University of Southern California Dissertations and Theses

An Experimental Study Of Two Moral Responses Of Two Socioeconomic Groups Based On Four Paradigms

(USC Thesis Other)

An Experimental Study Of Two Moral Responses Of Two Socioeconomic Groups Based On Four Paradigms

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 69-4541 PRICE, Edette Barnett, 1931- AN EXPERIMENTAL STUDY OF TWO MORAL RESPONSES OF TWO SOCIOECONOMIC GROUPS BASED ON FOUR PARADIGMS. University of Southern California, Ph.D., 1968 Education, psychology University Microfilms, Inc., Ann Arbor, Michigan AN EXPERIMENTAL STUDY OF TWO MORAL RESPONSES OF TWO SOCIOECONOMIC GROUPS BASED ON FOUR PARADIGMS fay Edette Barnett Price A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Educational Psychology) August 1968 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOO L UNIVERSITY PARK LOS ANGELES. CALIFO RNIA 9 0 0 0 7 This thesis, written by .............. under the direction of hex....Thesis Committee, and approved by all its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillment of the requirements for the degree of J ? 9 . 9 . 19 . ? . . . 9 . £ . . ? . U ? - . 9 . sophy D ate AHS^st,..1968 THESIS COMMITTEE nLn *ju&r TABLE OF CONTENTS Page LIST OF TABLES.................................... iv Chapter I. INTRODUCTION............................... 1 Statement of the Problem Significance of the Study Background of the Study Hypotheses Assumptions Limitations and Delimitations Definitions of Terms Used Organization of the Remainder of the Study II. REVIEW OF LITERATURE....................... 15 III. THE RESEARCH DESIGN.............. 38 Subjects Materials Paradigm Procedure Method of Analysis IV. RESULTS AND DISCUSSION..................... 61 Moral Judgments The Hypotheses The Experiment Social Classes IX Chapter Page Control Supplementary Findings V. SUMMARY, CONCLUSIONS, AND IMPLICATIONS . . . 115 Summary Conclusions Implications for Further Research BIBLIOGRAPHY...................................... 123 APPENDIXES........................................ 130 LIST OF TABLES Table Page 1. Socioeconomic Characteristics of the Families of the Students Tested........ 40 2. Agreement Between Two Judges for the Categories of Moral Responses.......... 60 3. Self-Criticism Responses for the Test Trial Game...................................... 63 4. Self-Criticism Responses for Test Stories . 64 5. Self-Criticism Responses for Follow-up Stories................................. 66 6. Reparation Responses for Test Trial Game . . 68 7. Reparation Responses for Test Stories . . . 69 8. Reparation Responses for Follow-up Stories 70 9. Subjects' Responses According to Social C l a s s ................................... 97 10. Story Stability Analysis for Test Stories Versus Follow-up Stories ................. Ill iv CHAPTER I INTRODUCTION Today it is widely accepted that the child is not born with the ability to use moral judgment but that he learns to use it (Sherif, 1948). Thus he is neither moral or immoral at birth but he grows into moral or immoral behavior as a result of environmental conditioning. The question which still remains unanswered despite research interest by psychologists, psychiatrists, anthropologists, sociologists, pediatricians, and educators is the nature of the process whereby the child learns or acquires his moral judgment. Statement of the Problem Before it is possible to state that moral judg ment is or is not being used, it is necessary to find ways of measuring responses to test situations designed to differentiate between moral and amoral or immoral 1 actions or beliefs. This study undertook to evaluate two categories of moral responses— self-criticism and repara tion— under controlled conditions. Four paradigms were incorporated into the experimental design to determine their influence on the subjects' responses: high cogni tion cuing with examiner control of reward, low cognition cuing with examiner control of reward, high cognition cuing with subject control of reward, and low cognition cuing with subject control of reward. Inherent in the experimental conditions was the factor of socioeconomic status, and for this reason one-half of the subjects were drawn from the upper middle class and one-half from the lower class according to the "Index of Status Character istics" of Werner, Meeker, and Eells (1960). In order to measure the moral responses of se lected subjects before and after a guided conditioning of moral judgmental ability, the experimental design con sisted of two different types of reactions developed in three test parts. First, the subjects' responses to an activity which might be judged as "careless," and which involved the breaking of a doll as the result of this carelessness, were recorded. Second, the subjects were told a story, asked to evaluate an act of carelessness, 3 and these responses were also recorded. The third part, nearly identical with the second part, occurred five days later in order to measure the moral judgments of the subjects after a brief lapse of time. The experimental design is presented in detail in Chapter III. Significance of the Study The immediate concern of this study was to see if competent observers could agree about moral responses of children. Aronfreed (1961), Miller and Swanson (1960), and their collaborator (Allinsmith [p. 157]) scored two types of responses to story stems involving stealing and cheating. These studies had interjudge reliability rang ing from 84 to 97 per cent. If morals are man-made, then can they be manipu lated? If they can be manipulated, then can they be brought out of the "mystique" in which they dwell? That moral judgments can be manipulated and con trolled in an experimental setting was an assumption on which this research was based. Similar research has been conducted by Jones (1924), Estes (1941), Skinner (1941), Brown (1960), Baer (1960), Goodstein (1965), and Skinner (1966), among others. Despite this previous research, it would appear that the acquisition of moral judgment abil ity is a subject which needs much additional clarification. This is currently being undertaken in many fields. One example is the psychoanalytical study of morals on a global basis; another is the behaviorists' study of observable activity. In order to achieve specificity in the measurement of moral judgments, this research selected two moral responses: self-criticism and reparation. It then selected variables which were controlled during the experiment. These are explained in Chapter III which details the research design. If moral concepts could be brought more into the realm of education, they could be dealt with as a process of learning. Certainly, since analysis of behavior is evolving as a science of behavior, if moral judgments can be manipulated this opens many avenues for the educator. Background of the Study The process by which a child acquires his moral judgments and values is, as a recent review of this sub ject pointed out (Pittel and Mendelsohn, 1966), still rather obscure. Contemporary conceptions of moral development tend to attribute a wide array of relevant 5 phenomena to a unity psychological structure established through a single process of acquisition. This assumption of an underlying unity in the forms and sources of moral behavior may obscure some important differences between specific responses and their pertinent antecedents. As Pressey and Kuhlen (1957) pointed out: There are not separate groups of saints and sinners. Most people are sometimes honest, sometimes not, sometimes helpful, sometimes not, average in virtue as in other traits, (p. 277) In the teaching of moral concepts, adults often tell the child what is "wrong" and what he should not do rather than what he should do. As a result, children are more certain of what they ought not to do in a situation where a moral decision must be made than of how they can react properly. As Durkin pointed out, "the path of goodness is left highly undefined for the child" (1959). Despite this form of negative instruction, it has been found that a relationship does exist between the develop ment of a child's moral judgment and his mental, social, emotional, and cultural development (Ugurel-Semin, 1952). Based on this type of finding, this research undertook to measure the relationship between a child's self-critical or reparative responses in moral judgment situations and the amount of external adult guidance and/or supervision at the time the judgments were made. In addition, it sought to inculcate in the child a sense of what action he should take to coincide with his moral judgment, thus helping him to learn to perceive positive activities to support his judgment. < Hypotheses In order to delineate the areas of a child's self-critical and reparative moral judgment responses, eight hypotheses are proposed: 1. Subjects exposed to high cognitive cuing will give significantly more self-critical re sponses than subjects exposed to low cogni tive cuing. 2. Subjects exposed to high cognitive cuing will give significantly more reparative responses than subjects exposed to low cognitive cuing. 3. Subjects exposed to examiner control of pun ishment will give significantly more self- critical responses than subjects exposed to self-control of punishment. 4. Subjects exposed to examiner control of punishment will give significantly more repar ative responses than subjects exposed to self- control of punishment. 5. Lower-class subjects will give significantly more self-critical responses than middle-class subjects. 6. Lower-class subjects will give significantly more reparation responses than middle-class subjects. 7. Lower-class subjects exposed to examiner con trol of punishment will give more self- critical responses than middle-class subjects exposed to examiner control of punishment. 8. Lower-class subjects exposed to examiner con trol of punishment will give more reparation responses than middle-class subjects exposed to examiner control of punishment. Assumptions The wealth of contemporary research which has been reported in the literature on children's moral devel opment forms the background of this experimental design. It would have been helpful if consensus had been reached by the various researchers in this field. Since this is not the case, the writer has formulated a couple of as sumptions based in part on past research findings, which are presented in order to clarify some of the aspects of the research being reported: 1. Moral responses can be measured, and these responses are indicators of moral judgment. 2. Moral judgment is not a unitary psycholog ical structure established through a single process of acquisition. Limitations and Delimitations This study was limited to the effects of manipu lative techniques upon the behavior of the subjects enrolled in five schools in sixth grade classes located in the Los Angeles City School District. These subjects were composed of 120 Caucasian girls, all of whom com pleted the experimental treatment. The girls were between the ages of ten years-seven months and eleven years-four months, with average or above average intelligence quo tients (90 to 117). They came from upper middle-class and lower-class backgrounds. The study was limited to behavior outside the classroom with a strange examiner for about one twenty- minute session, and for a one-half hour session with other members of the experiment five days after the first session. Certain restraints need to be recognized in interpreting the results of this experiment. While there was no punishment on the test trial or the follow-up stories, it would be hard to say that the moral responses observed showed internalization in the sense of being independent of the original negative reinforcements on which they were based. The reason is that the examiner had to be present in order to observe the subjects' re sponses. Further, the effect of the punishment itself, being compounded of disapproval and deprivation, is dif ficult to evaluate precisely. A certain amount of manip ulative prodding was present during the first two parts of the experiment, which in itself is not measurable with any degree of accuracy, and since it was to some extent contingent upon the responses given by the subjects it can not stand as a criterion. Finally, the treatments were not entirely divorced from the subjects' previous socialization and generalization of transfer may have entered into their inducement of moral responses. This 10 experiment was also limited in that no provision for follow-up study, beyond that of the five-day period be tween treatments, was made. Definitions of Terms Used Moral judgment.— The development of moral judg- ment consists of learning moral concepts, or the princi ples of right and wrong, in an abstract, verbal form (Jones, 1952; Mueller, 1953). As used herein, moral judgments of the subjects refer to their self-critical or reparative responses in the experimental situations. Self-critical response.— As used in this study, self-criticism is the use of a verbal referrent to one's behavior, thoughts, or feelings. It has been posited that self-critical tendencies are related to the extent of internalization of parental sanctions (Henry and Short, 1954), but as used here self-criticism is intended to indicate a response wherein an individual takes the role of another with respect to himself. It is a response in which an individual takes action with respect to him self as the object of action. Reparative response.— Reparation is the use of a 11 verbal expression to indicate that am individual feels he can correct or ameliorate the effects of a transgression by his own efforts. Although a reparative response may simply reduce unpleasant feelings without necessarily implying self-evaluation, it is constructive and resti- tutive, and therefore implies some kind of manipulation of one's own behavior as well as that of the external environment. Cognitive structure.— Although the usual use of cognition is in the sense of knowing or perceiving, the term cognitive structure is used here to indicate instruc tion given by the examiner to the subject. Two levels of cognitive structure were planned by the researcher in an attempt to differentiate the responses of the subjects. In the high cognition or high cuing paradigms, the exam iner verbalized extensively the explicit standards to be used in reference to the acts on which the subjects were expected to pass moral judgment. In the low cognition or low cuing paradigms, verbal instructions were minimal. Thus, cuing or the term cognitive structure, as used herein, refers to the amount of the examiner's verbaliza tion before and during the various parts of the research. 12 Control.— Inherent in the design of the experi ment was the reward or punishment factor related to the subjects' responses. This factor was incorporated to try to condition the subject to respond with good moral judg ment to either the actions involved in the first part of the experiment or the verbalization in the second part of the experiment. Control was one of the variables, and it had two levels: examiner control and subject control. Under examiner control, the amount of deprivation of reward— or punishment— was at the discretion of the exam iner. Under subject control, the child being tested determined the amount of deprivation of reward— or pun ishment— the action or the response merited. Reward.— English and English (1958) stated that reward is a "satisfaction-yielding stimulus or stimulus object that is obtained upon the successful performance of a task." In this experiment the reward for successful performance (correct responses) was measured by the amount of candy that was not taken away (see Chapter III). In other words, reward as used in this study is the absence of deprivation. ..................... 13 Organization of the Remainder of the Study This chapter has presented a statement of the problem and discussed significance to increased under standing of behavior. The hypotheses to be tested and the assumptions on which the research rests were stated. Since some of the terms used have implications not usu ally associated with those words, definitions were given. Chapter II reviews some of the more pertinent research. In view of the massive amount of investigation in the fields of behavior, learning, and development of moral judgment, it was necessary to select for presenta tion only those studies which appeared to have the great est amount of relevance to this research. Therefore, extreme selectivity was exercised. In Chapter III the research design is described. Included in this design were the criteria for the selec tion of the subjects, a description of the materials used, and a statement of the experimental variables. The pro cedure followed in each of the three parts of the research is then outlined and a brief description given of the method used to analyze the results of the experiment. Chapter IV presents the findings of the research ........14 from the point of view of (1) the moral judgment responses given by the subjects, (2) the hypotheses stated in this chapter, (3) the experimental design, (4) the two socio economic classes, and (5) the area of control of punish ment. Certain additional findings appeared in analyzing the results. Although these had not been anticipated, they are reported since they may be of interest to other researchers. Many of the elements of the research itself were omitted in the body of this report since they were too detailed to be of general interest. However, after the conclusions, and implications for further study, have been presented in Chapter V, these elements are included in the Appendix in order that this report may be complete. CHAPTER II REVIEW OF LITERATURE Current studies of moral development show strong influence from three theoretical traditions: behaviorism, psychoanalysis, and the cognitive theories of Piaget. Of these, behaviorism was the first to make a major impact through inspiring the monumental Character Education Inquiry (Hartshorne and May, 1928; Hartshorne, May, and Moller, 1929; Hartshorne, May, and Shuttleworth, 1930). The major conclusion of these researchers, in part pre ordained by the highly specific nature of their theoret ical concepts and of the tasks and tests they employed, is that moral qualities, such as honesty and deceit, "represent not general ideals but specific habits learned in relation to specific situations which have made one or the other response successful" (Vol. Ill, p. 372). The first finding leading to this conclusion is that of the low predictability of cheating in one situation 15 16 to cheating in another. Correlations between cheating in one type of situation and in others ranged from .00 to .45 (Hartshorne and May, 1928; Row, 1964). Service, self-control, and sympathy have been found to be equally specific to the particular situation (Hartshorne and May, 1930; Murphy, 1937). A second research finding is that children are not divisible into two groups: "cheaters" and "honest children." Children's cheating scores were distributed in bell-curve fashion around an average score of moderate cheating. A third finding is the importance of the expediency aspect of the decision to cheat; that is, the tendency to cheat depends upon the degree of risk of detection and the effort required to cheat. The non cheaters appeared to be primarily more cautious them more honest. Other aspects of the immediate situation were peer group approval and example. Some classrooms showed a high tendency to cheat, while other, seemingly identi cally-composed, classrooms in the same school showed little tendency to cheat. These research findings on situational varia tion suggest that moral conduct is in large part the result of an individual decision in a specific moral conflict situation. This aspect of moral conduct tends ........ 17 to be ignored by common views of moral character as a set of "general good habits" or as a "strong conscience." To some extent, however, Hartshorne and May's conclusion that moral conduct is specific to the situa tion must be qualified because the result of a recent factor analysis by Burton (1963) of the Hartshorne and May data indicated a small general factor in the various experimental tests of classroom cheating, while most of the variation in cheating seems to be due to reactions to the nature of the individual situation. Eysenck's (1953) further analyses of Hartshorne and May's data led him to conclude that there was an underlying trait of honesty which a person brings with him to a resistance to temptation situation. However, Eysenck's results strongly agree with Hartshorne and May's rejection of an "all or none" formulation regarding a person's character. The chief importance of this pioneering research was the precedent it set for studying moral character not merely through verbal response but through observation of concrete behaviors. This precedent was followed a few years later by MacKinnon (1933, 1938), who compared the personality characteristics of college studients yielding ■ ' 18 or not yielding to temptation of cheating. His concep tual framework, however, rejected Martshprne and May's theory of specificity and turned instead to structural and developmental hypotheses from psychoanalysis. With these as a guide, MacKinnon observed a number of consist ent relationships reflecting dynamic processes. In the light of his data, MacKinnon concluded that, in refuta tion of these conclusions of Hartshorne and May, "The findings of the present investigation lend support to a theory of the consistency of personality" (1938, p. 500). Murray (1938), as part of his attempt to define and study systematically the personality of normal indi viduals, developed three sets of questionnaire items designed to tap the dimensions of superego integration, superego conflict, and the sentiments of superego. Superego integration was defined as a "condition in which the dictates of 'conscience' have been so far accepted by the Ego that the subject wills the obligatory (the socially demanded action)" (p. 147), and was measured by items such as: “I have developed a good deal of self con trol," "I am seldom tempted to do anything wrong," "I carry a strict conscience about with me wherever 1 go," and "I prohibit myself the enjoyment of certain unprofitable pleasures." Superego conflict, the opposite of superego integration, represented the "condition of conflict in which asocial impulses are 'at war with conscience'" (p. 147). This was measured by items such as: "I often ask myself 'Have I done right,'" "I am apt to lower my eyes when someone looks me square in the face," and "Sometimes 1 feel— after 1 have done something- that I have not done it correctly and that I must repeat it to satisfy myself." Finally, the items designed to tap the sentiments of superego are aphorisms like: "The moral man watches diligently over his secret thought," "To starve is a small matter; to lose one's virtue is a great one," "It is better to be faithful than famous," and "Virtue is merely a struggle wherein we overcome our weaknesses." While Murray did not spell out exactly what val ues were considered as a part of the superego or con science, it is clear that implicitly they are the tradi tional values of conventional morality. However, the scale items themselves were manifestly concerned with impulse control, guilt, and self-esteem rather than with specific moral values. In an attempt to relate these measures to behavior, Christenson (1938) observed subjects in the Murray assessment program in experimental tempta tion situations involving cheating. He found that a com posite honest rating correlated + .55 with superego inte gration. Thus, there is some evidence that self-reported impulse control is associated with resistance to tempta tion, although this does not imply that moral values determine this behavior. This finding is particularly noteworthy considering previous unsuccessful efforts to predict morally relevant behavior from moral attitudes and knowledge alone. Similar methods which dealt individually with separate aspects with moral attitudes, moral behavior, and conflict resulting from disparity between attitudes and behavior were introduced by Beller (1945) and by Friedenberg and Havighurst (1948). Beller distinguished between two attitude components, "an interiorized norm or internal symbolic reaction, and a predisposition to the overt act"(Beller, 1949, p. 137). He asked subjects to respond to items pertaining to honesty by stating how they would behave in the situation described and how right it would be for them to act in such a manner. Friedenberg and Havighurst (1948) adapted Beller's technique to a more comprehensive study of moral values in a number of content areas. An inventory consisting of 115 acts divided into eleven categories of moral behavior was administered twice, first with instructions, then in turn in terms of "How would you feel?" In the latter administration, subjects were asked to anticipate their feelings if they were to commit each of the acts pre sented. Although the consistency of the subjects and of the derived score for strength of conscience were high, Friedenberg and Havighurst did not accept it as a measure of conscience. Instead, they argued for its use as a projective device from which more global personality dimensions might be inferred. Havighurst and Taba (1949) used "moral character" in its common sense meaning as the sum total of a set of virtue, which included honesty, loyalty, responsibility, and moral courage. The instru ments used in this investigation were similar to those previously mentioned. These investigators, like their predecessors, found no significant correlations between moral beliefs and attitudes and behavioral measures of morality or resistance to temptation. The problem of the prediction of morally relevant behavior has been approached by other means. Gough (1948), using items derived from a role-theory approach, devised a scale which successfully discriminated between subjects who consistently demonstrated high moral behavior and those who do not demonstrate high moral behavior. Gough and Peterson (1952) and Gough (I960), using this scale, were able to determine, even within the delinquent or criminal range of behavior, that it discriminated among various subgroups defined by severity of offense and recidivism. Thus, although this scale does not use any items having any direct bearing on conventional moral standards, the problem of prediction of at least certain forms of moral behavior is not impossible of solution. Wright (1942) designed an experiment deciding how to share a supply of nuts with a friend momentarily absent from the room. Responses were classified into three categories: selfish (giving the other less than one's self), equalitarian (sharing equally), and generous (giving more to the other). The major findings were: (1) selfishness tended to decrease with age; (2) it reached its maximum around eight years of age; and (3) fluctuated thereafter around more equalitarian decisions. Marked differences appeared among the three social class levels (poor, rich, and middle class). The poor children were least selfish and high on generosity; the rich tended to 23 to be generous rather than equalitarian; middle-class children were least generous and most selfish. Ugurel Semin (1958) also carried out a content analysis of the comments made by the children in the course of the experiment and related these comments both to the child's age and the particular decision he made. The analysis led the author to the following general con clusion: The process of moral thought is characterized by five different tendencies whose common trait is found in the change from centralization to decentralization. Moral thought moves (a) from external consideration of the moral understand ing, (b) from being linked to the present moment toward consideration of life as a whole, (c) from consideration.of a specific connection to the linking up of various connections, (d) from an individual and personal consideration of the moral action toward reciprocity and cooperation, and (e) from unilateral consideration of the moral rule toward its mutual understanding. (pp. 471-472) The similarities between Ugurel-Semin's concep tions and those of Piaget are even more pronounced than the final parenthesis suggests, for like Piaget, Ugurel- Semin interpreted her data primarily in cognitive, evaluative, and maturational terms, without regard for emotional, motivational, or social psychological implica tions. Noteworthy in this regard was the failure to ' " ' 24 consider the significance for the process of moral devel opment of the observed differences in response associated with socioeconomic status and. family size. Whiting and Child (1953) undertook to investigate the antecedents of guilt, the variable being defined operationally as the presence or absence in a society of "patient responsibility for illness." Although not all of their hypotheses were confirmed, they did find support for their prediction that guilt (as measured by patient responsibility) was more likely to occur in societies employing severe techniques of socialization, such as strict and frequent punishment or enforcing of rapid changes in habit patterns. Encouraged by these results, Whiting and his students (Faigin, 1952; Hollenberg, 1952; Chusdi, 1954) undertook a second study of internalization of moral values with the focus not only on the culture as a whole but on individual children within three contrasting cul tures in the Southwest: a Mormon community, a Texas town, and a Zuni pueblo. The general method devised by these investigators for measuring internalization of moral values is of some interest, inasmuch as variations of the same technique have since been adopted by most workers 25 engaged in the study of moral development. The method is but one step removed by Piaget's procedures, since it requires the child to complete a story in which the "hero" has committed some form of moral transgression. Contrasted with the Piagetian form of what the hero "should do," the child is free to deal with the implied moral conflict in whatever way he wishes. In line with theoretical prediction, the Mormon children showed the highest levels of internalization and the Zuni the low est. The major hypothesis of the study, however, had to do with the relation between the degree of internaliza tion and the type of discipline used with the child. The authors postulated that the degree of internalization varies positively with the extent to which denial of love is used as the principal technique of discipline for the child. This hypothesis was confirmed not only across the three cultures (with Mormon parents placing heaviest reliance on this method) but also within each of the three cultures (Chasdi, 1954). The projective technique most extensively used has been some variant of the story-completion approach. Allinsmith and Greening (1955) asked subjects to supply an ending to a story stem involving the violation of social prohibitions against transgressions. The severity of "moral needs,1 1 i.e., internalized moral standards or superego, were inferred from the intensity of guilt responses manifested in the story endings. Although their primary concern was with a generalized dimension of superego strength, Allinsmith and Greening concluded that such a dimension could not be properly inferred from guilt responses to stories involving only one content area. In a previous study, these authors had found that subjects who showed intense guilt in some areas, e.g. stealing, did not show similar responses to other content also involving the violation of moral prohibition, e.g. disobedience to parents. Research findings suggest that guilt (transgres sion) responses fall into three main categories or func tional types (Allinsmith, 1960; Bandura and Walters, 1959; Kohlberg, 1963). One type includes punishment, fear, and hiding— emotions which represent anxiety about punishment rather than guilt. Since the stories explic itly rule out the possibility of detection, such anxiety is unrealistic. These tendencies to express "moral anxiety" in terms of fantasies of harm to a transgressor also do not seem to serve a directly moral function, and 27 do not correlate with resistance to temptation, or moral ity of judgment (Badura and Walters, 1959; Burton, 1958; Kohlberg, 1963). A second type of response is the direct expres sion of self-blame and self-judgment. This response does seem to play a direct moral function, since it is corre lated with experimental resistance to temptation (Brinder and McMichael, 1963; MacKinnon, 1938), resistance to temp tation (Bandura and Walters, 1959; McCord and McCord, 1956), and development of moral judgment (Johnson, 1963). A third type of response is best represented by confession. This response seems to represent something in between the internal moral reaction of self-criticism sind the externally focused anxiety represented by a pre occupation with punishment. In some contexts, confession is correlated with resistance to temptation (Grinder, 1962; Rebelsky, Allinsmith and Grinder, 1963), in some context, it is not (Burton, Maccoby, and Allinsmith, 1961; Rau, 1964). Confession may be an expression pri marily of concern about the judgment of others rather than an act based on self-judgment, but it does represent a tendency to be upset by imagined disapproval of others even though one could get away with it (Rebelsky, 1963). ■ ' 28 Punishment by love withdrawal, ignoring mother's statement that she doesn't like the child when he is bad and so on, has been thought to be especially critical in producing guilt (Sears, Maccoby & Levin, 1957; Burton, Maccoby and Allinsmith, 1961); however, it has not been found to relate to self-critical responses (Aronfreed, 1961; Hoffman, 1963). The following quotation from Hoffman states the general conclusion: The use of psychological discipline (which includes techniques that appeal to the child's needs for affection and self-esteem and his concern for others) especially in the context of an affectionate parent-child relationship appears to foster the development of an inter nalized, moral orientation, especially with respect to one's reactions following the viola tion of a moral standard. (p. 26) With regard to guilt, the findings are ambigu ous. Kibbutz children display as much confessional guilt on story completion tests as family reared rural controls (Luria, Goldwasser, and Goldwasser, 1963). The kibbutz children, however, did not describe as intense guilt feelings as did the family-reared city children (Robin and Goldman, 1963). Since neither study used a measure with much claim to validity as an index of internal guilt, and since the control groups involved differed (urban versus rural), it is difficult to draw conclusions 29 from these studies. They do suggest that kibbutz children are not grossly lacking in guilt reactions or a "super ego," but that they react less intensely to some common transgressions. This, of course, may be a matter of group ideology rather than a quality of the parent-child relationship. The theory which has most directly inspired recent research on moral judgment has been that of the Swiss psychologist, Piaget (1932). His account of moral devel opment springs largely from his general theory of the development of the child's conception of the world (Piaget, 1928). Piaget presented a series of brief sto ries, each centering on a moral issue, to more than one hundred Swiss children and on the basis of their responses to his semistructured questions, distinguished two major stages of moral development. The first stage, which he called "moral realism," is based on "an ethic of authority." The child views moral rules and restraints as laid down from above; they must be interpreted literally and can not be altered. In accordance with the principle of "imminent justice," punishment follows inevitably upon violation, and its severity varies directly with the enormity of the consequences of action regardless of the motive which inspired it. In contrast, the more mature stage of "reciproc ity" or "cooperation" is characterized by the "ethics of mutual respect.” Punishment, instead of being generalized, is specific to the infraction, aimed at reciprocity in kind or restitution, and is guided by the principle of "equity" involving consideration of the motives underly ing the act and of the particular circumstances in which the transgression was committed. An impressive number of studies over a quarter of a century in both Europe and America have reported age differences consistent with Piaget's postulated shift from moral realism toward reciprocity and equity. Thus, evidence in support of one or another aspect of Piaget's theory is reported in studies from England (Harrower, 1934; Morris, 1943); Peel, 1959); Switzerland (Lerner, 1937); Belgium (Carueo, 1943); Italy (Ponzo, 1956); and the United States (Lerner, 1937; Dennis, 1943; Lui, 1950; MacRae, 1954; Havighurst and Neugarten, 1955; Medinnus, 1957, 1959; Kohlberg, 1958, 1959; Durkin, 1959; and Boehm and Ness, 1961). Two of the foregoing research ers (Dennis and Havighurst and Neugarten) were carried out with Indian children, and one with Chinese-American (Lui). On the basis of the review of virtually all of these studies, Boehm and Ness concluded that "age is the only consistently operative factor in the development toward maturity" (p. 10). At the same time, they call attention to "strong trends toward sociocultural influ ences ." The closest correspondence with Piaget's original results is found in studies of children from continental Europe (Lerner, 1937; Caruso, 1943; Ponzo, 1956). The farther one moves from the European mainland in distance and culture, the more frequent are the departures from Piaget's findings. One of the first attempts to verify Piaget's theory turned up with highly significant differ ences across both social class and culture. Harrower (1935), an English psychologist, sought to determine whether similar results would be obtained with children of the same age range in another country and from differ ent social class levels. The results from the lower socioeconomic groups were consistent with Piaget's the ory; however, a significant difference emerged at the higher social level. The author concluded with the fol lowing : 32 Either the stages of development which Piaget has been emphasizing are not a universal characteristic of development per se, but are to be found only within certain uniform groups, groups which are subject to certain constant conditions; or, in certain environments these stages can be so far accelerated that children exhibiting characteristics of the . . . most developed [stage] are to be found at the ages of the first, (p. 93) Confirmatory evidence for Harrower's findings and interpretation comes from a series of studies of class differences in moral development conducted in the United States (Lerner, 1937; MacRae, 1943; Kohlberg, 1959; Durkin, 1959; Boehm and Noss, 1961), and even more force fully from comparisons of cross-cultural data (Dannis, 1943; Lui, 1950; Havighurst and Neugarten, 1955; Boehm, 1957) . The most pronounced class differences are reported in the first American study carried out thirty years ago by Lerner who found middle-class children showed an earlier decrease of moral realism than working-class youngsters. Almost two decades later, lower but still reliable relationships in the same directed were cited by MacRae and Kohlberg. Boehm, in a recent unpublished study, showed no significant differences by class, but a trend in the predicted direction for specific stories. 33 This decrease in class differences is consistent with differences in child-rearing over a twenty-five year period. A second major line of evidence calling into question central assumptions in Piaget's theory of moral development appears in the work of MacRae (1954), Kohl berg (1958), Durkin (1959), and Maccoby (1959). Each of these investigators note from their data that the presumed components of morality as described by Piaget are not empirically consistent. The first and one of the most comprehensive study of this phenomenon is by MacRae, who did a cluster analysis of responses to moral judgment questions holding age constant and identified four rela tively independent factors. MacRae suggests further that the types of questions used by Piaget are concerned pri marily with cognitive emphasis is the pervasive use of "should" and "right" in the questions employed in Piaget's method. It was not until the 1950's that researchers shifted from Piaget's type of question to one designed to get at what the child might wish to do rather than merely what he thought was the appropriate answer. The Kohlberg studies stated that the same stages of judgment were found in the new type responses. ................................................... 34 Although Piaget conducted dozens of experiments on moral standards, presumably with children of both sexes, in only one instance does he pay systematic attention to the differences in the responses of boys and girls. For example, in answer to the question of what a smaller boy should do when struck by a bigger boy, the "tendency to consider it legitimate to give back the blows received" tended to increase more rapidly with age for boys than for girls (Piaget, 1932, pp. 301-302) . Since then only three studies have dealt with possible sex differences in moral responses. Boehm (1961, p. 10) found such a dif ference; Medinnus (1957) "found girls to be less advanced than boys in respect to the concept of imminent justice and punishment," and Morris (1958) "found no significant differences between boys and girls in age of decreasing moral realism but found a tendency for the values of girls to change earlier than those of boys." Thus, the find ings do not seem to be consistent. Nevertheless, Boehm and Noss (1961) describe the absence of sex differences in their own research as a contradiction to a trend "heretofore evident in previous studies, that girls show a more advanced development than do boys on an overall basis" (p. 10) . 35 Some of the conclusions that may be drawn from the work of Piaget and the studies which have tried to duplicate his work are: 1. Piaget unjustifiably minimizes the influence of the environment on the child's understanding of what is just (MacRae, 1954; Kohlberg, 1964; Johnson, 1963; Havighurst and Neugarten, 1955). Specific cultural fac tors appear to stimulate or retard age trends of develop ment. 2. The range of variation in moral development by age, class, and culture is so great as to call into question the dominant role of maturational factors, at least beyond the age of five. Aronfreed (1961), after reviewing many of the studies, concludes, "The recent work of Kohlberg (1958) and Peel (1959) tends to confirm, in a very general way, that the child's moral conceptions are age related" (p. 7). He continues: But it is noteworthy that the studies, when taken together with others in which age differ ences were not found (Boehm, 1957; Durkin, 1959; Lerner, 1937; MacRae, 1954), suggest as a group that the child's cognitive resources for moral judgment are more closely related with social states and other indicators of cultural expec tations. These findings are interesting because they imply that the validity of highly differen tiated cognitive-evaluative standards is not a .......... 36 necessary accompaniment of advancing age or experience. Apparently, the role of cognitive equipment in the child's moral behavior is dependent on the kinds of experience provided by its environment, (pp. 7-8) It would appear that under appropriate environ mental conditions, children as young as six years of age can learn equally well responses characteristic of Piaget's moral realism or the patterns reciprocation presumably indicative of moral maturity. This is consistent with Carmichael's (1939) finding that six-year old children are fully capable of recognizing misdeeds, forecasting the future, and learning a wide variety of adaptive re sponses, including avoidance, prevarication, confession, and restitution. The various manifestations of moral development posited by Piaget, while they may be corre lated in a particular cultural context are most appropri ately regarded as separate variables representing responses to specific social-situational influences. Since the literature relating to the assessment of moral values is extensive, it was necessary to limit consideration to those studies most relevant to this study. It is somewhat of a sobering thought on the state of our knowledge in the behavioral sciences that even the most reliable facts and theories about specific 37 phenomena, self-criticism, and reparation, can be seri ously challenged by asking a simple descriptive question: how does this phenomenon vary with sex, age, and social background of the person? As late as the 1940's, the notion underlying much of American child psychology was a normal maturational sequence that could be expected of all children everywhere. Although the theory of bad blood has been replaced, a definitive study of the acquisition moral specifics is lacking. The problem seems to be the result of an insufficient effort to conceptualize the nature of moral values and their relation to behavior. No one seems to be asking the question of how to combine a research design to investigate hypotheses regarding the interaction of the cognitive, affective, and behavioral dimensions of moral orientation. The fact that the child's maturational level may not be as deter mining an influence in moral development as was once believed does not, in any way, detract from the scientific importance that human morality is man-made rather than an inevitable product of organismic evolution. CHAPTER III THE RESEARCH DESIGN The purpose of this chapter is to present a description of the subjects and the material used, as well as a description of the variables, and to outline the procedures and the method of analysis used in this inves tigation. Subjects This study was conducted in three lower-class elementary schools in San Pedro and two upper middle- class elementary schools, one in Palos Verdes and the other in Universal City. Each of these schools had an enrollment of between 1100 and 1200 pupils and all schools belong to the Los Angeles City School District and were contacted by the District's administrative offices. These five schools were selected because the pupils in them most closely resembled a target population described to 38 the District by this researcher. All subjects were enrolled in the sixth grade and the District mandated that all girls in a class which was used be included in the testing. The cumulative folder for each of the girls was made available to this research er and these were consulted to establish the personal data of the subjects and some of the socioeconomic cri teria of the subjects' families. To supplement the material in the folders, a let ter was sent to the families of all -the subjects. From these two sources— the folders and the responses to the letters— comparable statistics were compiled for the two socioeconomic classes, which are summarized in Table 1. In the letter asking for the family information, permis sion was also requested to use the girls in the experi ment. Because of the nature of the schools and the com munities selected, many of these pupils are used fre quently for research projects, and both the students and the families are accustomed to this type of experimental testing. Perhaps for this reason, in no case was permis sion for the use of the child in the experiment denied. Of the 163 subjects in the five selected class rooms, a final test group of 120 subjects was desired. 40 TABLE 1 SOCIOECONOMIC CHARACTERISTICS OF THE FAMILIES OF THE STUDENTS TESTED Characteristic Upper Middle Class Lower Class Occupation of father Type of house Area lived in Amount of income Source of income Professional men, medical doctors, dentists, engi neers, account ants, business men, self-employed Single dwelling, 87% owned Skilled and unskilled laborers, service connected jobs, unemployed Single and multiple dwellings, 22% owned Residential Mixed business and district of homes, residential no multiple dwellings $15,000-$40,000 Earned (salary) $3,000-$5,450 Earned (salary) and welfare Amount of education Father Mother 13-17 yrs. school 8-13 yrs. school 10-18 yrs. school 4-12 yrs. school Place of birth Subjects native born Subjects native bora .........................................................41 equally divided between the upper middle-class and lower- class groups. The following reasons guided the discard ing of 43 subjects: 5 students were over the age group selected 9 students did not fall within the I.Q. range 20 students were found to be in a wrong grade placement 4 students were non-Caucasians 5 students did not fall within the reading achievement range Of the 120 girls selected, all of the subjects were native-born Caucasians currently in Grade 6. All fell within an age bracket of 10 years-7 months to 11 years- 5 months. According to the intelligence quotient as measured on the California Test of Mental Maturity, which had been administered during October, 1966, the subjects' scores ranged from 90 to 115, with a mean of 104. Read ing achievement grade placement scores, obtained in April, 1967, on the California Achievement Test, ranged from 5.9 to J.0.4, with a mean of 7.1. Of the 120 girls selected, thirty were from each of the two upper middle- class schools and twenty from each of the three lower- class schools, making sixty subjects in each 42 classification. Materials The materials used in this research were quite simple. The first part of the experiment consisted of a task on the part of the subject and for this a rectangular piece of composition board was used, approximately two by three and one-half feet in size (see Figure 1). This board rested on a small table and the subject was seated at one end of the board. Twenty-four small-zized plastic girl dolls were clustered thickly in a triangular forma tion at the far end of the board from where the subject was seated. Behind the base of the formation of plastic dolls, at the very edge of the board, stood an antique china doll about eight inches in height, dressed as an adult woman. A large cloth-padded cardboard box rested on a chair just about the far edge of the board and behind the doll. The box was below the board's surface so that the interior was not visible to the subject when she was seated. For the experiment, the child was given a small hoe, called a "pusher," which was approximately two feet long. In addition to the china doll on the table, the ......... 44 examiner had a second doll which was identical except that its head had broken off. This was concealed in the cloth-padded box in such a way that the child could not see it from where she sat. The "reward" element in the experiment was repre sented by thirty small pieces of individually-wrapped sugarless candy bars. The original experimental design had called for the rewards to be in the form of nickles, but it was learned that parental permission was needed if money was to be used, so the candy, which did not require permission, was substituted. The candy bars were neatly piled in front of the subject and within easy reach was the box into which the candy to be given up and taken away during the*experiment would be placed. The other materials needed for the research consisted of the stories to be used in the second and third parts of the research. In addition score sheets were provided on which the examiner could record the subject's responses. Samples of these materials may be found in the Appendix. Paradigm Four of the variables were used in a paradigm "...................................45.. including high or low cognitive cuing and self or examiner control of punishment, all of which could vary independently of each other. These may be described as follows. High cognitive cuing— self- control of punishment This treatment maximized cognitive structure through verbalization of an explicit standard in refer ence to the aggressive acts, and also gave the subject control over her own punishment. She was able to control the quantity of the reward she would relinquish after each trial of the game. High cognitive cuing— examiner control of punishment This treatment also maximized the cognitive struc ture through verbalization, but it gave the control over the punishment to the examiner. Under this paradigm the examiner decided on the quantity of the reward to be relinquished after each trial of the game. low cognitive cuing— self- control of punishment This treatment minimized the cognitive 46 instructions. Instructions were given only at the begin ning of the experiment and were not repeated between the trials. To support the minimal nature of the instructions, the examiner offered no comment on the "carelessness" or "roughness" of the subject's game manners as had been done in the high cognitive paradigm. The subject had control joyer her own punishment and could determine the quantity of the reward she would relinquish after each trial of the game. Low cognitive cuing— examiner control of punishment This treatment also minimized the cognitive instructions. The instructions were given by the exam iner only at the beginning of the experiment and were not repeated between each trial. Also the examiner did not comment on the "carelessness" or "roughness" in this treatment. As in the second variation under the high cognitive cuing treatment, the examiner determined the quantity of the reward to be relinquished as a penalty after each trial of the game. Another independent variable was designed to be i incorporated into the paradigms. This was the socio- ......47 economic class of the subjects. The two dependent vari ables were the moral judgment responses of the subjects. These responses were recorded by the examiner on score sheets and had to do with either self-critical responses or reparative responses by the subjects. Procedure The total testing to determine whether moral judgments can be measured experimentally had three separ ate and self-contained phases or parts. The first part involved the game-task utilizing one of the four para digms for each subject. At the conclusion of this part of the procedure, the examiner recorded the subject's responses for further analysis. The second part involved the responses the subjects made to each of ten stories read by the examiner. In the third part, additional stories were read and the subject's responses were re corded again. Each of these parts is described so as to clarify the activities involved. Part I Each subject was taken individually from her classroom to the experimental room by the examiner. This person was a female Caucasian, holding a California Pupil Personnel Credential, with a psychometrist designation, who had been trained in the use of the tests for the experiment but was not informed about the purpose of the experiment. After the examiner and the subject were seated at the table containing the board and the dolls, the exam iner resorted to one of the four paradigmatic procedures described above. The task in which the subject was re quired to engage was arranged so that some degree of so-called "carelessness" was inevitable. In other words, the dolls were arranged in such a tight triangle that, to get the pusher through, some dolls had to be knocked over. The examiner told a "let's pretend" story to the subject, explaining that the board represented a school playground. The subject with the pusher was to pretend that she had an urgent message for the china doll, who was the teacher. All of the plastic dolls were other pupils. Although the subject wanted to be courteous, in order for her to get to the teacher she would have to move her pusher through the cluster of pupil-dolls. When she finally got through, she could then gently shove the teacher-china doll into the soft box to indicate that she 49 had reached the teacher and delivered the message. The verbal instructions which were then given to the subject by the examiner varied with the paradigm being used. The instructions included here were for the high cognitive structure— self-control of punishment paradigm, but the other sets of instructions (see Appen dix) varied only in degree. I have something here for you to do. To do this work, you have to be very careful and gen tle. Being careful and gentle is the most impor tant thing. Here's the way to do it. Back here is an object which you must push back into this box [experimenter indicates box by lifting it up and showing it to subject]. Here's how it goes into the box. You push it off the board with this pusher. You can't lift the pusher; you must keep it down on the board when you push. The experimenter demonstrates and then hands the pusher to the subject. Of course, to push it off, you will have to push through the dolls, and so you will knock some of them over, even though you're trying to be careful and gentle. The idea is to see how many of the dolls you will knock down. If you knock down just a few, that's good. If you knock down a lot of them, that's not so good. When you use the pusher, try to be as careful and gentle as you can. Speaking in this same extremely explicit fashion, the experimenter went on to explain that there were thirty .........................................................50 pieces of candy in a pile placed in front of the subject and after each trial the child was to decide "how care less and rough" she had been and penalize herself accord ingly by giving up one, two, or three pieces of the candy and placing them in the empty box. If she had considered herself to be extremely "careless and rough," she would complete the ten trials with none of the original candy. Thus the subject was permitted to evaluate the amount of punishment she deserved for her own actions. In the second paradigm, that of high cognitive structure with the examiner determining the amount of punishment, the instructions to the subject were equally explicit. For the two paradigms where much less verbaliza tion was contributed by the examiner, the instructions were clear but were not reiterated nor was there any encouragement or caution expressed by the examiner during the trial. After the child had completed one trial, and either the subject or the examiner had assessed the pen alty for the "carelessness" of knocking down some dolls, the examiner simply said, "All right, go ahead." At the conclusion of the ten trials, the eleventh trial for all of the four paradigms was planned as an unexpected disruption caused by the apparent breaking of .......................... 51 t the large doll which, in reality, was made so that it would come apart. When the subject pushed the doll off the table, the experimenter looked into the box, assumed surprise, and remarked, "Oh my— it's broken 1" She then held up the broken doll and continued. "And we don't have another doll here to use. I wonder why it broke.” Up to this point, the trials had been designed to encourage the child to blame herself for the number of plastic dolls knocked down, and to believe that her care lessness, roughness, or impatience had caused the dolls to fall. With this kind of conditioning, the questions which the examiner now posed as a result of the "acci dent" were designed to elicit either self-critical or reparative responses from the subject. Although the first question was planned for self-criticism, if the child responded to the examiner's observation with any kind of remark which fit either category, the examiner noted the reply on the score sheet (see Appendix) and this part of the research was terminated. If the girl made no moral judgment of her actions, the examiner asked, "Why do you think it broke?" Regardless of the answer given, the examiner continued: "Well, now that it's broken, I wonder what we should do." This statement was intended primarily ......... ■ 52 : to elicit a reparative response, if the subject offered any suggestion that was relative to the question, the test trial was terminated. If the subject gave no re sponse or gave one not relevant to the statement, the experimenter posed the final question: "What do you think we should do now? Regardless of the subject's response to this, the test trial was terminated. Part II Immediately following the task with the dolls, the examiner went on with the second part of the experi ment, which included the reading of ten stories to each subject. The stories were simple incidents which might happen in the life of any of the girls. None of the stories showed that the subject deliberately committed an act of aggression, but they were designed to reproduce the kinds of situations around which self-critical and reparative responses might be elicited. In one case, for example, a child is playing up in a tree when she is called to come down and on the way down she breaks a bird feeder which was in the tree. In another, a sweater which a child has borrowed is lost. In none of the stories is the child caught or accused but the questions tended to evoke 53 an empathic response from the subject. The objective involved in using this method is to minimize the child's awareness that her own responses have importance. At the same timef it permits a precise structuring under which responses can be readily classi fied and objectively tabulated. The advantages and the validity of such a technique have been demonstrated in a number of previous studies (Hakanson and Gordon, 1958; Kagan, 1956; Purcell, 1958; Allinsmith and Aronfreed, 1960 and 1963). The instructions by the examiner to the subject varied to correspond with the requirement of high or low cognitive cuing. Before reading any of the stories, the examiner explained the procedure to the subject, and the following explanation is presented as an example of the instructions in the low cognitive cuing paradigm. The questions that I am going to ask you at the end of each story have no right or wrong an swers. The reason that I sun asking these ques tions is that I am interested in what young people think, because if we know more about what you think, we might be able to understand each other better. I will read the story to you and then, when I finish, I will ask you some ques tions. Do you understand? One of the stories used in this part of the experiment is reproduced here to illustrate the type of 54 stories told to the subjects: The teacher in your classroom has set up committees to prepare book reports. Your job is to see that no one takes home any of the books. (Me day after the committees have met to decide on what book to report on next, you are in a hurry to leave the room and do not check the books carefully. One of the students takes home a book and loses it. She tells the teacher about losing the book. At the end of each story the examiner asked the subject several questions. To avoid a wide variety of responses, highly structured questions were asked. As in Part I, candy was established to reward good answers by not having to relinquish any of it. Also as in the first part, the first question was designed to evoke a self-critical response, the second question either self criticism or an offer of reparation, and the third ques tion primarily reparation. In examiner control of pun ishment, she suggested the number of pieces of candy the child should give up as a result of her evaluation of the child's answers. In subject control of punishment, the examiner told the child to evaluate her own answers by giving up some of the candy. Again, since thirty pieces of candy were involved, the child penalized her self with from one to three pieces for each of the ten stories. ........................... 55 For the above story, the following questions were then asked: a. Why do you think the book was lost? b. Now that the book is lost, what do you think you should do? c. What do you think you would do? For all ten of the test stories, the examiner recorded the subject's answers verbatim on a recording sheet (see sample in the Appendix). After the subject was finished with the questions following the stories, she was asked not to discuss the experiment with the other children. Subsequently, the examiner checked with each classroom teacher and with the noon aids to see how well the security had been maintained. She reported that, as far as could be ascertained, the girls had not dis cussed the experiment with their classmates. Part III The examiner returned to each school five days after the administration of the first two parts of the experiment. In each of the three lower-class schools, twenty girls had taken part in the experiment and five had been assigned to each paradigm. Thus, for Part III, arrangements were made so that all five children in each paradigm were available to participate simultaneously. The fact that there were no absences among the subjects was extremely fortuitous for the experiment. Thirty girls had taken part in each of the two upper middle- class schools, so that either seven or eight girls had been assigned to each paradigm. For Part III, in order to keep the participating group small and intimate, no more than five students took part at any one time so that the size of the groups varied from two to five girls. Stories were again used, but this time no reward/ punishment variable was included. The children were told that the examiner was interested in group reaction this time instead of individual reaction, and was interested in written rather than oral responses. The children were also told that their written answers were on paper which did not contain their names so that no one would know how any individual answered. The girls were taken together into the speech room in each school. After the examiner and the subjects were seated, the girls were given a test sheet containing stories. Before beginning the reading of the stories, the examiner explained again to the subjects that there .......57 were no right or wrong answers but, as in Part II, the answers would be regarded as revealing what young people think. The instructions by the examiner to the different groups continued to vary, of course, to correspond to the requirements of high and low cognitive cuing. Following the explanation, each child read each of the ten stories to herself as the examiner read them aloud. Each subject then wrote a response to the ques tions which, this time, were printed on the form given to the child. These questions followed the same pattern as the questions in the test stories. When all of the stories had been completed, and the children had been given sufficient time to answer the questions, the answer sheets (see Appendix for the form) were gathered up by the examiner and the children were dismissed. Again, they were asked not to discuss the stories with any of their classmates and a check was made with the teachers and the noon aides to discover whether it appeared that any of the girls had talked about the experiment. So far as could be learned, there had been no communication about the stories. Method of Analysis Before the results could be subjected to statis tical analysis, it was necessary to categorize the re sponses given by the subjects. The two judges who undertook to evaluate the responses of the subjects were two doctoral students in the field of educational psychol ogy at the University of Southern California. The cri teria for judgments of the responses were explained in detail by the writer, each classification of moral responses being evaluated as self-criticism or repara tion. The two judges worked independently of each other to determine the categorization of the responses based on the following instructions. Self-criticism was to be judged as being present in the subject's response if her remark referred to her behavior as being the cause of the transgression. For example, any response in the test trial which referred to the fact that she had not pushed the doll the right way was judged to be self-criticism. Likewise, in the responses to the stories, if she said she was responsible for the lost book in the sample story given above, this was accepted as being self-critical. 59 Reparation was judged to be present if the sub ject indicated in her response that the effect of her transgression could be corrected or ameliorated through her own resources for constructive action. These re sponses usually referred to repair in the test trial, with answers such as "I can glue her head back on" being considered as a reparative answer. In the story example given for Part II, a response such as "I will take my allowance and help pay for the book," would be included in the reparation category. After all of the responses had been evaluated by the two judges, analysis of the data was done in accord ance with the procedure of a 2 x 2 x 2 factorial chi square design by Snedecor (1956) for testing the differ ences in the frequencies between groups. Percentages of agreement were computed over the total number of instances of classification in each of the two categories. These percentages of agreement, summarized in Table 2, indicate that the classification system had a high degree of objectivity. TABLE 2 AGREEMENT BETWEEN TWO JUDGES FOR THE CATEGORIES OF MORAL RESPONSES Moral Response Percentage of Agreement Self-criticism 98 Reparation 97 CHAPTER IV RESULTS AND DISCUSSION In reporting the findings of this research, the first step consists of an analysis of the judges' evalu ations of the moral judgments given during each of the three parts of the experiment; these were divided into the two categories of self-criticism and reparation. The responses were then tabulated and the results subjected to a chi square analysis. A discussion of the three parts follows based on examiner and researcher observa tion and analysis of the experiment and the recorded responses of the subjects. The findings that relate to the two social classes are then presented. The various aspects of all of these variables are discussed, finally, in relation to each of the hypotheses stated in Chapter I. Following the determination of the acceptance or rejection of the hypotheses, the stability of the response pattern is discussed. 61 62 Moral Judgments Self-criticism responses Before discussing the findings, it should be pointed out that although the total number of subjects participating was 120, the total number of responses is represented by different figures. With some subjects, both self-criticism and reparation were recorded and with some other subjects no reply could be considered either self-critical or reparative. Total number of subjects, therefore, is not a consideration in this part of the analysis of the results of this research. Table 3 reports the self-criticism responses for the test trial game. Examination of the table indicates that there appeared to be little difference in the self- critical replies regardless of the paradigm under which the subject was working. As shown in Table 4, it would appear that the examiner's control induced more self-critical responses than did the subject's control. The cognitive level of the instructions appeared to have less influence on the tendency toward self-criticism than did the control of the reward. It is quite apparent, also, that social TABLE 3 SELF-CRITICISM RESPONSES FOR TEST TRIAL GAME Social Class raraaigin Upper-Middle Lower xotai Low cognitive cuing- examiner control 10 12 22 Low cognitive cuing- subject control 11 9 20 High cognitive cuing- examiner control 9 11 20 High cognitive cuing- subject control 9 8 17 Totals 39 40 79 Source df X2 A. Cuing B. Control C. Social of punishment class 1 1 1 0.32 0.32 0.01 A X B 1 0.32 A X C 1 0.01 B X C 1 0.62 A X B X C 1 0.01 64 TABLE 4 SELF-CRITICISM RESPONSES FOR TEST STORIES Paradigm Social Class Upper-Middle Lower Total Low cognitive cuing- examiner control 86 119 205 Low cognitive cuing- subject control 84 89 173 High cognitive cuing- examiner control 91 126 217 High cognitive cuing- subject control 76 104 180 Totals 337 438 775 Source df id A. Cuing 1 0.47 B. Control of punishment 1 6.14* C. Social Class 1 13.16* A X B 1 0.03 A X C 1 0.81 B X C 1 1.58 A X B X C 1 0.57 65 class was important in the self-critical responses, with 337 upper middle-class group students offering a response that could be considered self-critical compared with 438 lower-class group students. This finding will be dis cussed further in the discussion of the two classes and also under Hypothesis 5. Table 5 reports the responses for the four para digmatic groups which corresponded to the cuing-control factor. Although no reward/punishment variable was incor porated into the experimental design, this fector appears because the four original groupings of the subject have been maintained throughout the reporting of"the data. Inspection of the table shows a notable increase in the self-criticism responses of the upper middle social class group, from 337 in Table 4 to 394 in Table 5. It is also interesting that the upper middle-class students in the low cognitive cuing paradigm increased their self-critical responses just under fifty per cent on the follow-up stories. Reparative responses As with the self-critical responses, the repara tive responses have been treated in their relationship to 66 TABLE 5 SELF-CRITICISM RESPONSES FOR FOLLOW-UP STORIES Social Class Paradigm Total Upper-Middle Lower 124 131 255 113 89 202 81 121 202 76 93 169 394 434 828 Source df X2 A. Cuing 1 8.95* B. Control of punishment 1 8.9.2* C. Social class 1 1.93 A X B 1 0.48 A X C 1 6.61* B X C 1 6.61* Low cognitive cuing- examiner control Low cognitive cuing- subject control High cognitive cuing- examiner control High cognitive cuing- subject control Totals A X B X C 1 0.08 the four paradigms and to the two social classes. These responses are presented in Tables 6, 7, and 8. Inspec tion of these tables indicates that the lower-class stu dents in general were more inclined to offer some form of reparation for their carelessness or roughness than were the upper middle-class students. While the ratio was 4 to 3.5 in the test trial game (see Table 6), for the lower class compared to the upper middle-class students, this ratio increased slightly in each of the other parts of the experiment. For the test stories it was 4.2 to 3.3, and for the follow-up stories it was 4.3 to 3.4. Another interesting comparison is the fact that reparative responses were higher for the groups with examiner control than subject control of punishment on both the test stories and, especially, on the follow-up stories. This ratio on the follow-up stories was 4.5 to 3.3, whereas the two control groups had been about even in the scores for the test trial game. This might tend to indicate that the examiner's approval or disapproval \ of the reparative responses as determined by the amount of punishment had significance to the pupils. As a result, this evaluative conditioning contributed to the 68 TABLE 6 REPARATION RESPONSES FOR TEST TRIAL GAME Paradigm Social Class Total Upper-Middle Lower Low cognitive cuing- examiner control 9 10 19 Low cognitive cuing- subject control 8 12 20 High cognitive cuing- examiner control 9 9 18 High cognitive cuing- subject control 9 9 18 Totals 35 40 75 Source df d A. Cuing 1 0.12 B. Control of punishment 1 0.01 C. Social class 1 0.33 A X B 1 0.01 A X C 1 0.33 B X C 1 0.12 A X B X C 1 0.12 69 TABLE 7 SEPARATION RESPONSES FOR TEST STORIES Paradigm Social Class Upper-Middle Lower Total Low cognitive cuing- examiner control Low cognitive cuing- subject control High cognitive cuing- examiner control High cognitive cuing- subject control Totals 92 89 72 76 132 101 122 69 224 190 194 145 329 424 753 Source df X* A. Cuing 1 7.47* B. Control of punishment 1 9.13* C. Social class 1 12.00* A X B 1 ' 0.30 A X C 1 0.01 B X C 1 9.60* A X B X C 1 1.12 70 TABLE 8 REPARATION RESPONSES FOR FOLLOW-UP STORIES Paradigm Social Class Total Upper-Middle Lower Low cognitive cuing- examiner control 102 139 241 Low cognitive cuing- subject control 80 94 174 High cognitive cuing- examiner control 83 121 204 High cognitive cuing- subject control 78 77 155 Totals 343 431 774 Source df X2 A. Cuing B. Control of punishment C. Social class 1 1 1 4.05* 17.38* 10.00* A X B 1 0.42 A X C 1 0.25 B X C 1 4.97* A X B X C 1 0.33 71' responses of these pupils on the final testing. Visual inspection of the six tables reporting the self-criticism and reparative responses thus has per mitted some rather interesting generalizations. The true value of these tables to this research, however, lies in their application to the hypotheses. The Hypotheses In discussing the eight hypotheses stated in Chapter I and reporting the findings relating to each, it should be noted that all of the pertinent data are incor porated into the six tables found under the subjects* responses presented above: Tables 3 through 8. Hypothesis 1 Subjects exposed to high cognitive cuing will give significantly more self-critical responses than subjects exposed to low cogni tive cuing. This hypothesis was tested by comparing the number of self-critical responses made by subjects exposed to high cognitive cuing with those exposed to the low cognitive cuing paradigm in three situations: test trial game, test ........... 72 stories, and follow-up stories. Table 3 shows that there was no significant difference between treatments for the trial game. This could be due to lack of reliability of the test since it gave the subjects only one chance to give a self-critical response. Although similar designs were used by Aronfreed (1963, 1965) and Allinsmith (1963), in which significant results between treatments were obtained, their subjects were younger than those used in this research and the story used for their game allowed the examiner to be both nurturant and informative in the treatment. In contrast, this examiner gave the instruc tions to the subjects in both treatments in a matter-of- fact manner. Table 4 shows that there were no significant dif ferences between treatments in the number of self-criti cal responses for the test stories. However, Table 5 shows that cuing did affect the number of self-critical responses given by the subjects in the follow-up stories, but not in the direction predicted. Subjects exposed to low cognitive cuing gave significantly more self-critical responses than did those exposed to high cognitive cuing. By way of summarizing the self-critical responses, it would have to be stated that for the test trial game, ....73 subjects exposed to high cognitive cuing gave fewer self- critical responses than subjects exposed to low cognitive cuing, but not significantly so. For the test stories, subjects exposed to high cognitive cuing gave more self- critical responses than subjects exposed to low cognitive cuing, but the difference was not significant. For the follow-up stories, the difference in the number of self- critical responses was significantly in favor of the subjects exposed to low cognitive cuing compared to those exposed to high cognitive cuing. Thus Hypothesis 1 is rejected. Hypothesis 2 Subjects exposed to high cognitive cuing will give significantly more reparative responses than subjects exposed to low cognitive cuing. This hypothesis was tested by comparing the number of reparation responses made by subjects exposed to high cognitive cuing with those in the low cognitive cuing paradigm, in three test situations: test trial game, test stories, and follow-up stories. Tables 6, 7, and 8 pre sent the analysis in tabular form. Table 6 shows that there was no significant 74 difference between treatments for the test trial game. Data for the test stories, presented in Table 7, show that a chi square of 7.47 was significant at the .05 level, but not in the expected direction. The subjects exposed to the low cuing treatment gave significantly more reparation responses than the subjects exposed to the high cuing treatment. Although the examiner cued the subjects in the high cognitive treatment after each story, she did so in a bland, matter-of-fact voice, not trying to be nurturant. In cuing, she may have emphasized what was "wrong" and what the subject should not do, rather than suggesting a constructive response, which would have provided a positive feedback to the subject rather than a negative one. The data in Table 8 shows that there was also a significant difference in the reparation responses of the subjects in the low cognitive cuing paradigm and the high cognitive cuing paradigm, but again it was not in the direction expected. In summary, while the test trial game indicated that Hypothesis 2 is tenable, the significance of the finding in the opposite direction to that hypothesized for both the test stories and the follow-up stories would indicate that Hypothesis 2 should be rejected. «T Hypothesis 3 Subjects exposed to examiner control of punishment will give significantly more self- critical responses than subjects exposed to self- control of punishment. This hypothesis was tested by comparing the number of self-critical responses given by the subjects who were in the paradigm of depriving themselves of a reward after each test trial game with those in the paradigm of being deprived by the examiner of a reward after each test trial game. Results were again analyzed by means of the 2x2x2 chi square statistic, and are reported in Tables 3, 4, and 5. Table 3's data support the hypothesis but at a level which fails to achieve significance. The data from Table 4 strongly support the hypoth esis in that the locus of control— subject or examiner— did significantly affect the number of self-critical responses. Thus, more self-criticism responses were given by the subjects whose punishment was controlled by the examiner than by subjects who controlled their own ................... "" 76 punishment. Although there was no concrete punishment for the test stories, punishment was represented by the deprivation of reward in the test trial games and the test stories situation. The data in Table 5 support the research hypoth esis for the follow-up stories. It should be remembered that there was no punishment or reward situation in the follow-up stories, but the scores made by the subjects were recorded to maintain the original paradigmatic grouping of students. Thus, subjects who had been under examiner control of punishment in the first two parts of the experiment gave significantly more self-criticism responses than those exposed to self-control. Since Tables 4 and 5 support Hypothesis 3 at a level which achieves significance at the .05 level of confidence, and Table 3 supports it even though not at a significant level, this study accepts Hypothesis 3. Hypothesis 4 Subjects exposed to examiner control of punishment will give significantly more repara tive responses than subjects exposed to self- control of punishment. ............... ‘ ....... 77 This hypothesis was tested by comparing the number of reparative responses given by the students who deprived themselves of a reward after each test trial game with those who were deprived by the examiner of a reward after each test trial game. As seen in Table 6, the reparation responses for the two examiner-control paradigms totaled 37 and for the two subject-control paradigms, 38. Although this indi cates a very slight difference between the two groups, the difference that does exist is in the direction not expected. Table 7 presents data which significantly support the research hypothesis that control of punishment did affect the number of reparation responses given by the subjects in the direction predicted. The data in Table 8 also support the research hypothesis for the follow-up stories at a significant level. Thus, Hypothesis 4 is accepted for Parts II and III at the .05 level of significance. Whether or not examiner control of punishment— which might, in effect, condition the subjects in the desired direction of moral judgment— was cumulative was not specifically a part of the experiment. The following 78 recapitulation of the reparation responses of the subjects in the three parts of the experiment isr therefore, presented as being of interest but is not interpreted as having validity: Examiner control Subject control Test trial game 37 38 Test stories 418 335 Follow-up stories 445 329 Hypothesis 5 Lower-class subjects will give significantly more self-critical responses than middle-class subjects. It should be pointed out first that the socioeconomic classification of the students involved in this research was, in both groups, at the upper level of the classes indicated, according to the "Index of Status Character istics" of Warner, Meeker, and Eells (1960). Character istics of the lower-class subjects as described in Chap ter III would indicate that this particular group of students would fall well up toward the top of this cate gory. The characteristics of the group referred to as "middle-class subjects" in the hypothesis would indicate 79 that a more accurate designation for this group would be "upper middle class," which is the nomenclature the group has been given in the tables. As a matter of fact, according to the group divisions of Warner, Meeker, and Eells, this particular group of students would probably fall toward the extreme upper limits of the upper middle- class group. The data in Tables 3, 4, and 5 indicate that in all three parts of the experiment, lower-class subjects did indeed give more self-criticism responses than did the upper middle-class subjects. In the test trial game the difference was so slight that it did not have signif icance. In both the test stories and the follow-up stories, however, the number of self-critical responses given by the lower-class students exceeded by a sizable amount the number of self-critical responses given by the upper middle-class students. As a matter of fact, for the test stories the chi square difference was 13.16, which was significant well above the .05 level. Hypoth esis 5 is therefore suppprted by the data. Previous studies which carried out comparisons within both social class and peer group have shown that children involved in the more extensive social 80 participation within their school activities and/or responsibilities within their home, on the whole, were judged to be more mature in their moral behavior (Boehm and Nass, 1962; Brink, 1963; Kohlberg, 1964). The subjects in this study were selected from two schools whose popu lation was rated 61 on the "Index of Status Characteris tics" of Warner, Meeker, and Eells (1960), and from three schools rated 36 on that scale. The subjects were equated on C.A. and I.Q> However, the mean I.Q. for the sample in this study from these schools was 104. The mean I.Q. for the lower class schools was 90, 88, and 93, and the mean I.Q. for this sample from these schools was 104. Included in the subjects tested from both schools were: Middle Lower 2 play leaders 14 play leaders 0 members of the safety 17 members of the safety committee committee 0 student government 25 student government members members 3 class presidents 1 class president Therefore, if social participation or responsibilities are valid predictors; of mature judgment, then the lower- class subjects in this study must be regarded as lower- 81 ; class subjects on a rating scale but as the "elite” group within their specific subculture. The effect of the participation in school activ ities is reflected in some of the responses of the lower- class subjects to the first story: in response to the question, "Why do you think the book was lost?" some of the replies were: "I was careless about my duty," "I neglected my responsibility," "I was lacking about my duty," "I disappointed my teacher by being lazy," or "I was responsible for the lost book because it was my duty to see that no books should be taken from the library." Some of the responses made by the lower-class subjects to several other stories exhibited a rather concrete approach to self-criticism. One story which showed a marked trend was the bird feeder story, in answer to the question, "Why do you think the bird feeder was broken?" Typical lower-class subjects' responses were, "I was clumsy coming down the tree, but Jane's mother should know kids are clumsy when they hurry," "I guess 1 should have looked out for the bird feeder, but her ma should have told me to look out for it." Middle- class subjects' responses, in comparison, made no references to specific warning instructions that they should or might expect from adults. They appeared to accept the blame in a matter-of-fact way and move on to the next story. The theme of those who made self-critical responses was to the effect that the subject was clumsy or was not thinking about anything but getting out of the tree. For those subjects who did not make self-critical responses, the trend revolved around the fact that the broken article itself was to blame: "Jane's mother ought to buy a better feeder," "If you buy cheap things you can't expect them to last,” or "Bird feeders are stupid anyway." Hypothesis 6 Lower-class subjects will give significantly more reparation responses than middle-class subjects. The comments made under Hypothesis 5 about the social classes of the subjects should be applied to this hypoth esis also, to the effect that both of the groupings tested were well toward the upper end of the broad class categori z ation. Tables 6, 7, and 8 present the data on reparation responses which are applicable to this hypothesis. As 83 examination of the tables reveals, in all three parts of the experiment the lower-class subjects did indeed give more reparation responses than middle-class sub jects. In the test trial game, the difference was not significant but in the test stories and in the follow-up stories the difference was significant above the .05 level. Hypothesis 6 is therefore accepted by this study. Examination of the responses of the subjects in both social class groups indicated that most of the stu dents' comments reflected three separate themes. The first was that the subject could manipulate her own behavior. "I can look for the book, and if I can't find it, I can pay for it." "I can replace the book, and if I can't I will have to pay a little at a time." "I can make a new bird feeder." "I can paste the bird feeder back together," or "I can paste the head back on the doll." The second theme was that of dependency on a family member, such as "I will tell my mother and she will buy a book," or "My dad will make a new bird feeder," or "If I tell mother, she will take the ink out of the capes." A third theme indicated the inability to make any 84 constructive reparative response. Most of these were to the effect that the examiner should decide what ought to be done. "If you tell me what to do I will do itr" or "I don't know what to do; why don't you tell me" or simply, "I don't know." Even with examiner prompting, although the prompted answer could not be counted for this study, these subjects seemed unable to think of any action to take. This writer compared each of the test protocols to look for one variable, such as an intact home, I.Q. range, participation in school activities, achievement scores, father's occupation, or educational level, which might shed light on this inability to act. None could be found. Since this third theme was found more often in middle-class subjects than in lower-class subjects, the protocols of the middle-class subjects were inspected separately. The one variable present in every middle- class protocol except two was that all of these subjects' I.Q.'s fell within the lower half of the I.Q. range. Hypothesis 7 Lower-class subjects exposed to examiner con trol of punishment will give more self-critical 85 responses than middle-class subjects exposed to examiner control of punishment. This hypothesis was tested by comparing the number of self-critical responses given by lower-class subjects who were deprived by the examiner of a reward with middle- class subjects who were deprived by the examiner of a reward in the three situations: test trial game, test stories, and follow-up stories. The data are reported in Tables 3, 4, and 5. A visual comparison of the self-criticism re sponses when control of punishment is vested in the exam iner in the three parts of the experiment is shown as follows: Lower Upper Middle Test trial game 23 19 Test stories 235 177 Follow-up stories 252 205 As can be seen readily, there was an increase in the number of self-critical responses for both the lower and upper middle-class groups between the test stories and the follow-up stories. Thus it could be reasoned that the children could have been conditioned by their pre vious exposure to examiner control of their responses in 86 the test stories. When they were retested five days later they were more inclined to be critical of them selves in the incidents incorporated into the stories. The hypothesis is accepted on the basis of the findings, although none of the figures reached the level of significance. Hypothesis 8 Lower-class subjects exposed to examiner control of punishment will give more reparation responses than middle-class subjects exposed to examiner control of punishment. This hypothesis was tested by comparing the number of reparation responses given by lower-class subjects who were deprived by the examiner of a reward with middle- class subjects who were deprived of a reward by the examiner in three situations: test trial game, test stories, and follow-up stories. The data are reported in Tables 6, 7, and 8. This same kind of visual comparison of the repar ation responses of the two groups is presented as was previously given for the self-criticism responses in Hypothesis 7. 87 Lower Upper Middle Test trial game 19 18 Test stories 254 164 Follow-up stories 260 185 These figures show that there was an increase in the number of reparative responses by both the lower and the upper middle-class groups between the test stories and the follow-up stories. Again, as in Hypothesis 7f no attempt was made to interpret the reason for the increase in reparation thinking on the part of the subjects during the five-day lapse of time between the two parts of the experiment. The facts are presented without an attempt at interpretive comment. was found that the difference in the number of reparation responses between the two social groups produced chi squares of 9.60 for the test stories and 4.97 for the follow-up stories, both of which were significant at the .05 level of confidence. Although the difference in the number of responses between the two groups did not reach a level of significance for the test trial games, Hypoth esis 8 is accepted. Subjecting the data to chi square analysis, it As was pointed out in the discussion of the fifth 88 hypothesis, these findings tended to be somewhat baffling in view of some other research which produced results showing that middle-class children are inclined to learn moral standards earlier than lower-class children. The Experiment The material presented in this section has not been subjected to any kind of statistical analysis. It is, rather, a compilation of some of the comments made by the examiner, together with a discussion of various aspects of the experimental design. Part I The reward incentive which was introduced into the experiment may not have been a consistent motivation for all of the subjects. From observations made by the children to the examiner, apparently only five girls were candy-motivated; all of them were from the lower class. Bruner (1960) stated that money was the reward which he had used because it has universality. In view of the restrictions placed by the schools on the use of nickles, and the substitution of candy, the reward/punishment aspect of the control may not have had the importance originally anticipated. The difference in the attitudes of the two social classes toward the dolls used in the experiment was interesting. Whereas the lower-class girls tended to comment that the antique china doll was "pretty" or "unusual," and evidenced a certain admiration of it, the upper middle-class girls tended to take the doll for granted. This difference could have been a precondition ing factor in the varied responses after the doll appar ently broke. The lower-class children were shocked or startled to a far greater extent than were the upper middle-class girls. The examiner reported that some of the lower-class subjects were noticeably stunned by the accident, some of them to a point of being incapable of responding readily with comments either of self-blame or self-criticism or of offering any constructive type of reparative suggestions. Those who were able to make suggestions tended to blame themselves much more harshly than did any of the upper middle-class girls. These latter children appeared to accept the accident in a far more casual manner. A few were somewhat disdainful, comment ing that it must have been a "cheap" doll to have broken so easily or, as one child remarked, "If you buy a cheap ..................... .......... ' " 90 doll, that's what you get!" In this part of the experiment, only five of the 120 subjects failed to give some kind of response to the examiner's questions. Three of these were in the lower social class and two were in the upper middle social class. Part II Since candy was still used as the reward/punish ment incentive, the comments under Part I apply equally to Part II of the experiment, which contained the ten stories. The examiner reported again that she felt that the candy was not meaningful enough to all of the subjects and therefore did not constitute a reinforcer. Motiva tion was apparently provided by the fact that the candy represented a goal. It would have made no difference if the incentive had been pebbles or toothpicks; the subjects who were motivated to be achievers accepted the candy as a measurement of success. The challenge lay in "how many" were left— not the fact that they^were able to keep the candy itself. Of the five lower-class girls who appeared to be candy-motivated in Part I, all gave correct responses in 91 all of the stories. In other words, for these few sub jects, the motivation of retention of the candy appeared to induce them to consider their answers carefully and to accept the blame so that they could accept the reward. An unanticipated result was evident in this part of the experiment. It had been expected that the girls would tend to discuss the stories with the examiner to the extent of offering either self-critical or reparative suggestions. The examiner reported, however, that many of the subjects appeared to have insufficient interest in the outcome of the situations posed in the stories to offer any kind of response at all. This was especially noticeable when there was high cognition cuing and where the amount of instruction given to the subjects seemed to inhibit an answer of any kind. Although not anticipated, this finding is consistent if the subject looked upon the cuing by the examiner as being punitive instead of instructive. For example, between each trial she said, "Be careful,” or "You are being rough; be more careful." Thus the subject could more readily accept the value of the adult examiner who did not criticize the subject after each trial or story (low cuing) them one who did criticize it (high cuing). .. 92 The examiner's comment was that the "kids seemed afraid" to make a suggestion. When prompted, the middle-class girls seemed less inclined than the lower-class girls toward any feeling of self-blame or self-recrimination and therefore did not appear to believe that they were at fault or should have to take any kind of reparative action. It should be pointed out that Story #7 had a good many "no answer" responses. This story had to do with a roast which was to be prepared by the subject and which was not properly cooked. Whether it was the story itself or the questions about it which confused the girls is not known; it is simply a fact that a trend of "no ans wers" was noted for this story, and this in turn has an effect on the overall compilation of responses. Part III The first observation that might be made about follow-up stories is that the five candy-motivated lower- class subjects continued to answer all story questions correctly in their responses. Every one of their obser vations would have to be ranked as being perfect. Another point that might be made is to the effect that the children appeared to be far more willing to respond to the questions about the stories when their answers were written and when only they knew the exact response they were making. Whereas many had hesitated to tell the examiner how they felt about the action deline ated in the original stories— thus creating a good many "no answer" notations on the examiner's report— almost all of the children gave some kind of answer in Part III. While the judges were unable to rank the responses as either self-criticism" or "reparation" in many cases, it should be pointed out that the children did make an at tempt at a response. Without the verbal guidance of the examiner, there were many responses that were not consid ered applicable and were thus not included in the statis tical computations. It was discovered that Story #1 of this series was improperly constructed. The story had to do with a ball that had rolled into the street and been ruined by a car. None of the children seemed to know what they should do, and further examination of the situation re vealed that the youngsters are not held responsible for such an accident. Instead, normally they would report the damage to the school and another ball would be 94 supplied. The responses given to this story, therefore, showed the poor story content rather than the inadequacy of the answers. The achievement motivation of some of the stu dents was indicated by the fact that the groups of sub jects falling into the examiner-control paradigms re sponded more readily and more frequently than did those in the subject-control paradigms. Summary Overall, it would appear that the subjects in the low cognitive cuing paradigms felt more free to respond with ease to the questions. It almost seemed as though the repetition of caution and care on the part of the examiner tended to "freeze" many of the subjects so that their answers were cautious and/or stilted. It would also appear that the children tended to be more punitive than the examiner, as measured by the amount of candy they gave up when they were asked to evaluate their own performances. The fact that some of the stories did not seem to evoke responsiveness on the part of some or many individuals might have been due to the experimental ..........95 design. Although much care had been given to developing situations to which the children could relate, in the full experiment it was found that they were not equally appli cable to the life situations of the different girls. When the pilot study had been conducted, all but one of the stories seemed to elicit comparable interest and this one story had been changed. In the larger population, however, this same amount of interest in the content did not appear to continue. Further comments as they relate to the various parts of the experiment are found in the next section covering the two social classes. Social Classes A number of the findings of interest bear a rela tionship to differences in the two social classes used. Some of these are mentioned in the section on hypoth eses; some, however, while not part of the statistical analysis, are pertinent to the attempt of this research to learn more about the moral judgments of children. It had been presumed that it would be possible to manipulate the responses of some of the subjects. One of the conclusions reached was that it did not appear that 96. examiner cuing influenced the moral judgment responses given by the subjects in either of the two social classes. While the reason for this may have differed slightly between the two classes, as will be described in more detail, this study showed that moral judgments could be somewhat manipulated through a reward/punishment stimu lus but not through the amount of verbalized support which was given. Another presumption, from the research reported by Turiel (1964) and Bronfenbrenner (1964), was that the upper middle-class students would be more cooperative and more articulate. Examination of the report sheets filled in by the examiner, however, led to a decision to check this factor. Table 9, which is a compilation of the total responses reported in Tables 3 through 8, shows that the reverse is true, and that the upper middle-class students gave 45 per cent while the lower-class students gave 55 per cent of the total responses. This finding might have been a result of the fact that it appeared from the children's comments to the examiner that the lower-class girls were more anxious to succeed, to score well, than the upper middle-class girls, which tends to substantiate Bronfenbrenner's (1961) 97 TABLE 9 SUBJECTS' RESPONSES ACCORDING TO SOCIAL CLASS Upper Middle Class Lower Class Responses High Cognitive Cuing Low Cognitive Cuing High Cognitive Cuing Low Cognitive Cuing Self-critical Game 18 21 19 21 Test stories 167 170 230 208 Follow-up stories 157 237 214 220 Total 342 428 463 449 Reparative Game 18 17 18 22 Test stories 148 181 191 233 Follow-up stories 161 182 198 233 Total 327 380 407 488 High cognitive cuing totals 1,538 Low cognitive cuing totals 1,745 Upper middle class totals 1,477 Lower class totals 1,807 earlier findings. He reported that it would appear that the values of the two classes are tending to come to gether. In this research into the moral judgments of the two classes, the responses of the subjects would appear to indicate that the upper middle-class girls had less incentive to criticize themselves, or to blame themselves for the broken doll or for the accidents in Mtfie stories, and therefore less responsibility to offer reparative suggestions than the lower class girls. It must be remembered that the so-named lower class which was in cluded in this study is not to be confused with a "ghetto" class. The subjects in the lower social class were at the upper end of the Werner lower-class scale (1960), and he has pointed out that his findings indicate that mem bers of this class are trying to adopt middle and upper class standards. Another observation of interest had to do with the difference in the attitude of the two social classes. It seemed that many of the girls in the upper middle class, although they took the research tasks seriously, looked at their part in it as a game which they were playing. They seemed glad to have the excuse not to be in their regular classes; they indicated that they were 99 accustomed to taking part in experiments and found them interesting; and at the same time they had only overt interest in the experiment. The lower-class girls, on the other hand, seemed to have been far more interestedly involved. They seemed to have a more friendly attitude toward the examiner than did the upper middle-class girls and chatted freely with her. They asked questions such as "Are you seeing how well we can read?" and "Is this an arithmetic test?" They tended to verbalize their reason ing for determining the number of pieces of candy they gave up, in subject control of punishment, and in general seemed to want to become an integral part of the research rather than mere onlookers. The amount of cognition involved also appeared to be an influence. In the high cognitive cuing paradigms, the upper middle-class girls seemed to listen and to respond without being unduly influenced; but the lower- class girls appeared to be overwhelmed by the responsi bility of being part of the experiment. Those in the high cognitive cuing paradigms lost some of their sense of freedom they had evidenced when they first came into the examination room. On the other hand, in the low cognitive cuing paradigms, the upper middle-class girls 100 talked freely to the examiner but did not appear to be quite as much at ease with her as did the lower-class girls. It could be said that the upper middle-class girls seemed almost to need help to make moral responses. It appeared that they tended not to be able to act on their own to make reparation for their actions. Both in the game situation and in their answers to the questions about the stories, they frequently tended to refer to having someone else help them. If mother would pay for it or father would fix it, then the girls would be ready to make amends. If someone else would do something, they would all have made reparation. This did not appear to be equally true for the lower-class girls. Their responses frequently included another person, especially a parent or a sibling, but they indicated that they needed help in making a reparation. If so-and-so would help "me" do something, then I could do something about the situation, appeared to be their solution. Almost all of the lower- class girls had some kind of reparative suggestion for handling an accident, whereas seventeen of the upper middle-class girls asked, in effect, "Should I invent an answer?" In other words, the lower-class girls felt that loi ' making amends was up to them, but the upper middle-class girls felt that it was not truly their responsibility unless someone else either told them what to do or did it for them. The broken china doll illustrates the above find ing. Most of the lower-class girls wanted to repair the doll or take it home and have a father, mother, or brother put the doll together again. They seemed to be genuinely concerned about their responsibility for breaking the doll. The upper middle-class girls, on the other hand, seemed to shrug their shoulders about the mishap and to assume that someone would repair the doll but that it was not up to them. "That's what happens," remarked one girl, "when you buy cheap and old things." While not expressed too frequently in so many words, this represents the apparent feeling of many of the upper middle-class sub jects. A concomitant aspect of this is that the upper middle-class children stated frequently that "Mother can get another one," or that either mother or father would replace the doll. It apparently did not occur to the lower-class children that the doll could be replaced easily. They had been more impressed with the doll originally than the upper middle-class girls and were more concerned about its breakage. There appeared to be some differences between the responses of the lower and upper middle-class subjects to the different stories. In Story #3, for example, having to do with a friend's broken eye glasses, both groups in the high cognition paradigm appeared to be afraid to answer when the examiner cautioned, "Be sure to be care ful" in considering the response. The upper middle-class children became quite defensive and attempted to justify their actions. The lower-class girls, on the other hand, tended to punish themselves as severely as possible by depriving themselves of the maximum amount of candy for their transgressions. Concerning the accidents that happened in both Stories #5 and #6, the upper middle-class girls again appeared to be defensive and many of them stated that a parent or some other person should have warned them to be careful. One remarked, in response to a question in Story #8, having to do with some misplaced money, "I'm too young to be responsible." Again, it was noted that the middle-class children tended to leave reparation to the parents to make good the accident, while the lower-class children tended to try to figure out what they themselves could do. In Story #5, ......103 I concerning the broken bird feeder, for example, "Mama will buy smother one," exemplifies many responses of the middle-class girls, whereas "My father will help me fix it," was indicative of the responses of the lower-class girls. Bronfenbrenner had noted this tendency and had attributed it to a lack of initiative and a trend toward oversocialization on the part of the middle classes and upper middle classes (1958). Miller and Swanson (1958) also referred to this type of attitude, pointing out that most members of middle-class society have somebody over them who makes final decisions. These investigators studied the differing patterns of behavior exhibited by middle-class families from "bureaucratic" and "entre preneurial" settings. They argue that the entrepreneurial- bureaucratic dichotomy represents a new change in American social structure that cuts across and overrides social class influences and carries with it itw own character istic pattern of family structure and socialization. Another study, this one by Bold and Slater (1958), shows that* in line with Miller and Swanson's hypothesis, par ents from these same two groups tend to favor rather different ends and means of socialization, with entrepreneurial families putting considerable emphasis on the development of independence and mastery. These differences appear at both upper and lower middle-class levels but are less pronounced in the lower middle-class socioeconomic strata (Miller and Swanson, 1960). Over 65 per cent of the fathers in the middle-class sample were employed in large corporations. Thus the subjects in this research in the upper middle-class social group had apparently been conditioned to look to someone else to decide on what action should be taken. The lower-class children, on the other hand, had been conditioned to solve their problems themselves rather than to expect someone else to step in and make such decisions. In Story #6, about a broken toy, the upper middle-class girls tended to feel that they would get someone else to "buy" a replacement for the broken birthday present; the lower-class girls, however, remarked that they would have to work to buy a new one. At the same time, the middle-class children did not apparently place as much value on a "model ship" as did the lower-class children. Buying a new replacement was not a matter of undue concern to them; the lower- class children, on the other hand, seemed to feel that a ....... 105" model ship must be something precious and they were ready to save up their money to buy a new one because they seemed unable even to conceive that it might be repaired. As a result of this somewhat extreme weight of responsi bility, they tended to turn defensive, stating that any thing "that valuable" should not have been left around. Because so many more of the lower-class subjects responded with moral judgment answers than had been expected, particularly in the examiner control of punish ment paradigm, the writer discussed the findings with the principals of the lower-class schools. One principal instantly pointed out the factor that had distorted the results, saying that the group selected happened to "hit the Mormon Bible Blet." She pointed out that the religion of these subjects trained them for internal control and for self-responsibility. With this in mind, the cumula tive folders were again examined. Since religion had not been part of the criteria established for this re search, no exact breakdown is possible, but it would appear that many of the lower-class girls from one of the schools were religiously oriented. As a result of this development, the writer went back over the score sheets of the girls in this school ' " “ "" 106' and found that as a group they had punished themselves much more severely for the number of dolls they had knocked down in the game and for the self-criticism and blame they felt for the accidents in the stories. A count of the amount of candy remaining for the girls in this group revealed that, with examiner control of pun ishment, they were about average. Those in the paradigm of self-control of punishment, it was discovered, in almost every case had deprived themselves of the maximum amount of candy allowed and so they ended the game or the series of stories with none of their candy left for them selves. The difference in the amount of candy left was not a finding for just the students in the "Mormon Bible Belt.1 1 In general it was found that the lower-class sub jects tended to have less candy remaining than the upper- class subjects. In discussing various cultures, Whiting and Child (1953) reported that the middle-class popula tion was more stingy than either the lower or the upper classes, and it could be that the middle class children tended to want to keep as much of their candy as possible. Other Whiting and Child findings may also apply. The upper middle-class girls tended to deprive themselves of 107 one piece o£ candy compared to three pieces for the lower class, pointing out, perhaps, that the lower class is more self-punitive. Thus, the lower class appears to have a higher desire to achieve and yet their self-pun ishment is greater. They appear to be aiming for perfec tion, and when this is missed the punishment is greater. At the same time, when the examiner had control of the amount of candy taken away, answers by both the lower and the upper middle-class girls were given more readily. When there was self-control of punishment, the lower class especially tended to give fewer verbal responses but to inflict more punishment on themselves by depriving them selves of the candy. The findings reported here have been randomly selected to point out some of the nonsignificant results of the research. These findings can not be entirely docu mented nor have they been subjected to statistical anal ysis. They are presented to furnish a report for refer ence by other researchers as to the kinds of results that may be expected. Many of the factors included here con firm research reported previously; others are contrary to previous reports. Where the latter has occured, an attempt has been made to find reasons for the ' ' " " 108 ! i discrepancies. In any event, some of the findings have been incorporated in some of the analyses of the hypoth eses. Control In view of the fact that control was neither critical nor especially important in this research, the only area which is discussed here relates to a comparison of the responses of those in the paradigm of self-control of punishment with those in the examiner-control of pun ishment. It was interesting to see that children working under examiner control gave more total responses than those under self-control, but also that more of them gave no response. It is possible that examiner control af fected the lower-class subjects more than the upper-class girls because the children are more apt to conform. This finding is consistent if the subjects were looking upon the cuing of the examiner as being punitive instead of instructive. She said between each trial, "Be careful," "You are being rough, be more careful." Therefore it would appear that the subject could more readily accept the value of the adult examiner who did not criticize the ......109 the subject after each trial or story than one who did criticize it. Of the middle-class girls who gave no response, when the examiner prodded for a reply, seventeen asked if they should invent an answer. The examiner said, "If you want to." In contrast, those in self-control para digms gave fewer self-critical responses than those in examiner control, but these subjects responded to most of the questions so there were few "no responses." As a matter of fact, in the game phase of the experiment, only five of the total sample gave no response. With subject control of punishment, both social classes tended to give defensive or "buck-passing" re sponses. For example, when the ship was broken in Story #6, several answered that "the person in charge" should have put the ship away. In test Story #7, when the sub ject was late placing a roast in the oven so that it was undercooked, the upper middle-class subjects especially tended to shrug off self-blame or guilt. One answered that her mother needed "an ultrasonic oven to cook the roast in four minutes"; and another that they could "eat hot dogs.” Thus, the upper middle-class girls sensed a need to respond under examiner control of punishment, but ~ 110 many of their responses showed no moral judgment of them selves or any indication that they felt they were to blame for the accident. In general it would appear that the girls who were in the examiner control of punishment paradigms gave some self-critical answers with a certain amount of self justification , but if the examiner prodded for additional comments, seeking a reparative response, many of the girls had no further comments, either because they "froze" or because they felt that they were not responsible and gave a "smart aleck" response. Supplementary Findings The picture of moral responses emerging from the findings of this study do not appear to substantiate the findings of Hartshome and May (1930) concerning preado lescent moral behavior. They found that the most influ ential factors determining resistance temptation are situational factors rather than a fixed individual moral character trait of honesty. As an inspection of Table 10 shows, some groups had nearly a 90 per cent carry over from the test story situation to the follow-up story five days later. Although the test trial game was not included Ill TABLE 10 STORY STABILITY ANALYSIS FOR TEST STORIES VERSUS FOLLOW-UP STORIES Stability Range in Percentagesa Reparation Self-criticism 80-89 1 1 70-79 - 1 60-69 4 4 50-59 - - 40-49 3 - 30-39 - 1 0-29 - 1 Total 8 8 aSupport for the percentages given may be found in the Appendix. because this part of the experiment had not reached statistical significance, the responses followed the same pattern. The analysis of the data from this study sug gests, rather, that they follow the more recent factor analysis by Burton (1963) of the Hartshome and May data, that there is a small general factor in the various exper imental tests. While most of the variation in responses seems to be due to reactions to the individual situations, part is a product of stable individual differences in attitudes toward moral response. Certainly, part of the stability in this study must come from the fact that only two responses— self-criticism and reparation— were taken into account. As shown in the preceding table, all of the groups scored in the 40 per cent or above range in repara tion responses. While six of the eight groups scored in the 60 per cent or above group in the self-critical responses. The conclusion that must be drawn is the rejection of an all-or-none formulation regarding a per son's moral behavior. This is not too different from what Hartshorne and May discovered, although their stabil ity range was much lower. The fact that this study used a well-defined ..... ".. 113'' scoring system to deal with the subjects' responses cer tainly limits the generalizability of the data to a global behavior pattern. The same limitation must be pointed out with regard to the extreme care used to delimit the heterogeneity of the two groups. As nearly as it could be arranged, all of the subjects in the lower-class group had social characteristics which fell into quite narrow bands. The same statement can be made of the subjects in the upper middle-class group. Regardless of the care exercised to bring all subjects in each of the two groups into as nearly comparable social characteristics as pos sible, it must be remembered that no two homes or no two backgrounds of the children can be the same. The data derived from this investigation, therefore, are as spe cific as to categories as the experimental design would permit, but they still can not be projected to include other subjects globally who meet the same limitations of the social characteristics. It has been previously pointed out that only two indices of moral judgment were considered: self-criticism and reparation. These, of course, are learned responses to transgressions but they may also be intervening proc esses to guilt. In any event, they are two components of ..... '.114 the superego and as such are likely to have different antecedents, be differently reinforced during socializa tion especially in different moral content areas, and can be studied separately as well as part of the global superego or conscience. The mere fact of the presence of indices of self-criticism and reparation does not provide an unambiguous measure of moral values, nor was this measure implied in the research design. CHAPTER V SUMMARY/ CONCLUSIONS, AND IMPLICATIONS Summary One of the initial purposes of this study was to learn more about the developing moral judgments of chil dren. Despite continuing research into the how and when and where of the education of children in the area of moral behavior, too little is yet known and confirmed. Thus, this study selected two aspects of morality— self- criticism and reparation— for experiment using a group of sixth grade girls, one-half of whom fell into a socioeco nomic level characterized as lower class and the other one-half into upper middle class. Variables were established so that the subjects were divided into four paradigms including high or low conditive cuing and self or examiner control of punish ment, all of which could vary independently of each other. These four groupings were maintained through the three 115 116 stages of testing. The first part was a test game where attention and care were required and this was followed by an accident to elicit subject reactions. The second part consisted of ten stories of accidents wherein the subjects were asked, first, to consider themselves as having caused or created a negative action, and, second, to make a decision as to the reason for the accident and what action could now be taken. The third part occurred five days later and duplicated the second part except that the stories themselves had been changed. After categorization of the subjects' responses, the data were subjected to statistical analysis and the findings were reported. The self-criticism and repara tion responses were discussed first, followed by an anal ysis of the three parts of the research design. The find ings concerning similarities and differences of the two social classes were reviewed as well as those having to do with examiner control versus subject control of the reward/punishment variable. These discussions about the experimental results were then synthesized for presenta tion in order to support or reject eight hypotheses which had been stated at the outset of the study. Several valid conclusions seemed to emerge from ....... 117. the analysis of the findings and these are presented at this time. Conclusions 1. Lower-class subjects in this research gave more moral responses in the categories studied than did the middle-class subjects. This was true for both self- criticism and reparation. 3. Examiner control of punishment tended to elicit more responses than did self-control of punishment. 3. It is possible that the cuing in the experi ment , which consisted of reminders by the examiner between each test trial to be more "careful" or not to be so "rough," could have engendered defeatism on the part of the subjects, since they were continuously being told about what they had done "wrong." In any event, cuing did not appear to influence moral judgments, although it was found that low cognitive cuing did tend to elicit more moral responses than did high cognitive cuing. 4. Lower-class subjects tended to have more self-drive than middle-class subjects. This conclusion is based on the fact that the lower-class girls exhibited as a group more initiative than did the middle-class 118' girls in answers to questions that elicited self-critical and, especially, reparative responses. The middle-class subjects appeared to lack the ability to respond con structively with reparative suggestions. 5. When oververbalization was present in the paradigm— that is, in the high cognitive cuing paradigm— lower-class subjects appeared to be more frightened by the examiner than did the middle-class subjects, as shown in their behavior. When the examiner continually cau tioned or warned the subjects during the experiment, this appeared to have a greater effect on lower-class than middle-class subjects, making them unwilling or unable to give any response. 6. Regardless of cuing or areas of control of punishment, the lower-class subjects exhibited in their response patterns a sense of responsibility regarding their personal part in making amends for the destruction of the property involved in the situation. This was not equally true of the middle-class subjects who frequently indicated that they did not feel a sense of responsibil ity in the situation. 7. Although the lower-class subjects tended to give more self-critical and/or reparation responses than 119 did middle-class subjects, the middle-class subjects tended to be more stable in their answers, especially in the reparation responses. Implications for Further Research There are formidable problems associated with any attempt to measure moral responses. Nevertheless, as a result of some of the particular features of this experi ment, some suggestions for further research were clari fied. These suggestions are: 1. Instruments should be developed which have been sufficiently standardized or validated by extensive use. At the same time, reliability data should include estimates as to internal consistency. In reviewing the literature, most tests appeared to be made up for a par ticular study and then abandoned; also, on many tests the experimenters included only a small number of moral or ethical areas, thus limiting the generality of their research. The content typically sampled is based on categories of conventional morality or on the author's theoretical preoccupation. Many dimensions of behavior which are salient are thereby excluded. More comprehen sive test instruments therefore appear to be greatly 12a needed if continuity and validity are to be established in evaluating moral responses. 2. A comprehensive study should involve subjects of different age groups, rather than only one. This once overworked area of developmental changes now seems to be given too little attention. The fact that a subject's maturational level may not be as influential in determin ing moral development as was once believed does not de tract from the scientific importance of developmental studies. On the contrary, as the evidence mounts in favor of the view that human morality is man-made rather than an inevitable product of organismic evolution, an understanding of the genetic process of moral development becomes more urgent and intriguing. Such changes over time, however, must be studied not against a purely chronological scale, but in the context of the ever- changing matrix of social relationships through which the subject is molded to man's estate. 3. A longitudinal study covering at least ten years of the subject's life should be undertaken. This would include not only responses of the subjects in both natural and experimental settings but also the responses of the subject's parents. This might take the form of ■ 121 interviewing both subjects and parents away from home as well as home visits in which trained personnel could evaluate the interaction of subject and subject's par ents. This could also help in the development of reli able instruments for the judging of moral responses. 4. A study such as this, with previously sug gested improvements, should be repeated with boys. Girls tend to exhibit a higher level of verbal development than boys on intelligence and achievement tests at the elementary age level (Goodenough, 1954), although there are exceptions (Bayley, 1957). Girls are also rated as being more honest than boys (Hartshorne.and May, 1928), and as developing a conscience earlier than boys (Sears, Maccoby, and Levin, 1957), but there are also a number of exceptions depending upon the definition of conscience (Miller, 1961). 5. A new study should attempt to view evaluative attitudes as subjective phenomena whose measurement is best achieved independently of a concern about the rela tionship of those attitudes to conventional or normative standards of moral valuation. Whether the attitudes be approved or disapproved by society is a subsequent ques tion which is not as revealing as the patterning of 122' subjective attitudes of evaluation per se. As Josselyn (1948) has suggested: "While from the standpoint of soci ety, behavior is either good or bad, from the standpoint of the individual it always has some positive value. It represents the best solution for his situation at this time" (p. 224). Moral values must be understood from the subject's point of view as well as from society's. 6. An experimental study should be developed and tested in which the subject must make reparation for the destruction of someone's reputation or name, as compared with the reparation responses for the destruction of property as investigated herein. t BIBLIOGRAPHY 123 BIBLIOGRAPHY Books Baldwin, J. M. Social and Ethical Interpretations in Metal Development. New York: MacMillan, 1906. Bruner, J. S. The Process of Education. Cambridge: Harvard University Press, 1960. Eysenck, H. J. The Structure of Human Personality. New York: Wiley, 1953. Fenichel, O. The Psychoanalytic Theory of Neurosis. New York: W. W. Norton, 1945. Freud, A. The Ego and the Mechanism of Defense. New York: International University Press, 1946. Freud, S. New Introductory Lectures on Psychoanalysis. New York: W. W. Norton, 1933. ______. The Problem of Anxiety. New York: W. W. Norton, 1936. Hartshorne, H., and May, M. A. Studies in the Nature of Character. Vol. I: Studies in Deceit. Vol. II: Studies in Service and Self-control. Vol. Ill: Studies in the Organization of Character. New York: Macmillan, 1928, 1929, and 1930. Havighurst, R., and Neugarten, B. American Indian and White Children. Chicago: University of Chicago Press, 1955. 124 125 Henry A. F., and Short, J. F., Jr. Suicide and Homicide: Some Economic, Sociological and Psychological Aspects of Aggression. Glencoe, 111..! Free Press, 1954. Hull, C. L. A Behavior System. New Haven: Yale Univer sity Press, 1952. Jones, M. R. (ed.). Nebraska Symposium on Motivation! 1959. Lincoln: Nebraska University Press, 1959. Martin, W. E., and Stendler, Celis (eds.). Readings in Child Development. New York: Harcourt, Brace, 1954. McClelland, D. C. (ed.). Studies in Motivation. New York: Appleton-Century-Crofts, 1955. Mead, G. H. Mind, Self and Society. Chicago: Univer sity of Chicago Press, 1934. Miller, N. E., and Dollard, J. Social Learning and Imitation. New Haven: Yale University Press, 1941. Miller, D. R., and Swanson, G. E. The Changing American Parent. New York: Wiley & Sons, 1958. ________, (eds.). Inner Conflict and Defense. New York: Holt, 1960. Mowrer, O. H. Learning Theory and Behavior. New York: Wiley, 1960. . Learning Theory and Symbolic Processes, New York: Wiley, 1960. ________ . Learning Theory and Personality Dynamics. New York: Ronald, 1950. Murray, H. A. (ed.). Explorations in Personality: A Clinical and Experimental Study of Fifty Men of College Age. New York: Oxford University Press, 1938. 126 Osgood, C. E. Method and Theory in Experimental Psychol ogy. New York: Oxford University Press, 1953. Piaget, J. The Moral Judgment of the Child. Glencoe, 111.: Free Press, 1948. Sears, R. R., Maccoby, Eleanor E., and Levin, J. Pat terns of Child Rearing. Evanston, 111.: Row, Peterson, 1957. Warner, Lloyd W., Meeker, Marcia, and Eells, Kenneth. Social Class in America: A Manual of Procedure for the Measurement of Social Studies. Chicago: Science Research Associates, 1960. Whiting, J. W., and Child, I. L. Child Training and Personality: A Cross-Cultural Study. New Haven: Yale University Press, 1953. Articles and Periodicals Allinsmith, Wesley A. "The Learning of Moral Standards," in Inner Conflict and Defense. Edited by D. R. Miller and G. E. Swanson. New York: Holt, 1960, pp. 141-176. Aronfreed, J. "The Nature, Variety and Social Patterning of Moral Responses to Transgression," Journal of Abnormal Social Psychology, LXIII (1961), 223-241. Boehn, L. "The Development of Independence: A Compara tive study," Child Development, XXVIII (1957), 85-92. Bronfenbrenner, W. "Socialization and Social Class Through Time and Space," in Readings in Social Psychology. Edited by Eleanor Maccoby, P. M. Newcomb, and E. L. Hartley. 3rd ed. New York: Holt, Rinehart, 1958, pp. 400-425. "Freudian Theories of Identification and Their Derivatives," Child Development, XXXI (1960). 127 Bronfenbrenner, W. "Moral Development of Students in Protestant, Catholic, and Public Schools," Amer ican Psychologist, XXVII (1964), 550-565. Bruner, J. S., and Goodman, C. C. "Value and Need as Organizing Factors in Perception," Psychological Review, LVI (1949), 200-207. Church, R. M. "Transmission of Learned Behavior Between Rats," Journal of Abnormal Social Psychology, LIV (1957) . Davis, A., and Havighurst, R. J. "Social Class and Color Differences in Child-Rearing," American Sociolog ical Review, XI (1946) Dennis, W. "Animism and Related Tendencies in Hopi Chil dren ," Journal of Abnormal Social Psychology, XXXVIII (1943). Durkin, D. "Children's Concepts of Justice: A Comparison with the Piaget Data," Child Development, XXX (1959). Harrower, M. R. "Social Status and Moral Development," British Journal of Educational Psychology, 1935, pp. 75-95. Hill, W. F. "Learning Theory and the Acquisition of Values," Psychological Review, LXVII (1960), 317-331. Lerner, E. "The Problem of Perspective in Moral Reason ing," American Journal of Sociology, XLIV (1937), 249-269. Littman, R. A., Moore, R. A., and Pierce-Jones, J. "Social Class Differences in Child Rearing: A Third Community for Comparison with Chicago and Newton, Massachusetts," American Sociological Review, XXII (1957). 128 McClelland, D. C., Rindlishacher, A., and de Charners, R. "Religious and Other Sources of Parental Atti tudes Toward Independence Training," in Studies in Motivation. Edited by D. C. McClelland. New York: Appleton-Century-Crofts, 1955. Maccoby, E. E., and Gibbs, P. K., et al. "Method of Child-Rearing in Two Social Classes," in Readings in Child Development. Edited by W. E. Martin and Celis Stendler. New York: Harcourt, Brace, 1954. MacKinnon, D. W. "Violation of Prohibitions," in Explor ations in Personality: A Clinical and Experi mental Study of Fifty Men of College Age. Edited by H. A. Murray, et al. New York: Oxford Univer sity Press, 1938, pp. 491-501. MacRae, D. "A Test of Piaget's Theories of Moral Devel opment,'1 Journal of Abnormal Social Psychology, XXIX (1954), 14-18. Peel, E. A. "Experimental Examination of Some of Piaget's Schemata Concerning Children's Percep tion and Thinking and a Discussion of Their Edu cational Significance," British Journal of Edu cational Psychology, XXIX (1959), 89-103. Pittel, Stephen M., and Mendelsohn, Gerald A. "Measure ment of Moral Values: A Review and Critique," Psychology Bulletin, LXVI (1966). Russell, W. H., and Storms, L. H. "Implicit Verbal Chaining in Paired-Associate Learning," Journal of Experimental Psychology, XLIX (1953).. Sanford, N. "The Dynamics of Identification," Psychology Review, LXV (1955). Siegel, A., and Kohn, L. "Permissiveness, Permission and Aggression: The Effect of Adult Presence or Absence on Aggression in Children's Play," Child Development, XXX (1959). 128 ca- Siegel, 1. E., Hoffman, M., Dregor, A., and Torgoff, I. "Toward a Theory of Influence Techniques: Prelim* inary Report," Merri11-Palmer Quarterly, 1954, pp. 113-126. Ugurel-Semin, Refia. "Moral Behavior and Moral Judgment of Children," Journal of Abnormal Social Psychol ogy, XLVII (1952), 463-474. Whiting, J. W. "Sorcery, Sin and the Superego: A Cross- Cultural Study of Some Mechanisms of Social Con trol ," in Nebraska Symposium on Motivation: 1959, Edited by M. R. Jones. Lincoln: University of Nebraska Press, 1959. Wright, B. A. "Altruism in Children and the Perceived Conduct of Others," Journal of Abnormal Social Psychology, XXXVII (1942), 218-233. Unpublished Material Boehm, L., and Nass, M. L. "Social Class Differences in Conscience Development: Progress Report." Mimeographed, 1961. Heinecke, C. M. "Some Antecedents and Correlates of Guilt and Fear in Young Boys." Unpublished Doc toral dissertation. Harvard University, 1953. (Dissertation Abstracts, 1954.) Kohlberg, L. "The Development of Modes of Moral Thinking and Choice in the Years 10-16." Unpublished Doctoral dissertation, University of Chicago, 1958. Liu, C. “The Influence of Cultural Background on the Moral Development of Children." Unpublished Ph. D. dissertation, Columbia University, 1950. MacKinnon, D. W. "The Violation of Prohibitions in the Solving of Problems." Unpublished doctoral dis sertation, Harvard University, 1933. 129 Turiel, Edward C. "An Experimental Analysis of Develop mental Stages in the Child's moral Judgment." Unpublished Ph. D. dissertation, Yale University, 1964. A P P E N D I X E S 130 APPENDIX A THE EXPERIMENTAL DESIGN 131 THE EXPERIMENTAL DESIGN Part I These instructions were given verbally to the subjects by the examiner at the time of the test game part of the experiment. High cognitive cuing— subject control I have something here for you to do. To do this work, you have to be very careful and gentle. Being care ful and gentle is the most important thing. Here's the way to do it. Back here is an object which you must push back into this box (experimenter indicates box by lifting it up and showing it to subject). Here's how it goes into the box. You push it off the board with this pusher. You can't lift the pusher; you must keep it down on the board when you push (experimen ter demonstrates, then hands pusher to the subject). Of course, to push it off, you will have to push 132 133'' through the dolls, and so you will knock some of them over, even though you're trying to be careful and gentle (this is true, as the spacing between the dolls is much too narrow to permit the pusher's blade to pass without knocking some over). The idea is to see how many of the dolls you will knock down. If you knock down just a few, that's good. If you knock down a lot of them, that's not so good. When you use the pusher, try to be as careful and gentle as you can. Here is a pile of thirty candy bars for you (experimenter places pile and empty box in front of the subject). When we're all through you may keep however many you have left. Each time, after you push off the big doll, you look at the number of dolls you knocked down, and you decide how careless and rough you've been. Then you decide how many candy bars you think is right and put them in the box here. Those are the bars you lose. You might take one candy bar, or two, or three, but not more than three. The amount depends on how care less and rough you think you've been. (The subject will then go through ten training trials. The experimenter will begin each trial by saying:) Remember to keep the pusher down on the board. Push off " ........ 134 the object and knock down as few doll as as you can. And be careful and gentle as you can. All right, go ahead. (On each trial, after the object has been pushed over into the box, the experimenter will say:) All right, you knocked down some dolls. You decide how careless and rough you've been. Take as many candy bars as you think is right and put them into the box. (While the subject is making the payment, the experimenter will reset the board.) High cognitive cuing— examiner control I have something here for you to do. To do this work, you have to be very careful and gentle. Being ’ careful and gentle is the most important thing. Here's the way to do it. Back here is an object which you must push back into this box (experimenter indicates box by lifting it up and showing it to the subject). Here's how it goes into the box. You push it off the board with this pusher. You can't lift the pusher; you must keep it down on the board when you push (experimenter demonstrates, then hands the pusher to the subject). ........135 Of course, to push it off, you will have to push through the dolls, and so you will knock some of them over, even though you're trying to be careful and gentle. The idea is to see how many of the dolls you will knock down. If you knock down just a few, that's good. If you knock down a lot of them, that's not so good. When you use the pusher, try to be as careful and gentle as you can. Here is a pile of candy bars for you. When we're all through you may keep however many you have left. Each time, after you push off the doll, I will look at the number of dolls you pushed down, and I will decide how careless and rough you've been. Then I will decide how many candy bars I think is right and put them in the box here. Those are the bars you lose. I might take one candy bar, or two, or three, but not more them three. The amount depends on how careless and rough I think you've been. (The subject will then go through ten training trials. The experimenter will begin each trial by say ing:) Remember to keep the pusher down on the board. Push off the big doll and knock down as few dolls as you can. And be careful and gentle as you can. All right, go ahead. (On each trial, after the big doll has been pushed over into the box, the experimenter will say:) All right, you knocked down some dolls, so I will decide how care less and rough you've been. I will take as many candy bars as I think is right and put them into the box. Low cognitive cuing— subject control I have something here for you to do. Here's the way to do it. Back here is a large doll which you must push back into this box. Here's how it goes into the box. You must push it off the board with this pusher. You can't lift the pusher; you must keep it down on the board when you push. Of course, to push it off, you will have to push through the dolls, and so you will knock some over. The idea is to see how few you will knock down. Here is a pile of thirty candy bars for you. When we're all through you may keep however many you have left. Each time, after you push off the object, we will look at the number of dolls you knocked down. Then you decide how many candy bars you should take from your pile. ' 137' You take the number of candy bars you think is right and put them in the box here. Those are the candy bars you lose. You might take one candy bar, or two, or three, but not more than three. (The subject will then go through ten training trials. The experimenter will begin the first trial by saying:) Remember to keep the pusher down on the board. Push off the large doll and knock down as few dolls as you can. All right, go ahead. (The experimenter will begin all subsequent trials by simply saying:) Go ahead. (On each trial, after the object has been pushed over into the box, the experimenter will say:) All right, you knocked down some dolls. You take one candy bar, two, or three, not more than three. Low cognitive cuing— examiner control I have something here for you to do. Here's the way to do it. Back here is a large doll which you must push back into this box. Here's how it goes into the box. You must push it off the board with this pusher. You can't lift the pusher; you must 138 keep it down on the board when you push. Of course, to push it off, you will have to push through the dolls, and so you will knock some over. The idea is to see how few you will knock down. Here is a pile of thirty candy bars for you. When we're all through you may keep however many you have left. Each time, after you push off the object, I will look at the number of dolls you have knocked down, and I will take the number of candy bars I think is right and put them in the box here. Those are the candy bars you lose. I might take one candy bar, or two, or three, but not more than three. (The subject will then go through ten training trials. The experimenter will begin the first trial by saying:) Remember to keep the pusher down on the board. Push off the large doll and knock down as few dolls as you can. All right, go ahead. (The experimenter will begin all subsequent trials by saying simple:) Go ahead. (On each trial, after the object has been pushed over into the box, the experimenter will say:) All right. You knocked down some dolls. I'll have to take— let's see (one, two, three) candy bars. 139 Test trial The eleventh trial of all of the four experimental trials was planned as an apparently unexpected disruption due to the breaking in half of the large doll. It was made so it would come apart simply. Since the experi menter had to be present during the test trial to observe the subject's responses, some change in the situation had to occur to indicate to the subject that the previous punitive consequences of transgressions were no longer forthcoming. Accordingly, the experimenter terminated his role as a socializing agent on this trial and used nondirective, apparently casual verbal stimuli to make the situation more appropriate for the child to show her own overt reactions. When the subject pushed off the object on this trial, the experimenter looked into the box with a sur prised expression and said: "Oh my— it's broken." While saying this, the experimenter holds up the broken large doll and looking at it, says (stimulus 1): "And we don't have another object here to use. I wonder why it broke." 140." I This was to elicit a self-critical response in the subject. If the subject gave any response to stimulus 1 that was relevant to the cause of the large doll's breaking, whether or not it was self-critical, the exper imenter went on to stimulus 3. If the response was not clearly relevant to the large doll's breaking, or if the subject gave no response, the experimenter presented stimulus 2: "Why do you think it broke?" This was meant to provide a stronger eliciting stimulus for self-criticism. Stimulus 3: "Well, now that it's broken, I wonder what we should do." This was meant to elicit reparative responses. If the subject offered any response that was relevant to the implied question, whether or not it was reparative, the experimenter terminated the test trial procedure. If the subject gave no response, or one not relevant to the question, the experimenter then presented stimulus 4: "What do you think we should do now?" Then, regardless of the subject's response to this last question, the test trial was terminated. The subject's responses to all of the experimenter's verbal 141 stimuli were recorded. / Part II The instructions given by the examiner to the subject were as follows: Now I'm going to read you some short stories. I'll read the questions while you listen, then you answer the questions I will ask you. Are there any questions? (If so, they will be answered.) Okay, let's begin. 1. The soft ball team on which you play has asked permission to practice on an empty lot next to Mr. Smith's house. When it is your turn at bat you hit the ball straight to the third baseman but he is not paying attention to the game and does not catch the ball. The ball breaks one of Mr. Smith's windows instead. Mr. Smith comes out to look at the broken window. a. Why do you think the window was broken? b. Now that the window is broken, I wonder what you should do? c. What do you think you should do now? 2. The teacher in your classroom has set up com mittees to prepare book reports. Your job is to see that 142 no one takes home any of the books. One day after the committees have met to decide on what book to report next, you are in a hurry to leave the room and do not check the books carefully. One of the students takes a book home and loses it. She tells the teacher about losing the book. a. Why do you think the book was lost? b. Now that the book is lost, what do you think you should do? c. What do you think you should do? 3. One day you were wrestling with a friend of yours. This friend was wearing glasses and they fell off and broke. When your friend goes home to tell his mother what happened to his glasses, his mother looks unhappy, but she doesn't say anything. a. Why do you think the glasses got broken? b. Now that the glasses are broken, I wonder what you should do? c. What do you think you should do? 4. Both you and Susan play in the school orches tra. One Wednesday when the orchestra is practicing you and Susan get the giggles. The music teacher starts to ” T" 143"' walk back to quiet both of you, but she trips and knocks another child's instrument out of her hands and it breaks. The teacher dismisses the practice session but doesn't say anything about what happened. a. Why do you think the instrument was , _ broken? b. Now that the instrument is broken, I wonder what you should do? c. What do you think you should do now? 5. You and several of your friends like to climb a large fig tree in Jane's back yard, but this one day when you were all there, Jane's mother tells you to be careful not to break the bird feeder that she just hung in the tree. All of you are still up in the tree when her mother comes out to tell you to hurry up and get down because she is going to take you to the movies. All of you.shinny down quickly but just as you get down the bird feeder falls and breaks. Jane's mother comes out to see what happened. a. Why do you think the feeder got broken? b. Now that it is broken, X wonder what you should do? ...... ' ..144..: ’ c. What 4o you think you should do now? 6. You are admiring one of your cousin's birth- day gifts, a model ship, when he tells you it's time for ice cream and cake. You put the ship down on top of a low table in the living room. After eating, everyone at the party is playing blind man's bluff over the living room when the person who is "it" knocks over the table on which you put the model ship and the ship is smashed. Your cousin looks at the smashed ship sadly, but says, "Let's go on with the game." a. What do you think is the real cause of the ship's being broken? b. Now that it is broken, I wonder what you should do? c. What do you think you should do now? 7. One afternoon when your mother had to go out, she had told you to start dinner before she got home. She told you to be sure to put the roast in the oven at 4 o'clock in order for it to be cooked thoroughly by dinner time. She also told you at what temperature to set the oven. After school you stopped at a friend's house and forgot about watching the time. When you did .......145.. remember it was nearly 5 o'clock, so you rushed home to start dinner. You decided that if you set the oven tem perature higher, the roast could cook quicker. When your mother cut the roast, the outside was all right but the inside was not thoroughly cooked. a. Why do you think the roast was not proper ly cooked? b. Now that it is not cooked properly, I wonder what you should do? c. What do you think you should do now? 8. Your friend Susan has collected all the money for the Girl Scout cookies and she asks you to give the money to your neighbor because she is Susan's troop leader. You take the money for the cookies, but when you go to your neighbor's house, she isn't home so you put the money in your room. Several weeks later when Susan goes to pick up her order of cookies, the troop leader tells Susan that she never received any money from her, and so she never ordered any cookies for Susan. a. Why do you think Susan has no cookie order? b. Now that she is without her cookies, 1 wonder what you should do? ........... 146"' c. What do you think you should do now? 9. When the school chorus sings for special occasions it furnishes robes for the students to wear. When the singing is finished, your job is to fold the robes, put them in boxes, and put them away in a closet, and lock the door. This time you are in a hurry to leave school so you just lay some of the robes in the closet without putting them in the boxes. When the music teacher checks the robes later she finds that some ink has been knocked over in the closet and spilled on the robes that were not in the boxes. a. Why do you think the robes were stained with ink? b. Now that the robes are stained, 1 wonder what you should do? c. What do you think you should do? 10. One day when you and your friend go to the show, she gives you her money to keep because she does not have a place to put it, so you put the money in your purse. During the movie your purse falls on the floor and everything is dropped out. It has spread so far that you do not get back all the money that was in the purse. ......... ... '......... 147' When you go to give back the money to your friend you see that all of her money is not there. a. Why do you really think the money was missing? b. Now that some of the money is missing, I wonder what you should do? c. What do you think you should do now. Part III 1. While you are on safety duty, a friend of yours asks you for a pass to go outside the school yard to get a ball which had been thrown outside the fence during a game. Since you could not see a teacher on the yard and you could see that the ball was only a few feet outside the gate, you gave her permission to go to get the ball, but before she could get there the ball rolled out into the street and was run over by a car. a. Why do you really think the ball was ruined? b. Now that the ball is ruined, I wonder what you should do? c. What do you think you should do now? ...... 148 2. It is your responsibility to collect all of the school library books in your room and to reshelf them in the library correctly. One day you were in a hurry to get finished with this job so you let a friend help you. The next day when your class went into the library to check out new books, the teacher found several books that were placed on the wrong shelf. She started to put them back in the correct order, but didn't say anything to you. a. Why do you think the books were reshelved incorrectly? b. I wonder what you should do? c. What do you think you should do? 3. The music teacher assigns you the job of seeing that all of the glee club robes have numbers in them. Each robe is assigned to two students, a boy and a girl, because there are two glee clubs. All the robes but three have numbers so you decide you can remember who gets the unnumbered robes. At the end of the semester there are two robes missing, but all those with numbers are returned. The teacher asked you why only 32 had numbers in them when there were 35 robes. She says she " ..... ' ....149 can't hold anyone responsible for the robes because every one says they returned theirs. a. Why do you think the two robes can not be found? b. Now that the two robes are missing, I wonder what you should do? c. What do you think you should do now? 4. Your school decides to sell donuts from door- to-door to raise money so each room can have its own television set. Sandy is sick one day and asks you to take her money for the donuts to school that day because it is the last day for handing in money. You agree, but forget all about it. When all the class rooms are totaled your class is next to the top for selling the most donuts. a. Why do you think your class did not get the top place? b. Now that your class did not get the top place, I wonder what you should do? c. What do you think you should do now? 5. For her birthday party Linda receives a beautiful new sweater. The Saturday afternoon after her ...... 150' birthday you both go to the show and since her mother insisted that you both take sweaters, Linda lets you wear her new one. The show is warm inside and you put the sweater on the seat next to you, but when you get up to go, the sweater is gone. You ask the usher and the manager but no one has the sweater. You check back for the next few days but the sweater is not found. a. Why do you really think the sweater got lost? b. Now that it is lost, I wonder what you should do? c. What do you think you should do now? 6. You and your friends like to climb an apricot tree in Mrs. Ellen's back yard. This one day she tells you to be careful of the nest a bird has built in the apricot tree. When your mother calls for you to come home, you forget about the warning. Just as you jump out of the tree, you see the nest on the ground turned upside down. a. Why do you think the nest fell out of the tree? b. Now that the nest is down, I wonder what ............................ ' . 151 ; you should do? c. What do you think you should do now? 7. Your mother asked you if you would like to baby-sit for your little cousin while her mother goes out to lunch. You agree to take the job, and your cousin's mother tells you how much she will pay you. She also gives you instructions about keeping the little girl out of the living room because she is afraid something will get broken. Later, during the afternoon, you want to watch a program on television and the television is in the living room. Just as the program is over the mother comes home and your little cousin runs to greet her. As she is running, she bumps into a table and knocks it over and a lamp is broken. Her mother doesn't say anything; she just pays you. a. Why do you really think the lamp was broken? b. 1 wonder what you would do? c. What do you think you would do? 8. Your scout troop is preparing for a cook-out. Each member is to bring something for the cook-out. Your job is to bring one-half of the wieners, smother girl 152 will bring the rest. The day of the cook-out, your family decides to go to Disneyland. You have not told your mother about the cook-out or about the wieners you are responsible for bringing. On the way to Disneyland you ask your Dad to stop at your friend's house because you want to give her something. You intend to give her the money to buy the wieners, but she isn't home so you leave the money in an envelope that you put in the mailbox. At the next Girl Scout meeting the troop leader asks you why you didn't provide the wieners. You tell about giving the money to your girl friend. She says, "All right" to you. Then she goes over to talk to your friend. a. Why do you think the wiener's were not available at the cook-out? b. Now that the cook-out is over, I wonder what you would do? c. What are you going to do now? 9. The school that you attend has a rule which says that no textbooks can be taken home. However, one day you decide you need to study for a test the next day so you take your geography book home. The next morning when you start off for school it is raining so your 153 mother drives you to school. When you get out of the car at school you forget the book in the car. That day when you get home from school you see your little brother playing with the book and he has torn out pages and scribbled all over it with a pencil. a. Why do you think the book is ruined? b. Now that the book is ruined, I wonder what you would do? c. What do you think you would do now? 10. One day when you and your friend go to the show, she gives you her money to keep because she does not have a place to put it, so you put the money in your purse. During the movie your purse falls on the floor and everything is dropped out. It has spread so far that you do not get back all the money that was in the purse. When you go to give back the money to your friend, you see that all of her money is not there. a. Why do you really think the money was missing? b. Now that some of the money is missing, I wonder what you should do? c. What do you think you should do now? APPENDIX B THE SCORE SHEET 154 THE SCORE SHEET Student ____________________ Treatment: Number of reinforcers taken: High evaluative— high control _______________________ High evaluative— low control _______________________ Low evaluative— high control _______________________ Low evaluative— low control _______________________ Answers: Stimulus 1. Oh my,— it's broken. And we don't have another doll. I wonder why it broke? Stimulus 2. Why do you think it broke? Stimulus 3. Well, now that it's broken, I wonder what we should do? 155 - ... -■..... 156 Stimulus 4. what do you think we should do now? Answers to stories: 1. a. ___________ b. 2. a. b. c. 3. a. b. c. 4. a. b. c. 5. a. b. c. 6« a • b. c. 157 7. a. b. c. 8. a. b. c. 9. a. b. c. 10. a. b. c. APPENDIX C TABULATION OF RESPONSES 158 159 REPARATION RESPONSE Stories Number of Responses Class con trol evalu ation Exper iment Follow Up 0-2 3-7 8-10 Total Lower High High X 4 11 15 X — 7 8 15 Lower Low High X 5 5 5 15 X 1 11 3 15 Lower High Low X 5 7 3 15 X 1 13 1 15 Lower Low Low X 3 6 6 15 X 2 8 5 15 Middle High High X 2 13 15 X 2 13 15 Middle High Low X _ 8 7 15 X — 12 3 15 Middle Low Low X 1 9 5 15 X 2 11 2 15 Middle Low High X 1 8 6 15 X - 9 6 15 160 Class Lower Lower Lower Lower Middle Middle Middle Middle SELF-CRITICISM RESPONSE Stories Number of Responses Con- Evalu- -------------- ------------------- trol ation Exper- Follow Q.2 3_? 8.1# xment up High High X 4 11 15 X — 4 11 15 Low High X 2 7 6 15 X 4 3 8 15 High Low X 3 2 10 15 X 1 9 5 15 Low Low X 4 7 4 15 X 3 9 3 15 High High X 2 2 11 15 X — 3 12 15 High Low X 2 7 6 15 X 2 9 4 15 Low Low X 3 7 5 15 X - 6 9 15 Low High X 3 8 4 15 X - 4 11 15 161 TEST STORIES Class Control Test Stories Response- N° Total ( ation Response Lower High High Lower Low High Lower High Low Lower Low Low Self Crit. Reparation Self Crit. Reparation Self Crit. Reparation Self Crit. Reparation 11 9 9 9 8 9 9 9 4 6 6 6 7 6 6 6 15 15 15 15 15 15 15 15 Middle High High Middle High Low Middle Low Low Middle Low High Self Crit. Reparation Self Crit. Reparation Self Crit. Reparation Self Crit. Reparation 12 10 9 12 11 8 10 9 3 5 6 3 4 7 5 6 15 15 15 15 15 15 15 15 MIDDLE CLASS— LOW EVALUATION-HIGH CONTROL Stories Reparations V. L. M Total g 1 0 •r l O M U *H M a ) cn V.L. 0 2 0 2 S 0 4 3 7 M 0 2 4 6 Total 0 8 7 15 Follow-up Stories V.L. Reparations S M Total V.L. g 1 0 •H O •H •P •H O «M rH ( U CO M Total 1 15 MIDDLE CLASS-HIGH EVALUATION— HIGH CONTROL Stories 6 a •H o •H -P •H U U H < u w V.L. Reparations S M Total V.L. 0 0 2 2 S 0 1 1 2 M 0 0 11 11 Total 14 15 Follow-up Stories E u •H o •H +J •rl M U •H H < u (0 Reparation V.L. S M V.L. 0 0 0 0 S 0 1 2 3 M 0 0 12 12 Total Total 14 15 MIDDLE CLASS— LOW EVALUATION— LOW CONTROL Stories V.L. g t o •H U •H •H U o iH 0 ) CO M Total Reparation V.L. S M 8 Total 15 Follow-up Stories V.L. g m •H 3 s ■H U iH a ) w M V.L. Reparation S M Total 1 0 0 1 2 1 0 3 1 5 5 11 Total 4 6 5 15 Self Criticism m Self Criticism MIDDLE CLASS— HIGH EVALUATION— LOW CONTROL Stories V.L. Reparation V.L. S M M Total Total 15 low-up Stories V.L. Reparation S M Total V.L. 0 1 0 1 S 0 2 0 2 M 1 2 9 12 Total 15 LOWER CLASS— LOW EVALUATION— HIGH CONTROL Stories e u •H o •H +> •rt U < U CO V.L. Reparation S M Total V.L. 1 1 1 3 S 1 1 0 2 M 3 5 2 10 Total 15 Follow-up Stories V.L. 6 M •H 3 s -H U U ^ M < u co V.L. Total Reparation S M 11 Total 15 LOWER CLASS-HIGH EVALUATION— HIGH CONTROL Stories 6 W •H 0 3 s •H U u 1 < U CO V.L, M Total V.L. Reparation S M 11 Total 0 0 0 0 0 2 2 4 0 2 9 11 15 Follow-up Stories 6 ( 0 •H O •H •P •H P O ( U CQ V.L. Reparation S M Total V.L. 0 0 0 0 S 0 2 2 4 M 0 2 9 11 Total 11 15 LOWER CLASS— HIGH EVALUATION— LOW CONTROL Stories V.L. e m •H O 3 s •H u u m M < u w Total V.L. Reparations S M Total 15 Follow-up stories V.L, Reparations S M Total e w •H o •H ■P •H M U <W rt Q ) O) V.L. M Total 10 15 Self Criticism h* Self Criticism LOWER CLASS— LOW EVALUATION— LOW CONTROL Stories V.L. M Reparations V.L. S M Total Total 4 15 low-up stories Reparations V.L. S M Total V.L. M Total 3 15

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Effects Of Two Types Of Experimenter Intervention And Schedules Of Reinforcement On Verbal Operant Conditioning Of Affective Self-References

PDF

An Experimental Study Of The Degree Of Affective Response Elicited By Several Mediated And Nonmediated Instructional Methods

PDF

Effects Of Success And Failure On Impulsivity And Distractibility Of Three Types Of Educationally Handicapped Children

PDF

Confirmation Of Expectancy And Changes In Teachers' Evaluations Of Student Behaviors

PDF

Auditory Perceptual Si Factors As Non-Predictors Of Reading Achievement In An Upper-Class And Upper-Middle-Class Population

PDF

Transfer Of The Partial Reinforcement Extinction Effect Across Tasks In Normal And Retarded Boys

PDF

An Empirical Investigation Of F-Test Bias, Disproportionality, And Mode Of Analysis Of Variance

PDF

Anxiety, Physiologically And Psychologically Measured, And Its Consequences On Mental Test Performance

PDF

The Relative Efficiency Of Prompting And Confirmation Learning Paradigms

PDF

Delay Of Feedback And The Acquisition And Retention Of Verbal Material Inthe Classroom

PDF

An Investigation Of The Use Of Focused Video Tape Feedback In High Schoolgroup Counseling

PDF

Ethnicity And Measures Of Educability: Differences Among Navajo, Pueblo And Rural Spanish-American First Graders On Measures Of Learning Style, Hearing Vocabulary, Entry Skills, Motivation And H...

PDF

Clinical Implications Of Verbal Learning Without Awareness: An Experimental Study

PDF

An Application Of A Two-Stage 'Attention' Model To Concept Formation In The Mentally Retarded

PDF

The Modification Of Maladaptive Behavior Of A Class Of Educationally Handicapped Children By Operant Conditioning Techniques

PDF

Testing Attitudes Toward Physical Handicap: Stereotyping And Objectivityin A Partially Integrated Elementary Program

PDF

Prediction Of First Grade Reading Achievement From Selected Structure-Of-Intellect Factors

PDF

The Association Of Cnv (Contingent Negative Variation) And P3 Evoked Brain Waves With Human Learning

PDF

Selected Characteristics Of A Children'S Individual Test Of Creativity

PDF

Prescriptive Teaching As A Supplement To Behavior Modification In The Remediation Of Learning Disorders

Asset Metadata

Creator Price, Edette Barnett (author)

Core Title An Experimental Study Of Two Moral Responses Of Two Socioeconomic Groups Based On Four Paradigms

Degree Doctor of Philosophy

Degree Program Educational Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag education, educational psychology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Wolf, Richard M. (committee chair), Davis, Daniel J. (committee member), McIntrye, Robert B. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-646269

Unique identifier UC11361015

Identifier 6904541.pdf (filename),usctheses-c18-646269 (legacy record id)

Legacy Identifier 6904541.pdf

Dmrecord 646269

Document Type Dissertation

Rights Price, Edette Barnett

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA