Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Children’s ego functioning and their stage of moral reasoning Matsuba, Michio K. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_fall_matsuba_michio.pdf [ 4.5MB ]
JSON: 831-1.0086258.json
JSON-LD: 831-1.0086258-ld.json
RDF/XML (Pretty): 831-1.0086258-rdf.xml
RDF/JSON: 831-1.0086258-rdf.json
Turtle: 831-1.0086258-turtle.txt
N-Triples: 831-1.0086258-rdf-ntriples.txt
Original Record: 831-1.0086258-source.json
Full Text

Full Text

CHILDREN'S EGO FUNCTIONING AND THEIR STAGE OF MORAL REASONING by MICHIO KYLE MATSUBA B.Sc., The University of Toronto, 1989 B.A., The University of British Columbia, 1990 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1993 © Michio Kyle Matsuba, 1993  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature  Department of ^Pc+/CH01_0(./ The University of British Columbia Vancouver, Canada Date Cct^1q43  DE-6 (2/88)  ii  Abstract It has been claimed that both peer and family interactions are important in facilitating moral development. As well, there has been evidence suggesting that children's ego functioning is an important consideration in understanding moral functioning. This study investigated the relationship among these variables and how they relate to children's moral reasoning level. Forty target children (Grade 5 and 10) participated along with a same-sex friend and a parent (total N = 120). Each participant's stage of moral development was assessed in a moral judgment interview. Target children also participated in two discussion sessions (one with a friend and one with a parent). In each discussion, three moral conflicts were discussed (two real-life and one hypothetical). Target children's ego functioning in these discussion sessions was rated by observers using a Q-sort procedure. Results revealed that older children tended to use more complex ego processes than younger children. As well, children generally "coped" more in discussing hypothetical dilemmas and "defended" more with real-life dilemmas. The predicted differences in ego functioning when discussing dilemmas with a peer versus a parent were not evident. However, consistent with expectations, a strong relation was found between ego functioning and level of moral reasoning with moral development being predicted by both cognitive coping and attention-focusing coping ego functions. The results are discussed in terms of the factors that foster moral growth in children.  iii  Table of Contents Abstract ^  ii  Table of Contents ^  iii  List of Tables ^ Acknowledgements ^  vi  Introduction ^  1  Theoretical Issues in Moral Reasoning Development ^2 Kohlberg's Philosophical Roots^  2  Kohlberg's Psychological Roots^  4  Empirical Work Using Kohlberg's Theory ^  8  Intellectual, Social Perspective-taking, and Moral Reasoning Development ^ 8 Cognitive Conflict ^  10  Moral Reasoning in the Context of Peer Relations ^ 14 Moral Reasoning in the Context of Friendship ^ 16 Moral Reasoning in the Context of Parent-Child Relationship ^ 19 The Different Roles of Friends and Parents ^23 Ego Functioning ^  29  Defense Mechanisms and Cognitive Abilities ^ 31 Defense Mechanisms and Social Perspective-taking Abilities ^  33  Defense Mechanisms and Gender ^  34  Defense Mechanisms and Social Circumstances ^35 Ego Functioning and Moral Reasoning^  37  The Present Study ^  43  Method ^ Participants^  48 48  iv Procedure ^  48  Scoring ^  50  Moral Reasoning ^  50  Ego Functioning ^  51  Results ^ Ego Functioning ^  53 53  Post hoc Analyses ^  53  Planned Analyses ^  56  Moral Reasoning ^  62  Moral Reasoning and Ego Functioning ^  62  Discussion ^  67  References ^  83  Appendix A ^  89  Appendix B ^  92  Appendix C ^  96  List of Tables Table 1^Children's Ego Functioning across Grade ^57 Table 2^Children's Ego Functioning Across Dilemma Discussions ^ 60 Table 3^Multiple Regression Analysis of Moral Reasoning Scores ^64 Table 4^Ego Functioning Correlation Matrix ^  66  vi Acknowledgements This study would not be possible without the assistance of a number of people, each of whom have played a vital role in the undertaking of this project. Before acknowledging these people, I first wish to thank SSHRC for their grant to Dr. Walker which has helped to fund this study, and to the Vancouver School Board for its help in the recruitment of participants. I wish to begin by thanking my committee members: Dr. Charlotte Johnston for her cheerful willingness to be on my committee, her friendly assistance which has made the thesis experience more enjoyable, and her knowledge on methodological matters which has help to strengthen the quality of this study; and to Dr. Michael Chandler for his wisdom from which came challenging questions and insightful comments. As well, I would like to thank the members of my research team who have made this study possible: To Jody Peters and Esther Sweetman who have put in timeless hours transcribing audiotapes quickly and accurately; to Russ Pitts and Karl Hennig who have sacrificed their vision by watching and coding numerous video-taped discussions; and to many fellow colleagues who have provided helpful comments and criticisms along the way to completing this thesis. I also need and want to acknowledge my friend and supervisor Dr. Larry Walker. From his balanced dedication to his research, his students, his family, and God, I have learned a lot about what is important in life. He has taught me a great deal! I would also like to thank my parents for their belief in my abilities, their never ending support of my academic pursuits, and their patience with me during the stormy times. Finally, I give by humble thanks and praise to God for all the things that He has done, especially in providing me with supportive brothers and sisters in Christ. I pray that all my efforts in uncovering the mysteries of His people are acceptable to Him.  1 Introduction Over the past 20 years, many psychologists have invested a great deal of their resources in extending our understanding of the moral domain. This work has been important since it has led to the identification of numerous variables that are related to the development of children's moral reasoning. Yet despite the inroads made into the moral domain, there is much left unknown about children's moral reasoning development. The purpose of this study was to further explore those roads less travelled. More specifically, the challenge which lies ahead is to determine how two areas of human functioning, moral reasoning and ego functioning, relate to each other. Attempts to bridge these two domains is not entirely new. Haan (1977; Haan, Aerts, & Cooper, 1985) thought that individuals' ego functioning should relate to their level of moral reasoning. She reasoned that people who face moral dilemmas experience a certain amount of stress, and that the ego processes, or adaptational styles, people employ to deal with this stress may be related to how they reason about these moral issues, and ultimately to how they develop in regards to their moral reasoning. Yet (for reasons to be outlined later) the studies conducted by Haan have not been entirely successful in demonstrating a clear relationship between children's moral reasoning and ego functioning. Thus, the goal of this study was to reexamine this relationship in hopes of obtaining a better understanding of how these psychological domains may be linked, and how other variables, such as age, gender, and social context may influence this relationship. In order to understand the significance of this study, a discussion of the theoretical and empirical foundations upon which it is grounded will be presented. The areas to be reviewed include; (a) moral reasoning development, (b) ego functioning, (c) the importance of age, gender and social context to both these psychological domains, and (d) a list of the specific predictions of how moral reasoning and ego functioning may relate to one another as well as to the other variables under study.  2  Theoretical Issues in Moral Reasoning Development In recent times, exploration in the moral domain has been profoundly influenced by the work of Lawrence Kohlberg (1981, 1984). Kohlberg's theory of moral reasoning development is unique in its explicit attempt to construct a bridge between moral psychology and ethics. Kohlberg saw that such a bridge was possible because of the parallel features (isomorphism) common to both cognitive-developmental psychology and formalistic philosophy. These commonalties will become apparent in the ensuing discussions.  Kohlberg's Philosophical Roots In the construction of any psychological model, there is no atheoretical starting point, and Kohlberg's model is no exception. Often researchers are oblivious to the theoretical underpinnings of their work. Not so with Kohlberg. Kohlberg (1981) made explicit his efforts of grounding his work in a philosophic tradition. He believed that a psychological theory of ethics is incomplete if its philosophic implications are not spelled out. As a result, Kohlberg was clear to state the meta-ethical assumptions upon which he built his theory (see Kohlberg, Levine, & Hewer, 1983). One of the key assumptions in Kohlberg's theory is that of cognitivism. This assumption claims that the rightness or wrongness of a judgment depends solely on the reasons offered and not on any affective component. By emphasizing this assumption of cognitivism Kohlberg was placing primary importance on people's rational nature in making moral judgments. A second important assumption in Kohlberg's theory is that of universalism. Kohlberg believed "that there is a universalistically valid form of rational thought process which all persons could articulate, assuming social and cultural conditions suitable to cognitive-moral stage of development" (Kohlberg et al., 1983, p. 75). Here Kohlberg is rejecting cultural relativism and arguing for moral judgments to be viewed from a perspective of methodological non-relativism. He was claiming that the ontogenesis of rational moral thinking occurs in all cultures in the same stepwise,  3 invariant stage sequence. A third assumption in Kohlberg's theory is that of prescriptivism. Kohlberg claimed that "in moral judgment, there is an implicit commitment to action by the speaker and by others who share his or her principle, a commitment specifiable as a rule or principle" (Kohlberg et al., 1983, p. 77). These last two assumptions have often been stressed by formalist philosophers as what is necessary in forming an adequate moral judgment. Thus, these last two assumptions lead to the fourth assumption of formalism. Formalism refers to a discipline within meta-ethics that defines moral judgments, methods, or points of view according to their formal character as opposed to their content. By formal character, one is referring to the moral judgments that are impersonal, ideal, universalizable, preemptive, etc. For instance, take Baier's (1965) concept of the moral point of view. He posits that the moral point of view is an abstracted position where one can make moral decisions as an independent, unbiased, impartial, objective, dispassionate, disinterested observer. By taking such a perspective, the decisions reached are expected to be ones that everyone should abide by since it is in the interest of everyone. This allows the procedural form to be separated from any content which may bias the outcome. In addition, people can rationally agree upon the process of taking the moral point of view without necessarily reaching an agreement upon the content or substantive principles of morality. For Kohlberg, this formalist position seems to parallel the distinction between form and content as it relates to his theory of moral development: A given stage has certain formal characteristics which may generate various pro or con moral contents, all of which can be consistent with its form. Thus, Kohlberg envisioned the development of moral reasoning to be the movement towards constructing a moral point of view and using moral judgments which meet formalistic criteria like universality and prescriptiveness. Finally, the fifth assumption deals with the primacy of justice in defining the moral domain. Kohlberg seemed to define the concept of justice in terms of fairness.  4 Kohlberg believed that the resolution of moral disputes in a fair manner involve the operations of equality and reciprocity. The reason for defining the moral domain in this way is because justice reasoning seems to conform with the Piagetian structure system (Piaget, 1970) while maintaining the integrity of the meta-ethical assumptions already made. Therefore, framing the moral domain in terms of justice best represents Kohlberg's conception of morality as being universal, rational, and structural. This will become more apparent later.  Kohlberg's Psychological Roots From a psychological perspective, Kohlberg's theory of moral reasoning development stems from the "cognitive-developmental" tradition (Kohlberg, 1981). One of the key figures within this tradition was Jean Piaget, on whose work Kohlberg relied in the development of his own theory of moral reasoning. Piaget (1975/1985) argued that people construct mental structures in order to make sense of information they receive from their environment. According to Piaget, a mental structure is a system of transformational laws that organize and govern cognitive operations, and are reflected in individuals' actual responses to conflicts. These operations enable the subject to organize external information in some coherent order. Moreover, cognitive structures are continually being altered in order to achieve a better fit with experience. On the occasions where structures have changed their formation, children are described as having moved from a lower stage to a higher stage. The process which explains such development is called equilibration. According to Piaget (1975/1985), the process of equilibration is triggered when a child's cognitive structure is no longer able to make sense of the external reality. This state is referred to as intrapsychic conflict, and leads to the deformation of the individual's existing cognitive structure. This, in turn, is followed by the construction of a new cognitive structure which operates to alleviate the intrapsychic conflict by allowing the person to make sense of external reality (Chapman & McBride, 1992).  5 In many ways Kohlberg's research can be considered an extension of Piaget's work in the moral domain. Kohlberg (1981) writes: A cognitive-developmental theory of moralization holds that there is a sequence of moral stages for the same basic reasons that there are cognitive or logico-mathematical stages; that is, because cognitive-structural reorganizations toward the more equilibrated occur in the course of interaction between the organism and the environment. In the area of 14c, Piaget holds that a psychological theory of development is closely linked to a theory of normative logic. Following Piaget, I claim the same is true in the area of moral judgment. (p. 133) Moreover, as mentioned previously, Kohlberg thought that justice reasoning, in particular, involved the cognitive operations most amenable to structuraldevelopmental stage analysis. Kohlberg describes, rather ambiguously, two general operations that are crucial to moral development. The first operation is social perspective-taking, or role-taking. Kohlberg hypothesized that as individuals progress through the stages they are able to broaden their social perspective, and coordinate the increasingly divergent viewpoints when reasoning about moral dilemmas. The second general operation involves two justice operations: equality and reciprocity. Each stage organizes, or structures, the duty and rights reasoning through the use of these cognitive, justice operations. One can think of these operations as "interiorized actions" involved in forms of distribution and exchange. According to Kohlberg, in the distribution of goods one can either carry this out by the operation of equality (i.e., equity, distributive equality proportionate to circumstance and need), or reciprocity (i.e., merit or desert, reward in return for effort, virtue, or talent). Thus, each stage in Kohlberg's model reflects a justice structure which is able to organize both the patterns of role-taking and the use of the central operations of reciprocity and equality in moral conflict situations. Cognitive operations are important concepts in Kohlberg's theory. The reason why they are important is due to the properties they possess. One important property  6 of cognitive operations is that of reversibility. Reversibility allows for equilibrium to take place; that is, the balancing of conflicting value claims. Equilibrium and disequilibrium are two states that occur during the process of equilibration. A person is in a state of equilibrium when his or her present stage structure is able to resolve moral conflicts. However, there comes a time when the stage structure is no longer able to resolve new moral problems which the person faces. When this occurs, the person undergoes a state of disequilibrium, and remains in this state until the person's justice operations undergoes reorganization. This reorganization enables the person to make sense of concrete situations he or she experiences, and to resolve their moral conflicts. Once this reorganization is established, the person returns to a state of equilibrium. As a person reaches higher stages, the cognitive operations become more reversible in the sense that they are able to balance a broader array of conflicting value claims. In addition to describing the underlying processes involved in moral development, Kohlberg (1981) sketched out a six stage model of moral reasoning development (see Appendix A). This stage model is hierarchic in that each advancing stage represents an increasingly complex, abstract, general, and reversible structure over the preceding stage. Moreover, Kohlberg labelled his stages as "hard stages" since they meet the following Piagetian stage criteria: (a) stages follow an invariant sequence in individual development; (b) there is a "structural wholeness" to each stage, that is, underlying the responses to different tasks at a given stage is an underlying thoughtorganization; and (c) there is a hierarchical integration of stages which means that higher stages reintegrate the structures at lower stages (see Walker, 1988). At the pinnacle of his model, Kohlberg proposed a developmental end-point labelled Stage 6, and that each of the stages in the hierarchy represents an increase in correspondence with this end-point standard. Having an end-point is believed to be important because "the construction of any stage hierarchy requires a normatively designated point of reference from which the developmental process in question can be described  retrospective as learning" (Habermas, 1990, p. 224). Thus, for Kohlberg, Stage 6 represents the end-point in his rational reconstruction of the ontogenesis of justice reasoning (Habermas, 1983). This final stage is important because it helps to define the area of human activity under study (Kohlberg et al., 1983). At Stage 6, the structure is able to organize patterns of role taking and cognitive operations such that people are able to formulate moral principles that provide some consistency to the different moral decisions they make. Because role taking at this stage involves grasping the perspective of all parties involved (what Kohlberg called "moral musical chairs"), the solutions derived from this process should be acceptable to all (Kohlberg et al., 1983). Further, the equilibrium of operations occurring at this structural stage has reached the ideal and there is complete reversibility in perspective-taking abilities. As a result, there should be universal agreement among people who have reached Stage 6 in moral development assuming agreement on the "facts" of the conflict. From a philosophical perspective, Stage 6 has many features common to some of the normative ethical theories. Stage 6, as Kohlberg admitted, is distinctly Kantian in flavour with the emphasis on the principles of respect for persons and of justice. Furthermore, the Piagetian process of equilibrium seems naturally allied with the formalistic traditions of Kant (1785/1964) and Rawls (1971). Take, for instance, Rawls' reflective equilibrium: This involves the process of "tossing" "unpruned" justice principles between intuition and imagined situations. That is, justice principles based on intuition are applied to concrete situations for validation and then revision (Rawls, 1971). The revised principles are then checked with intuition and revised accordingly. This procedure continues until the principles derived "fit" both one's intuition and imagined concrete situations. Once this state is reached, equilibrium is said to be established. As well, Rawls' "veil of ignorance" in the "original position" is a manifestation of the formalistic criteria of universalizability and impartiality. The veil of ignorance is suppose to exemplify the formalist idea of reversibility, which is a  8 property of the equilibrium state. To summarize the connection between psychology and philosophy, Kohlberg (1981) wrote: Piaget's theory is explanatory of psychological; it explains (1) why justice is a compelling, obligatory, "natural" norm and (2) why concepts of justice change, moving to greater equilibrium. Rawls's theory is justificatory; it undertakes to prove that certain principles of justice held at Stage 6 (and important at Stage 5) are the ones that would be chosen in a condition of complete reflective equilibrium; that is, the ones that would be chosen in the original position. (p. 201) By drawing parallels between Piaget's and Rawls' theory, Kohlberg does not mean to imply that Rawls' theory of justice should or will be accepted by philosophers as the most morally adequate. Instead, Kohlberg's psychological claim is that something like Rawls' principles of justice should be chosen at Stage 6 since they are "more reversible," or in better equilibrium than justice principles used at previous stages. This summary of the psychological and philosophic traditions from which Kohlberg's theory has emerged is important for the reader to keep in mind, for this historical information acts as a marker indicating the road which Kohlberg has cleared into the moral domain. It is this road which I have entered. However, while Kohlberg's research has taken me a great distance, I believe it falls short of where research in the moral domain is currently at. Thus, I also have relied on the research of others working in the Kohlbergian tradition to help me in the formulation of this study. The presentation of this research work is of considerable importance in understanding the significance of the present study.  Empirical Work Using Kohlberg's Theory Intellectual. Social Perspective-taking, and Moral Reasoning Development In the preceding section, a discussion on the philosophical and psychological traditions from whence Kohlberg's theory arose was presented. Of particular interest to Kohlberg were the parallel features between cognitive-developmental psychology and formalistic philosophies. In this section, we shift our focus and examine the parallel  9 features across three content domains within the cognitive-developmental paradigm; (a) intellectual development (Piaget, 1956), (b) social perspective-taking development (Selman, 1980), and (c) moral reasoning development (Kohlberg, 1984). According to Kohlberg (1981), there is a parallelism, or isomorphism, between the development of the forms of logical and moral judgments. That is, in order to attain the next stage of moral judgment, one requires a new set of logical operations not present at the previous moral judgment stage. Unless the person possesses these new logical operations, he or she cannot develop beyond his or her current moral judgment stage. Thus, while it is possible that a person can be at a given logical stage and yet not have reached the parallel moral stage, the reverse is not possible; a person cannot have reached the moral stage before the having attained parallel logical stage. One implication of this is that "moral development is its own sequential process rather than the reflection of intellectual development in a slightly different content area" (pp. 137138). In his attempt to explain the asymmetry between the level of logicomathematical judgments and the level of moral judgments, Kohlberg (1981) raised the issue of social role-taking. According to Kohlberg, a great deal of variance in level of moral judgment after the intellectual variance is taken into account can be explained by social environmental factors. More specifically, Kohlberg interpreted these factors in terms of the amount of opportunities the social environment provided for role-taking. Kohlberg saw these opportunities as important because he believed they stimulated moral development. That is, through role-taking people adopt views which may differ from their own in dealing with moral situations. Therefore, moral development can be considered as "fundamentally a process of the restructuring of modes of role-taking" (Kohlberg, 1984, p. 74). In terms of role-taking opportunities, Kohlberg (1984) saw a sequence of groups in which the developing child participates beginning with the family into which he or  10 she is born. However, Kohlberg did not consider this to be an extraordinary group in terms of role-taking opportunities. On the other hand, as the child matures, he or she becomes incorporated into a peer group which Piaget (1965) considered to be a unique source of role-taking opportunities. Finally, the third type of group refers to secondary institutions such as law, government, school and work. Hence, Kohlberg argued that generally it is the participation in all these various groups which help to stimulate the development of basic moral values. These claims of Kohlberg, that both intellectual and social perspective-taking stages precede the parallel moral stage, have been demonstrated (e.g., Walker, 1980). Walker showed that only those participants who had reached beginning formal substage of cognitive development, as measured by a set of Piagetian tasks (Piaget, 1924/1928, 1941/1952; Inhelder & Piaget, 1955/1958; Piaget & Inhelder, 1941/1974), and perspective-taking Stage 3, as measured by Selman's perspective-taking interview (Selman & Byrne, 1973), were able to benefit from a program designed to stimulate moral reasoning to Stage 3. However, participants who had not reached beginning formal operations and perspective-taking Stage 3 did not evidence moral reasoning at Stage 3 even after undergoing the intervention condition. Walker concluded that his findings support Kohlberg's claim that both intellectual and perspective-taking development are necessary, though not sufficient, conditions for moral development. Cognitive Conflict  In addition to role-taking opportunities, Kohlberg (1984) considered another factor important to the stimulation of moral development. This factor he labelled cognitive-moral conflict. Cognitive-moral conflict occurs when there becomes a sense of contradiction arising in the use of one's current moral stage structure. Ideally, this sense of contradiction triggers the movement of the person's moral structure to the next higher stage, provided the individual has attained the intellectual and social perspective-taking prerequisites. Moreover, experiences of cognitive-moral conflict can  11  happen "either through exposure to decision situations that arouse internal contradictions in one's moral reasoning structure or through exposure to the moral reasoning of significant others which is discrepant in content or structure from one's own reasoning" (Kohlberg, 1984, pp. 202-203). The common technique in stimulating moral development as noted in the literature is through exposing people to the moral reasoning of significant others which somehow differs from the participant's own reasoning. In Walker's (1980) study, participants in the experimental condition took part in an intervention program This program involved exposing children to Stage 3 moral reasoning, which was one-stage above their current level. This exposure came in the form of role-playing moral dilemmas with two adults. "Each child was asked to imagine that he or she was the central character in each of six dilemma situations and then the adults provided advice suggesting how to 'resolve' the dilemma and why" (Walker, 1980, p. 135). As has been stated, only those children with the intellectual and perspective-taking prerequisites benefitted from this intervention program. Another study by Arbuthnot (1975) also incorporated a role-playing experimental condition as a means to stimulate moral reasoning development. In this study, college students role-played either a protagonist or an antagonist in a moral dilemma, and then tried to convince the other person that his or her own position was more justifiable. Arbuthnot found that the role-playing proved to be an effective means of producing immediate and longer-term (1 week) increases in level of moral reasoning for those participants who role-played opposite an opponent who employed higherstage reasoning (i.e., higher than subjects' own stage). This change was not found in the group where both members were at the same moral reasoning level or in the control group where members did not engage in role-playing, but rather engaged in a superfluous task believed to have no influence on development. Based on this finding it seems that role-playing in these conflictual situations appears to be an effective method  12 of promoting moral development (presumably because of the positive influence roleplaying has on one's perspective-taking abilities in morally conflictual situations). However, role-playing seemed to benefit only those who were at a lower stage of moral reasoning compared to their partners. This may be due to peers at higher stages of moral reasoning having already formed cognitive structures which are able to accommodate both their own and their partner's perspectives which the mental structures of partners at lower stages are not yet able to do. A common feature to Arbuthnot's (1975) and Walker's (1980) studies is the use of moral discussions as an experimental context in which to stimulate moral reasoning. The rationale for the use of moral discussion in studies originated from Piaget (1932/1965). Piaget viewed discussions as a means to create cognitive conflict by having two or more discussants who hold incompatible opinions confront each other. In order to resolve their difference of opinions, each person is required to consider the other's perspective. Further, the more willing the person is to consider the other's opinion, the more likely he or she will experience cognitive conflict (Berkowitz, Oser, & Althof, 1987). Therefore, moral discussion seems to be a reasonable method in which to foster moral reasoning development. While it seems to be true that moral discourse is an effective context in which to create cognitive conflict, it is yet uncertain why exactly this is the case. Berkowitz and Gibbs (1985) proposed numerous variables that may be operating through discussions to promote moral reasoning development. Through reviewing a series of studies, Berkowitz and Gibbs identified many possible variables and classified them into two general sets. In one set they placed the variables which focus on the processes involved in moral discussions and their developmental effectiveness. This area of research has received much attention. For instance, Berkowitz and Gibbs (1983) identified interaction styles through analyzing the moral discussion process. In their study, Berkowitz and Gibbs instructed pairs of college students to discuss moral issues and  13 then try to reach agreement on these issues. These researchers then analyzed the discussions by highlighting the occasions where "transactive discussions" took place. A transactive discussion was broadly defined to occur when one of the individuals' reasoning interacts in any way with the reasoning of his or her partner during their discussion. These transactive discussions were divided into two categories: (a)  representational -- those discussion segments, or transacts, which simply elicit or paraphrase another's reasoning instead of transforming it, and (b) operational - those statements which transform or operate on the reasoning of the other person. To analyze their data, Berkowitz and Gibbs (1983) created a regression equation entering the following variables: "type of dyadic stage disparity on pretest scores (no disparity, minor disparity, or major disparity), percentage of total statements that were operational transacts, percentage of total statements that were representational transacts, and pretest score of the lower stage subjects in the dyad" (p. 407). By doing this, Berkowitz and Gibbs (1983) were able to show that operational transacts were the best predictor of moral stage increases, even better than match/mismatch stage disparity between peer discussants (which was the second best predictor). Furthermore, they reported that there was very little shared variance between these two predictor variables. As a result, the findings from this study seem to show that how people discuss moral conflicts is an important predictor of moral development regardless of whether there is some stage disparity between the interacting partners. The other category identified by Berkowitz and Gibbs (1985) is labelled  person/context variables, and includes "the stage(s) of moral reasoning, experimenter/teacher instructions, stage match/mismatch, discussant relationship, degree of disagreement on problem solutions, and so on" (p. 75). According to Berkowitz and Gibbs, all of these variables have been associated with the developmental effectiveness of moral discussions. For example, Berkowitz, Gibbs, and Broughton (1980) reported the dyadic stage mismatch of people to be an effective  14 technique in moral discussions to facilitate lower-stage subjects' moral growth. That is, in discussion sessions, higher stage moral "reasoners" seem to facilitate the moral development of their lower-stage partners. Further, Berkowitz and Gibbs included discussant relationships as another subset of variables. Of particular interest to the present study are parent-child relationships as well as those between friends. It is to these two types of relationships I now turn.  Moral Reasoning in the Context of Peer Relations Kohlberg (1984) claimed that relationships with peers are important in a child's moral reasoning development. In these relationships, children are provided with many opportunities to take the perspective of their peers in order to resolve moral conflicts which may arise. Ideally, these role-taking opportunities create cognitive-moral conflict between a child's view of the situation and the peer's view. Numerous studies have investigated the importance of role-taking opportunities in the context of peer relations. In addition to Arbuthnot (1975), Berkowitz, Gibbs and Broughton (1980), and Berkowitz and Gibbs (1983), there have been studies by Maitland and Goldman (1974), and Damon and Killen (1982) which have focused on moral reasoning development in a peer relation context. Damon and Killen, for instance, had children discuss moral-type problems under three different conditions. They discovered that children who actively participated in resolving the actual problems with their peers scored higher in their moral reasoning level at post-test compared to the two control group conditions: (a) children who discussed a hypothetical problem with the experimenter, and (b) children who received only the pre- and post-test. This finding supports Kohlberg's claim that peer relations are an important contributor to children's moral reasoning development. In addition, upon further investigation of the discussions between peers, Damon and Killen (1982) found that children who were initially at lower stages of moral reasoning and who later reached higher stages at post-test were involved in discussions where there was a "reciprocal quality of acceptance of transformation of one another's  15 ideas" (p. 365). These findings suggest that those children who had morally developed were participating in two-way, collaborative interactions where an understanding of their partner's perspective was necessary. In contrast, no change in moral stage was found in children whose interactions were asymmetrical in nature (i.e., where one child contributed more ideas while the other child accepted more ideas in order to resolve the conflict). These interactions were described as not being reciprocal in nature because of the lack of co-construction toward a resolution of the conflict. Unfortunately, Damon and Killen were unable to specify an interaction style characteristic of morally advanced children. Rather, these children were more varied in their interaction styles than lower stage children. Further investigation of the interaction between these "morally mature" children is warranted. Nevertheless, for "morally immature" children, Damon and Killen's results are consistent with those of Berkowitz and Gibbs (1983) in demonstrating that interaction styles involving the exchange of ideas in order to resolve problems among peers are predictive of higher stages of moral reasoning than other styles. Thus, what seems to be important to moral development are both the opportunities and the desire to understand another's perspective. Whether this occurs through procedures of confrontation (Arbuthnot, 1975) or collaboration (Damon & Killen, 1982) may be dependent on the social context, the experimental task, and the age of the members within the sample. The general results of the studies mentioned here strongly support the claim that peer relations provide children with perspective-taking opportunities important to moral reasoning development. Moreover, the studies by Berkowitz and Gibbs (1983) and Damon and Killen (1982) suggest that how people interact with their peers when discussing moral issues is an important predictor of moral reasoning development. Hence, Kohlberg seems to have been right in emphasizing the importance of peer relations. However, the focus of this study is on a subset of the peer group known as close friends.  16 Moral Reasoning in the Context of Friendship  Given that researchers have shown the significant impact peer interaction has on children's moral reasoning growth, it seems somewhat intuitive that interaction with friends in particular may have a greater impact, and maybe even a different one, on children's moral reasoning development compared to interaction with strangers or acquaintances. In fact, there is evidence which suggests that there may be some truth to this notion. For instance, Newcomb and Brady (1982) studied Grade 2 and Grade 6 boys as they worked on a task together. There were two types of pair-groupings: One group was comprised of mutual friends while the second group consisted of acquaintances. Each pair was assigned to one of three task conditions: (a) cooperative (equally shared rewards), (b) competitive (proportional-to-work-accomplished rewards), and (c) no rewards. These task sessions were videotaped to capture the communicative exchanges, affective expressions, task performance, and synchronism of task-oriented behavior. In general, Newcomb and Brady reported that friendship interactions for both the second and sixth graders were characterized by greater mutuality than interaction between acquaintances. That is, compared to acquaintances, friends more frequently exchanged information through discussions, and were more likely to offer mutually beneficial suggestions and to comply with each other's directives. Furthermore, friends worked in conjunction with each other more extensively than acquaintances. Thus, mutual exchanges were characteristic of boys' friendship in task-oriented interactions. As well, in a study by Berndt and Perry (1986), the researchers identified certain characteristics which were unique to friendships. They reported that children in Grades 2, 4, 6, and 8 perceived their friends as more supportive than acquaintances. Included in this label of support were such features as play or association, prosocial behavior, intimacy, loyalty, and attachment or self-esteem enhancement. Based on the results from this study, Newcomb and Brady's (1982) study, as well as other studies, Berndt  17 (1987) was able to draw a number of conclusions about the differences in the nature of relations between friends as compared to peers. First, the conversations between friends are marked by a greater sense of mutuality than conversations between nonfriend peers. That is, as Berndt reports, friends talk more, show more intimate selfdisclosure, and explain their own view in a debate compared to non-friend peers. Second, in conversations, there is greater evidence of affect, both positive and negative, between friends as opposed to non-friend peers. In conversations, friends seem to smile more, express more agreement and disagreement, and be more critical compared to non-friend peers. Berndt interpreted the disagreements and criticisms between friends as an indication of the honesty and openness in the friendship. Third, friends show evidence of greater concern with their equality in contrast to non-friend peers, particularly in situations were they are being compared in their performance. That is, Berndt reported that friends may try to achieve equality through sharing, helping, or competing with each other, depending on the situation. Therefore, Berndt's (1987) review has been important in highlighting some of the important differences between friendships and peer relations. Moreover, Berndt suggests that since friends have more "connected conversations" where friends are more responsive to each other's request and more often give an explanation of their own views than non-friends, discussions with friends "may facilitate greater advances in moral reasoning than do non-friends' conversations" (p. 297). Interestingly, Kohlberg never distinguished between close friend and peer relationships in terms of their influence on children's moral growth. Unfortunately, there has been a paucity of research directly investigating the importance of friendships to moral reasoning development. One study linking friendship to moral reasoning was performed by Thoma and Ladewig (1991). Thoma and Ladewig assessed 156 college students' stage of moral reasoning. In addition, by giving these students a questionnaire whereby they listed those people who provide them with social and emotional support, the researchers were able to divide the  18 students into two general groups: (a) students who had formed close friends in college, and (b) students who had not. Thoma and Ladewig found that moral reasoning scores were higher for those students in the close friendship group than students who were not. This was true independently of the students' college level (i.e., freshman, sophomore, junior, and senior). Furthermore, in the friendship group, there appeared to be a positive correlation between the stage of moral reasoning and the year the student was enrolled: Upper-year students scored higher in moral reasoning compared to the lower-year students. Such an effect was not seen in the non-friendship group. As well, other variables such as gender, age, grade-point average, place of residence, gender of close friends, and number of friends listed were uncorrelated with moral reasoning scores. Hence, friendship through the college years seems to be related to moral reasoning development. However, the reason for the relationship between friendship and stage of moral reasoning in this study is uncertain. One possible explanation is that higher stage moral "reasoners" are attracted to each other which enables them to establish close friendships. An alternative explanation is that friendships provide the forum where the individual is exposed to new ideas and interests which may challenge his or her own ideas, thereby producing cognitive conflict. The claim that friendships are important to moral reasoning growth has garnered further support from a second study by Thoma and Ladewig (1993). In this study, Thoma and Ladewig isolated three friendship variables: "a) friendship density (i.e., the degree to which one's friends are also friends with one another); b) friendship boundary density (i.e., the degree to which college and non-college friends represent distinct friendship networks), and c) perceived close college friend support" (Thoma & Ladewig, 1993, pp. 2-3). The results of the study show that moral reasoning scores were positively correlated with social support which is consistent with Thoma and Ladewig's (1991) earlier findings. However, moral reasoning scores were negatively correlated with both friendship density and boundary density. That is, the less dense or complex  19 the network of friends the higher the moral reasoning score. Therefore, these students with less dense networks (where they have a greater proportion of friends who do not know each other) may be exposed to a greater variety of perspectives on issues. This may be compared to those students who have friends who all know each other. In this situation, there may be less of a variety of perspectives since the denser network of friends may share many perspectives in common. However, this is only one way to interpret the results. Further studies are needed to determine the causal direction in which this relationship operates. As well, additional studies investigating age-related differences in the importance of friendships to moral reasoning would also be of interest. Thoma and Ladewig studied college students only. Yet conceptions of friendship in late adolescence differ markedly from those in childhood (Berndt, 1981; Youniss, 1980). Nevertheless, Thoma and Ladewig's studies have been important in bringing to light the unique contribution close friends may provide to one's moral reasoning development. Moral Reasoning in the Context of the Parent-Child Relationship As mentioned, Kohlberg (1984) argued that peer interaction is important to moral development because it requires that children understand other perspectives when deciding what is best for the group. In contrast to peer interaction, Kohlberg claimed that children's interaction within the family "is not unique or critically necessary for moral development.... [T]here is no evidence that the family is a uniquely necessary setting for normal moral development" (p. 73). The reasoning behind Kohlberg's bold statement is that in a family, most of the decisions are made unilaterally by parents with little input contributed by children. This is due to the asymmetry in the parent-child relationship where parents hold most of the decision-making power. Since children do not have much say in the family decision-making process, it is not necessary for them to understand how other family members think about issues.  20 However, many researchers (Buck, Walsh, & Rothman, 1981; Hart, 1988; Holstein, 1972; Jurkovic & Prentice, 1974; Parikh, 1980) have challenged this belief. These and other studies have attempted to demonstrate the existence of a relationship between parents' and children's stage of moral reasoning. Yet when Powers (1988) and Walker and Taylor (1991) considered all these studies, they reported that, at best, there only were modest and inconsistent correlations between children's stage of moral reasoning and that of their parents. Powers explained that these weak results may be due to inadequacies in Kohlberg's moral judgment scoring system. Powers (1988) wrote: "One important problem in comparing results from these studies is that the majority were completed before the manual for the Standard Issue Scoring method (Colby & Kohlberg, 1987) was available" (p. 215). Alternatively, Walker and Taylor explain these inconsistencies by suggesting that what may be important to a child's moral development is not the parents' moral competence but their moral performance. That is, the manner in which parents interact with their children in discussing moral dilemmas may be crucial. So far, research seems to support this hypothesis. For example, Edwards (1980), in her review, reported studies which found positive correlations between moral judgment levels in children and parental affection and encouragement of discussion about morality. Edwards (1980) stated: Reading these studies, one gets a definite impression of the kind of parents believed most likely to have children at high moral judgment stages. The most successful parents are expected to be these verbal and overtly rational people who encourage warm and close relations with children, and who promote a "democratic" style of family life. That is, they foster discussions oriented towards a fair consideration of everyone's viewpoint. (p. 517) Further support for the importance of parental moral performance on children's moral development comes from Powers (1982; in Powers, 1988). Using a transact classification system similar to Berkowitz and Gibbs' (1983), Powers was able to identify interaction styles used in discussions of moral issues. Powers reported a positive  21 correlation between maternal affective support and adolescent level of moral judgment, and a negative correlation between a family's affective conflict and level of moral judgment. Conflict in Powers' study refers to speeches which, for example, attack another's reasoning or personality in a hostile or sarcastic fashion. Powers believed that these types of behavior may not be conducive to moral development since they may lead to an unwillingness to consider another point of view. In addition, none of Powers' parental behavior (performance) categories believed to be cognitively stimulating (focusing, challenging and sharing perspectives) were related to moral development as Powers hypothesized. Thus, Powers' results suggest that in a parent-child interaction, affectively stimulating behavior from parents is more important to adolescent moral development than is cognitively stimulating behavior. In an extension to Powers' study, Walker and Taylor (1991) investigated the importance of parent-child interactions to the child's moral development. In their study, Walker and Taylor video-taped and transcribed moral dilemma discussions involving the mother, father, and child. These discussions were then coded using a modified version of Powers' Developmental Environments Coding System (DECS). The modified DECS included the following categories: (1) operational (speeches that operate on the reasoning of another): critique, competitive request, counterconsideration, concession, clarification; (2) representational (speeches that elicit or re-present the reasoning of another): paraphrase, request, comprehension check; (3) informative (speeches that entail; sharing of opinions): opinion, competitive opinion, agreement, disagreement, request for change, intent for closure; (4) supportive (speeches that indicate positive affect and encouragement (including listening responses), humor; (5) cognitively interfering (speeches that interfere with sustained and coherent discussion): distracting, refusal, devalue task, distortion; (6) conflictual (speeches that indicate negative affect): resist/threaten, hostility; and (7) miscellaneous: unclear, incomplete statements whose meaning could not be discerned. (Walker & Taylor, 1991, p. 269)  22 The rationale behind revising the DECS was to separate Powers' transactive statements (speech which illustrates a family member coordinating another's moral perspective with his or her own) into operational and representational categories. The reason for this distinction was because Walker and Taylor believed that one of these transactive styles may be more important to children's moral reasoning development than the other, and Powers' system did not allow them to investigate this. By modifying Powers' coding system, Walker and Taylor found that families characterized as having a high level of moral reasoning disparity between the parents and the child, and high levels of representational and supportive interaction by parents predicted the greatest moral advancement by their children over a 2-year period. Thus, parents who listen and encourage their child, who make the effort to ensure they understand their child's opinion, and who help to create an affectively positive environment appear to have children who show greater moral development. In contrast, families with moderate moral reasoning disparity and high levels of operational and informative parental discussion styles predicted the least moral development. These findings are noteworthy since it was believed that an operational style of interaction would induce the intrapsychic conflict necessary to stimulate moral development. In addition, families with high moral reasoning disparity, high levels of informative, cognitively interfering parental discussion styles, and high levels of conflictual interactions also predicted little moral development. This result seems to suggest that family environments where there is a direct conflictual and confrontational approach to discussing moral issues may not be the most ideal in fostering children's moral development. As a result, this study seems to demonstrate that parental behavior, in terms of discussion styles, seems to be an important contributor to the moral development of children. Further, Walker and Taylor's study adds to the growing body of research findings which suggests that parents play an influential role in their children's moral development and that a positive affective parental style of interaction may facilitate this development.  23  The Different Roles of Friends and Parents In the two previous sections, the research findings presented have shown that parents and friends seem to have an important influence on children's moral development. In addition to these results, there has been evidence suggesting that both parents and friends may play valuable, albeit different, roles in children's moral development. These results come from the social development literature which has provided illustrations of the unique importance of both parental relations and friendships to children. A number of these studies are discussed below. In a study by Furman and Buhrmester (1985), the researchers obtained fifth- and sixth-grade children's perception of their relations with significant others. These children reported parents to be important sources of affection, enhancement of selfworth, a sense of reliable aid, and instrumental aid. Furthermore, mothers were seen by children to be more important for companionship than fathers, and girls reported more intimacy with mothers than fathers. Friends, on the other hand, were seen by children to be providers of companionship as well as intimacy. Moreover, girls reported more intimacy, affection, and enhancement of worth with their best friends compared to boys. Thus, children appear to view relations with friends quite differently than relations with parents. In another study, Buhrmester and Furman (1987) recorded children's perceptions of their relations with significant others in three grades. They found that the second-, fifth-, and eighth-grade students reported that their same-sex friends were important providers of companionship. However, relative to same-sex friends, ratings on companionship with parents were lower for second graders, equal for the fifth graders, and lower for the eighth graders. As for intimacy ratings, there was a clear sex difference. Girls' intimacy ratings for parents were significantly higher in Grade 2, equal in Grade 5, and significantly lower in Grade 8 compared to the ratings for samesex friends. Boys' intimacy ratings for parents were equal in Grade 2, significantly  24 higher in Grade 5, and equal in Grade 8 relative to same-sex friends. Again, these results suggest that children perceive their relations with parents as qualitatively different than those with same-sex friends. In addition to Furman and Buhrmester, Youniss (1980; Youniss & Smollar, 1985) has devoted much research effort investigating differences in children's relations with parents and friends. In his first book, Youniss (1980) was able to paint the two social worlds, adult and peer relations, in which children are involved by asking them to describe their interactions with their friends and parents. Youniss found that in childadult relations, adults are perceived by children as having unilateral authority. As a result, children tend to conform to the requests made by adults. However, as children approach adolescence, they begin to see adults less as authority figures and more as people with their own assets and shortcomings similar to themselves. This reflects children's movement towards a position of equality between them and adults where both are free to take initiatives, and where both can provide help when the other is in need. In contrast, the developmental course which peer relations follows is different than that of adult relations. At the age of 8 years, children understand peer relations as the pragmatic practice of equality and direct reciprocity. Direct reciprocity is seen in concrete terms as the sharing or exchange of skills or tangible objects. At around the age of 9 years, children appear to have a different understanding of reciprocity. First, there is an emerging understanding of cooperation as an important principle in the process of reciprocity, and thus, important in friendship formation and maintenance. Second, there is a growing awareness of friends being psychologically different from oneself. Children described their friends "as being in separate psychological states and as making adjustments to their states. Actions subsequently taken appeared to be directed toward achieving equality or equity between the states" (Youniss, 1980, p. 230). During early adolescence, the principle of equity is expanded to take into account  25 similar personalities and shared identity. As well, the understanding of cooperation is elaborated to include the openness of coming to one another and revealing problems and admitting difficulties. The outcome of subscribing to these principles of direct reciprocity is the establishment of enduring intimate relationships. To further explore the nature of adolescent relations with parents and friends, Youniss, along with Smollar, performed an additional series of studies. In these studies, Youniss and Smollar (1985) interviewed adolescents asking them to describe their relations with their mothers, fathers, and same-sex close friends. Youniss and Smollar discovered that adolescent daughters perceived their fathers to be authority figures who provided advice on practical matters as well as guidelines on how their daughters should behave. In general, the contact between the daughter and father was infrequent, and when it did occur it usually lacked intimacy, understanding, and acceptance. Mothers also were considered authority figures. However, unlike with fathers, daughters confided in, fought with, and disobeyed their mothers more frequently. Further, the mother-daughter relationship was perceived as meeting each other's emotional and material needs. Therefore, relations with mothers was seen as more mutually equitable than with fathers. Sons, on the other hand, saw their father as a person with whom they shared recreational or work activities, and discussed objective issues or practical problems. This relationship was perceived to be unilateral in that the father met the material needs of the son, while the son reciprocated by being obedient to and respectful of the father. Thus, the relationship can be described as distant but respectful. However, sons' relations with mothers were closer than their relations with fathers in terms of openness and confidence. Sons tended to tell their mothers about their activities and about the problems they were having. Nevertheless, the mother was still considered the rule-maker who demanded respect and obedience. Hence, mothers appeared to be  26 more open in their relations with their sons while still maintaining some unilateral authority. The majority of female relations with same-sex close friends can be characterized by shared activities, mutual intimacy, mutual understanding, acceptance of respect for differences of opinions, and a perception of self as relaxed, open, natural, outgoing, accepted, and accepting. This also applied to about 40% to 45% of the male same-sex friendships. However, for another 30% of male friendships, this characterization did not apply. For this subgroup, close friends were characterized by shared activities that involved a sense of guardedness in communication rather than intimacy, intolerance rather than acceptance and respect, and non-understanding rather than mutual understanding. Moreover, this subgroup of young men could involve themselves in only a limited range of discussion topics when interacting with their friends, and were more likely to see themselves in close friendships as criticized, distrustful, insensitive, and selfish, than as open, relaxed, and accepting. The studies by Furman and Buhrmester (1985; Buhrmester & Furman, 1987) and Youniss (1980; Youniss & Smollar, 1985) have been important in demonstrating that children perceive their relations with friends quite differently than their relations with parents. They also show that there are clear differences in how females, compared to males, view their relations with parents and friends. The fact that children perceive friends and parents in two distinct "lights" may lead one to suspect that children will behave differently when interacting with a parent as opposed to a friend. There has been some evidence suggesting that, in fact, this may be the case. For instance, Kruger (1992) had "focal" girls (mean age = 8.6 years) undergo an intervention designed to stimulate moral growth. In this intervention program, the focal girls discussed moral conflicts with either a same-sex friend or with their mother. In addition to the pre- and post-test measures that monitored changes in moral reasoning, Kruger observed and categorized the girls' styles of interaction with parents and friends, and related these  27 styles to the girls' level of moral reasoning at post-test. While at pretest there was no difference in level of moral reasoning between the focal girls with a friend versus those with their mother, upon post-test a difference was evident. Focal girls with a friend were reasoning at a higher level on the post-test measure given immediately following the discussion compared to focal girls with their mothers. There was no difference in level of reasoning between pre- and post-test for the girls who interacted with their mothers. Furthermore, in trying to predict level of moral reasoning based on dyadic styles of interaction, Kruger reported that the Leadership Style in peer interaction was one of two best predictors of post-test reasoning: "Leadership Style features the focal's spontaneous control of the interaction by way of questioning and passive compliance by the partner" (p. 204). The Egalitarian Style in the mother-daughter interactions was the other best predictor of post-test reasoning: "Egalitarian Style represents the focal's active and spontaneous collaboration with the partner in the transactive dialogue, ... [and] features the focal's and partner's equal status" (p. 204). One interpretation of the findings is that the girls who felt comfortable enough to express their critical thoughts in an uninhibited fashion may be the common denominator underlying moral development in both social situations. For instance, in the cases where the "uninhibited types" were interacting with their mothers, they may have considered themselves as equals in trying to resolve the conflict, with both mother and daughter "speaking their mind." On the other hand, when these same type of girls were interacting with their peers, they may have taken greater responsibility and control of the situation relative to their friends. That is, compared to these girls, the friends may have felt less able to openly criticize ideas and to feel comfortable in taking control of the situation. As well, the reason why the egalitarian style was not predictive of higher stages of moral reasoning in peer interaction may be because most of the girls in Kruger's sample felt uncomfortable in expressing their thoughts or in taking control, and that the uninhibited types may have been the exception rather than the norm. Whether this interpretation  28 of Kruger's findings is accurate is uncertain, and further scrutiny of her data would be helpful. There are caveats to Kruger's study which need to be mentioned. First, to generalize the results to all children would be inappropriate since only girls from a single age group (7 - 10 years) were included in the sample. Second, Kruger did not employ the appropriate experimental control groups to determine whether the mere exposure to additional moral problems would promote development independent of any discussion with a partner. As well, a general control group that received only the pre- and post-tests should have been included in order to determine whether higher reasoning on the post-test would have occurred regardless of exposure to additional moral issues. These points are important because by including the proper control groups alternative interpretations of Kruger's results may have been eliminated. For example, if the changes in level of moral reasoning take place independently of any interaction with partners, Kruger's findings could be interpreted to mean that while interaction with friends may not contribute anything unique to increasing level of reasoning, interaction with mothers may act to prevent moral reasoning growth. Despite the flaws in Kruger's study, her findings do suggest that there are important differences in how children interact with parents compared to friends, and that these differences may be important in understanding variability in children's level of moral reasoning. Thus far, a major focus of this paper has been on understanding the importance of relationships on children's moral reasoning development. However, there is another area of human functioning that may be important to consider in order to further our understanding of moral reasoning development. This area is known as ego functioning (Haan, 1977), or adaptational styles (Hart, 1992; Vaillant, 1977). Like moral reasoning development, there is a history behind this concept of ego functioning. Knowledge of  29 this history may be of benefit to the reader, and so I temporarily retreat from the moral frontiers to present a brief discussion on ego functioning. Ego Functioning  The historical roots of the study of ego functioning can be traced to Freud's concept of a defense mechanism (1894/1962). For Freud, the function of defense mechanisms was to keep conflicting thoughts or feelings out of awareness. Moreover, Freud described five important properties of defense mechanisms: a) Defenses were major means of managing instinct and affect; b) they were unconscious; c) they were discrete from one another; d) although often the hallmarks of major psychiatric syndromes, defenses were reversible; and, finally, e) defenses could be adaptive as well as pathological. (Vaillant, 1986, pp. viii-ix) However, while Freud introduced and explained defense mechanisms and some of their properties, he provided no empirical evidence of their existence (Paulhus, Fridhandler, & Hayes, in press). This evidence would come later. The seed planted by Freud a century ago has produced many studies on defense mechanisms with research branches growing in a number of theoretical directions (see Paulhus et al., in press). One of these branches represents the work of Norma Haan (1977; Haan, Aerts, & Cooper, 1985). In terms of the importance of this branch, Paulhus et al. wrote: "The work of Norma Haan has had the single strongest influence on contemporary work--both on defense mechanisms and coping" (p. 15). As Paulhus et al. (in press) have demonstrated, researchers hold strikingly different conceptions of the nature of defense mechanisms. In her writing, Haan (1977) redefines defense mechanisms into the term ego functioning. Ego functioning refers to the processes people use to solve life problems and to achieve a better understanding of who they are as "persons". These processes help people organize their lives in order to create a sense of self-consistency. The reason for investigating ego functioning in the present study is derived, in part, from Haan's belief that understanding individuals' ego  30 functioning may be important in explaining some of the variability across people in stage of moral reasoning. Haan's interest in ego functioning is the result of her dissatisfaction with both the psychoanalytic tradition and the cognitive-developmental paradigm in explaining development. From one viewpoint, the Freudian system is oversold on the omnipresence and centrality of pathological functioning but undersold on the importance of rational determination in everyday life. The Piagetian system is oversold on the omnipresence of rationality and undersold on the facile willingness of people to twist, bend, and forego rationality when it suits them. (Haan, 1977, p. 6) As a result of this dissatisfaction, Haan sought to establish a middle ground that would integrate these two different theoretical perspectives. She was able to accomplish this through her model of ego processes. Haan stated that there are many ego processes people may use in dealing with everyday life situations. However, in order to make some sense of all of them, Haan created a taxonomy with three general modes of which two, coping and defending, will be discussed. Haan drew a clear distinction between coping and defending ego processes. She believed that coping ego processes are normative in that people regularly employ them in everyday situations. The use of coping ego processes is considered important to an individual's moral development because they "permit" the assimilation and accommodation of new information which may be critical to this development (Haan, 1977). However, under stressful circumstances a person may perceive these situations as threats to his or her sense of self-consistency. In order to deal with these threatening events, the person may employ non-normative defensive ego processes in order to endure the stressful situation. Defensive processes operate by intervening between a person's moral competence and his or her moral performance and distorting, but not destroying, his or her moral stage achievement. Hence, through this model, Haan believed that she would be able to maintain the Piagetian insistence  31 that structures are irreversible in development, and are directly expressed in action, while preserving the Freudian insistence that defensive actions are typically used by people when a threatening situation warrants. In addition to the general distinction between coping and defending, Haan specifically described 10 ego processes of defense and 10 ego processes of coping, with each defense process having a coping counterpart (see Appendix B). Haan further grouped the 20 ego processes into four categories according to their general function: (1) Cognitive processes generally represent the instrumental aspects of one's problemsolving efforts and involve the accommodation of other perspectives; (2) Intraceptivereflexive processes reflect a person's effort to assimilate his or her thoughts, feelings, and  intuitions; (3) Attention-focusing processes describe the effort to be aware of and to focus on important problems; and (4) Affective-impulse Regulating processes describe efforts to transform primitive feelings and emotions into forms which are accommodating to the social context. One example of a coping ego process under the cognitive function is Objectivity. People using this process are able to separate ideas from feelings, and ideas from each other in order to achieve objective evaluations when the situation warrants it. Isolation, on the other hand, is the counter-part defensive ego process under the cognitive function. People using this process are unable to relate their feelings with their ideas, and unable to link their ideas together. There has been a small body of research investigating the relationship between defense mechanisms and cognitive and social circumstances. However, one should note that most of this work has been carried out by researchers working outside Haan's ego functioning paradigm. Nevertheless, the findings of these studies are helpful in the general understanding of the nature of defense mechanisms as it relates to this study. Defense Mechanisms and Cognitive Abilities One of the few studies relating defense mechanisms to cognitive abilities was conducted by Chandler, Paget, and Koch (1978). These researchers attempted to  32 explore the "relationships between children's developing cognitive abilities and their success or failure in interpreting and explaining various mechanisms of psychological defense" (p. 197). They began by explaining that "affective interchanges" people experience in life are comprised of three elements forming a subject-affect-object complex: "An author or subject (S) initiates an impulse or affect (A) towards a particular target or object (0)" (p. 198). During stressful circumstances, defense mechanisms operate by transforming this complex in three ways. One way is to transform any single element in the S-A-O complex into its logical inverse. Two defense mechanisms which do this simple inverse are repression and denial. Repression focuses on the affect element. An example of repression consists of the following transformation: "'I feel anger towards you' to the alternate 'I don't feel anger at all'" (p. 199). Denial, on the other hand, focuses on the object element. An example of denial involves transforming the initial statement "I feel anger towards you" to "I feel anger, but not at you." The second form of transformation involves substituting any single element with its reciprocal and the end result being the neutralization of that element. Four defense mechanisms which fall into this form are rationalization, displacement, turning against the self, and reaction formation. The first three mechanisms substitute the object term with some more acceptable alternative. Reaction formation, on the other hand, substitutes the affective element with some more supportable alternative. Finally, the third form involves the whole complex as opposed to an individual element. This operation "focuses on more encompassing or overarching joint propositional statements which combine into a single superordinate unit the separate elements previously considered" (p. 199). Two defense mechanisms which involve operating on two or more elements are projection and introjection. An example of projection would be when "I am angry at you" is transformed to "You are angry at me." In the case of introjection, "You are angry at me" is changed to "I am angry at you."  33 Chandler et al. (1978) hypothesized that since simple inverses and reciprocals involve reversibility in thought children who have attained concrete operational thinking will have little trouble understanding these two types of defense mechanisms. Further, since children employ inverse before reciprocal operations (Piaget, 1970), Chandler et al. proposed that children's understanding of mechanisms using inverse operations would precede those using reciprocal operations. And in order to understand the more complex mechanisms like projection and introjection, Chandler et al. argued that formal operational thought is a necessary prerequisite. Formal operational reasoning ... is assumed by Piaget to entail a propositional form of logic which operates on metastatements or second-order propositions about elementary propositions and allows a kind of reversibility of thought which incorporates the flexibilities of both simple inverse and reciprocal operations. (Chandler et al., p. 200) The findings Chandler et al. report support their hypotheses. That is, preoperational children were unable to understand the defense mechanisms presented in this study. Concrete operational children were able to understand those mechanisms using simple inverse and reciprocal transformations, and had an easier time with the former compared to the latter. However, concrete operational children were generally unable to understand the more complex mechanisms which formal operational children were able to understand. Thus, these results suggest that children's understanding of psychological defenses is dependent, in part, on their level of cognitive development. Defense Mechanisms and Social Perspective-taking Abilities In addition to cognitive abilities, there is some evidence that children's perspective-taking abilities relate to their use of defense mechanisms. This evidence comes from a study by Dollinger and McGuire (1981). These researchers separated children into three age groups (youngest, mean age = 6-4; intermediate, mean age = 86; oldest, mean age = 11-7). All the children were presented with stories of academic and social conflict. These stories depicted children using the defense mechanisms of  34 repression, denial, displacement, projection, rationalization, somatization, and selfblame (turning against self). After the story was read, children were asked to explain why the story character acted the way he or she did. Children's responses were recorded and scored for their understanding of the stories. In addition, children were given a social perspective-taking measure designed by Chandler (1973). This measure involved children having to look at a related sequence of cartoon pictures and tell a story as the pictures were presented. Once this was done, one or two "significant" pictures were removed from the series, and, in a modification of the standard procedure, a puppet was put in place of these pictures. The children were then asked to retell the story from the puppet's perspective with the significant pictures missing. Children's responses were scored based on the egocentric intrusions the children made. Egocentric intrusions consisted of the attribution of privileged information (information known to the children) to the puppet. The results of this study showed that of the low egocentric children, a majority (75%) of them had few failures in defense understanding. However, approximately half (47%) of those children who were moderately egocentric and none (0%) of the very egocentric children passed the understanding-of-defense criterion. Hence, Dollinger and McGuire's findings suggest that the reason why children have a difficult time explaining another's hidden or covert motives is related, in part, to the difficulty they have in taking another child's perspective in conflicts. Defense Mechanisms and Gender As well as focusing on the relationship between defense mechanisms and cognitive and social perspective-taking abilities, there have been a few studies which have looked at gender differences in the use of defense mechanisms (see Cramer, 1979). These studies have reported that when choosing defense mechanisms, male and female adults differ. Adult males score higher on the mechanisms turning against the object and projection, whereas females score higher on turning against the self. There  35 has also been a tendency for females to score higher on reversal, which includes negation, denial, reaction formation and repression, and principalization, which includes intellectualization and rationalization. According to Cramer (1979), "these findings support Erikson's (1964) position that males are psychologically oriented toward the external world, whereas females are oriented internally" (p. 476). In Cramer's (1979) study, she focused her attention on the defense mechanisms of adolescents. Her subjects comprised two age groups, Grade 9-10, and Grade 11-12 students. To assess the functioning of their defense mechanisms, students were given Gleser and Ihilevich's (1969) Defense Mechanism Inventory (DMI). After analyzing the data, Cramer found that turning against the object and projection were chosen significantly more often by males compared to females. Turning against the self and principalization, on the contrary, were chosen significantly more often by females than males. Moreover, planned comparisons of the simple interaction effects within each age group show that the Sex by Defense interaction was stronger in the older group than in the younger group. Cramer interpreted these findings as suggesting that the externalizing of conflict for males and the internalizing of conflicts for females begin sometime prior to early adolescence.  Defense Mechanisms and Social Circumstances In addition to gender differences in the operation of defense mechanisms, there is some evidence that in different social circumstances children exhibit the use of different defense mechanisms. This evidence comes from Cramer (1983), who studied children's defense mechanisms through their responses to videotaped vignettes depicting unpleasant child-related situations. In this study, children in two age groups (Grades 1-2, and Grades 4-5) were asked to watch four peer situations and four adultchild situations. Upon completion of the video presentations, the experimenter explained to the children that the ending of the movie had not been portrayed, and were then asked what they would do in a situation like the one that was presented.  36 Children's responses were recorded and scored for the defensive or coping nature of the reaction. In addition to the open-ended question, the experimenter gave each of the children four possible endings to the movie, and then asked each child to select the one they would do if they were the character in the video. The four possible endings represented one of four defense responses: turning against the object, turning against self, projection, and reversal. In support of previous findings, Cramer (1983) reported that the younger boys chose turning against the object significantly more often than did girls, while the younger girls chose reversal significantly more often than boys. Since both of these defenses are logically simple, Cramer's results are consistent with those of Chandler et al. (1978). As well, the findings offer some evidence that boys became externally oriented, while girls internally oriented. However, gender-related differences did not emerge for the older group. Moreover, in terms of differences between the two age groups, little was found. It seems that the more complex mechanisms have not fully developed by Grades 4 and 5. This is supported by the fact that the complex mechanisms of projection and turning against the self are not found in the response repertoire of the older children in this study. Another set of findings in Cramer's (1983) study that is of interest was that children seem to make different defense choices in unpleasant situations involving peers compared to those situations involving adults. While reversal is the mechanism of choice in adult-child situations, turning against the object, turning against the self, and projection are chosen more often in peer situations. Moreover, the most frequent coping response recorded from the open-ended question to the peer situation was to negotiate the problem. In the adult situation, children's coping response was acquiescence to the adult's position. Thus, when faced with a conflict involving a larger, more powerful parental figure on whom the child is dependent, it may be more adaptive for children to give up their wishes. And in order to prevent unpleasant  37 feelings of doing this, they adapt by using the mechanism of reversal to change their mental thoughts. On the other hand, when conflicts with peers arise, children need not abandon their wishes or feelings. Instead, they may feel free to "express these directly (turning against the object), or may misrepresent the agent (as in projection) or the object (as in turning against the self) of the negative feeling" (Cramer, 1983, p. 93). Hence, from these studies, one obtains a better understanding of the relationship between such variables as cognitive and social perspective-taking abilities, gender, and social circumstances and the use of particular defense mechanisms. This information will prove to be helpful in the formulation of hypotheses for the present study. Yet before these hypotheses are laid out a final discussion remains. In the following section, we return to the moral frontier armed with our knowledge of defense mechanisms and ego functioning. The hope of this study is to use this knowledge in order to further our understanding of children's moral reasoning development. Ego Functioning and Moral Reasoning Much of the research investigating the relationship between ego functioning and moral reasoning has been carried out by Haan (1977; Haan, Aerts, & Cooper, 1985). Haan's work is important because of its potential of making a unique contribution to our understanding of moral development. While Kohlberg's (Kohlberg et al., 1983) model emphasizes the importance of cognition and rationality to the moral domain, Haan's model attempts to explore beyond rationality by including other significant factors, such as emotions and possible psychopathologies, which are wrapped in an individual's ego functioning, or adaptational style as Hart (1992) labelled it. This personality variable of ego functioning may be an important influence on the moral reasoning process. Certain ego processes may lead some children to reason morally at higher levels. Generally speaking, Haan (1977) believed that coping processes would allow people to reason at their level of competence. On the other hand, there may be some ego processes which lead children to reason at lower levels. Haan generally  38 thought that defending processes would impede people from reasoning at their competence level. To determine the validity of her beliefs, a number of studies have been conducted. These studies which have investigated the connection between children's ego functioning and their moral reasoning, at first sight, provide no clear consensus on the nature of this relationship. Haan, Stroud, and Holstein (1973), for example, compared stages of moral reasoning and ego functioning based on extensive psychiatric interviews of "hippies." Haan et al. reported that there was a positive correlation between moral stages and several coping processes. Those processes most positively associated with moral stage included objectivity, intellectuality, logical analysis, and concentration. In addition, the intraceptive coping functions were positively correlated with stage of moral reasoning, but to a lesser extent. Defensive functions, however, were independent of stage of moral reasoning. In another study, Haan (1977) had 47-year-old adults -- subjects and spouses -complete a series of measures to obtain the following information: IQ, SES, cognitive stage, moral stage, and ego functioning score. In predicting moral stage, the researchers formed a regression equation. They reported that neither IQ, SES, or cognitive level made a significant contribution. Ego functioning, on the other hand, did make a significant contribution. Specifically, denial and intellectualizing made positive contributions to predicting moral stage scores, whereas repressiveness made a negative one. In explaining these findings in light of Haan et al.'s (1973) study, Haan (1977) states that, when "general factors" such as SES and IQ are not controlled, coping functions appear to be positively correlated with level of moral reasoning. When these general factors are controlled people who reach higher levels of moral reasoning seem to engage in the defending processes of denial and intellectualization which suggests that these people are socially and emotionally "illiterate" to some degree.  39 Hart and Chmiel (1992), on the other hand, reported results which differ from those of Haan (1977). In their study, Hart and Chmiel utilized Kohlberg's data from his longitudinal study involving boys and men (Colby, Kohlberg, Gibbs, & Lieberman, 1983). In re-analyzing the data, Hart and Chmiel reported that adolescent ego functioning is related to concurrently measured stage of moral reasoning as well as stage of moral reasoning in adulthood. When stage of moral reasoning in adolescence is controlled, Hart and Chmiel found that a significant amount of variance in stage of moral reasoning in adulthood is accounted for by ego functioning in adolescence. Further, in cross-lagged correlations between adolescents' ego functioning and their stage of moral reasoning in adulthood, the researchers report that these correlations are consistently higher than the cross-lagged correlations between adolescents' stage of moral reasoning and their ego functioning in adulthood. This infers a causal link between ego functioning in adolescence and stage of moral reasoning in adulthood. In terms of the importance of specific ego processes, while adolescent coping processes from all four of Haan's (1977) ego function groupings generally were predictive of higher stages of moral reasoning compared to defending processes, in particular objectivity and intellectuality were positively associated, whereas isolation, regression, rationalization, and denial were negatively associated with moral reasoning advancement. Finally, Hart and Chmiel created a regression equation to predict adult moral reasoning development based on the following variables: adolescents' ego functioning, IQ, and stage of moral reasoning. They discovered that stage of moral reasoning in adulthood (age range 24 - 34) was predicted by adolescents' stage of moral reasoning, IQ, and ego functioning. Furthermore, stage of moral reasoning for adults at ages 32 - 34 years was best predicted from their ego functioning in adolescence as well as their IQ. Stage of moral reasoning in adolescence, however, did not significantly improve this prediction.  40 Hart and Chmiel's (1992) findings are important because they suggest that the ego processes people employ may determine, in part, how they develop morally. In addition, these results indicate that some coping processes, or "mature processes" as these authors label them, may facilitate moral development while some defending processes, or "immature processes", may impede it. Therefore, defensive processes may not only intervene between a person's moral competence and performance in a particular situation by distorting his or her moral stage achievement as Haan (1977) claimed, but constant use of these defensive processes across situations may impede the structural development taking place. Likewise, as well as allowing the individual to reason at his or her competence level in specific situations, regular use of coping processes across situations may act to facilitate moral reasoning development. These points are elaborated by Hart and Chmiel: Adaptational styles are likely to influence moral judgment by regulating the extent to which individuals are (1) sensitive to the weaknesses of their own moral principles, and therefore experience cognitive disequilibrium, and (2) capable of understanding, and being empathic to, the perspective of others. It seems clear that individuals who are using mature adaptational styles will be more likely than those who rely heavily on immature ones to develop sophisticated moral judgment. (pp. 173-174) An additional study by Haan (1978) seems to support Hart and Chmiel's (1992) findings. In this project, Haan had a sample of adolescents ranging in age from 12 to 17 years from which six friendship groups were formed. These participants were pretested first to obtain an assessment of their moral stage of reasoning based on Kohlberg's (1984) work. This was followed by an intervention which involved the participants engaging in five group sessions of games and simulations which involved moral issues. These sessions involved different tasks and these tasks varied in the general amount of stress produced within the participants. For instance, in one session, participants played the game "NeoPD":  41 Two subgroups were constituted to play a version of Prisoner's Dilemma for a penny a point, with each subgroup's final earnings being equally divided among its members. After the first three blocks of trials, negotiations between representatives of each subgroup began. The oscillations between competition and cooperation usually found in Prisoner's Dilemma occurred. (Haan, 1978, p. 292) Another game called "BaFa" involved randomly dividing the group into two subgroups each subgroup representing different cultures: Each culture privately practiced its designated mode of communication and expressed its underlying values; one culture was egalitarian, but highly competitive, whereas the other was sexist, but noncompetitive. Each culture observed the other; then the players visited the opposite culture one at a time and attempted to participate in it with varying degrees of help from their hosts. After all had visited, the entire group discussed the meanings and worth of each culture, the 'cultural shock' they felt when they visited, and so on. (Haan, 1978, p. 292) Upon termination of these group sessions, the participants were tested again 1 and 4 weeks later. Results suggested that, compared to a control group which received only pre- and post-test, subjects in the experimental group experienced significant advancements in stage of moral reasoning 1 and 4 weeks after termination of the sessions. As well, through a factor analysis procedure, Haan was able to specify 14 factors underlying the ego processes. When ego factor scores were averaged across group sessions, objectivity and cognitive coping factors contributed positively to the regression equation predicting stage of moral reasoning at the first post-test. Pretest scores of stage of moral reasoning also explained a significant amount of the variance in post-test scores. Hence, Haan has shown in this study that cognitive coping processes seem to be an important predictor of post-test moral reasoning scores. The results of Haan's (1973, 1978) and Hart and Chmiel's (1992) studies clearly illustrate the importance of cognitive coping processes to moral development. Yet Haan's (1977) study seems to suggest that the defending processes of denial and intellectualization positively correlate with stage of moral reasoning. Why this discrepancy exists is uncertain. One explanation could be due to differences in the  42 situational context between the studies. Situational factors have been shown to be an important influence on both the stage of moral reasoning obtained as well as the ego processes used. In Haan's (1978) study, for instance, in addition to forming regression equations with post-test scores as the dependent variable, Haan created regression equations for each of the individual group sessions. The dependent variable in these equations was individuals' stage of moral reasoning which was scored based on individuals' responses in each of the group sessions. Haan reported that there was a high degree of variability in level of moral reasoning across the five sessions. For example, moral reasoning scores were significantly higher in the BaFa condition compared to the NeoPD condition. Thus, production of moral reasoning seems to vary with the nature of the situation. Moreover, which ego factor contributed significantly to each of the five regression equations also differed. For example, in the conflictual NeoPD session, defensive doubting, cognitive coping, interpersonal accuracy, and not denying factors contributed to the regression equation. In contrast, in the BaFa session, affective regulation, suppression, interpersonal logic, and cognitive empathy factors contributed positively to the regression equation. Hence, different patterns of ego processes seemed to coordinate levels of moral reasoning in different situations. Similar findings were reported in another study by Haan (1986). Haan's (1978) study has been important in illustrating how specific ego processes used in particular situations can predict moral stage scores obtained from responses made within that situation. Further, Haan has shown that when ego processes are averaged across situations, the use of cognitive coping processes are predictive of posttest moral stage scores. However, what remains uncertain is whether the people with whom individuals interact in discussion or game situations influence the ego processes that individual employs, and whether these selected ego processes relate to stage of moral reasoning in some way. Recall that in Haan's study, she had groups consisting of friends. Yet there is evidence (Cramer, 1983) suggesting that individuals would react  43 differently if the group members were unfamiliar peers, unfamiliar adults, or family members. Further, as has been shown, there is a substantial body of evidence which suggests that interaction with both friends and parents are important contributors to a child's moral reasoning development. Given the moral importance of these people, it seems possible that how children interact with them may be a good predictor of the children's moral reasoning level. For instance, if children use predominantly defending ego processes in their interaction styles with friends and family, they may be more likely to reason morally at lower levels compared to children who use predominantly coping processes. As a result, the present study was conceived in order to investigate this issue. In particular, this study attempts to investigate whether children's ego functioning in moral discussions with parents and friends can be used to predict children's stage of moral reasoning based on Kohlberg's Moral Judgment Interview (MB). One reason why individual interviews are used to obtain data for scoring moral stage is because Kohlberg's individual interview is the standard measure of moral reasoning competence. Another reason is because the use of these interviews to obtain moral stage scores allows these scores and ego functioning scores to be obtained from separate data sources. Ego functioning scores are obtained from observing dyadic interactions between target children and their partners. This is done to prevent the claim that relations between level of moral reasoning and ego functioning may have been artificially produced since the two variables may be tapping the same underlying factor within the same data. The Present Study  To investigate the relationship between children's ego functioning and their stage of moral reasoning, the following study was conducted. In this study, participants' ego functioning was observed in two social contexts: parent-child and friend-child interactions. Further, participants from two grades (Grade 5 and 10) were included in this study in order to assess age differences. In each interaction, the two partners  44  discussed one hypothetical moral dilemma as well as two real-life moral conflicts involving the two partners. To generate these real-life conflicts, each person revealed a moral conflict they were having or have had with their partner. In light of the theoretical and empirical literature presented, some preliminary predictions are made regarding how the variables of social context, age, gender, and ego functioning relate. These predictions are followed by the main prediction involving the relationship between moral reasoning, ego functioning, age, and gender. To begin with, age differences in children's ego functioning were predicted. Recall that Chandler et al.'s (1978) study investigated the developmental trends in defense mechanisms and showed that there is a correlation between children's cognitive abilities and their understanding of more "complex" defense mechanisms. Since children's understanding of defense mechanisms depends on their level of cognitive development, it was thought that children's use of defense mechanisms would depend, to some extent, on their level of cognitive development as well. As a result, it was predicted that younger children will use less complex ego processes compared to the adolescents. For instance, the cognitive coping processes seem to require formal operational abilities in order to stand back from the situation, reflect, and evaluate it objectively. Thus, these processes are more likely to be used by children in Grade 10 compared to those in Grade 5. Likewise, the coping intraceptive processes require the ability to deal with cognitive complex situations, and to understand the other person's perspective. Such thinking requires complex cognitive abilities more often found in adolescents than in younger children. As well, based on the work of Chandler et al., one can predict that the less complex defending processes of rationalization, denial, repression, displacement, and reaction formation may be more often used by the younger children, while the more complex defending process of projection would be expected to be used more often by adolescents.  45 Secondly, differences were predicted in the use of ego processes when children interact with their friend as opposed to their parent. As the work of Youniss (1980; Youniss & Smollar, 1985) and Furman and Buhrmester (1980) have shown, children do perceive their relations with parents differently compared to their friends. Furthermore, Cramer's (1983) study has shown that children choose different defense mechanisms in conflictual situations involving their parents as opposed to their friends. One explanation for these reported differences may be due to the power structure in these two relationships. In friendships, both members are considered to be of equal status, and therefore, each person's opinion should be considered of equal value. In parent-child relationships, since the parent is considered the authority figure, there may be a tendency on the part of the child to undervalue his or her own opinion while viewing the parent's point-of-view as necessarily correct. As a result, it is reasonable to expect that children will employ the cognitive and intraceptive coping processes to a greater extent in their interaction with friends than with parents. These coping processes require the consideration of feelings and ideas generated by themselves and others in order to negotiate and solve the problem. Relationships where there is this balance of power seem to suit these type of processes. On the other hand, the affective coping processes will be used more often with parents than with friends. When interacting with parents, children may be more careful in how they express themselves in the presence of their parents who are often acting as instructors in how to behave socially. Finally, in terms of the defending processes, in interacting with friends, children will use projection and the cognitive processes more often than when interacting with their parents. These processes may be important to the child in order to cope with the threatening feeling of inequality in the relationship. However, the defending attention-focusing and affective processes, as well as the regression and doubt, were predicted to be employed in interaction with parents as opposed to friends.  46 Such defense processes may be used by children when interacting with parents who they perceive to be more knowledgeable about moral issues. Thirdly, gender differences in the use of ego processes in children's social interaction are predicted. Since girls report more intimacy, affection, and enhancement of worth from their friends relative to boys (Furman & Buhrmester, 1985), one may expect girls to be characterized as using the coping processes of empathy and sublimation in their interaction with friends in comparison with boys. As well, according to Cramer (1979), girls are considered to be more internally oriented, whereas boys are more externally oriented. Therefore, it may be expected that the use of the defense mechanisms of rationalization and intellectualization would be more typical of girls, while projection would be more typical of boys. Fourthly, differences in the use of ego processes as a function of the type of dilemma (i.e, hypothetical vs. real life) is predicted. The results from Hart and Chmiel's (1992) and Haan's (1978) studies seem to suggest that in hypothetical moral dilemmas, people tend to use predominantly cognitive processes. In real-life dilemmas, a broader array of ego processes are expected to be used. In Walker, de Vries, and Trevethan's (1987) study, real-life dilemmas generated by children (age range: 6 - 15 years) revolved around issues relating to friendship, honesty, theft, and fighting. Since these dilemmas may be personally significant to the participants, affective-impulse regulating processes and intraceptive processes may be more common-place here than in hypothetical dilemmas. While these first four predictions are aimed at improving one's understanding of the ego functioning domain, this last prediction attempts to further uncover the relationship between ego functioning and moral reasoning, which is the main goal of this study. Based on the literature reviewed there are a number of predictions which can be made about the nature of this relationship. With the exception of Haan's (1977) study, researchers have found that the use of coping processes is correlated with higher  47 levels of moral reasoning (Haan, 1978; Haan, Stroud, & Holstein, 1973; Hart & Chmiel, 1992). In particular, these researchers found that cognitive, intraceptive-reflexive, and attention-focusing coping functions are positively correlated with level of moral reasoning. In contrast, the researchers reported that the four defending functions were negatively correlated with level of moral reasoning Thus, in general, it was predicted that the coping functions will be positively correlated, and the defending functions will be negatively correlated, with level of moral reasoning. Yet to what extent do each of these ego functions explain some unique part of the variance in children's level of moral reasoning? In order to determine this, multiple regression equations will be formulated with level of moral reasoning acting as the dependent variable, and age, gender and the eight ego functions being the independent variables. While age differences have been found in level of moral reasoning (see Colby et al., 1983), it is important to determine whether ego functions are able to explain additional variance of the dependent variable. Although the cognitive and intraceptivereflexive coping functions, and the cognitive, attention-focusing, and the affectiveimpulse regulation defending functions are predicted to share some of the explained variance with age, each of these functions are predicted to explain unique components of the variance as well. That is, although these functions are believed to be correlated with age, they, nevertheless, may account for portions of the variance not accounted for by age. While age differences in level of moral reasoning have been reported in the literature, clear gender differences have not surfaced (see Walker, 1991). Yet it is predicted that there are gender differences in the use of cognitive and intraceptive defending functions. The role that gender may play in the regression equation is to act as a suppressor variable, thereby enhancing the effects of these and other ego functions.  48 Method  Participants With the cooperation of the Vancouver School Board, 40 children were recruited to participate in this study through letters sent to parents of the students. Only children from two-parent families were asked to participate. The children were selected from two age groups with 10 boys and 10 girls in each group. One age group was comprised of Grade 5 students (M = 10.9 years, SD = .34) and the other age group was comprised of Grade 10 students (Ig = 15.5, SD = .31). In addition, a parent and a same-sex friend of each child (the friend was chosen by the child) were asked to take part in the study thus creating a total N of 120 participants. Generally, the fathers were older than the mothers (fathers' M = 47.9 years, SD = 5.7, range = 36-60, vs. mothers' M = 44.1 years, SD = 6.8, range = 32-54). The friends who participated averaged the same age as the target children although there was greater variability (Grade 5 friends' M = 11.1, SD = 2.3, range = 9.4-12.7, and Grade 10 friends' M = 15.4, SD = .83, range = 13.8-17.5).  Procedure Each target child participated in two sessions involving an interview and a dyadic discussion (i.e., parent-child & friend-child), and each session lasted approximately 2 hours. The order in which these sessions occurred was randomized. The two sessions were spaced approximately a week apart. Each session involved having two participants (either parent and child or friend and child) report to the Department of Psychology at the University of British Columbia. After being briefed about the study and having provided written consent, both participants were taken to separate rooms for an individual interview of approximately an hour in length. These interviews were audio-taped for later transcription and scoring. Each interview involved children reacting to four moral dilemmas; three were hypothetical dilemmas from Kohlberg's Moral Judgment Interview  49 (MJI) and one involved a conflict that the participant reported having or already had with their partner. Although there are three different forms of the MJI (A, B, and C), only two were used, A and B. The reason for this is that forms A and B are exactly parallel to each other, whereas Form C is not. The form chosen for the interviews was randomly selected and the same form was given to both participants. The three hypothetical dilemmas from the randomly selected MJI form were read to each participant followed by a series of questions from the WI manual about the dilemma. The fourth dilemma was generated by the participant and included responses to standardized questions (see Appendix C). These questions allowed the participants to describe the conflict, why they considered it to be a moral conflict, and how they felt the conflict should be resolved. Following the individual interviews, the two participants convened for a dyadic discussion. There were three parts to this section. The first part involved the discussion of one of the hypothetical dilemmas which they had previously heard in the individual interview. In order to promote discussion, the experimenter chose a dilemma in which there was some disagreement in responses between the participants. The second part involved the discussion of a real-life conflict the child was having or already had with the partner (i.e., parent or friend) as revealed in the individual interview. The third part involved the discussion of a real-life conflict the partner was having or already had with the child as revealed in the individual interview. The order in which these three moral conflicts were discussed was randomized. It should be noted that each participant was asked in the individual interview to consent to revealing his or her conflict to the partner. In the rare event that the participant did not generate a conflict or refused to discuss the conflict with the partner, he or she was asked to generate an example of a conflict that may occur between a child and a parent (or friend depending on who the partner was).  50 For each discussion, the two participants were instructed to review (in the case of hypothetical dilemma) or reveal (in the case of real-life dilemmas) the dilemma and then discuss a series of questions similar to the ones provided in the individual interviews. Further, they were asked to discuss the questions such that each person had the opportunity to express his or her position. For questions where there were disagreements, the participants were asked to try and resolve the disagreements. After the instructions were given, the experimenter left the room. The only time the experimenter reentered the room was at the end of each of the dilemma discussions in order to inform the participants which conflict they were to discuss next and to provide them with the appropriate set of discussion questions. The discussion section lasted approximately an hour and was video-recorded. The video-recording was used for subsequent coding of ego functioning. The entire session lasted approximately 2 hours. Upon completion of the last discussion, the experimenter entered the room to thank the participants for their time and provided them each with the $10 remuneration. This process was repeated approximately a week later with the child and his or her other partner.  Scoring Moral reasoning. Using Colby and Kohlberg's (1987) Standard Issue Scoring Manual, the MJIs for each target child were blindly scored by the author. Based on these interviews involving hypothetical dilemmas, responses were analyzed such that every moral judgment which matched a criterion judgment in the appropriate section of the manual was scored. The moral judgments obtained were then summarized into a single score by calculating Weighted Average Scores (WAS). The WAS is calculated by multiplying the percent usage of each stage by the stage number and then summing the totals (ranges from 100 - 500). Since each target subject completed two MJIs (one for each session), a single score was calculated based on the six dilemmas from the two interview sessions. Some justification for averaging across sessions comes from the fact  51 that there was no significant difference between sessions in terms of WAS. When a 2 (grade) X 2 (sex) X 2 (session) Analysis of Variance (ANOVA) was computed, no session effect was found a: (1, 36) = 2.031, p = .163). As well, the correlation between the two forms was a substantial r = .65. Inter-rater reliability was determined through a second rater who independently scored 10 randomly selected interviews (25% of the sample). There was a high correlation between the two raters for the WAS ( = .93). As well for GSSs, there was high reliability with Cohen's Kn = .87. Cohen's Kn statistic was used because it corrects for chance agreement between the raters (Brennan & Prediger, 1981).  Ego functioning. Haan's (1977) Q-sort method was used to code for the use of specific ego processes by each child in the discussion sessions. The advantage of using the Q-sort procedure is that it allows for easy comparison of results across samples with very different backgrounds. The Q-sort is comprised of 60 different items, three of which represent each of the ten defending ego processes and three of which represent each of the ten coping ego processes. 0-sorts were completed for each of the three dilemma discussions (one hypothetical and two real-life). Upon viewing a video-taped discussion, coders sorted through and placed the 60 items into a forced quasi-normal distribution of nine steps. Items most uncharacteristic of the child were placed at the low-scoring end and items most characteristic of the child were placed at the highscoring end of the distribution, with neutral items in the middle. The target's two discussion sessions were coded by different observers to preclude the possibility of coder bias. Each target child was involved in six discussions (two sessions X three dilemmas), and two well-trained coders performed Q-sorts on these discussions. The inter-rater reliability between these two coders was assessed by having them score 42 randomly selected video-taped discussions (17.5% of the sample). Similar to past research (Haan, 1977), the Q Correlation was used as the index of inter-rater reliability.  52 The standard criterion for adequate inter-rater reliability is a Q Correlation of .40. The mean correlation between the coders was .56, with 83% of all the correlations being above the standard criterion. Based on this Q-sorting procedure, two types of ego functioning scores were obtained. The first set of scores are those obtained for each of the 20 individual ego processes. This is calculated by summing the three item scores within each process. The second set included eight summary scores representing each of the four ego functions in two modes, coping and defending. These summary scores are obtained by averaging the scores for each of the processes within a particular function mode. For example, the summary score for the cognitive coping function is obtained by averaging the coping scores for objectivity, intellectuality, and logical analysis.  53 Results The results to be presented are organized into three sections. The first section addresses the issue of how ego functions and processes differ as a function of age, gender, social circumstance (parent-child vs. friend-child interaction), and type of dilemma. The second section examines patterns of moral reasoning. Finally, the third section reports the relationship between ego functioning (as well as demographic variables) and moral reasoning. Ego Functioning Two paths were taken in exploring the data on children's ego functioning. The first path involved conducting exploratory post hoc analyses. Given that relatively little research has been performed in this area, it is of interest to determine what significant effects would surface when ego functioning was examined as a function of all variables in general. The second path involved investigating the specific a priori hypotheses made. Post hoc Analyses To perform the post hoc analyses, eight 2 (grade: 5, 10) X 2 (gender) X 2 (session: parent-child, friend-child) X 3 (dilemma: hypothetical, target real-life, partner real-life) ANOVAs were conducted, one for each ego function. In each of these ANOVAs, grade and sex were between-subject factors, while session and dilemma were within-subject factors. Since eight univariate ANOVAs were conducted post hoc, a Bonferroni inequality procedure was implemented to prevent an inflated Type I error. Thus, a family-wise error rate criterion was set at p < .08, and the p criterion for each individual ANOVA was set at a conservative .01. By setting a more liberal family-wise .p criterion than the traditional .05 and using the conservative Bonferroni inequality procedure, it was hoped that a reasonable compromise between Type I and Type II errors would be achieved.  54 From the set of eight ANOVAS, significant effects were found for three ego functions. For the cognitive coping ego function, there was a main effect for grade, F (1, 36) = 7.45,9 < .01. Older children scored higher on cognitive coping compared to younger children (Ms = 5.96 vs. 4.76) 1 . This effect is consistent with the prediction. Given that the use of cognitive coping processes require more mature cognitive abilities, it was expected that younger children would be less likely to possess these processes in their repertoire compared to older children. As well, for cognitive coping, there was a main effect for dilemma, F (2, 72) = 4.90, p < .01. Differences across dilemmas here and subsequently in this section were examined by the Tukey method of multiple comparisons test with .p < .01. This comparison revealed that on hypothetical dilemmas, children evidenced more cognitive coping compared to real-life target and real-life partner dilemmas (s = 5.68 vs. 5.23 and 5.17). This effect is again consistent with my prediction. That is, children were expected to use predominantly cognitive processes on hypothetical dilemmas in which they may have little personal investment. An additional finding was the dilemma effect for the attention-focusing coping function, F (2, 72) = 4.96,9 < .01. Here the use of this function was more characteristic of children in the hypothetical dilemma discussions compared to both the real-life target and partner dilemma discussions (Ms = 5.80 vs. 5.34 and 5.25). Although this effect was not predicted, it seems congruent with the dilemma effect for the cognitive coping function. That is, if children are more logical, objective, and intellectual in hypothetical dilemmas compared to the real-life dilemmas, they are more likely to be task-focused (i.e., concentration) as well. For attention-focusing defending ego function (i.e., denial), there was a main dilemma effect, F (2, 72) = 6.95, p < .002. In the hypotheical dilemmas, the attention1. Note. The mean scores are based on a 1 to 9-point scale with 1 representing "most uncharacteristic", 9 representing "most characteristic", and 5 representing the neutral point.  55 focusing defending ego function was more uncharacteristic of children than in the reallife dilemmas (Ms = 4.25 vs. 4.73 and 4.80). This finding does not fit with predictions, and therefore, suggests that there may be more to the dilemma effect than initially thought. Moreover, there was a significant grade X dilemma effect, F (2, 72) = 7.24, .p < .001. Further investigation into this interaction was conducted by analyzing the simple main effects of grade for each dilemma. When this was done, a marginally significant grade effect was discovered for the real-life partner dilemma only, F (1, 38) = 6.98, .p < .012. Grade 5 children tended to use more denial in their real-life partner dilemmas compared to Grade 10 children (Ms = 5.30 vs. 4.30). In addition, for the attention-focusing defending function, there was a significant session X dilemma effect, F (2, 72) = 5.30, p < .007. This interaction was further explored by analyzing the simple main effects of dilemma for each session. This analysis revealed that there was a dilemma effect only in the parent-child session, F (2, 72) = 9.25, _co < .001. In parent-child sessions, children were using more denial in the real-life dilemmas compared to the hypothetical dilemmas (s = 3.96 vs. 4.97 and 5.01). While the exploratory analysis confirmed some of my predictions, they also provided new information which was not anticipated, yet may be important in understanding the nature of children's ego functioning. One important finding is that older children do tend to use more of the "complex" cognitive coping processes compared to younger children. However, the dilemma effects which have emerged suggests that my initial theorizing may be wrong. In the hypothetical dilemmas, children tend to use more cognitive and attention-focusing coping processes compared to the real-life situation. These results generally can be considered to be consistent with my predictions. However, the fact that there was a dilemma effect for the attention-focusing defending function (i.e., denial) was not predicted. Why is it that  56 children use more denial in real-life dilemma discussions than in hypothetical dilemma discussions? An answer to this question will be proposed later. As well, other findings in need of explanation are the interaction effects for denial. First, younger children were found to use more denial in their real-life partner dilemmas compared to Grade 10 children. Second, in the parent-child sessions, children were using more denial in the real-life dilemmas compared to the hypothetical dilemmas. A discussion of these results will be presented later.  Planned Analyses In addition to post hoc analyses, planned analyses were conducted to investigate the specific a priori predictions made (pp. 43-47). In these analyses, the conventional .05 significance level will be used. The first issue deals with whether older children (Grade 10) use more complex ego processes compared to younger children (Grade 5). To examine this, ANOVAs were conducted for each of the following ego functions and processes: the cognitive coping function, the intraceptive-reflexive coping function, rationalization, denial, and the affective-impulse regulating function (repression, displacement, and reaction formation). These processes were examined with a 2 (grade: 5, 10) X 2 (session: parent-child, friend-child) X 3 (dilemma: hypothetical, target real-life, partner real-life) ANOVA. Session and dilemma were included as factors since they are repeated measures, but only grade effects will be reported at this point. These analyses revealed four significant main effects for grade (see Table 1). First, as the post hoc analyses found, there was a grade effect for cognitive coping function, F (1, 38) = 7.14, p = .011. As predicted, use of the "complex" cognitive coping function was more characteristic of older as compared to younger children (Ms = 5.96 vs. 4.76). Second, there was a grade effect for intraceptive-reflexive coping function, F (1, 38) = 5.43, p < .05. Older children used the complex intraceptive coping function more often than younger children (Ms = 5.87 vs. 5.32). Third, there was a grade effect  57 TABLE 1 Children's Ego Functioning across Grade  Grade  Ego Function/Processes  10  5  Cognitive Coping  4.76  (1.56)  5.96  (1.28)  Intraceptive Coping  5.32  (.65)  5.87  (.81)  Affective Defending  5.10  (.66)  4.68  (.45)  Repression  4.98  (1.18)  4.30  (.70)  Note. Standard deviations are indicated in parentheses. All functions reported show a  significant grade difference at the .05 level.  58 for the affective-impulse regulating defending function, F (1, 38) = 5.58, 9 < .05. Younger children tended to use this less complex defending function more than did older children (Ms = 5.10 vs. 4.68). Fourth, there was a grade effect for repression, F (1, 38) = 5.01,.p < .05. In this case, younger children were found to use the less complex process of repression more frequently than older children (Ms = 4.98 vs. 4.30). Hence, these findings lend support to the prediction that younger children use the less complex processes and functions more than older children, while older children tend to use the more complex processes and functions to a greater extent than the younger children. However, there were no significant grade effects found for the two remaining defending processes of rationalization and denial. The second set of predictions concerns differences in children's use of ego processes when interacting with a parent as opposed to a friend. It was predicted that children would use cognitive and intraceptive-reflexive coping functions, cognitive defending function, and projection to a greater extent when interacting with friends versus parents. On the other hand, it was predicted that children would use affectiveimpulse regulating coping and defending functions, denial, regression, and doubt more often when interacting with parents versus friends. For each of these processes and functions, a 2 (session) X 3 (dilemma) ANOVA was conducted. Dilemma was included as a factor since it is a repeated measure, but only session effects will be reported at this point. The results show no significant main effects for session were found for any of the ego processes and functions (cognitive and affective-impulse regulating functions, intraceptive-reflexive coping function, projection, denial, regression, and doubt). That is, there was no significant differences in children's use of these ego processes and functions when they interacted with their friends versus their parents. The third set of predictions dealt with gender differences in the use of ego processes. It was predicted that girls would use more empathy and sublimation in their  59 interaction with friends in comparison with boys. Moreover, girls, in general, were predicted to use the defensive mechanisms of rationalization and intellectualization more often than boys. In contrast, boys were predicted to use more projection than girls. To examine these issues, 2 (sex) X 2 (session) X 3 (dilemma) ANOVAs were conducted for the above ego processes. Again, session and dilemma were included as factors because they were repeated measures, but only gender effects will be reported at this point. The ANOVAs conducted showed a significant gender effect for the projection process, F (1, 38) = 5.08,D < .05. Females were characterized as using this externally oriented process more than males which is contrary to the prediction made (Ms = 4.75 vs. 4.04). However, none of the other functions showed a gender effect. Finally, differences in ego processes were predicted between types of dilemma. That is, in hypothetical dilemmas, cognitive processes were hypothesized to be more frequently found, while the real-life dilemmas were thought to invoke intraceptivereflexive and affective-impulse regulating processes. To determine the validity of these claims, 2 (session) X 3 (dilemma) ANOVAs were computed for each of the six ego functions (cognitive coping and defending, intraceptive-reflexive coping and defending, and affective-impulse regulating coping and defending). Session was included as a factor in the ANOVAs since it is a repeated measure, but only dilemma effects will be reported at this point. From this set of ANOVAs, three significant dilemma effects emerged (see Table 2). Differences across dilemmas in this section were examined using the Tukey method with < .05. First, as reported in the post hoc analyses, a significant main effect for dilemma was found on the cognitive coping function, F (2, 78) = 4.94,D < .01. It appears that children evidence more cognitive coping on hypothetical dilemma discussions compared to real-life target and real-life partner dilemma discussions (Ms = 5.68 vs. 5.23 and 5.17). As well, there was a significant main effect for dilemma on the intraceptive-  60 TABLE 2 Children's Ego Functioning across Dilemma Discussions  Dilemma Target  Partner  Ego Function  Hypothetical  Cognitive Coping  5.68a  (1.61)  5.23b  (1.64)  5.17c  (1.74)  Intracep. Defending  4.17a  (1.38)  4.57 b  (1.40)  4.57c  (1.57)  Cognitive Defending  3.85d  (.99)  4.12  (.98)  4.20e  (.87)  Note.^Standard deviations are indicated in parentheses.  a differs from b and c, p < .05 d differs from e, p < .05  61 reflexive defending function, F (2, 78) = 3.95,.p < .05. Use of the intraceptive defending function was more characteristic of children in real-life target and partner dilemma discussions than in hypothetical dilemma discussions (s = 4.57 and 4.57 vs. 4.17). These two findings are consistent with the predictions that children use predominantly cognitive processes in hypothetical dilemma discussions where they may have little personal investment, and use more intraceptive and affective functions in real-life dilemma discussions where they have more personal investment. However, there is evidence which raises doubts about the generality of this finding. For instance, a main effect for dilemma on cognitive defending function was found, F (2, 78) = 3.85, D < .05. Upon further exploration, it was discovered that children's use of the cognitive defending function was more uncharacteristic on hypothetical dilemma discussions than in the real-life partner dilemma discussions (s = 3.85 vs. 4.20). This is contrary to the prediction that children in hypothetical dilemma discussions are characterized as generally using more cognitive functions than in real-life dilemma discussions. Thus, while the grade effects in the use of complex ego processes and functions emerged, which are consistent with the hypotheses and post hoc findings, other predictions did not bear out. In the case of ego functioning between the two sessions, no significant differences arose. That is, children did not appear to use different ego processes when interacting with their friends as opposed to their parents. Moreover, in the case of ego functioning between gender, only one significant effect emerged. Females were characterized as using projection more frequently than males. This is contrary to my hypothesis which predicted that males would use more projection compared to females. Finally, the results for the dilemma effect suggest that the reasoning behind the proposed hypothesis may be more complicated than initially believed. While use of cognitive coping was more characteristic of children in hypothetical dilemma discussions, and use of intraceptive-reflexive defending was more  62 characteristic of children in real-life dilemma discussions as predicted, there were other findings not consistent with the predictions. Use of cognitive defending was more uncharacteristic in hypothetical dilemmas than in real-life dilemmas. Moreover, the findings from the post hoc analyses show a dilemma effect for the two attentionfocusing functions which was not predicted. Denial was found to be more characteristic of children in real-life dilemmas than in hypothetical dilemmas. Concentration, on the other hand, was found to be more characteristic of children in the hypothetical dilemmas versus the real-life dilemmas. These latter findings which were not predicted are informative in that there may be more to the dilemma effect - the situational context - than previously thought. Moral Reasoning As mentioned previously, children's overall level of moral reasoning was obtained by combining the scores yielded by the two interviews (Form A & B; i.e., six dilemmas). Level of moral reasoning was expressed in weighted average scores (WAS). To determine grade and gender effects, a 2 (grade) X 2 (gender) ANOVA was conducted. Two significant results emerged. First, there was a strong main effect for grade, F (1, 36) = 52.49,D < .001. Older children tended to score higher in their level of moral reasoning compared to younger children (s = 311.5 vs. 256.0). This finding is consistent with past research showing age to be positively correlated with children's level of moral reasoning (Colby et al., 1983). As well, there was a main effect for gender, F (1, 36) = 8.96,.p < .005. Females tended to score lower in their level of moral reasoning compared to males (Ms = 272.3 vs. 295.2). This finding is surprising given that there generally is a lack of gender differences using Kohlberg's measure, especially in childhood (Walker, 1984, 1986b, 1991). Moral Reasoning and Ego Functioning This final section attempts to identify the relationship between ego functioning and moral reasoning. In order to do this, a multiple regression procedure was  63 performed with level of moral reasoning acting as the dependent variable. Age, sex, and the eight ego functions (both coping and defending on the cognitive, intraceptivereflexive, attention-focusing, and affective-impulse regulating functions) were the predictor variables. A summary score for each of the ego functions was obtained by averaging over dilemmas and sessions. To form the regression equation, the stepwise selection method was performed. The probability of F-to-enter was set at a default of .05. Therefore, a variable was allowed to enter the equation only if the probability associated with the F test was less than or equal to the default .p < .05. Based on the stepwise selection procedure, three predictors variables emerged significant (see Table 3). The results indicate that level of moral reasoning can be predicted from age, the cognitive coping function, and the attention-focusing coping function. These three variable were able to account for 68% of the variance in level of moral reasoning (R 2 = .683). Age, not surprisingly given the strong developmental character of moral reasoning, accounted for the greatest amount of variance, 53%. The partial correlation for age (controlling for the other variables in the equation) demonstrated a strong, positive relationship with moral reasoning. Cognitive coping accounted for an additional 11% of the variance when age was controlled. The partial correlation for cognitive coping showed a strong, positive relationship with moral reasoning as well. Attention-focusing accounted for a further 4% of the variance when age and cognitive coping were controlled. The partial correlation for attention-focusing coping illustrated that there is a negative relationship with moral reasoning. That is, when age and cognitive coping are controlled, attention-focusing is negatively correlated with moral reasoning. After controlling for these variables, none of the others added significantly to the regression equations; and when the remainder of the variables were forced into the equation, they, in combination, only accounted for an additional 5% of the variance.  64 TABLE 3  Multiple Regression Analysis of Moral Reasoning Scores  Predictor Variables  ^  R^R2^R2^Partial Change  Age  .73  .53  .53 ***  .66  Cognitive Coping  .80  .65  .11**  .54  Attention Coping  .83  .68  .04*  -.32  Note. *** p < .001 ** p < .005 * p < .05.  65 Further, an examination of the zero-order correlations (Table 4) illustrates the fact that most of the coping ego functions were positively correlated, and most of the defending ego functions were negatively correlated with level of moral reasoning. Also, there were strong inter-correlations among the ego function variables (see Table 4). That is, the coping functions tended to be positively correlated with each other, the defending functions tended to be positively correlated with each other, and the coping functions were negatively correlated with the defending functions, as would be expected. Yet despite these high correlations, the tolerance level for each of the functions did not reach a low enough level to warrant concerns about the problem of multicollinearity (Dillion & Goldstein, 1984; Tabachnick & Fidell, 1989).  66 TABLE 4  Ego Functioning Correlation Matrix  MJI Cognitive Coping Cognitive Defending Intraceptive Coping Intraceptive Defend Attention Coping Attention Defending Affective Coping  Cog  Cog  Intr  Intr  Attn  Attn  Aff  Aff  Cop  Def  Cop  Def  Cop  Def  Cop  Def  .59  -.41  .41  -.51  .25  -.40  .19  -.48  -.78  .74  -.94  .76  -.76  .44  -.89  -.79  .76  -.72  .67  -.64  .67  -.77  .52  -.58  .36  -.79  -.83  .74  -.58  .86  -.67  .63  -.72  -.64  .60 -.35  67 Discussion The purpose of this study was to explore two areas of human functioning, children's ego functioning and their moral reasoning, and to determine how they may be linked with each other. While there has been a substantial amount of research devoted to extending the moral frontiers, there has been much less work investigating children's ego functioning, and even fewer studies bridging the two areas. Moreover, the prior research attempting to connect these two areas has produced some conflicting results. Thus, this study sought to reexamine these areas in hopes of improving our understanding of children's ego functioning and its relation to moral reasoning development. The initial task of merging these areas of study was undertaken by Haan (1977). Her reason for doing this was based on her dissatisfaction with the psychoanalytic approach's overemphasis on pathological functioning and underemphasis on the importance of rationality. As well, Haan was equally unhappy with the cognitivedevelopmental approach's overemphasis on rational functioning and underemphasis on other aspects of functioning such as affect. Thus, Haan attempted to integrate these paradigms into her model of ego processes. Since Kohlberg's model is considered a cognitive-developmental model, the charge that it tends to emphasize rational, objective thought holds some legitimacy. In fact, Kohlberg (Kohlberg et al., 1983) himself built his model on the assumption that cognitivism was an important component in moral reasoning. However, there have been some who have claimed that there is more to the moral domain than rational thought (Blasi, 1990; Gilligan, 1982; Haan, Aerts, & Cooper, 1985; Puka, 1991). The purpose of drawing the area of ego functioning into the moral domain was to broaden it beyond rationality by focusing on other components, like affect, which manifest themselves in individual's ego functioning. Thus, it was hoped that how children "coped" in moral dilemma discussions would reveal further insight into the nature of the  68 moral domain. Yet before discussing these domains jointly, I first will examine each of them individually. The investigation of children's ego functioning in this study was unique in its attempt to identify factors which may influence the use of these coping and defending processes. As a result, gender and age (grade) differences were predicted in the use of children's ego processes and functions. These predictions were made based on past research (e.g., Chandler et al., 1978; Cramer, 1979). Moreover, differences in social circumstances were predicted in the use of ego processes and functions as well. Past research (Cramer, 1983; Haan, 1978) suggests that the social situation in which a child perceives him or herself to be affects how the child reacts in those circumstances. In this study, two situational variables were isolated, types of dilemma and dyadic discussion sessions. Two discussion sessions were selected, parent-child interactions and friend-child interactions. The reason these two discussion contexts were chosen is based on a large body of research illustrating the importance of friends and parents to children's social and moral reasoning development (i.e., Powers, 1988; Thoma & Ladewig, 1993; Walker & Taylor, 1991). In addition, the effect of dilemma type on children's use of ego processes and functions was considered. Haan's (1978, 1986) studies show that college students discussing different topics and playing different games appear to elicit different ego processes and functions. This study attempted to consider both discussion session and dilemma together. The results from this study provides further enlightenment in children's use of ego processes and functioning. To begin with, children's ego functioning was investigated for differences across grade. There has been evidence suggesting that the understanding of defense mechanisms differs as a function of level of cognitive development (Chandler et al., 1978). Based on this study, it was thought that use of the ego functions and processes would vary as a function of grade (age), which is believed to be correlated with level of cognitive development (Piaget, 1970). Consistent with  69 predictions, the results from this study indicate that there were differences found between grade (age) in the use of specific ego processes and functions. In particular, for older children, the use of cognitive and intraceptive-reflexive coping functions were found to be more characteristic compared to younger children. As for younger children, the use of the affective-impulse regulating defending function and the ego process of repression were seen as being more characteristic than older children. Therefore, not only do older children understand these complex ego functions, as Chandler et al. have shown, but older children appear to be using these complex ego functions in their interaction with others as well. Younger children, however, do not have a firm understanding of these complex ego functions and, therefore, are likely to be perceived as using the less complex ego functions and processes. However, the use of the less complex defending processes of rationalization and denial was found not to differ between the two grade (age) levels. While this finding does not support the predictions made, it does not necessarily refute the reasoning behind the predictions either. It is possible that both older and younger children are using these defense processes to the same extent since the two groups possess the required cognitive abilities. While grade differences generally were found in children's use of ego processes and functions as predicted, the same was not true for gender. Boys and girls were not perceived to differ in their use of empathy, sublimation, rationalization, and intellectualization as predicted. This is an interesting finding given the results of some of the research in this area. For instance, Cramer (1979) reported that males tend to choose the externally oriented defense mechanisms of turning against the object and projection. In contrast, females tend to choose the internally oriented defense mechanisms of turning against self and the cognitive coping processes. Moreover, the only significant gender difference found was in the use of projection. Contrary to prediction, females were characterized as using more projection than males. That is,  70 compared to boys, girls appear to be projecting on to others objectionable tendencies rather than recognizing it as part of themselves. A number of reasons for the discrepancy in results between this study and that of Cramer (1979) can be postulated. It is possible that the process of measurement may influence the implementation of ego functions. Thus, it may be true that girls generally prefer internally oriented ego processes while boys generally prefer externally oriented ego processes. However, in specific situations, such as a moral discussion with a friend or parent, these preferences may not manifest themselves in their use. Moreover, the discrepancy could relate to the fact that Cramer's method of assessment was by selfreport while this study assessed children's ego functioning based on observation. Hence, children may perceive themselves as acting differently than the way an outside observer is perceiving them. Future studies investigating the source of this discrepancy would be interesting and worthwhile in order to improve our understanding of the differences between children's perception of their ego functioning and the perception of an objective observer. As well as investigating gender differences, differences in the use of ego functions and processes were explored under various social circumstances. One of the social context variables included in this study was discussion sessions; that is, do children's use of ego processes and functions differ when they are interacting with a parent versus a friend? The interest in this variable was derived from Kohlberg's (1984) claim that peer relations are an important contributor to children's moral reasoning development. As well, Powers (1988) and Walker and Taylor (1991) have shown that interaction with parents is important to children's moral growth. Hence, we know that parents and friends are important to children's moral development. However, in this study, I wanted to determine whether children's use of ego processes and functions would differ between these two dilemma discussion sessions.  71 As discussed, there is a substantial body of research (Buhrmester & Furman, 1987; Furman & Buhrmester, 1985; Youniss, 1980; Youniss & Smollar, 1985) which clearly shows that children perceive their relations with parents differently than their relations with friends. Kruger's (1992) study suggests that children behave differently when interacting with parents versus friends. Finally, in her study, Cramer (1983) reported that children seem to choose different defense mechanisms in situations involving peers compared to situations involving adults. Part of the reasoning behind these findings relates to the balance of power within these relationships. While parents tend to be perceived as possessing a greater amount of power in the parent-child relationship, in the friend-child relationship there is a greater sense of equality. Therefore, it was predicted that there would be strong differences in the use of ego processes and functions between the session conditions. Contrary to previous findings and the stated predictions, there were no significant differences in the use of ego processes and functions between the two sessions. That is, there were no processes or functions more characteristic of children in their interaction with parents than in their interaction with friends. One explanation for this finding is related to the fact that parents who participated in this study were not randomly sampled. Rather, it was necessary to ask parents to volunteer to participate in this study. As a result, these parents may tend to be in relations with their children which can be characterized as more egalitarian than is typical; that is, where parents and children are treated as relatively equal in status. Thus, if this is the case, then one might expect fewer differences between the parent-child interaction and the friendchild interaction. Another explanation for the lack of findings here is that in the parent-child interactions the gender of the parent with whom the child interacted was not considered. In this study, target children were match with friends of the same-sex. However, with parents, target children interacted with either their mother or father.  72 The reason this was done was because of the small cell size that would be produced if an additional variable was taken into account in the statistical analyses of the data. As a result of this decision, however, there may be an interactive effect between gender of the parent and the fact that it is a parental figure. That is, any parental effect expected could be muddied by gender effects since interactions with mothers were not distinguished from interactions with fathers. Thus, such an effect may be obscuring differences that would otherwise surface between parent-child and friend-child interactions. An addition consideration for this unexpected finding is that in past research children's perception of their relations with parents and friends was measured. As well, in Cramer's (1983) study, children were allowed to choose which defense mechanisms they would use in hypothetical situations involving their parents and friends. However, in this study, children's use of ego processes and functions were measured by observing children in interaction with their partners. It is possible that in these real-life interactions, children may not show as much variability in their use of ego processes in different situations as they may expect or as they may perceive themselves. That is, while children may anticipate using specific processes with specific people, when actually interacting with these people children may be employing different processes all together. Thus, this could explain, in part, why session effects were not present. While no session effects were discovered, there were a number of dilemma effects. Recall that in hypothetical moral dilemma discussions, cognitive processes were hypothesized to be more frequently found, while in real-life dilemma discussions, a broader array of processes were anticipated, particularly the use of intraceptivereflexive and affective-impulse regulating processes. The reasoning behind these predictions is that because real-life dilemmas are more personally significant to the target children, intraceptive-reflexive and affective-impulse regulating processes may be more commonly used here than in hypothetical dilemmas. Discussing hypothetical  73 dilemmas may be seen as more of an intellectual exercise requiring predominantly cognitive processes. The results from this study provide evidence that the use of the cognitive coping function is more characteristic of children in hypothetical dilemma than in the real-life dilemma discussions. Moreover, the use of the intraceptive-reflexive function is less characteristic of children in hypothetical dilemmas than in real-life dilemmas. These findings are consistent with the hypothesis. However, there were a number of findings which were not consistent with prediction. For instance, the use of the cognitive defending function was more characteristic of the partner real-life dilemmas compared to the hypothetical dilemmas. This finding suggests that there may be a different explanation (to be discussed later) to the dilemma effect seen. Furthermore, from the post hoc analyses, additional dilemma effects were discovered in children's use of the attention-focusing functions. First, there was a significant dilemma effect for the use of the attention-focusing coping function. In hypothetical dilemma discussions, children were characterized as using more concentration than in the real-life dilemmas. As well, the post hoc analysis revealed a dilemma effect for attention-focusing defending function. Use of this function was more characteristic of children in real-life dilemma discussions compared to the hypothetical dilemma discussions. Thus, the results from both sets of analyses suggested that in the hypothetical dilemmas, children appear to be coping better in terms of their use of cognitive and attention-focusing functions, and defending less in their intraceptive-reflexive and attention-focusing functions. In contrast, in both real-life situations, children appear to be functioning poorly relative to the hypothetical dilemma discussions. Children in real-life discussions are characterized as coping less well in their cognitive function, and defending to a greater extent in their cognitive, intraceptive-reflexive, and attention-  74 focusing functions (relative to hypothetical discussions). Hence, the general picture appearing is that children are functioning relatively poorly in real-life dilemma discussions in terms of coping with conflictual situations. One explanation for these results is that since hypothetical dilemmas tend not to be of any personal significance to the discussants or to the relationship between the discussants, they are perceived to be less threatening, and therefore, less stressful to the target child. As a result, they feel more relaxed in discussing these issues with whomever their partner may be. In contrast, it is likely that children perceive the discussion of the real-life dilemmas as more threatening to the relationship than the discussion of hypothetical dilemmas. That is, the target children may be reluctant to discuss these real-life dilemmas for fear of how their partner may react. These children may feel some stress in discussing these issues, especially when they were aware that they were participating in a research project. Therefore, they are more likely to use defending processes and functions. Such as explanation would be consistent with Haan's (1977) theory. Recall Haan's claim that in stressful circumstances where people perceive these situations as threats to their sense of self-consistency, people may employ defensive ego processes. In addition to the main dilemma effects, an interaction between grade and dilemma was found for one ego function. It seems that Grade 5 children tend to use more denial in real-life partner dilemmas compared to Grade 10 children. This effect is consistent with the prediction that younger children would be more likely to use denial than older children. Moreover, what is of interest is the fact that this effect is only evident in situations where the partner's moral dilemma is being discussed. That is, the target child is denying present or past facts and feelings in regards to the moral conflict the partner has raised. Another interesting finding was the interaction effect between session and dilemma. Upon further investigation, it was discovered that in their interaction with  75 their parents, children were characterized as using more denial in the real-life dilemmas compared to the hypothetical dilemmas. This suggests that children are having a difficult time discussing real-life conflicts involving their parents with their parents. Children seem to want to de-emphasize the importance of moral conflicts they are having with their parents, while emphasizing the more pleasant aspects of the relationships. In summary, children were found to use a variety of ego functions in both hypothetical and real-life dilemma discussions. Moreover, children were found to cope better and defend less in hypothetical discussions than in real-life discussions. In hypothetical dilemma discussions, these children tended to rely more on the cognitive and attention-focusing coping functions than in real-life dilemmas. As well, the use of the defending functions of attention-focusing and intraceptive-reflexive were less characteristic of these children in hypothetical dilemmas than in real-life dilemmas. These findings suggest that social context can have a substantial effect on the processes children use when interacting with others. In situations where they feel uncomfortable, such as discussing moral dilemmas involving their partner with that partner, children may resort to defending processes. However, in situations where they feel more relaxed, such as discussing hypothetical, impersonal dilemmas, children may be more likely to use the cognitive and attention-focusing coping processes. Further, this dilemma effect may be so strong as to negate any session effects. That is, a discussion of a moral conflict the children are having with a partner whom they know well may be so stressful that it does not matter whether the partner is a friend or a parent. What may be relevant is the fact that their partners are people with whom they are well acquainted and to whom they are closely related. As a result, these children may be fearful that discussing a moral conflict between them and their partners may strain the relationship.  76 Whether this is a valid interpretation of the results can only be determined with further research. One issue that should be explored is whether there is a direct correspondence between stress and the use of defending ego functions. There are a number of measures that could be used to assess the stress level of a person in a situation, one being galvanic skin response (GSR). Assuming that children are experiencing stress in these real-life dilemma discussions, would they experience the same amount of stress outside the lab setting? That is, do children experience stress, and thereby, use defending processes and functions when they discuss moral conflicts with their parents or friends in their everyday interactions? To determine this would require field observations. Another issue worth pursuing is to identify other "moral" scenarios besides hypothetical moral dilemma discussions where children may use coping functions. In these alternative situations, are there specific coping functions and processes other than cognitive and attention-focusing which would be more typical? A third issue is to determine the stability in children's use of ego functioning across situations and time. If their use is stable then they may be considered a personality variable. These are some of the important issues worth addressing in future research. In the area of children's moral reasoning, two effects were found. First, there was a grade (age) difference in level of moral reasoning. Grade 10 children scored higher in their level of moral reasoning than Grade 5 children. This finding is consistent with the results from Colby et al.'s (1983) longitudinal study which demonstrated that as their participants grew older, there was a gradual increase in stage of reasoning. Thus, this grade (age) difference was of no great surprise. The second effect emerging was a difference in level of moral reasoning for gender. Females tended to score lower in their moral reasoning compared to males. This finding raises a much debated issue of gender differences in moral reasoning, and has led to a bantering between Baumrind (1986) and Walker (1984, 1986b) in the  77 developmental literature. This debate has revolved around the claim that Kohlberg's stage theory is biased against females (Gilligan, 1982; Haan, 1977). Gilligan, for instance, has argued that Kohlberg's theory does not fairly reflect the feminine concerns of welfare, caring, and responsibility. As a result, in Kohlberg's scoring system, females will tend to score lower than males since Kohlberg's theory emphasizes the masculine attributes of rationality, objectivity, and individuality. However, doubts regarding Gilligan's claims have been raised by Walker (1984, 1986b, 1991). Walker (1991) reviewed the literature covering "all studies using Kohlberg's measure in which sex differences in the development of moral reasoning was examined" (p. 350). Of the 152 samples collected from some 80 studies, Walker reported that 85.5% of these samples showed no significant sex difference. For 5.9% of the samples, females scored higher, and for 8.6% of the samples, males scored higher. Furthermore, Walker performed a meta-analysis with these samples to determine whether there is an overall significant gender difference. The meta-analysis showed that there were no significant difference for gender. Given the substantial body of evidence refuting the gender difference in moral reasoning claim, it is difficult how to make sense of this study's finding. However, two points can be made. First, the sample size in this study is not very large (target children N = 40). This fact needs to be taken into account when interpreting the results of this study. When dealing with a small sample size, there is always the danger that one's sample is not a completely fair representation of the population. Second, the gender difference found is not a very large one (1/5 of a stage). Therefore, although the difference reached statistical significance, the actual numbers are not all that impressive. Certainly these findings would not provide much support to Gilligan's claim that Kohlberg's model tends to place the masculine values of rights, rationality, individuality, etc., at higher stages, while placing feminine values of care and connectedness at lower stages. The validity of such claims would require a more  78 substantial gender difference in level of moral reasoning with a particular focus on Stage 3 and above reasoning (WAS of 300 or more). In this study, the WASs for both genders were below 300. While the discussion thus far has focused on the two areas of human functioning separately, this final section attempts to tie these two domains together by asking: What is the nature of the relationship between children's moral reasoning and their ego functioning? In particular, can our knowledge of which ego functions children's use in moral dilemma discussions aid us in predicting their level of moral reasoning? While Haan (1977, 1985, 1986; Haan et al., 1985) has explored college students' use of ego processes in group discussions and in game situations, she did not compare children's use of ego functions in didactic discussions with parents versus friends. As has been illustrated in the review of the literature, parents and friends are believed to play important, but potentially different, roles in children's moral development. To address these questions, a stepwise multiple regression procedure was performed. The eight ego functions, along with age and sex, were entered into a regression equation to determine whether they predicted children's level of moral reasoning. The results showed that there were three predictor variables which reached significance. They were age, cognitive coping, and attention-focusing coping. Combined, these three variables accounted for 68% of the variance in the level of moral reasoning: (a) age was found to explain a majority of the variance, 53%; (b) cognitive coping explained another 11%; and (c) attention-focusing coping explained an additional 4%. The fact that age was found to be an important contributor is of no surprise given that a significant grade (age) difference was found in level of moral reasoning. Of greater interest is that cognitive coping explained a substantial amount of variance even after controlling for age. Past research (Haan, 1978; Hart & Chmiel, 1992) has found that cognitive coping at pretest tends to predict level of moral reasoning at post-test  79 better than the other ego functions. The findings of this study are consistent with these results. Children's ability to be logical, intellectual, objective, and task-focused seems to be positively correlated to their ability to reason about moral issues. These results seem to substantiate Kohlberg's (Kohlberg et al., 1983) claim that his model places emphasis on cognition: People rely on their rational nature when making moral judgments. Moreover, this coping function may be important in the assimilation and accommodation of new information critical to moral reasoning development. That is, these coping functions may play an important role in the facilitation of children's level of moral reasoning by allowing them to experience a broad array of perspectives on moral issues. Hence, these findings provide some support for Hart and Chmiel's (1992) claim that children using coping functions would be more likely to develop sophisticated moral judgments than those using defending functions. However, the finding that attention-focusing coping explained a significant amount of variance provides an interesting twist to this study. Recall that attentionfocusing became negatively correlated with moral reasoning after age and cognitive coping were controlled. Children at higher levels of moral reasoning were characterized as concentrating less than children at lower levels of moral reasoning. An explanation for this finding is that children at higher levels of moral reasoning may not need to concentrate as hard as children at lower levels when discussing moral dilemmas. These children, having scored high on cognitive coping, may have the cognitive coping abilities necessary to perform the task easily. In contrast, children at lower levels of moral reasoning, and who have scored lower in cognitive coping, may need to concentrate more in the dilemma discussion since they may not possess the full complement of cognitive coping abilities necessary to easily complete the task. Thus, this finding suggests that the specific use of the attention-focusing coping function may not lead to sophisticated moral judgments. However, it should be noted that this  80 finding was a weak one, barely attaining significance. In addition, the zero-order correlation with moral reasoning was found to be positive. Therefore, further research is necessary before any strong conclusions can be drawn. The fact that a relationship exists between moral reasoning and ego functioning may be vulnerable to the criticism that it is, to some extent, spurious. The rationale behind this claim is that since ego functioning and moral reasoning were obtained from contexts of moral deliberations, one should not be surprised to find a relationship since they are both tapping the same type of activity. Therefore, in order to show that a legitimate relationship exists, one could have observed children's ego functioning in situations other than moral deliberations. However, I feel this claim has minimal validity. While it is true that both the individual interviews and dyadic discussions dealt with moral issues, the procedures in the two situations differed substantially. In one situation, level of moral reasoning is obtained based on an individual interview technique which is designed to tap people's reasoning about hypothetical moral dilemmas. In the other situation, use of ego processes is obtained based on observations of people's behavior in a dyadic discussion of both hypothetical and reallife moral issues. These issues covered a wider variety of topics and are more "relational" in nature compared to the hypothetical dilemma discussions in the individual interview. Hence, the underlying data sources used to understand these areas of human functioning do differ both in their breadth of topics as well as in their personal importance to the individual. Moreover, the decision to use moral dilemma discussions as the context in which to observe children's ego functioning was selected for a reason. Researchers belief that these types of discussion scenarios are important in fostering moral reasoning development (Berkowitz, Oser, & Althof, 1987; Piaget, 1932/1965). As a result, it was hoped that how children behaved in these situations may enlighten our understanding of factors important to children's moral growth.  81 The finding that cognitive coping and attention-focusing coping are predictors of level of moral reasoning is further interesting in light of the fact that the use of these functions were two which distinguished real-life from hypothetical dilemma discussions. Recall that these functions were found to be more characteristic in the hypothetical and less characteristic in the real-life dilemmas. More specifically, in discussing moral dilemmas with parents and friends, target children used a variety of ego functions depending on the dilemma context. In hypothetical dilemma discussions, children were characterized as using less defensive functions (cognitive, intraceptive-reflexive, and attention-focusing) and more coping functions (cognitive and attention-focusing) compared to when they discussed real-life dilemmas. The impression yielded by these results was that children found it difficult and stressful to discuss real-life moral conflicts. These target children were asked to discuss with their partner real-life moral conflicts which they were having or had recently had with that partner. While hypothetical moral conflicts were impersonal, real-life conflicts were personal issues which were not always easily discussed. On many discussion occasions it appeared that children were minimizing the importance of the conflict in order to maintain a "smooth" relationship in front of the camera in the lab. Hence, these findings suggest that while discussing real-life moral conflicts may not be an important social context in which to stimulate moral development, discussing hypothetical moral conflicts may be important by encouraging children to use their cognitive coping abilities. However, observing children discussing moral conflicts is only one scenario in which to determine their use of ego functions. There are many other situations in which to obtain this type of information. If cognitive and attention-focusing coping functions are important to children's moral reasoning development, would their use in other social situations and across time continue to be predictive of their level of moral reasoning? Thus, observing children's ego functioning in a variety of scenarios may prove useful in addressing this issue. In addition, doing this will further aid us in  82 determining whether the relationship between ego functioning and moral reasoning is in fact spurious. Thus, from this study further inroads into the moral domain has been accomplished through the revelation of a clear link between personality ego functioning and moral reasoning level. Other than age, the strongest predictor of children's level of moral reasoning was cognitive coping. This is not surprising given the cognitive, rational constitution of Kohlberg's model. However, it should be noted that the majority of the coping functions, not just the cognitive ones, were positively correlated, while most of the defending functions were negatively correlated with level of moral reasoning. These findings suggest that other components, such as affect, are integral parts of the moral domain. The difficulty for psychologists is trying to tease these different components apart. There have been many researchers (e.g., Blasi, 1990; Puka, 1991), who have suggested broadening the moral domain beyond Kohlberg's cognitive conception by identifying and understanding the role these other components play within the area of morality. Yet how one goes about doing this in some systematic fashion, as Kohlberg's theory did, remains a daunting challenge.  83 References Arbuthnot, J. (1975). Modification of moral judgment through role playing. Developmental Psychology, 11, 319-324. Baier, K. (1965). The moral point of view: A rational basis of ethics. New York: Random House. Baumrind, D. (1986). Sex differences in moral reasoning: A response to Walker's (1984) conclusion that there are none. Child Development. 57, 511-521. Berkowitz, M. W., & Gibbs, J. C. (1983). Measuring the developmental features of moral discussion. Merrill-Palmer Quarterly. 29, 399-410. Berkowitz, M. W., & Gibbs, J. C. (1985). The process of moral conflict resolution and moral development. In M. W. Berkowitz (Ed.), New directions for child development: Peer conflict and psychological growth (No. 29, pp. 71-84). San Francisco: Jossey-Bass. Berkowitz, M. W., Gibbs, J. C., & Broughton, J. M. (1980). The relation of moral judgment stage disparity to developmental effects of peer dialogues. MerrillPalmer Quarterly. 26, 341-357. Berkowitz, M. W., Oser, F., & Althof, W. (1987). The development of sociomoral discourse. In W. M. Kurtines & J. L. Gewirtz (Eds.), Moral development through social interaction (pp. 322-352). New York: Wiley. Berndt, T. J. (1981). Relations between social cognition, nonsocial cognition, and social behavior. In J. H. Flavell & L. D. Ross (Eds.), Social cognitive development (pp. 176-199). New York: Cambridge University Press. Berndt, T. J. (1987). The distinctive features of conversations between friends: Theories, research, and implications for sociomoral development. In W. M. Kurtines & J. L. Gewirtz (Eds.), Moral development through social interaction (pp. 281-300). New York: Wiley. Berndt, T. J., & Perry, T. B. (1986). Children's perceptions of friendships as supportive relationships. Developmental Psychology. 22, 640-648. Blasi, A. (1990). How should psychologists define morality? or, The negative side effects of philosophy's influence on psychology. In T. Wren (Ed), The moral  domain: Essays in the ongoing discussion between philosophy and the social sciences (pp. 28-70). Cambridge, MA: MIT Press.  Brennan, R. L., & Prediger, D. J. (1981). Coefficient kappa: Some uses, misuses, and alternatives. Educational and Psychological Measurement. 41, 687-699. Buck, L. Z., Walsh, W. F., & Rothman, G. (1981). Relationship between parental moral judgment and socialization. Youth and Society. 13, 91-116. Buhrmester, D., & Furman, W. (1987). The development of companionship and intimacy. Child Development. 58, 1101-1113.  84 Chandler, M. J. (1973). The picture arrangement subtest of the WAIS as an index of social egocentrism: A comparative study of normal and emotionally disturbed children. Journal of Abnormal Child Psychology. 1, 340-349. Chandler, M. J., Paget, K. F., & Koch, D. A. (1978). The child's demystification of psychological defense mechanisms: A structural and developmental analysis. Developmental Psychology. 14, 197-205. Chapman, M., & McBride, M. L. (1992). The education of reason: Cognitive conflict and its role in intellectual development. In C. U. Shantz & W. W. Hartup (Eds.), Conflict in child and adolescent development. Cambridge, England: Cambridge University Press. Colby, A., & Kohlberg, L. (Eds.). (1987). The measurement of moral judgment: Standard issue scoring manual (Vol. 1-2). New York: Cambridge University Press. Colby, A., Kohlberg, L., Gibbs, J., & Lieberman, M. (1983). A longitudinal study of moral judgment. Monographs of the Society for Research in Child Development, 48 (1-2, Serial No. 200). Cramer, P. (1979). Defense mechanisms in adolesence. Developmental Psychology. 15, 476-477. Cramer, P. (1983). Children's use of defense mechanisms in reaction to displeasure caused by others. Journal of Personality. 51, 79-94. Damon, W., & Killen, M. (1982). Peer interaction and the process of change in children's moral reasoning. Merrill-Palmer Quarterly. 28, 347-367. Dillon, W. R., & Goldstein, M. (1984). Multivariate analysis: Methods and applications. New York: Wiley. Dollinger, S. J., & McGuire, B. (1981). The development of psychological-mindedness: Children's understanding of defense mechanisms. Journal of Clinical Child Psychology. 10, 117-121. Edwards, C. P. (1980). The comparative study of the development of moral judgment and reasoning. In R. H. Munroe, R. L. Munroe, & B. B. Whiting (Eds.), Handbook of cross-cultural human development (pp. 501-528). New York: Garland Press. Erikson, E. H. (1964). Inner and outer space: Reflections on womanhood. In R. J. Lifton (Ed.), The woman in America. Boston: Beacon Press. Freud, S. (1962). The neuro-psychoses of defense. Standard Edition. 3, 43-61. (Original work published in 1894) Furman, W., & Buhrmester, D. (1985). Children's perception of the personal relationships in their social networks. Developmental Psychology. 21, 1016-1024. Gilligan, C. (1982). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press.  85 Gleser, G. C., & Ihilevich, D. (1969). An objective instrument for measuring defense mechanisms. Journal of Consulting and Clinical Psychology. 33, 51-60. Haan, N. (1977). Coping and defending: Processes of self-environment organization. New York: Academic Press. Haan, N. (1978). Two moralities in action contexts. Journal of Personality and Social Psychology. 32, 255-270. Haan, N. (1986). Systematic variability in the quality of moral action, as defined in two formulations. Journal of Personality and Social Psychology. 50, 1271-1284. Haan, N., Aerts, E., & Cooper, B. (1985). On moral grounds: The search for practical morality. New York: New York University Press. Haan, N., Stroud, J., & Holstein, C. (1973). Moral and ego stages in relationship to ego processes: A study of "hippies". Journal of Personality. 41, 596-612. Habermas, J. (1983). Interpretive social science vs hermeneuticism. In N. Haan, R. N. Bellah, P. Rabinow & W. M. Sullivan (Eds.), Social science as moral inquiry (pp. 251-270). New York: Columbia University Press. Habermas, J. (1990). Justice and solidarity: On discussion concerning Stage 6. In T. Wren (Ed.), The moral domain: Essays in the ongoing discussion between philosophy and the social sciences (pp. 224-251). Cambridge, MA: MIT Press. Hart, D. (1988). A longitudinal study of adolescents' socialization and identification as predictors of adult moral judgment development. Merrill-Palmer Quarterly. 34, 245-260. Hart, D. (1992). Becoming men: The development of aspirations, values, and adaptational styles. New York: Plenum Press. Hart, D., & Chmiel, S. (1992). The influence of defense mechanisms on moral judgment development: A longitudinal study. Developmental Psychology. 28, 722-730. Holstein, C. B. (1972). The relation of children's moral judgment level to that of their parents and to communication patterns in the family. In R. C. Smart & M. S. Smart (Eds.), Readings in child development and relationships (pp. 484-494). New York: Macmillan. Inhelder, B., & Piaget, J. (1958). [The growth of logical thinking form childhood to adolescence]. (A. Parsons & S. Milgram, Trans.). New York: Basic. (Original work published in 1955) Jurkovic, G. J., & Prentice, N. M. (1974). Dimensions of moral interaction and moral judgment in delinquent and nondelinquent families. Journal of Consulting and Clinical Psychology. 42, 256-262. Kant, I. (1964). [The groundwork to the metaphysics of morals] (H. J. Paton, Trans.). New York: Harper & Row. (Original work published in 1785) Kohlberg, L. (1981). Essays in moral development: Volume I. The philosophy of moral development. New York: Harper & Row.  86 Kohlberg, L. (1984). Essays in moral development: Volume II. The psychology of moral development: Moral stages. their nature and validity. San Francisco: Harper & Row. Kohlberg, L., Levine, C., & Hewer, A. (1983). Moral stages: A current formulation and a response to critics. New York: Karger. Kruger, A. C. (1992). The effect of peer and adult-child transactive discussions on moral reasoning. Merrill-Palmer Quarterly. 38, 191-211. Maitland, K. A., & Goldman, J. R. (1974). Moral judgment as a function of peer group interaction. Journal of Personality and Social Psychology. 30, 699-704. Newcomb, A. F., & Brady, J. E. (1982). Mutuality in boys' friendship relations. Child Development. 53, 392-395. Parikh, B. (1980). Development of moral judgment and its relations to family environmental factors in Indian and American families. Child Development. 51, 1030-1039. Paulhus, D. L., Fridhandler, B., & Hayes, S. (in press). Psychological defense: Contemporary theory and research. In S. R. Briggs, R. Hogan, & W. Jones (Eds.), Handbook of personality psychology. San Diego: Academic Press. Piaget, J. (1928). [Judgment and reasoning in the child] (M. Gabain, Trans.). New York: Harcourt, Brace. (Original work published in 1924.) Piaget, J. (1952). [The child's conception of number] (C. Gattegno & F. M. Hodgson, Trans.). London: Routledge & Kegan Paul. (Original work published in 1941.) Piaget, J. (1956). The origins of intelligence in children. New York: W. W. Norton & Co. Piaget, J. (1965). The moral judgment of the child. New York: Free Press. (Original work published in 1932.) Piaget, J. (1970). Piaget's theory. In P. Mussen (Ed.), Carmichael's manual of child psychology (Vol.1, pp. 705-732). New York: John Wiley & Sons. Piaget, J. (1985). The equilibration of cognitive structures. University of Chicago Press. (Original work published in 1975) Piaget, J., & Inhelder, B. (1974). [The child's construction of quantities] (A. J. Pomerans, Trans.). London: Routledge & Kegan Paul. (Original work published in 1941). Powers, S. I. (1988). Moral judgement development within the family. Journal of Moral Education. 17, 209-219. Puka, B. (1991). Towards the redevelopment of Kohlberg's theory: Preserving essential structure, removing controversial content. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development: Theory (Vol.1, pp. 373394). Hillsdale, NJ: Lawrence Erlbaum Associates.  87 Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard University Press. Selman, R. L. (1980). The growth of interpersonal understanding. New York: Academic Press. Selman, R. L., & Byrne, D. F. (1973). Manual for scoring social role-taking in social dilemmas. Unpublished manuscript. Thoma, S. J., & Ladewig, B. H. (1991, April). Close friendship and moral judgment development during the college years. Paper presented at the meeting of the American Educational Research Association, Chicago, IL. Thoma, S. J., & Ladewig, B. H. (1993, March). Close friendships, friendship networks, and moral judgment development during the college years. Paper presented at the meeting of the Society for Research in Child Development, New Orleans, LA. Tabachnick, B. G., & Fidell, L. S. (1989). Using multivariate statistics (2nd ed.). New York: Harper & Row. Vaillant, G. E. (1977). Adaptation to life. Boston: Little-Brown. Vaillant, G. E. (Ed.) (1986). Empirical studies of ego mechanisms of defense. Washington, DC: American Psychiatric Press. Walker, L. J. (1980). Cognitive and perspective-taking prerequisites for moral development. Child Development, 51, 131-139. Walker, L. J. (1984). Sex differences in the development of moral reasoning: A critical review. Child Development, 55, 677-691. Walker, L. J. (1986a). Experiential and cognitive sources of moral development in adulthood. Human Development, 29, 113-124. Walker, L. J. (1986b). Sex differences in the development of moral reasoning: A rejoinder to Baumrind. Child Development, 57, 522-526. Walker, L. J. (1988). The development of moral reasoning. Annals of Child Development, 5, 33-78. Walker, L. J. (1991). Sex differences in moral reasoning. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development: Research (Vol.2, pp. 333-364. Hillsdale, NJ: Lawrence Erlbaum Associates. Walker, L. J., de Vries, B., & Trevethan, S. D. (1987). Moral stages and moral orientations in real-life and hypothetical dilemmas. Child Development, 58, 842858. Walker, L. J., & Taylor, J. H. (1991). Family interactions and the development of moral reasoning. Child Development, 62, 264-283. Youniss, J. (1980). Parents and peers in social development: A Sullivan-Piaget perspective. Chicago: University of Chicago Press.  88 Youniss, J., & Smollar, J. (1985). Adolescent relations with mothers. fathers. and friends. Chicago: University of Chicago Press.  89 APPENDIX A KOHLBERG'S MORAL STAGE MODEL PRECONVENTIONAL Preconventional individuals are those who do not understand and uphold shared systems of moral rules, roles, and norms. Individuals at this level are responsive to cultural rules and labels of good and bad, right and wrong, yet interpret these rules and labels in terms of the physical consequences to their action, or in terms of the physical power of those who enforce the rule labels. Stage 1: Heteronomous Morality  Some of the features of this stage is that: (1) there is the ability to see situations from only a single perspective; (2) rightness and wrongness are defined by authority figures; (3) moral judgments are self-evident requiring no justification beyond citing the rules; and (4) wrong actions are identified with punishment. Stage 2: Individualistic, Instrumental Morality  Stage 2 can be characterized in the following ways: (1) individuals have the ability to recognize and respect more than one perspective in the situation; (2) individuals aim to maximize their own desires; (3) reciprocity is important at this stage in that individuals coordinate their actions in order for themselves to benefit.  90 CONVENTIONAL Individuals at the conventional level see maintaining the expectancies of their family, group, or nation as valuable in its own right. At this level, there is an attitude of conformity and loyalty to maintaining, supporting, and justifying personal expectations as well as the social order. Further, individuals at the conventional level are able to identify with other persons beyond themselves and understand morality to be shared systems of moral rules, roles, and norms.  Stage 3: Interpersonal Normative Morality The features of this stage are as follows: (1) there is the understanding of a third-person perspective; (2) there is the subscription to mutually trusting relationships among people; and (3) there is a belief in shared moral norms which emphasize people being good, altruistic role-occupants. Stage 4: Social System Morality Stage 4 has these characteristics: (1) individuals taking a members-of-society perspective; (2) a belief in the existence of a social system of sets of codes and procedures which apply equally to all members; and (3) the pursuit of one's interest as the legitimate goal only if it maintains the social system.  91 POSTCONVENTIONAL At this level, people understand the social systems of moral norms, rules, and roles, but accepting these societal systems is dependent on whether they accept the general moral principles which underlie these systems. Individuals at this level strive to define moral values and principles which have validity and application apart from the authority of groups. Stage 5: Human Rights and Social Welfare Morality At this stage, individuals consider: (1) the universal values and rights to be a part of any moral society; and (2) the societal laws and other social systems to be legitimate only to the extent that they either represent, preserve, or protect fundamental human rights and values. Stage 6: Morality of Universalizable, Reversible, and Prescriptive General Ethical Principles At this stage, individuals operate based on the principles of justice and respect for persons, and are able to take on all perspectives in a situation and balance these different points of view. Note. adapted from Colby & Kohlberg (1987, pp. 25-35)  92 APPENDIX B DEFINITIONS OF EGO PROCESSES coping cognitive processes  Objectivity: Subjects separate their ideas and feelings from each other so that they achieve objective evaluations when situations require this sort of behavior. Intellectuality: Subjects are capable of detachment in an affect-laden situation which requires impartial analysis and awareness and is so detached from restrictions of the environment and self that they are able to give their thoughts free rein. Logical Analysis: Subjects are interested in analyzing thoughtfully, carefully, and cogently the causal aspects of situations, personal or otherwise.  defending cognitive processes  Isolation: Subjects' affect seems not to be related to their ideas, and/or they seem not to be able to put their ideas together. Intellectualization: Subjects with high ratings retreat from affect to formulations of words and abstractions. Subjects think and talk on a level of abstraction not quite appropriate to the situation, use jargon, and do not specify how these ideas relate to context. Rationalization: Subjects offer superficially plausible reasons to explain their behavior and/or intentions, which allows their sub rosa self-gratification to escape attention, but they omit crucial aspects of situations, or are otherwise inexact.  93 coping intraceptive processes  Tolerance of ambiguity: Subjects are able to cope with cognitive and affective  complexity or dissonance. Subjects are capable of qualified judgment; they are able to think in terms of grays rather than blacks and whites.... They tolerate inevitably complex negative and positive feelings toward others. Empathy: Subjects sensitively put themselves in the other person's boots; they take the  other's role; they are able to imagine how the other person feels and thinks. In their interpersonal relationships they take account of others' feelings and ideas. Regression in the service of the ego: Subjects utilize feelings and ideas that are not  directly ordered or required by the practical immediate elements of the situation to add to their understanding of problems, their handling of situations, and their enjoyment of life.  defending intraceptive processes  Doubt: Subjects are unable to resolve ambiguity. They doubt the validity of their own  perceptions or judgments, are unable to make up their mind, and are unable to commit themselves to a course of action or presentation of incidents. Projection: Subjects attribute an objectionable tendency to another person, or persons,  instead of recognizing it as part of themselves. Regression: Subjects resort to evasive, wistful, demanding, dependent, ingratiating, non-  age appropriate behavior to avoid responsibility, aggression, and unpleasant demands from others and self.  94 coping attention-focusing process  Concentration: Subjects are able to set aside disturbing or attractive feelings or thoughts in order to concentrate on the task at hand.  defending attention-focusing process  Denial: Subjects deny present or past facts and feelings that would be painful to acknowledge and focus instead on the benign or pleasant.  95 coping affective processes  Sublimation: Subjects find alternate channels and means, which are self-satisfying, socially accepted, and tempered for the expression of affect which can sometimes be basically "primitive." Substitution: Subjects express tempered, domesticated feelings. Suppression: Subjects' infeasible and inappropriate feelings and affective responses are held in abeyance and controlled until the proper time and place and with the proper object. At the same time, affect can be expressed when it is appropriate.  defending affective processes  Displacement: Subjects temporarily and unsuccessfully attempt to control unacceptable affects or impulses in relation to their original objects or situations, and then expresses them in a situation of greater internal or external tolerance. Reaction formation: Subjects appear to have transformed their impulses and affects into their opposites, with resulting alteration of behavior which may, nevertheless, occasionally break down so that the original impulse is in evidence. Repression: Subjects unconsciously and purposefully forget. They have gaps in recall of the past and just can't remember or elaborate. Their constriction in thinking are due to a naive, oblivious, unthinking attitude. Note. adapted from Haan (1977)  96 APPENDIX C PROTOCOL TO THE REAL-LIFE DILEMMA INTERVIEW  1. The stories we've been discussing are examples of moral conflicts. It's quite normal for people to have conflicts or disagreements with others - it's part of everyday living. I'm wondering if you are presently experiencing, or have recently experienced, a moral conflict with your [mother/father/friend/child]? A moral conflict is a situation where you have to make a decision about what is right or wrong, and you're not sure what to do. Could you please describe the most recent moral conflict you experienced with your ... that was not resolved to the satisfaction of both of you? 2. What was the conflict for you in that situation? 3. What was it about this conflict that made it a problem of right and wrong? 4. What was it about this conflict that made it so important to you? (What was at stake for you in this conflict?) 5. What was your point of view, your ....'s point of view, and how were these points of view in conflict? 6. What were you feeling or experiencing at the time? What was your .... feeling? 7. In deciding what to do, what did you consider, what factors were involved? Why are those considerations important? How did you weigh those considerations?  97 8. What options do you have in this situation and what do you think about each? Which seem right? Why? Which seem wrong? Why? 9. Considering your position and your ....'s position, which points in each are right and which are wrong? Why? 10. What do you think is the most important thing to be concerned about in this relationship? Why is that important? 11. What action did you consider or are you considering? 12. How do you think this conflict should be resolved? 13. Do you think your actions were actually right or wrong? How do you know? 14. What do you think should be the right thing to do? Why?  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items