Marilyn Kourilsky
Jack Hirshleifer

ABSTRACT

The present study compares the effects of two teacher-training programs designed to emphasize one of two forms of behavior modification in teaching economics to 4th, 5th, and 6th grade pupils. Of 20 teachers trained in Mini-Society techniques (Socially Emergent Behavior Modification-SEBM), half were instructed to convert their classes into Token Economies (Imposed Behavior Modification-IBM), while half were trained to prevent such conversion. Instruments were employed to test effects of the different techniques upon improvements in economic understanding and upon autonomy. While both techniques of instruction yielded significant improvements in economic understanding, the SEBM (pure Mini-Society) results were significantly superior to those achieved by IBM (Mini-Society converted to Token Economy). In addition, only SEBM technique yielded increased scores for autonomy.

THERE HAS BEEN some confusion concerning two methods of influencing learning and other desired classroom behaviors via economic motivations: Mini-Society (10) versus Token Economy (14). Both methods represent forms of behavior modification in the broadest sense of that term. The key difference between them lies in the generation of the incentive structure. Token Economy influences subjects' choices through a dictated schedule of rewards and penalties, a prescribed structuring of the choice environment that may be termed Imposed Behavior Modification (IBM). In contrast, Mini-Society aims at reproducing, in the classroom as microcosm, those forms of influences on behavior that tend to arise naturally out of the cooperative/competitive co-existence of individuals in a society of scarce resources. In Mini-Society, rewards and penalties are determined by the freely emergent social institutions of the classroom - which might take on more or less recognizable forms such as property, majority voting, contract, etc. - and which may vary from one classroom community to another. A choice environment structured in this way may be termed Socially Emergent Behavior Modification (SEBM).

The two approaches distinguished here correspond with two underlying conceptions of economics, which may be called the "reactive" and the "creative" views. The former, the standard approach of modern analytical economics, regards the individual as a utility-maximizing mechanism aiming to achieve his/her desired ends within the constraints of his environment. This reduction of the human being to "economic man," while an admittedly over-simplified picture, has proved extremely useful in scientific work in economics. It permits the construction of mathematical models of behavior (i.e., of reactive response to changes in environment) whose predictions can then be tested against reality. Nevertheless, many great figures in economics - in the recent period including such names as Schumpeter (16), Knight (9), and Boulding (5) - have insisted that something very important is lost in such schema. The role of self-determination and creativity in economic decision-making, epitomized by the business entrepreneur, is not successfully captured in models of formal optimization. The entrepreneur does not merely optimize in a reactive way to environmental changes. Rather, his/her innovative decisions help create the environment in which (s)he and others live.

Token Economy corresponds to the reactive approach in economics. If, for example, punctuality is desired by the experimenter(s), (s)he offers an increased premium for on-time behavior, which changes the environment so as to make a higher degree of punctuality the optimal choice of the subject. In a Mini-Society, on the other hand, the question of inducing punctuality rarely comes up. And yet the typical Mini-Society may well achieve more punctuality, not because it is rewarded directly, but because punctuality provides more time for desired creative activities. (It is, of course, also possible that a particular Mini-Society might evolve in such a way as to explicitly reward punctuality.)

We propose that the IBM/SEBM distinction made above - corresponding to treating the individual as a fundamentally reactive mechanism on the one hand, or on the other hand as a socially influenced yet potentially creative and self-motivated personality - explains some of the puzzling discrepancies encountered in the literature on behavior modification.

The ordinary common sense of mankind tells us that individuals will reactively respond so as to engage more in an activity if rewarded, less if penalized. Practically the entire empirical corpus of economics elaborates this theme; "Law of Demand," as perhaps the best-known example, has been validated in innumerable circumstances. More recently, economists have gone on to show that what might be called "reactive rationality" has explanatory power even outside the narrowly-defined market sphere, for example, in the study of crime (3) or fertility (2). And such familiar psychological theories as Thorndike's Law of Effect and Skinner's operant conditioning, with their supporting bodies of evidence (8), parallel this view: positive reinforcement elicits more of a given behavior, negative reinforcement suppresses it. Behavior modification in the usual (IBM) sense is thus firmly grounded on both economic and psychological tradition.

On the other hand there have been some seeming paradoxes and failures of behavior modification programs. According to one review, IBM in the form of a Token Economy was more successful in controlling such behaviors as getting out of one's seat or talking out of turn, and less successful in inducing academic improvement (14:385). Of course, academic success requires a kind of self-motivated creative input on the part of the subject - something more than simply reactive adaptation. Also, social influences such as teacher praise and teacher expectations have an important effect upon academic behavior (14:386).

Among the known problems of Token Economy is evidence that upon withdrawal of an IBM reward the desired behavior will decline (15:5). This, of course, makes perfect sense on a reactive "economic-man" approach, though psychological conditioning theory might have led one to hope for a result less vulnerable to extinction. There are even some data suggesting that providing "extrinsic" incentives via a Token Economy may undermine the subjects' intrinsic incentive (self-motivation), so that ultimately less of the desired behavior will be forthcoming than if the program had never been instituted at all (7, 11, 12). The underlying explanation, we conjecture, is the psychologically invasive or reductive character of IBM. When subjects do not share in the decision-making process that generates their society's structure of rewards and penalties, they are in effect being treated as mere reactive mechanisms. Reduction of desired behavior represents a rebellion against the limitation upon self-motivation and creativity - even though the behavior in question may have been made "profitable," and even if the behavior is one in which the subjects originally had a positive intrinsic interest (15:6). In contrast, subjects who have participated in the development of the incentive structure achieve better in areas like academic progress, where self-motivation and some creative input are necessary for long-term success (13).

These problems and difficulties reported in the literature were paralleled in the senior author's experience by some of the results of past Mini-Society programs (10). The distinction between Mini-Society and Token Economy is, of course, not an absolute one. Even under Mini-Society the teacher, as an authority figure in the classroom, cannot entirely avoid imposing on students some of the personal reward-punishment system he or she is in the habit of implementing. And even under Token Economy systems, the particular personalities of the children involved cannot help but have some effect upon the structure of the actual social system that emerges. The particular problem observed was a tendency for Mini-Society as implemented in the classroom to slip in the direction of Token Economy. There may have been failure on the part of the teacher to grasp entirely the rationale of the instructional technique, impatience with the fumbling creative efforts of children, or simply inability to refrain from imposition of the instructor's aims and values. Slippage most commonly took the form of payment for non-productive activities or student obligations such as punctuality, completion of homework assignments, etc. Whatever the reason for slippage, the authors' informal observation was that children in Mini-Society classes that had slipped in the direction of Token Economy tended to perform below levels achieved by those in an uncontaminated Mini-Society. No formal test of the differences had as yet been conducted, however.

The above considerations led to the design of an experiment intended to elicit the hypothesized differential effects of Token Economy as an instance of Imposed Behavior Modification. The criteria to be considered were: (1) academic understanding, as measured by scores on an instrument testing economic knowledge at a cognitive level higher than recall, and (2) perceived autonomy, as measured by scores on instruments testing achievement responsibility. In addition, the differential impact upon assertiveness - a character trait that might be at least conjecturally associated with autonomy - was also tested. The underlying hypotheses were that Mini-Society subjects would (a) exhibit greater academic achievement, as a result of the environment being more congenial to self-motivation and to the creative child inputs necessary for learning at higher levels of the cognitive domain, and (b) exhibit a greater degree of perceived autonomy as a result of experience in situations which, rather than calling simply for reactive adaptation, encourage creative participation in the structuring of their environment.

Method

The study included 387 fourth, fifth, and sixth grade children in 20 different classes from Los Angeles County. The students were mostly from middle-class backgrounds and represented various ethnic groups. All 20 teachers attended a 24-hour economic education workshop, half were randomly assigned to the experimental group, the Mini-Society treatment, and the other half were assigned to the control group, the Token Economy treatment. The 20 teachers received instruction on the fundamentals of the Mini-Society strategy, and each subsequently passed a test in economics and education at or above the 90 percent competency level. However, the control group received additional training on how to convert a Mini-Society into a Token Economy whereas the experimental group received additional training on how to prevent a Mini-Society from becoming a Token Economy.

Instruments

Both the control and the experimental groups were given the following three tests: (I) The Test on Economic Decision-Making (TED), designed by the authors and the Joint Council on Economic Education, (2) The Intellectual Achievement Responsibility Questionnaire (IAR), designed by Crandall, Katonsky and Crandall (6) and (3) The Assertiveness Index, designed primarily by R.B. Porter and Cattell.

The Test on Economic Decision-Making (TED) consists of 30 multiple-choice items (four choices per question) on economic concepts related to decision-making. All questions involve the ability to apply a specific concept or to utilize an economic principle in performing an analysis. All of the questions are beyond the recall level of the cognitive domain (4). Three of the items deal with scarcity, four of the items test the three basic economic problems faced by all societies, four questions deal with cost-benefit analysis, four questions examine supply and demand, four questions test competition and monopoly, three items examine inflation, five items test money and banking, and three questions test for the mathematics of economics.

The Intellectual Achievement Responsibility Questionnaire (IAR) is an autonomy index. Specifically, it is an instrument for assessing children's belief that they, rather than other people, are responsible for their intellectual/academic successes and failures. The short form utilized by the authors consists of 16 questions of the original 30 that were found to be the most sensitive indicators of high versus low autonomy individuals. A typical question is as follows: "Suppose you study to become a teacher, doctor, or scientist, and you fail. Do you think this would happen (a) because you didn't work hard enough, or (b) because you needed some help and other people didn't give it to you?"

A child's "I+" score is obtained by summing all positive events for which (s)he assumes responsibility (credit), and the "I-" score, by summing all negative events for which (s)he assumes responsibility (blame). The possible scores for both "I+" and "I-" range from 0-8. A pupil's "I" score is the sum of the "I+" and "I-" scores and ranges from 0-16.

The Assertiveness Index consists of the five questions most germane to a child's world chosen from Porter and Cattell. Assertiveness is the quality of being appropriately emotionally direct, self-enhancing, and expressive, whereas nonassertive qualities include being emotionally indirect, self-denying, and inhibited (1). The assertiveness instrument is typified by the following question: "I am afraid to talk to most adults: (a) very true, (b) somewhat true, (c) somewhat untrue, (d) very untrue." The scoring on each question ranges from 1 to 5, with 3 being eliminated since no middle-of-the-road responses were provided. The higher the score, the higher the assertiveness level.

Procedure

Twenty teachers each received 24 hours of training in the Mini-Society Instructional System. They were then randomly separated into two groups of ten, allegedly for purposes of small-group interaction. The groups met in different locations to avoid mutual influence upon one another. Two additional hours were devoted to training each group.

More specifically, the procedures prescribed for the Mini-Society classes (experimental group) had the following major features: (a) identification of scarcity situations ("not enough pencils to go around"); (b) enactment of alternative solutions to the problem of scarcity including, in every case, exchange; (c) design of currency; (d) regular auctions; and (e) apart from the preceding, teacher self-restraint (participation in activities as paid consultant only, and even then in competition with student consultants) The teachers were told that they could expect to observe the evolution of business firms (including banks and insurance companies), government institutions like taxation and welfare, and even international (i.e., inter-classroom) trade. (See Appendix)

One compromise of the Mini-Society principle was made. An initial payment for short-term nonproductive activities (e.g., punctuality for the first week) was permitted solely for the purpose of infusing funds into the system. The effect of this initial short-run IBM-type perturbation was felt to be negligible relative to the cumulative effects of the Mini-Society program, and it is a requisite for initiating the Mini-Society. At the discretion of the teacher, a second such payment was permitted if it appeared necessary. Apart from this, the teachers were urged not to allow slippage, intentionally or unintentionally, from Mini-Society to Token Economy techniques, which also were carefully illustrated.

In the Token Economy classes (control group), the teachers were encouraged to select at least two or three behaviors for which children would receive weekly payments. Examples of such behaviors provided in the training sessions consisted solely of obligations normally incumbent upon students simply by virtue of being members of the class, e.g., neatness, not disturbing others, handing in homework assignments. The teachers were not instructed either to include or to omit academic achievement as a rewarded behavior; good grades were, in fact, rewarded in some, but not all, classes. The tokens earned by students could be spent in auctions or in buying goods and services from classroom enterprises.

Observation of all classes verified that in fact the treatments assigned to the experimental and control groups respectively were being carried out.

At the beginning of the semester, students in each classroom were given three pretests: the TED, the IAR, and the Assertiveness Index. They were posttested with the same instruments at the end of the semester. The classroom, as opposed to each subject, was considered the experimental unit, so that 20 observations were available.

The specific questions examined in the study were as follows:

1. At the end of the semester do students in Mini-Societies or Token Economies demonstrate more economic knowledge?
2. At the end of the semester is there a significant difference between Token Economy participants and Mini-Society participants with respect to (a) autonomy in general, (b) autonomy regarding positive events, (c) autonomy regarding negative events, and (d) assertiveness?
3. Do either of the instructional interventions - Mini-Society vs. Token Economy - have a significant effect on (a) economic understanding, (b) autonomy, or (c) assertiveness? We also tested for sex differences with respect to economic knowledge, autonomy, and assertiveness.

Results

Study question 1 was concerned with whether participation in Token Economy versus Mini-Society resulted in different levels of economic knowledge. Study question 2 involved the relative effects of the two economies upon autonomy and assertiveness. For both questions, t-statistics were computed. Study question 3 examined whether, for either of the economic curricula, significant gains from pre- to posttest were obtained for economic understanding, or for the attitude measures. A correlated t test was the most appropriate measure to investigate this issue. Possible differences between boys and girls with respect to the above variables were ascertained through use of a two-way analysis of variance blocking on sex.

Differences at high levels of statistical significance (p<.05) in favor of Mini-Society participants were obtained for economic understanding, autonomy, and assertiveness (study questions 1 and 2), as reported in Table 1.

Table 2 summarizes the correlated t results for the pre- to posttest comparisons on all five measures (study question 3). Although it was found that, on the basis of posttest comparison, the Mini-Society type economy was the favored curriculum for mastery of economic principles and increased autonomy and assertiveness, it was still possible that Token Economy students might have advanced significantly from the pre- to the post-testing on some variables. The results indicate that Mini-Society students scored significantly higher on the posttest for three measures - economic understanding total autonomy, and assertiveness. By contrast, the Token Economy students significantly advanced from pre- to posttest on only a single measure - economic understanding. Thus, it appears from these data that the Token Economy curriculum has a capability (although significantly inferior to that of the Mini-Society) of transferring skills in the general area of economic knowledge, but does not lead to higher degrees of autonomy or assertiveness.

As mentioned previously, sex differences with respect to Mini-Society and Token Economy were explored through use of a two-way analysis of variance blocking for sex; sex of the child was not found to be a significant factor for any of the measures.

Discussion

Mini-Society classes in this study significantly surpassed Token Economy classes in improvement of scores for economic knowledge and for autonomy and assertiveness. Academic achievement and attitudinal changes in the direction of autonomy thus appear to be favored by an environment that promotes self-motivation, creativity, and participation in the structuring of incentives - an environment characterized by Socially Emergent Behavior Modification (SEBM). In contrast, where the incentive structure is dictated from above - Imposed Behavior Modification (IBM) - these activities and qualities tend to be suppressed. The two environments both reward the rational calculation of advantages, but under different "rules of the game." It is possible for a subject in a dictatorial society, or even for a slave, to engage in rational calculation of the costs and benefits to him or her of alternative courses of action. But a free society calls for a higher order of calculation, one involving creative social participation and innovation. It was this quality that the present study tried to measure in the scores for autonomy and assertiveness. And as a by-product, academic achievement, requiring as it does some creative input on the part of the student, also seems to respond better to the SEBM-type environment of Mini-Society than to the IBM-type environment of Token Economy.

One important caution must be included, however. The system contrasted with Mini-Society here is not a "pure" Token Economy system, but one developed by teachers who had previously been instructed in the principles of Mini-Society.

Token Economy might have scored better under the direction of teachers who had not previously been "contaminated" by exposure to Mini-Society. The opposite may also be the case, however; if the author's conceptual interpretation is valid, a "pure" Token Economy - one in which the incentive structure is totally imposed on students from above, rewarding adaptive obedience to rules rather than creative innovation and social participation - would have scored more poorly than reported here. The resolution of this question must wait upon additional studies, preferably to be performed by a variety of authors with differing points of view and methods of attack.

APPENDIX

Summary of the Mini-Society

In the Mini-Society, the teacher is presented with a set of procedures for helping the children generate a working economic system in their classroom. The system is initiated by focusing on several scarcity situations in the classroom, such as "not enough felt-tip pens to go around." The children will enact several solutions to the problem of scarcity and will eventually try out a system of exchange where, in their words, "you pay for what you want." Initially they identify a few activities for which they will be paid in the currency of their Mini-Society. The currency will be used to bid for the felt-tip pens and other scarce resources. Immediately, the class is motivated to develop, design, and print its own currency. As the system swings into action, children begin to buy and sell goods and services such as pencils, crafts, posters, logos for businesses, and their time. They also rent "space" in the classroom to conduct business.

Situations later arise which challenge the children to develop institutions to facilitate the smooth operation of their system. Unequal distribution of wealth and varying spending patterns create a demand for lending institutions, banks, and insurance companies.

Some entrepreneurs may sell stock in their films in order to raise capital. As the children seek advice on how to establish various businesses, a system evolves in which the teacher and other members of the community play the role of paid consultants. In response to consumer demand, children develop their own firms ranging from simple billfold factories to elaborately conceived insurance companies. The growing activity encourages the teacher, as government, to hire class members to do certain "government" jobs. Competent "civil servants" are often bid away from the government by firms owned by the children. Trade is further enhanced through weekly auctions in which children bid for goods and services provided by other children. International trade often emerges in schools where there are several classes with Mini-Societies. Some children become specialists in the buying and selling of currency. An income tax, often begrudgingly instituted by the children, leads to the formation of a representative type of government, so that students gain appreciation not only of economics but also of political realities. As a result, the Mini-Societies developed in each classroom are different - not by design, but because children are different. To sum it up, if you asked the teacher, "How do the children know what to do in their Mini-Society economy?" the teacher might answer you by asking, "How do you know what to do in your economy?"

REFERENCES

1. Ashby, N., "Stop Putting Up With Put Downs," Today's Health, July-August 1975. pp. 15-19.

2. Becker, G.S., "An Economic Analysis of Fertility," Demographic and Economic Change in Developed Countries, National Bureau of Economic Research, Princeton University Press, 1960.

3. Becker, G.S., "Crime and Punishment: An Economic Approach ," Journal of Political Economy, 76, March-April 1968.

4. Bloom, B.S. (ed.), Taxonomy of Educational Objectives: The Classification of Educational Goals, Handbook I: Cognitive Domain, David McKay Company, New York, 1956.

5. Boulding, K.E., "Economics as a Moral Science," American Economic Review, 59, March 1969.

6. Crandall, V.; Katonsky, W.; Crandall, V., "Children's Beliefs in Their Own Control of Reinforcements in Intellectual-Academic Achievement Situations," Child Development, 6: 91-109, 1965.

7. Deci, E.L., "Effects of Externally Mediated Rewards on Intrinsic Motivation," Journal of Personality and Social Psychology, 18: 105-115, 1971.

8. Kimble, G., "Learning: Introduction," International Encyclopedia of the Social Sciences, Vol. 9, Macmillan and Free Press, 1968.

9. Knight, F.H., Risk, Uncertainty, and Profit, Houghton Mifflin, Boston, 1921.

10. Kourilsky, M., Beyond Simulation: The Mini-Society Approach to Economics and other Social Sciences, Educational Research Associates, Los Angeles, California, 1974.

11. Kruglanski, A.W.; Friedman, I.; Zeevi, G., "The Effects of Extrinsic Incentive on Some Qualitative Aspects of Task Performance," Journal of Personality, 39:606-617, 1971.

12. Lepper, M.R.; Greene, D.; Nisbett, R.E., "Undermining Children's Intrinsic Interest with Extrinsic Reward: A Test of the 'Overjustification' Hypothesis," Journal of Personality and Social Psychology, 28: 129-137, 1973.

13. Lovitt, T. C.; Curtiss, K. A., "Academic Response Rate as a Function of Teacher- and Self-Imposed Contingencies," Journal of Applied Behavioral Science, Vol. 2, 1969.

14. O'Leary, K.D.; Drabman, R., "Token Reinforcement Programs in the Classroom: A Review." Psychological Bulletin, 75: 379-398, 1971.

15. O'Leary, K.D.; Poulos, R.W.; Devine, V.T.. "Token Reinforcers: Bonuses or Bribes?" Journal of Consulting and Clinical Psychology, 38: 1-8, 1972.

16. Schumpeter, J.A., The Theory of Economic Development, Harvard University Press, Cambridge, Mass., 1936.

17. Williams, V.L.; Fish, M., "Rehabilitation and Economic Self-Interest," Crime and Delinquency, Oct. 1971.