others had none. This procedure is reminiscent of a quote from Skinner in Walden Two,4 which discusses the use of grade levels in schools: “The grade is an administrative device which does violence to the nature of the developmental process.”
The instructor would show us her computer screen using a liquid-crystal display (LCD) projector and walk us through each step. She would then check each person’s computer to see whether the student had done it correctly. Once everyone had completed that assignment satisfactorily, she would show us another feature. While in the classroom, I thought I had a clear understanding of how to do things. However, by the time I got back to my office, I could not do most of what she covered in class without referring to the manual or asking someone in the office for help. With so little practice, retention of the material was impossible.
Many people would respond to my situation by saying, “Of course. Isn’t that the way it is supposed to be done? She taught you; now the rest is up to you.” I remember my college history professor telling my freshman class, “Going to college is like going to the drugstore and ordering a soda. The soda jerk can fix you the very best soda he knows how, but whether you drink it is up to you.” He then proceeded to read from the same old musty yellow notes that he had first written back in 1912.
This model of teaching is inefficient in both business and academic settings. To be sure, most students eventually learn, but the problem is, How many learn, how much do they learn, and at what rate do they learn? To put the problem in some perspective, let’s revisit Morningside Academy, a private school in Seattle, Washington, that I discussed in Chapter 16.
Dr. Kent Johnson, founder of Morningside Academy, estimates that core academic content from kindergarten through the twelfth grade could be taught easily in six years. In his school, which is designed primarily for children who have had problems in regular schools, students achieve at a rate four to six times the national average.
Dr. Johnson opened Morningside Academy in 1980. He has maintained meticulous records of actual learning since that time. He has to do so because he gives parents a written money-back guarantee that their child will advance the equivalent of two grade levels in their worst subject every year. Since 1980, less than 1 percent of parents have requested their money back because the average achievement at Morningside is 2.4 grade levels per year. Such gains have been increasing since 1980, and it is now common for the school average on some subjects to increase over three grades per year, as measured by the California Achievement Test. Contrast those figures with the national average of less than one grade level per year!
The school’s results are equally impressive with adults. Adults who tested between second- and eighth-grade levels in reading on entering Morningside training gained at a rate of 1.7 grades per 20 hours of instruction. The U.S. standard is one grade per 100 hours of instruction. How does Morningside do it?
In 1967, the U.S. government funded a program, Project Follow-Through, to compare a wide range of teaching methods with the goal of finding effective methods for educating disadvantaged children. The results showed that the methods known as direct instruction and behavior analysis came in first and second, respectively, among 22 different approaches tried.5 The traditional classroom method came in dead last. As incredible as it might seem, the outcome of this research was that funds were provided to improve the traditional method rather than adopt any of the methods that demonstrated academic superiority!
At Morningside, Dr. Johnson combined direct instruction and behavior analysis to form a teaching method that he calls the Morningside model of generative instruction.6 Dr. Engelmann, mentioned previously, developed the direct instruction (DI) technique.7 DI is carefully scripted both in the order of presentation of the academic material and in the words the teacher uses during instruction. Through systematic analysis, Engelmann has been able to determine the most effective presentation and sequencing of the material.
The behavior analysis approach added measurement, feedback, and positive reinforcement to Engelmann’s material. In addition, Dr. Johnson used a technique developed by Dr. Ogden Lindsley of the University of Kansas (see Chapter 18). Dr. Lindsley named this method fluency.
Fluency is automatic, nonhesitant responding. If people are fluent in a particular subject, they do not have to think about what they are doing, and they can respond quickly and accurately after long periods of no instruction. They are also able to respond for extended periods with less fatigue and can generalize what they have learned to novel situations. This is what Dr. Johnson calls learning for free.
To attain fluency, responding at high rates is required. Because the method deals with measurable, observable behavior, the rates necessary to attain fluency can be determined for any academic subject. For example, certain basic math concepts such as addition, subtraction, multiplication, and division may require a student to be able to do as many as 80 math facts a minute with no errors. Some students can do up to 120 a minute. Although Morningside concentrates on basic skills, it also teaches such subjects as problem solving and fluent thinking skills.
Because practice is the hallmark of fluency training, a Morningside classroom hour is divided as follows: 10 minutes of instruction, 40 minutes of practice, and 10 minutes of break. By the way, students have no homework. The average teacher may not be able to relate to this model of classroom instruction because it is so different from the way traditional classroom teaching is done. This model seeks to determine the minimum amount of information that needs to be given to allow the maximum time for active involvement of the student with the material.
Dr. Johnson’s methods also have been applied in regular schools with impressive results. He has developed a model of individualized instruction that is manageable and effective for public education. However, the move to adopt it by the traditional educational community is very slow.
The work of Johnson and others has made one thing very clear: to learn efficiently, learners must have the opportunity for high rates of interaction with the material to be learned or the skill to be mastered. We know this is true because a high frequency of learner response dramatically increases the amount of positive reinforcement the learner can receive for her or his efforts—and positive reinforcement will accelerate learning.
Where does all this reinforcement come from? Clearly, no teacher or instructor can provide positive reinforcement for every learner behavior. In fact, the reinforcement comes from the students seeing or knowing that they got the correct answer. In most of our learning histories, external reinforcers have been paired with correct answers to the point that getting an answer correct is reinforcing to almost everyone. Therefore, students given many opportunities to answer questions correctly will have an opportunity for high rates of reinforcement quite independent of the social reinforcement they might get from the teachers, parents, or other students. A big part of the fluency model depends on high rates of correct responding. Nothing is more reinforcing than doing well, and students of all ages who get the opportunity to perform successfully at high frequency will be highly reinforced. This high level of positive reinforcement will accelerate learning to rates that surprise most students.
One of the criticisms of fluency techniques that has come from traditional educators is that it smacks of “rote learning,” and as I have been told many times by school staff, “Rote learning went out 30 years ago.” My response to them is that maybe it is not coincidental that learning rates in schools have been going down for the last 30 years as well! Of course, in my day, rote learning was accomplished by drill, and the drill was accomplished by negative reinforcement. Under such conditions, students would not be excited about learning. However, when drills and other rote-learning methods are coupled with positive reinforcement, students beg to do them.
A number of years ago, one of our industrial clients asked me if I would share what his workers were learning about fluency and positive reinforcement with a local school. Of course, I said that I would. The school selected six teachers to pilot positive reinforcement methods in their classes. To give the method a fair test, we chose teachers who varied in their enthusiasm to try these techniques. At a review I conducted after several weeks, one of the older teachers was still a little skeptical about this new approach. As I was talking to the group, I said, “You will really have it made when your students beg you to test them.” Toward the end of that school year, that teacher sent me a card. It read simply, “Dr. Daniels, I really have it made!”
None of the criticisms of this approach that I have seen advanced by traditional educators has been supported by credible research. However, research support for fluency as an effective method of learning is coming from several unrelated sources.
Richard A. Schmidt, in Motor Learning and Performance,8 says that in most motor learning, about 300 repetitions under the most favorable conditions are required to produce automaticity. He says, “Automatic responding can develop with several hundred trials of practice under the most favorable conditions, but it appears that many more trials than this are needed in less optimal real-world settings.”
Schmidt notes advantages to automaticity similar to those discovered quite separately by researchers working on fluency. He writes, “This capability to produce skills automatically means that component parts (e.g., dribbling and running in basketball) do not require much attention, which then frees attention for other important elements in a game (where to pass) or for style or form in dance. Also, automatic processes are faster than controlled processes, allowing the performer to respond more quickly and with certainty.”
Walter Schneider (1985), mentioned in the last chapter, made similar discoveries. In training air-traffic controllers, he found that in training them to rapidly visualize flight patterns, “we compress simulated time by a factor of 100; for example, making a judgment of where an aircraft should turn (a maneuver that normally takes about 5 minutes) would take about 0.5 second. By compressing time in this way, we can provide the trainee with more trials at executing this particular component in a single day of training than he or she could get in a year of training with conventional methods” (emphasis added). He further states, “Active participation is enhanced if subjects need to respond every few seconds.”
Fortunately, technology and the science of behavior have combined to make possible the fluent training of employees in a fraction of the time it now takes to learn something new.
Anders Ericsson9 at Florida State University has conducted extensive research on the acquisition of expertise. Over the last 20+ years, he has investigated expertise in every field in which there are objective ways to separate experts from others. His list of fields includes chess, music, sports, medicine, engineering, and ballet.
There is an amazing consistency to what Ericsson has discovered about what it takes to become an expert in any field of endeavor. He has discovered that it takes a minimum of 10 years of deliberate practice to become an expert at anything. Ericsson’s investigations have revealed that anyone who has attained expert status has engaged in at least 10,000 hours of deliberate practice. This amounts to an average of 20 hours of practice for 50 weeks a year over the course of 10 years.
One of his findings that flies in the face of popular notions about high performers is that at least 96 percent of the variance in performance is accounted for by deliberate practice alone. Another way of saying this is that less than 4 percent of an expert’s performance is due to all other factors, including something often referred to as “natural talent.” This is a very important finding for anyone who wants to become an expert at anything. This means that almost anyone who wants to excel at something needs to engage in deliberate practice for 10,000 hours. In his new book, Peak,10 Ericsson points out that this must be deliberate practice—that is, a concerted effort to improve one’s skill. (Unfortunately for me, four hours on the golf course doesn’t count as deliberate practice.)
It becomes clear from the research in all these areas that the number of reinforced repetitions is a very important element in teaching and learning. Although the relationship between speed and number of trials is not fully understood, the fact is clear that more repetitions lead to more reinforcement, which leads to faster learning, greater retention, the ability to generalize to novel situations, and the ability to perform under stressful and distracting circumstances. What manager in any organization would not want that?
As noted earlier, the research of Ericsson, Engelmann, and Johnson and Layng suggests that most failures in learning are actually failures in teaching methods, not the result of student motivation and/or ability. This finding has tremendous implications for business, industry, and government. It means that most performance problems are not because the performer is incapable of performing at a high level but are caused by either poor training, lack of a reinforcing environment, or both.
We see in most organizations that many performance problems can be addressed successfully and economically by employing teaching methods that involve high rates of responding and frequent reinforcement. Some of this is being done already.
For an example of what fluency techniques can accomplish in business, let me tell you about an application at a Delta Faucet plant in Tennessee. Delta opened a new plant, and it was important for the company to train a large number of people in a short period of time to assemble faucets quickly and correctly. By developing a computer training program based on the fluency technology, the company was able to reduce learning time from 24 hours plus extensive OJT to fewer than 10 hours plus only 1 hour of OJT. The first assembly teams went to the line having never seen a faucet part except on the computer training system, and they successfully built their first faucet in 42 minutes! Bill Hampton, human resources manager, said that each one-half hour spent in fluency training was the equivalent of three to four hours of hands-on training on the assembly line.
Blue Cross Blue Shield of Georgia used fluency-building techniques to teach clerks medical terminology. Previous classroom training lasted for three days with no real concept of what students learned or retained from the training. Deficiencies in training were soon discovered when the trainees returned to the workplace, and the company opted for a fluency implementation. Using a computer programmed for fluency training, all the students performed to criteria in fewer than five hours. These results are typical, not extraordinary.
Recently, a large healthcare provider used an accelerated learning program my team developed and reduced total training time for billing clerks from six to three months. Also, the new clerks achieved fluency in their tasks during the training, which allowed them to generate correct billing statements the first day on the job. The company expects that this training will result in a savings of $4.5 million.
Accelerated learning is not a thing of the future. It is available now. Organizations faced with the need to train more people more often should avail themselves of this powerful educational method. The educational community, having less pressure to teach more in a shorter period of time will be, unfortunately, slower to adopt these methods. In the meantime, businesses will have to continue to do the educators’ jobs for them. Because of the demonstrated effectiveness and efficiency of these techniques, business and government not only should use them—but they should encourage local schools to do likewise.
The following observation, written in 1988, supports the need for revamping our education methods:
There must be an “industrial revolution” in education, in which educational science and the ingenuity of educational technology combine to modernize the grossly inefficient and clumsy procedures of conventional education. Work in the schools of the future will be marvelously though simply organized, so as to adjust almost automatically to individual differences and the characteristics of the learning process. There will be many laborsaving schemes and devices, and even machines—not at all for the mechanizing of education, but for the freeing of teacher and pupil from educational drudgery and incompetence.11
This paragraph could have been written today. It is no less applicable to schools and business. The current model of training in business is to spend days or weeks in a classroom and months of training on the job. OJT is fast becoming a luxury that business can’t afford, and in the near future, the ability of a business to transfer knowledge efficiently will be a critical measure of survival.
1. A. D. Miller and W. L. Heward, “Do Your Students Really Know Their Math Facts?” Intervention in School and Clinic 28: 98–104; and P. Wingert, “The Sum of Mediocrity,” Newsweek, December 6, 1996, p. 96.
2. Siegfried Engelmann, War Against the Schools’ Academic Child Abuse (Portland, OR: Halcyon House, 1992).
3. Cathy L. Watkins, Project Follow-Through: A Case Study of Contingencies Influencing Instructional Practices of the Educational Establishment (Cambridge, MA: Cambridge Center for Behavioral Studies, 1997).
4. B. F. Skinner, Walden Two (Indianapolis: Hackett Publishing Company, 2005).
5. Watkins, Project Follow-Through.
6. K. R. Johnson and T. V. J. Layng, “Breaking the Structuralist Barrier: Literacy and Numeracy with Fluency,” American Psychologist 47 (1992): 1475–1490.
7. S. Engelmann and D. W. Carnine, Theory of Instruction (New York: Irvington, 1982).
8. Richard A. Schmidt, Motor Learning and Performance: From Principles to Practice (Windsor, Ontario, Canada: Human Kinetics, 1991).
9. K. Anders Ericsson and Neil Charness, “Expert Performance: Its Structure and Acquisition,” American Psychologist, August 1994, pp. 725–747.
10. Anders Ericsson and Robert Pool, Peak: Secrets from the New Science of Expertise (New York: Eamon Dolan/Houghton Mifflin Harcourt, 2016).
11. L. T. Benjamin, “A History of Teaching Machines,” American Psychologist 43(1988): 703–712.