Synthesis of Research on Moral Development

Larry Nucci
Educational Leadership
February 1987 - Pages 86-92

Children learn easily to conform to the conventions of the classroom: raise your hand, use your inside voice.

Highlights of Research on Moral Development

Educators can help children differentiate between the norms and conventions of their culture and the universal moral concerns for justice (fairness) and human welfare. Five educational practices enable teachers to engage in moral education that is neither indoctrinative nor relativistic.

Moral education should focus on issues of justice, fairness and human welfare.
Effective moral education programs are integrated within the curriculum, rather than treated separately as a special program or unit.
Moral discussion promotes moral development when the students use "transactive" discussion patterns, are at somewhat different moral levels, and are free to disagree about the best solution to a moral dilemma.
Cooperative goal structures promote both moral and academic growth.
Firm, fair, and flexible classroom management practices and rules contribute to students' moral growth. Teachers should respond to the harmful or unjust consequences of moral transgressions, rather than to broken rules or unfulfilled social expectations.

We have known for some time that the overwhelming majority of parents expect teachers and other school authorities to contribute to children's moral development (Gallup 1976). There is, however, considerable confusion and discord among people about what it is they mean when they talk about morality. In such a context teachers who want to provide moral education have difficulty even deciding what parents want them to teach, let alone how best to teach it.

One aim of this review is to help clarify what constitutes the moral domain. Recent research suggests that the apparent public confusion is not about what is moral but about what is "proper." Both children and adults generally agree about what is moral; this overall agreement can be seen by differentiating the sphere of action governed by moral precepts from that governed by consensus or social convention. I have attempted to pull together research and theory on children's social development that helps clarify the distinction between the moral and conventional domains. In light of that distinction, this review presents a synthesis of research finding on how to foster moral development.

The Distinction Between Morality and Convention

Children in any society should learn to conform to a number of social rules and expectations if they are to become participants in the culture. In our society, children need to learn that certain classes of adults (such as teachers and doctors) are addressed by titles. They are also expected to learn that it is unacceptable to be naked in public even if it is 90 degrees and sunny outside, and so forth.
Actions of this sort are examples of social conventions. Conventions are shared, uniform behaviors determined by the social system in which they are formed (Turiel, 1983). Over time, through accepted usage, these standards serve to maintain social organization. While conventions are important, they are arbitrary. This is because there is nothing inherently right or wrong about the actions they define. For example, dresses are worn only by women in American society. The social norm governing this behavior is arbitrary in that another form of dress (pants for women and dresses for men) could be designated to differentiate between the sexes.

In contrast with convention, moral considerations stem from factors intrinsic to actions: consequences such as harm to others, violations of rights, effects on general welfare. Moral issues are, thus, neither arbitrary nor determined by cultural precepts or by consensus. The individual's moral prescriptions (i.e., "It is wrong to steal from others") are determined by factors inherent in social relationships, as opposed to a particular form of social, cultural, or religious structure (Turiel 1983).

The following excerpt illustrates the distinction children make between moral and convention issues. The excerpt is from an interview with a three-year-old girl regarding her perceptions of spontaneously occurring transgressions at her preschool (from Nucci et al. 1983).

Moral Issues: Did you see what just happened? Yes. They were playing and John hit him too hard. Is that something you are supposed to do or not supposed to do? Not so hard to hurt. Is there a rule about that? Yes. What is the rule? You're not to hit hard. What if there were no rule about hitting hard; would it be all right to do then? No. Why not? Because he could get hurt and start to cry.

Conventional Issue: Did you see what just happened? Yes. They were noisy. Is that something you are supposed to do or not supposed to do? Not do. Is there a rule about that? Yes. We have to be quiet. What if there were no rule; would it be all right to do then? Yes. Why? Because there is no rule.

Cutting, charing and consuming a piece of pie are fraught with moral issues: What is fair? Who decides?

This kind of distinction between morality and convention is at variance with the accounts of moral development that have had the greatest impact on moral education (Piaget 1932, Kohlberg 1984). Within those earlier views, it is only at the higher stages of moral development that morality (fairness) is differentiated from the displaces convention as the basis for moral judgments. Over the past decade, however, 27 published accounts have reported research demonstrating that morality and convention are differentiated at very early ages and constitute distinct conceptual and developmental systems. These studies are summarized in several recent reviews (Nucci 1982, Turiel 1983, Turiel et al. in press). In brief, these studies have found the following.

Moral transgressions are viewed as wrong, irrespective of the presence of governing rules, while conventional acts are viewed as wrong only if they violate an existing rule or standard.
Individuals view conventional standards as culturally relative and alterable, while moral prescriptions are viewed as universal and unchangeable.
The forms of social interaction is the context of moral events differ qualitatively from interactions in the context of conventions. Specifically, children's and adults' responses to events in the moral domain focus on features intrinsic to the acts (such as harm or justice), while responses in the context of conventions focus on aspects of the social order (rules, regulations, normative expectations).
Individuals tend to treat moral transgressions as more serious than violations of convention and tend to view prosocial moral acts as better and more positive than adherence to conventions.

While the majority of these studies were conducted in the United States, essentially the same results have been obtained in Nigeria (Hollos et al. in press), Taiwan (Song et al. 1985), and the Virgin Islands (Nucci et al. 1983).

Finally, a series of studies involving several hundred Catholic, fundamentalist Christian, and Jewish children showed that children make distinctions between matters of morality and religious doctrine similar to the distinctions secular children draw between morality and convention (Nucci 1985). Most agreed that moral transgressions such as stealing, hitting, or slander would still be wrong even if there were no religious rules against them, because they are harmful to others. However, work on the Sabbath, women preaching in church or synagogue, and the use of contraceptives, for instance, would be all right in the children's view if there were no religious rules or scriptural injunctions concerning them. This research indicates that conceptions of morality (justice and beneficence) are independent of religion.

Question: When does the school rule, "Take turns using library books," move from a conventional issue to a moral issue?

Development Within the Moral and Conventional Domains

While a young child has an intuitive grasp that actions such as hitting and stealing are prima facie wrong, the child's moral concepts do not reflect a fully developed moral system. For example, although young children view it as wrong to keep all of the classroom toys to oneself and not share any of them with the other children (Damon 1977, Nucci 1981, Smetana 1981), preschoolers think it is quite all right to keep all of the favored toys to oneself as long as one shares the remainder (Damon 1977, 1980). Thus, while the young child's morality is structured by concepts of justice, it reflects a rather egocentric moral perspective.

With development, the child's moral perspective gives way to progressively more objective and inclusive notions of equality and reciprocity. With respect to sharing, for example, the four-year-olds' premise - whoever wants the most should get it - is replaced by the idea that distributive decisions should be based on strict equality or reciprocity - everybody should get the same. This strict reciprocity is replaced in turn by a recognition that there can be multiple valid claims to justice by different individuals and that persons with special needs, the poor or the handicapped, deserved special consideration (Damon 1977, 1980; Enright et al. 1980).

The changes observed with respect to distributive justice reflect the more general structural changes in the child's moral understanding. Moral development entails the construction of progressively more adequate conceptualizations of justice and beneficence (Berkowitz and Nucci, 1986; Damon 1977, 1980; Turiel 1983).

Just as children's conceptions of morality undergo development, so also do their concepts of social convention. Through observation and communication with others, children learn their society's conventions. However, the societal functions of conventions are usually quite complex, and even when children have learned what is "expected," they do not fully understand the reasons why such behaviors are considered reasonable and right. Indeed, to understand the importance of social conventions, children need to understand interpersonal relationships, social systems, and the role of behavioral norms in maintaining both. Such complex constructs take time to develop.

Conceptions of social convention progress through seven development levels reflecting underlying concepts of social organization (Turiel 1983). Development follows an oscillating pattern between periods affirming the importance of convention and phases negating it. This oscillation indicates the difficulty children have in accounting for the function of arbitrary social norms and illustrates the slow process of reflection and construction that precedes the adolescent's view of convention as important to the maintenance of the social system.

Answer: When a student's refusal to share some books violates another student's sense of justice.

Fostering Moral Development

If even very young children differentiate between actions in the moral and conventional domains and reason differently about the two, then moral or values education should clearly reflect this distinction. Moral education should move students through progressively more adequate forms of resolving conflicting claims to justice or human rights. Teaching about convention should move students toward an understanding of the role conventions play in establishing social organization, and the importance of convention for organizing and coordinating interactions within social systems.

The first step toward such an approach entails the teacher's analysis and identification of the moral or conventional nature of social issues employed in values lessons. Such an analysis would be necessary to ensure that the issues discussed are concordant with the values domain they are intended to affect. A discussion of dress code, for example, would constitute a poor issue from which to generate moral discussion since mode of dress is primarily a matter of convention. Likewise, consideration of whether it is right to steal to help a person in need would be a poor issue with which to generate a lesson intended to foster students' understanding of the function of social conventions.

Of course, not all issues of social right and wrong fall neatly into one domain or the other. In many cases one can find issues that overlap the two domains. Such issues afford rich opportunities with which to involve students in reasoning, necessitating the coordination of knowledge from more than one social dimension (See Turiel et al. in press for a full discussion of domain overlap).

Moral Development and the General Curriculum

Moral education should be integrated within the curriculum and not take the form of a "special" program or unit. A program that is simply inserted into the curriculum carries with it an inherent artificiality and discontinuity that renders such interventions incompatible with the more general aims of teachers and students. The life of such programs is generally brief. Regarding one of his own early programs of this genre, Lawrence Kohlberg (1985) quipped, While the intervention operation was a success, the patient died. When we went back a year later, we found not a single teacher had continued to engage in moral discussion after the commitment to the research had end (p. 80).

Moral Discussion: The "Plus One" Myth

As the Kohlberg quote implies, the central method used to generate moral development has been moral discussion. The use of discussion acknowledges that social growth is not simply a process of learning society's rules and values, but a gradual process in which students actively transform their understanding of morality and social convention through reflection and construction. That is, students' growth is a function of meaning-making rather than mere compliance with externally imposed values.

Despite the widespread and long-standing use of discussion of moral dilemmas as an educational method, it is only in the past five years that careful research of the mechanism underlying the effectiveness of moral discussion has been conducted (Berkowitz 1982, Berkowitz and Gibbs 1983, Berkowitz et al. 1980, Gibbs et al. 1983). On the basis of that research we can discard some long-held notions about moral discussion and focus our efforts on more effective interactional patterns.

The central myth uncovered in the research is that advances in the moral judgment of children are aided by teacher statements one stage above the modal reasoning level of the children (Blatt and Kohlberg 1975). Not only are such statements difficult to generate and therefore rare in classroom discussions, including those conducted by trained experts, but they seem far less relevant to changes in moral reasoning than statements by peers. The research by Berkowitz and his colleagues suggests that teachers serve less as instruments for direct intervention that as agents for the facilitation of peer discussion.

From the research we can identify three characteristics of effective moral discussion.

Conflict. According to Berkowitz (1982), stage change occurred most readily in students who disagreed about the moral solution to a dilemma. Consensus on the outcome reduced the likelihood that students would challenge or otherwise respond to one another's reasoning and thus reduced the impact of the discussion on students' existing notions of morality. The educational implication of this finding is that the issues or problems teachers select as the basis for moral discussion should be ones likely to generate disagreement.

A note of caution regarding the use of conflict comes from research with young children. Damon and Killen (1982) found that social conflict tended to retard and not to promote stage change in children under eight years of age. Development took place in contexts where children could resolve problems through cooperation and conciliation. Their findings are concordant with other research suggesting that moral development in young children occurs through co-construction rather than argumentation or passive withdrawal (Younnis 1980, 81).
Stage disparity. The optimal distance in developmental level among students participating in moral discussion is on the order of one-half stage. This stage disparity is about what one finds among students in a typical classroom and implies that normal heterogeneity among students is sufficient for effective moral discussion.
Transactive discussion. In their analyses of student discourse, Berkowitz and his colleagues identified several forms of student statements that are related to moral development. They labeled such statements transacts. Transacts are characterized by listeners' efforts to integrate the speaker's statements into his own framework before generating a response. Transacts are responses that attempt to extend the logic of the speaker's argument, refute the assumptions of the speaker's argument, or provide a point of commonality or resolution between the two conflicting positions. Listener behavior that was not found to be associated with the moral development includes forms of discourse in which the listener restates the speaker's argument (in the style of Carl Rogers) or engages in collective monologue in which the listener's statements seem not to have reflected those of the speaker.

Cooperative Goal Structure

David Johnson (1981) has suggested that successful moral discussion is more like to take place in classrooms employing cooperative goal structures in a democratic atmosphere than in the traditional classroom environment. There is a considerable body of evidence to support Johnson's claim that cooperative goal structures contribute to moral development. In a cooperative goal structure, students perceive that they can obtain their goal (e.g., learn a given body of material, complete a project, obtain a course grade) if and only if the other students with whom they are cooperatively linked obtained theirs (Johnson 1981, p. 280).

In addition to being linked to positive social outcomes (such as increased perspective-taking and moral stage, decrease in racial and ethnic stereotyping), cooperative goal structures have been associated with increases in student motivation and academic achievement (Slavin 1980, Slavin et al. 1985). Thus, the use of cooperative education may serve the dual purpose of promoting moral development and linking moral education to the broader curriculum.

Classroom Management

Each aspect of moral education discussed thus far is embedded within the more general social climate of the classroom; the rules, structure, and sanctions that make up what Philip Jackson (1968) calls the "hidden curriculum." While specific classroom management practices may vary, the overall features of classrooms that contribute to moral development are as follows:

Firm. Classroom rules and expectations are known and upheld by school authorities.
Fair. Rules are limited to those necessary for learning and are evenly applied; consequences are moderate rather than severe.
Flexible. There is room for negotiation between students and teachers regarding the establishment, removal, and enforcement of school and classroom rules.

In addition to the above characteristics of classroom and school climate, practices associated with moral development include the use of reasoning to respond to transgressions (Lickona 1983, Rohrkemper 1984). Research indicates that students are sensitive to whether teacher responses are concordant with the domain (moral or conventional) of the breach. Students evaluate not only their judgments of teacher responses but also the teachers as respondents. Students rated highest those teachers who responded to moral transgressions with statements focusing on the effects of acts ("Joe, that really hurt Mike"). Rated lower were teachers who responded with statements of school rules or normative expectations ("That's not the way for a Hawthorne student to act"). Rated lowest were teachers who used simple commands ("Stop it!" or "Don't hit").

As one would expect, students rated highest those teachers who responded to breaches of convention with rule statements, or evaluations of acts as deviant, and rated lower those teachers who responded to such transgressions in terms of their effects on others ("When you sit like that, it really upsets people"). As with moral transgressions, the use of simple commands was rated the least adequate.

This research suggests that students attend to the informational content of teacher responses to transgressions. It also suggests that the domain of teacher responses to transgression may prove to be an important variable for future studies of the relations between classroom management techniques and social development in children.

The Universal Nature of Morality

The philosopher Alasdair MacIntyre (1982) has characterized the current historical period as one of moral dissention. Yet in the midst of this moral Babel, the majority of parents expect schools to contribute to moral development of children. The research indicates that morality is centered on a set of universal concerns for justice, fairness, and human welfare that are available even to young children. Those findings provide a basis for moral education that is both nonindoctrinative and nonrelativistic. The universal and prescriptive nature of morality means that educators can do more than merely clarify student values. At the same time, the developmental and constructivist basis of moral knowledge is commensurate with interactive rather than directive educational practices.

Berkowitz, M. "The Role of Discussion in Moral Education." Paper presented at the International Symposium on Moral Education, Fribourg, Switzerland, September 1982.

Berkowitz, M., and J. Gibbs. "Measuring the Developmental Features of Moral Discussion." Merrill-Palmer Quarterly 24 (1982): 399-410.

Berkowitz, M., J. Gibbs, and J. Broughton. "The Relation of Moral Judgment Stage Disparity to Developmental Effects of Peer Dialogues." Merrill-Palmer Quarterly 26 (1980): 341-357.

Berkowitz, M., and L. Nucci. "What's Moral About Conventional Moral Reasoning?" Paper presented at the biennial meeting of the Society for Research on Adolescence, Madison, Wisconsin, March 1986.

Blatt, M., and L. Kohlberg. "The Effects of Classroom Moral Discussion Upon Children's Level of Moral Judgment." Journal of Moral Education 4 (1975): 129-161.

Damon, W. The Social World of the Child. San Francisco: Jossey-Bass, 1977.

Damon, W. "Patterns of Change in Children's Social Reasoning: A Two-Year Longitudinal Study." Child Development 51 (1980): 1010-1017.

Damon, W., and M. Killen. "Peer Interaction and the Process of Change in Children's Moral Reasoning." Merrill-Palmer Quarterly 28 (1982): 347-367

Enright R., L. Franklin, and L. Manheim. "On Children's Distributive Justice Reasoning: A Standardized and Objective Scale." Development Psychology 16 (May 1980): 193-202.

Gallup, G. Eighth Annual Gallup Poll of Public Attitudes Toward the Public Schools. New York: George Gallup, 1976.

Gibbs, J., S. Schnell, M. Berkowitz, and D. Goldstein. "Relations Between Formal Operations and Logical Conflict Resolutions." Paper presented at the biennial meeting of the Society for Research in Child Development, Detroit, April 1983.

Hollos, M., P. E. Leis, and E. Turiel. "Social Reasoning in Ijo Children and Adolescents in Nigerian Communities." Journal of Cross-Cultural Psychology 17 (September 1986): 352-374.

Jackson, P. W. Life in the Classroom. New York: Holt, Rinehart and Winston, 1968.

Johnson, D. "Social Psychology." In Psychology and Education: The State of the Union, edited by F. Farley and N. Gordon. Berkeley: McCutchan, 1981.

Kohlberg, L. Essays on Moral Development, Vol. II: The Psychology of Moral Development. San Francisco: Harper and Row, 1984.

Kohlberg, L. "Resolving Moral Conflicts Within the Just Community." In Moral Dilemmas: Philosophical and Psychological Issues in the Development of Moral Reasoning, edited by C. Harding. Chicago: Precedent Press, 1985.

Lickona, T. Raising Good Children. New York: Bantam Books, 1983.

MacIntyre, A. After Virtue. Notre Dame, IN: University of Notre Dame Press, 1982.

Nucci, L. "Conceptions of Personal Issues: A Domain Distinct from Moral or Societal Concepts." Child Development 52 (1981): 114-121.

Nucci, L. "Conceptual Development in the Moral and Conventional Domains: Implications for Values Education." Review of Educational Research 49 (1982): 93-122.

Nucci, L. "Children's Conceptions of Morality, Societal Convention and Religious Prescription." In Moral Dilemmas: Philosophical and Psychological Issues in the development of Moral Reasoning, edited by C. Harding. Chicago: Precedent Press, 1985.

Nucci, L., E. Turiel, and G. Encarnacion-Gawrych. "Children's Social Interactions and Social Concepts: Analyses of Morality and Convention in the Virgin Islands." Journal of Cross-Cultural Psychology 14 (1983): 469-487.

Piaget, J. The Moral Judgment of the Child. Glencoe, IL: Free Press, 1932, 1948.
Rohrkemper, M. "The Influence of Teacher Socialization Style on Students' Social Cognitive and Reported Interpersonal Classroom Behavior." The Elementary School Journal 85 (1984): 245-275.

Slavin, R. "Cooperative Learning." Review of Educational Research 50, 2 (1980): 315-342.

Slavin, R. S. Sharan, S. Kagan, R. H. Larowitz, C. Webb, and R. Schmuck. Learning to Cooperate, Cooperating to Learn. New York: Plenum, 1985.

Smetana, J. "Preschool Children's Conceptions of Moral and Social Rules." Child Development 52 (1981): 1333 - 1336.

Song, M., J. G. Smetana, and S. Kim. "Korean Children's Conceptions of Moral and Conventional Transgressions." Paper presented at the biennial meeting of the Society for Research in Child Development, Toronto, 1985.

Turiel, E. The Development of Social Knowledge: Morality and Convention. Cambridge: Cambridge University Press, 1983.

Turiel, E., M. Killen, and C. Helwig. "Morality: Its Structure, Functions and Vagaries." In The Emergence of Moral Concepts in Young Children. Chicago: University of Chicago Press, in press.

Younnis, J. Parents and Peers in Social Development. Chicago: University of Chicago Press, 1980.

Younnis, J. "Moral Development Through a Theory of Social Construction: AN Analysis." Merrill-Palmer Quarterly 27 (1981): 385-403.

Larry Nucci is a Professor of Educational Psychology at the University of Illinois at Chicago. He can be reached at lnucci@uic.edu.

back to Featured Articles