Turning Learning Into A Business: Concerns About ‘Quality Management’ at School (*)

Share this article

September 1993

Turning Learning Into A Business

Concerns About “Quality Management” at School

By Alfie Kohn

Every few years a new idea captures the imagination of educators and suddenly seems to be everywhere at once.  The journals are filled with breathless accounts of its importance, a cadre of consultants materializes to offer workshops and trainings, and a new vocabulary insinuates itself into discussions of familiar topics.  The idea that currently fits that description is Total Quality Management (TQM), based largely on the work that the renowned statistical consultant W. Edwards Deming has done in corporate settings.[1]

Because this essay will raise concerns about the application of TQM to schools, I should begin by acknowledging not only my admiration for Dr. Deming and his ideas but also my involvement with the movement he helped to set into motion.  Some of my own writing is apparently regarded as relevant in these circles, and I have been invited to lecture at a number of Quality conferences as well as to exchange ideas with Dr. Deming as part of a videotape series on his work.  I mention these things only by way of emphasizing that I have some connection to the field I am discussing.

Not only am I enthusiastic about the precepts of TQM in a business context, but I believe many of its underlying values resonate with some of the best work in educational theory.  Deming and his followers offer an essentially positive view of human nature, emphasizing people’s fundamental desire to learn and challenge themselves.  They promote democratic environments and shared decision-making.  They stress the benefits of cooperation and the destructive consequences of competition.  They insist that a climate of trust and safety must replace one based on fear.  And they urge the abolition of systems of rating, ranking, and behavioral manipulation, including grades.

While none of these ideas is new — indeed, each has been articulated by a variety of educational writers over the decades — we should be pleased to see them corroborated by people from other fields.  Anyone keen to promote the use of cooperative learning, for example, may find it useful to mention that some corporate consultants have come to understand the value of having people work together instead of trying to defeat each other.  But pointing out parallels in passing is very different from what is now happening with TQM.  Educators are appropriating a model native to the business world and attempting to transplant it, along with its methods and metaphors, to the classroom.

Part of the problem, to be sure, is that some educators waving the Quality banner have either misunderstood Deming’s work or drawn from it selectively to lend credibility to their own objectives.  If TQM offers any imperative to people in schools, it is the need to abolish standardized tests (Holt 1993; Rankin 1992) and grades (Deming, 1991, p. 24; 1992, p.111).  Yet higher test scores are sometimes cited by self-styled proponents of TQM as the measure of their success at transforming a school or district (McLeod et al. 1992; Schmoker and Wilson, 1993) — or even used as the basis for redesigning the curriculum (Abernethy and Serfass 1992, p. 16).  Likewise, there is reason to believe that the abolition of grades, admittedly a difficult feat, is not even on the agenda of many educators who see themselves as champions of Quality.  (Some of them have moved to allow students to participate in determining their own grades [American Assn. of School Administrators 1992, p. 31] but seem to regard this modification of current practice as sufficient.)  Moreover, the use of extrinsic motivators such as stickers, verbal rewards, and other Skinnerian inducements appears to be widespread among those who invoke Deming’s name.

As we have seen so often before, then, some people fail to understand the implications of the idea in question or to implement it properly.  But my complaint goes deeper than that.  I believe that a marketplace model, even correctly interpreted and applied, simply does not belong in the classroom.[2]  The gist of my argument is that the difference between two management approaches (old and new, top-down and participative, Taylor and Deming) is less significant than the difference between any method for managing workers, on the one hand, and, on the other hand, what happens — or should happen — in a classroom.

Notwithstanding Deming’s admonition to eliminate slogans and exhortations, some articles about TQM and education consist of just that:  a string of shibboleths about systems theory, profound knowledge, the need for vision and leadership, and so on.  Of course, the repetition of favored phrases tells us nothing about how to make meaningful change, but at least there is no danger that anyone will disapprove.  By contrast, there is cause for grave concern when we come across pointed assertions that “quality is achieved…by acknowledging that education is a business” (McLeod et al. 1992, p. 40) or “the modern school should look less like a factory and more like our best high-tech companies” (David Kearns, quoted in Blankstein 1992, p. 71).  Likewise, when this philosophy is put into practice by enterprising teachers, the results verge on self-parody:  second-grade classrooms turned into replicas of manufacturing plants, with children being instructed in the fine points of filling orders, pleasing customers, and turning out products to rigid specifications.

Something has gone terribly wrong here.  But, again, the problem is not that proponents of Quality fail to understand Deming’s work so much as that they fail to understand where that work belongs.  In the rush to introduce a business model into the schools, it is easy to lose sight of the very nature and purposes of education.


What one notices straight off in reading discussions about TQM in the school is a penchant for importing the nomenclature of the workplace.  When the process of teaching children is forced onto the Procrustean bed of management theory, it is necessary to designate an educational counterpart for each role in a business setting.  Since TQM is about satisfying customers, the first order of business is to figure out who should be called the “customer” of education.  Some writers nominate students for this honor; some vigorously disagree, insisting that the word is best applied to parents or other adults in the community; some say everybody is a customer; some say it depends on the situation; some attempt to finesse the question by reminding us that this is an important matter to decide and letting it go at that.

Rather than quibbling about who fits into this slot, though, we would do better to consider whether the model is really appropriate in the first place.  That it is so difficult to agree on the educational correlative to a company’s customer should alert us to the possibility that the question is misconceived.  Attempting to answer it is about as sensible as trying to figure out which member of a family is most like a colonel and which is most like a lieutenant.

The dictionary defines “customer” as one who purchases a good or service.  To talk about learning in terms of buying and selling not only reflects a warped view of the activity but also contributes to the warping.  The words we use, of course, have the power to change how we think and behave.  To the best of my knowledge, only one individual has expressed misgivings in print about this use of language (Sztajn 1992).  Most educators who have addressed themselves to the topic of TQM adopt the terminology uncritically — and, in one case, even suggest that anyone who resists it is suffering from “a fear of industrial models” (Rhodes 1990b, p. 24).

Other words borrowed from the manager’s lexicon and applied to education are no less distasteful than “customer.”  Students are sometimes said to be “raw materials” when they arrive in school — a metaphor with definite and, to my mind, disturbing pedagogical implications — while parents are supposed to be the “suppliers.”  USX ships steel to General Motors to be made into cars; Mom and Dad (having shed their status as customers) ship their child (ditto) to school to be processed into some other sort of commodity.  Alternatively, the job of teachers has been described as “delivering instruction” to students (Willis 1993, p. 4), a formulation that suggests a one-way transfer of information or skills.

The deafening clash of metaphors gets louder when students are also referred to as “workers” (e.g., Bonner 1982; Bonstingl 1992, p. 28; Schmoker and Wilson 1993, p. 393).  This locution, perhaps the most troubling of all, represents the logical conclusion of a well-established trend.  Many business leaders and public officials, after all, have argued that the primary objective of schools is to turn out willing, skilled employees whose labors will help corporations to triumph over their counterparts in other countries.  Educating children is seen not as something valuable in its own right but as an investment, something “important primarily because of its capacity to rescue, perpetuate, or enhance US business profitability” (Ray and Mickelson 1990, p. 123).  It is probably not a coincidence that the rationale for adopting TQM in particular is often couched not in terms of how students can be helped to become self-directed, lifelong learners but rather in terms of improving corporate competitiveness in global markets (e.g., Leonard 1991; McLeod et al. 1992; Schmoker and Wilson 1993).

Implicit in this perspective is the view of students as potential workers.  But in the last few years, we have witnessed a shift to something even more ominous:  a view of students as actual (that is, current) workers.  The activities in which they are already engaged are described in economic terms.  Moreover, businesspeople and politicians are not the only ones who describe schoolchildren this way.  Albert Shanker (quoted in Perry 1988, p. 44) declares that “we must start thinking of students as workers.”  William Glasser (1990, p. 6), whose books have supported many educators in transforming schools into more humanistic places, inexplicably insists that “the industrial analogy of workers and managers…is both accurate and appropriate.”

Need we enumerate the respects in which the analogy is neither?

* Workers are adults; most students are children, whose capacities and limits demand a developmentally appropriate set of strategies.  Age differences not only inform what and how we teach, but also affect our response to inappropriate behavior.  When a worker acts in such a way as to cause a problem for others, the supervisor may try to be understanding, but his or her attitude is altogether different from that toward a child who misbehaves.

* If workers are helped to acquire skills, this is generally intended as a way to build an effective organization; it is rarely a goal in its own right.  By contrast, helping students to acquire skills, to become good learners and good people, is the very point of school.

* While it is undesirable for many reasons to manipulate workers with incentives (Kohn 1993a), they still have to be paid; money is needed to survive.  Nothing, including grades, is analogous in a school setting.

* Most importantly, workers spend their days producing goods like automobiles and houses; they are hired to make things.  The only thing students should be making is meaning.  To turn the classroom into a workplace, through our practices or our parlance, is to put at risk the intellectual exploration and development that ought to be taking place there.


If we are serious about improving the quality of education, we need to think about the theories of learning that lie behind what goes on in schools.  Do we assume that children of different ages learn in pretty much the same way?  Do we see our task as feeding discrete bits of knowledge into passive receptacles?  Is the learning process teacher-centered or learner-centered?  These are only a few of the questions that matter, but it is unreasonable to expect a theorist or practitioner of corporate management to provide the answers to any of them.  When someone in the business world asks me for guidance, I unhesitatingly recommend the work of Deming.  Educators, though, would do better to turn first to Dewey.

What is striking about the dozens of articles and books on the application of TQM to education is their failure to address any fundamental questions about learning, per se.  The implication seems to be that these issues are less important than matters of management.  Even more striking is the absence of any discussion about curriculum.  Incredibly, page after page of talk about Quality schools typically contains not a single reference to whether the things we are teaching are worth learning — whether the curriculum is (or can be made) engaging, important, and relevant to children’s experiences.  In fact, one influential introduction to the topic explicitly promotes TQM as “content-free, applicable to any instruction or structural reform” (American Assn. of School Administrators 1992, p. 6).[3]

This is hardly the first time educators have been offered a package said to be compatible with any curriculum — one whose appeal lies precisely in the fact that educators need not reconsider what they are asking students to learn.  For example, behavior modification techniques are designed to increase compliance without regard to the value of the task or behavior in question (Kohn 1993b).  Certain versions of cooperative learning are also marketed as being adaptable to any curriculum, a strategy that Sapon-Shevin (1991) has wryly described as the “hamburger helper” model.  That TQM takes the same approach should not be surprising, perhaps, since its use in a corporate context is not dependent on what people in an organization are doing:  whether a company manufactures medicine or missiles is of no consequence to consultants who talk about moving toward Quality.

At best, then, TQM offers little challenge to the curricular status quo and leaves students with the same material to learn.  But if, as I suspect, it turns out not to be content-neutral in practice, its impact may actually be counterproductive.  The premise of TQM, at least as it is described by many of its proponents, seems to be that the trouble with organizations of all sorts, including schools, is lack of data.  Great importance is placed on the use of statistical control diagrams, flow charts, and a variety of other pictorial representations of quantitative information about performance.  New converts to TQM frequently display examples of these tools, along with pride at having mastered them.

Is it really true that the principal problem with American classrooms is a lack of data?  I think not, but we can put off this question to another day.  Instead, let us consider the possible effect on the curriculum of this way of thinking.  In academic social science, the questions addressed in a given discipline often seem to be selected on the basis of how well they lend themselves to the methodologies native to that field: the method drives the content.  Similarly, it is entirely possible that TQM classrooms will incline toward the analysis of data and the assignment of tasks where progress can be easily quantified and analyzed.  While Deming has said unequivocally that “the most important things we need to manage can’t be measured” (quoted in Rhodes 1990a, p. 34; also see Deming 1991, p. 27), an emphasis on statistical analysis, which was developed in the context of workplace production, is likely to accelerate the tendency toward teaching by the numbers.

There is evidence, in fact, that this is already happening.  In one case, proponents of systems thinking have talked about computer simulations of the actions of Shakespeare’s characters in order to maximize “the use of precise numbers to talk about psychological motives” (Senge and Lannon-Kim, 1991, p. 11).  In another case, people committed to Quality have built into assessment instruments such measures as “the number of poems a student [has] memorized” (Harris and Harris, 1992, p. 20).  David Langford, a pioneer in the field, was quoted recently as dismissing “the argument that much of what happens in the classroom is intangible and hence not subject to the statistical scrutiny TQM prescribes.  ‘It can all be measured,’ he maintains” (Willis, 1993, p. 5).

It was e.e. cummings who declared that “nothing measurable matters a good goddam.”  Even if we are not willing to go quite that far, it is hard to deny that much of what is worth learning cannot be usefully reduced to numbers without doing violence to its substance.  In fact, anyone sensitive to the tyranny of standardized testing should be just as skeptical about turning the process of children’s intellectual creation and discovery into bars on a Pareto chart.

At the heart of Deming’s approach is a repudiation of the use of numerical goals and mass inspection at the end of production.  What we choose to quantify, and how we do so, is highly relevant within the Quality framework.  But what are we to make of the assertion that “improving the system” for educators as well as managers means trying to “narrow the amount of variation within it” (Blankstein 1992, p. 72)?  Or that it means “removing inefficiencies” in the way students study (Tribus, 1990, p. 2) or in the curriculum itself (Schmoker and Wilson 1993, p. 392)?  Or that children should be subjected to “frequent assessment of quality at every stage” (Schmoker and Wilson 1993, p. 392) — indeed, “on a daily basis” (McLeod et al. 1992, p. 36)?

Assuming that these writers have not misinterpreted the goals of TQM, the issue once again is whether such a system is consistent with our goals as educators.  If we prefer to focus on process rather than product — if efficiency is a goal incommensurate with the act of discovery — then we have reason to question the appropriateness of TQM in a classroom setting.  Nor will our concerns be allayed when some of the particular recommendations being offered to us are modified a bit.  Even if students’ performance isn’t scrutinized every day, even if it isn’t construed in terms of reducing variation, and even if it isn’t relentlessly quantified, there is no getting around the fact that the Quality model is designed to draw attention to performance in order to improve it.

Unfortunately, a performance focus is inherently problematic in a classroom.  The work of leading motivation researchers (Ames, 1992; Dweck, 1986; Nicholls, 1989), none of which has been mentioned by defenders of TQM’s role in education, converges on a single crucial distinction concerning how to think about what happens in schools.  Variously framed as “mastery vs. ability,” “learning vs. performance,” and “task vs. ego,” the basic point is that there is an enormous difference between getting students to think about what they are doing, on the one hand, and how well they are doing (and therefore how good they are at doing it), on the other.  The latter orientation does a great deal of harm.

Students who must spend their time focusing on the quality of their work are less likely to succeed.  This may seem paradoxical, but the fact is that children overly concerned about how well they are doing come to see learning as a means to an end and start to think that their performance, especially when they fail, is due to innate intelligence (or its absence):  I screwed up, therefore I’m stupid.  That, in turn, leads them to assume there isn’t much point to trying harder next time, which means they are unlikely to improve.

This is not to say that when an assignment has been completed its worth cannot be judged.  There is a time to think about whether what one has done is any good, and it is important to talk with others about the results of one’s work.  But the research is clear that getting children to become preoccupied with their performance can interfere with their ability to remember things about the challenging tasks they just worked on (Graham and Golan, 1991).  It can undermine their ability to apply scientific principles to new situations (Dweck 1986).  It can reduce the quality of their work as measured on tests of creativity (Butler 1992) and their “readiness to contemplate diverse ideas” (Nicholls, 1989, p. 130).  Moreover, it can increase their fear of failure (Heyman and Dweck 1992), undermine their interest in what they are doing (Ryan 1982; Butler 1987) and lead them to choose the easiest possible task in an effort to avoid the sort of challenge that might lead to failure (Maehr and Stallings 1972; Harter, 1978; Pearlman, 1984; Elliott and Dweck, 1988).

Getting students to think about their performance on tests, or the grade they will receive, is a particularly potent way to inhibit creative thinking and conceptual learning (Grolnick and Ryan 1987; Butler and Nisan 1986; Butler 1988; Benware and Deci 1984), but the point here is that any sort of performance orientation can be detrimental — and TQM is unavoidably based on just such an orientation.

In another setting, with different people doing different things for different reasons, the Deming model is without equal.  But before we jump on the bandwagon carrying that model into our classrooms, we need to ask some hard questions about the nature of learning and the purposes of schools.  No one can object to “quality” as an abstract concept, but if a set of practices revolving around that word requires us to treat students as customers or workers, then we should leave management practices to managers and get on with the challenge of educating.


Postscript:  After this article was published, I learned that Deming himself was more reasonable than many Demingites:  At a March 1993 seminar in Detroit, presumably in response to a question, he growled, “We don’t have customers in education.  Don’t forget your horse sense.”  (Quoted by Maurice Holt, “Why Deming and OBE Don’t Mix,” Educational Leadership, September 1994, p. 85.)


1.  It is worth noting that Deming has pointedly distanced himself from much of what goes by the name of TQM.  Nonetheless, I refer here to Deming’s principles and the Quality movement inspired (if not always approved) by him more or less interchangeably.  I do so for the sake of simplicity and because my concerns about using Deming’s work in the classroom go well beyond the issue of whether the people citing that work are using it imprecisely.

2.  Whether it belongs in any educational setting — for example, whether it is appropriate to managing the work of adults in a school or district — is a question worth pondering, but one that I will not attempt to deal with here.  My concern is with teaching children.

3.  In his book on Quality schools, Glasser has written, “Actually, there is little wrong with the curriculum.  What is wrong is our inability to persuade more than a small percentage of students to learn it well” (1990, p. 118).  More recently, though, he seems to have changed his mind, declaring, “No matter how well the teachers manage them, if students do not find quality in what they are asked to do in their classes, they will not work hard enough to learn the material.  The answer is not to try to make them work harder; the answer is to increase the quality of what we ask them to learn” (1992, p. 691).


Abernethy, P. E., and R. W. Serfass.  “One District’s Quality Improvement Story.”  (November 1992).  Educational Leadership: 14-17.

American Association of School Administrators.  (1992).  Creating Quality Schools.  [Arlington, Va.:] AASA.

Ames, C.  (1992).  “Classrooms: Goals, Structures, and Student Motivation.”  Journal of Educational Psychology 84: 261-71.

Benware, C. A., and E. L. Deci.  (1984).  “Quality of Learning with an Active Versus Passive Motivational Set.”  American Educational Research Journal 21: 755-65.

Blankstein, A. M.  (March 1992).  “Lessons from Enlightened Corporations.”  Educational Leadership: 71-75.

Bonner, J. S.  (June 1982).  “Japanese Quality Circles: Can They Work in Education?”  Phi Delta Kappan: 681.

Bonstingl, J. J.  (1992).  Schools of QualityAn Introduction to Total Quality Management in Education.  Alexandria, Va.: ASCD.

Butler, R.  (1987).  “Task-Involving and Ego-Involving Properties of Evaluation: Effects of Different Feedback Conditions on Motivational Perceptions, Interest, and Performance.”  Journal of Educational Psychology 79: 474-82.

Butler, R.  (1992).  “What Young People Want to Know When: Effects of Mastery and Ability Goals on Interest in Different Kinds of Social Comparisons.”  Journal of Personality and Social Psychology 62: 934-43.

Butler, R., and M. Nisan.  (1986).  “Effects of No Feedback, Task-Related Comments, and Grades on Intrinsic Motivation and Performance.”  Journal of Educational Psychology 78: 210-16.

Deming, W. E.  (1991).  “Foundation for Management of Quality in the Western World.”  Paper delivered in Osaka, Japan, 1989; revised 1991.  Reprinted in An Introduction to Total Quality for SchoolsA Collection of Articles on the Concepts of Total Quality Management and W. Edwards Deming, edited by the American Association of School Administrators.  Arlington, Va.: AASA.

Deming, W. E.  (1992).  “The New Economics for Industry, Education, Government.”  Unpublished manuscript.

Dweck, C. S.  (1986).  “Motivational Processes Affecting Learning.”  American Psychologist 41: 1040-48.

Elliott, E. S., and C. S. Dweck.  (1988).  “Goals: An Approach to Motivation and Achievement.”  Journal of Personality and Social Psychology 54: 5-12.

Glasser, W.  (1990).  The Quality SchoolManaging Students Without Coercion.  New York: Harper and Row.

Glasser, W.  (May 1992).  “The Quality School Curriculum.”  Phi Delta Kappan: 690-694.

Graham, S., and S. Golan.  (1991).  “Motivational Influences on Cognition: Task Involvement, Ego Involvement, and Depth of Information Processing.”  Journal of Educational Psychology 83: 187-94.

Grolnick, W. S., and R. M. Ryan.  (1987).  “Autonomy in Children’s Learning: An Experimental and Individual Difference Investigation.”  Journal of Personality and Social Psychology 52: 890-98.

Harris, M. F., and R. C. Harris.  (November 1992).  “Glasser Comes to a Rural School.”  Educational Leadership: 18-21.

Harter, S.  (1978).  “Pleasure Derived from Challenge and the Effects of Receiving Grades on Children’s Difficulty Level Choices.”  Child Development 49: 788-99.

Heyman, G. D., and C. S. Dweck.  (1992).  “Achievement Goals and Intrinsic Motivation: Their Relation and Their Role in Adaptive Motivation.”  Motivation and Emotion 16: 231-47.

Holt, M. (January 1993).  “The Educational Consequences of W. Edwards Deming.”  Phi Delta Kappan: 382-388.

Kohn, A. (September-October 1993a).  “Why Incentive Plans Cannot Work.”  Harvard Business Review: 54-63.

Kohn, A. (1993b).  Punished by Rewards The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes.  Boston: Houghton Mifflin.

Leonard, J. F. (1991).  “Applying Deming’s Principles to Our Schools.”  Reprinted in An Introduction to Total Quality for SchoolsA Collection of Articles on the Concepts of Total Quality Management and W. Edwards Deming, edited by the American Association of School Administrators.  Arlington, Va.: AASA.

Maehr, M. L., and W. M. Stallings.  (1972).  “Freedom from External Evaluation.”  Child Development43: 177-85.

McLeod, W. B., B. A. Spencer, and L. T. Hairston.  (1992).  “Toward a System of Total Quality Management: Applying the Deming Approach to the Education Setting.”  ERS Spectrum 10, 2: 34-41.

Nicholls, J. G.  (1989).  The Competitive Ethos and Democratic Education.  Cambridge, Mass.: Harvard University Press.

Pearlman, C. (1984).  “The Effects of Level of Effectance Motivation, IQ, and a Penalty/Reward Contingency on the Choice of Problem Difficulty.”  Child Development 55: 537-42.

Perry, N. J. (7 November 1988).  “Saving the Schools: How Business Can Help.”  Fortune: 42-56.

Rankin, S. C.  (September 1992).  “Total Quality Management: Implications for Educational Assessment.”  NASSP Bulletin: 66-76.

Ray, C. A., and R. A. Mickelson.  (1990).  “Business Leaders and the Politics of School Reform.”  In Education Politics for the New CenturyThe Twentieth Anniversary Yearbook of the Politics of Education Association, edited by D. E. Mitchell and M. E. Goertz.  London: Falmer Press.

Rhodes, L. A.  (November 1990a).  “Why Quality Is Within Our Grasp … If We Reach.”  The School Administrator: 31-34.

Rhodes, L. A.  (December 1990b).  “Beyond Your Beliefs: Quantum Leaps Toward Quality Schools.”  The School Administrator: 23-26.

Ryan, R. M.  (1982).  “Control and Information in the Intrapersonal Sphere: An Extension of Cognitive Evaluation Theory.”  Journal of Personality and Social Psychology 43: 450-61.

Sapon-Shevin, M.  (1991).  “Cooperative Learning:  Liberatory Praxis or Hamburger Helper?”  Educational Foundations 5: 5-17.

Schmoker, M., and R. B. Wilson.  (January 1993).  “Transforming Schools Through Total Quality Education.”  Phi Delta Kappan: 389-395.

Senge, P., and C. Lannon-Kim.  (November 1991).  “Applying the Systems Approach in the Classroom.”  The School Administrator: 10-11.

Sztajn, P.  (November 1992).  “A Matter of Metaphors: Education as a Handmade Process.”  Educational Leadership: 35-37.

Tribus, M.  (November 1990).  “The Application of Quality Management Principles in Education, at Mt. Edgecumbe High School, Sitka, Alaska.”  Unpublished manuscript.

Wills, S.  (February 1993).  “Creating ‘Total Quality’ Schools.”  ASCD Update: 1, 4-5.

Copyright © 1993 by Alfie Kohn. This article may be downloaded, reproduced, and distributed without permission as long as each copy includes this notice along with citation information (i.e., name of the periodical in which it originally appeared, date of publication, and author’s name). Permission must be obtained in order to reprint this article in a published work or in order to offer it for sale in any form. Please write to the address indicated on the Contact Us page.
 www.alfiekohn.org — © Alfie Kohn

Share this article