A h“V¢B-'U v- - Thesis {as “as D s-qi'se 0‘ M. A. MICHELE} AN STATE UNIVERSWY Mary Susan 1966 ABSTRACT PATTERNS OF PARENT-CHILD RELATING FOLLOWING TRANSGRESSION BY THE CHILD: A CROSS-NATIONAL STUDY by Mary Susan Faaborg The Anderson Incomplete Stories, a projective technique, were developed to elicit children's perceptions of social conflict situations. These stories were ad- ministered under the supervision of H. H. and Gladys L. Anderson to over 10,000 children in eight countries. The present study randomly selected and analyzed 100 of the 350 protocols obtained from 7th grade children by the Andersons in their 1954 sample from Munich, Germany and 87 of the 299 protocols obtained in 1957 from 7th grade children in Benton Harbor, Michigan. -Completions to four of the eleven Anderson Incom— ,plete Stories were utilized in this report: A—2, the.§g§§ Mgap story; A—4, the Ink Spot story; B—4, the-ggmaqed Axe story; and B—5, the Soiled Dress story. Each of these stories depicts a hypothetical transgression situation where- in one character (the transgressor) has commited or been in- volved in an act of carelessness and/or disobedience, a loss of or damage to the property of another (the property owner). Mary Susan Faaborg .Aims- The purpose of the present study was to uti— lize content analyses of the completions to these four stories in order to investigate cultural differences in the following areas: 1. the level of conscience development revealed by the actions and thoughts attributed to the transgressor 2. the relative amount of domination or egalitarianism depicted in the interaction between transgressor and property owner 3. the relationship between the level of conscience manifested by the transgressor and the amount of domination or egalitarianism depicted in the transgressor-prOperty owner interaction Although conscience development can be placed on a develOpmental continuum, the present study considered only the relative extremes of this continuum, i.e., the primitive, immature level of conscience based on the fear of authority and the more mature type of conscience based on internal- ized values. Procedure. One Codinq Manual,containing six cate- gories, was designed for the analysis of all four stories. Three categories served to assess the relative maturity of conscience. These three categories are as follows: I. Transgressor's initial feelings II. Transgressor's communication regarding transgression VI. Transgressor's affective/behavioral response Mary Susan Faaborg Three additional content categories focused upon the relative amount of domination or egalitarianism present in the transgressor-property owner interaction: III. Interpersonal context of communication IV. Property owner's initial feelings V. PrOperty owner's action/communication The mutually exclusive and exhaustive items comprising these three categories were ranked in terms of the relative amount of domination or egalitarianism suggested by the item. Only the first two of the four stories analyzed in this study, i.e., the Lost Meat and the Ink Spot stories, were coded in each of the six categories above. The sibling stories, the Damaged Axe and the Soiled Dress, were coded in only Category IV, PrOperty_owner's initial feelings, and Category V, Property owner's action1communication. Hypotheses. The present study hypothesized that the culture group containing a predominance of domination- submission themas would also show a relative immaturity of conscience, i.e., a moral orientation based on fear of authority. A culture group containing a predominance of egalitarian themas would instead show a relative maturity of conScience, i.e.,- a moral orientation based on internalized values. Furthermore, it was hypothesized that cultural differences in conscience develOpment would be most clearly revealed when context effects were considered. Mary Susan Faaborg In other words, the transgressor's response must be con— sidered in relation to the interpersonal context in which it occurred before subtle differences between the two culture groups can be assessed. Findings. Chi-square tests assessed the relation- ship between cultural origin and level of conscience. Of the three content categories designed to differentiate the level of conscience, two of the categories (Category I, Transgressor's initial feelings, and Category II, Trans— gressor's communiation regarding transgression) revealed differences between the culture groups on one or both of the two parent-child stories considered. On both categories, Munich completions reflected a preponderance of responses indicating a primitive or immature level of conscience, based on fear of authority. The majority of Benton Harbor completions, on the other hand, reflected a mature level of conscience, based on internalized values. Cultural differences in the amount of domination or egalitarianism expressed in the transgressor-prOperty owner interaction were also assessed by chi—square tests. Of the three content categories designed to assess domination or egalitarianism, two of the categories (Category IV, Property owner's initial feelings, and Category V, Property_owner's actionzcommunication) revealed cultural differences on one or more of the four stories considered. In each case where Mary Susan Faaborg significant differences were obtained, Munich completions reflected significantly more domination than did Benton Harbor completions. Cultural differences in the relationship between level of conscience (primitive versus mature) and transgressor-prOperty owner interaction (domination versus egalitarianism)were also assessed by chi—square tests of significance. The relationship between three specific cate— gory pairs was tested: Category II, Transgressor's communi- cation, and Category III, Interpersonal context; Category II and Category V, property owner's action/communication; and Category V and Category VI, Transgressor's affective/ behavioral response. .0nly two category pairs, II and V, and V and VI, proved to be differentially related in the two culture groups. In each case, however, Munich completions showed a statistical dependence between the transgressor- prOperty owner interaction and the level of conscience de- velopment, while Benton Harbor completions did not. Maturity of conscience was originally defined as a moral orientation, based on internalized values, psycho- logically independent of the immediate environmental conse- quences. That Munich completions revealed a statistical de— pendence between transgressor's response and the amount of environmental domination supports the previous conclusion that Munich completions revealed greater immaturity of Mary Susan Faaborg conscience, i.e., that their responses were dependent upon the immediate environmental consequences. Because Benton Harbor completions did not reveal such a dependence, i.e., because their responses were not dependent upon the immediate environmental consequences, we conclude that they revealed greater maturity of conscience. Furthermore, the nature of the inter—category de— pendence revealed in the Munich completions gave additional support and clarification to this study's major hypothesis. Of the two category pairs which proved to be differentially related in the two culture groups, both pairs supported the hypothesis that parent—child relations characterized by domination and submission, foster in the child a dependency relationship which hinders the development of independent internalized value systems, i.e., maturity of conscience. Hence, the study supports the more general hypothesis that an analysis of context effects would clarify cultural differ- ences in the relationship between conscience development and parent—child interaction. I Approved 7]!) “(J )4 phalflfimy Chairman Date (] /4’\LAA% ! at 4:4. PATTERNS OF PARENT-CHILD RELATING FOLLOWING TRANSGRESSION BY THE CHILD: A CROSS-NATIONAL STUDY BY Mary Susan Faaborg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1966 DEDICATION To HAROLD H. ANDERSON who teaches, by example, the principles of confrontation, growth, and harmony. ACKNOWLEDGMENTS The writer greatly appreciates the many hours of guidance and support which she received from Dr. Harold H. Anderson during the preparation of this thesis. The writer also wishes to acknowledge Dr. Terrence M. Allen for his helpful suggestions in the statistical manipulation of these data, and to thank Dr. John R. Hurley for his direction and advice. For coding story completions to establish inter-coder reliability, this writer is grateful to Miss Karen Sue Kamerschen. iii TABLE OF CONTENTS Chapter Page I. INTRODUCTION . . . . . . . . . . . . . . . . . 1 Materials 5 Administration 8 Translation . . . . . . . . . . 8 The Andersons' Sample . . . . . 9 Protocols Used in the Present Study 9 Content Analysis of Story Completions 10 Reliability of Coding . . . . . . l3 Reliability of Story Instrument . . . . . 14 II. COMPARISON OF GROUPS ACCORDING TO CATEGORY ITEM FREQUENCIES . . . . . . . . . . . . . . 20 Procedure . . . . . . . . . . . . . . . . 21 Hypotheses . . . . . . . . . . . . . . . . 22 Findings . . . . . . . . . . . . . . . . . 24 Discussion . . . . . . . . . . . . . . . . 41 III. COMPARISON OF GROUPS ACCORDING TO THE CONTEXT IN WHICH CATEGORY ITEMS OCCUR . . . . . . . 49 Procedure . . . . . . . . . . . . . . . . 50 Hypotheses . . . . . . . . . . . . . . . . 50 Findings . . . . . . . . . . . . . . . . . 51 Discussion . . . . . . . . . . . . . . . . 58 IV. SUMMARY AND CONCLUSIONS . . . . . . . . . . . 64 REFERENCES . . . . . . . . . . . . . . . . . . . . . . 70 APPENDICES . . . . . . . . . . . . . . . . . . . . . . .A-l iv LIST OF TABLES Table l. Inter-coder reliability computed by Method I and Method II . . . . . . . . 2. Correlations of each category on Story Ar2 with the same category on Story A-4 and correlations of each category on Story 3—4 with the same category on Story B—5 ac- cording to Method I and Method II . . 3. Sex and culture group differences on Cate- gory I 4. Sex and culture group differences on Cate- gory II . . . . . . . . . . . . . 5. Sex and culture group differences on Cate- gory III . 6. ,Sex and culture group differences on Cate- gory IV . . . . . . . . . . . . . . . . . 7. Sex and culture group differences on Cate— gory V . . . . . . . . . . . . . . 8. Sex and culture group differences on Cate- gory VI . . . . . . . . . . . . . . '9. Sex and culture group differences on the relationship between Categories II and III 10. Sex and culture group differences on the relationship between Categories II and V 11. Sex and culture group differences on the relationship between Categories V and VI Page 15 18 26 29 32 34 37 4O 52 54 57 LIST OF APPENDICES Appendix Page A. CODING MANUAL . . . . . . . . . . . . . . . . A—l B. DIRECTIONS FOR ADMINISTERINGANDERSON INCOM- PLETE STORIES . . . . . . . . . . . . . . . B-l C. ANDERSON INCOMPLETE STORIES . . . . . . . . . C-l vi I . INTRODUCTION The term "conscience" has been used to designate a number of different phenomena. Piaget and his followers used it to refer to moral values, i.e., what the individual believed to be right or wrong. The present study, however, was less concerned with the contents of conscience and more concerned with the functions of conscience. The functions of conscience under investigation in this study were: 1) the setting of limits concerning what one should or should not do, think, or feel relatively independent of the immediate environmental consequences of such acts, thoughts,or feel- ings; and 2) the dispensing of internal rewards (increased self-esteem) and punishments (guilt) depending on one's ability to stay within the forementioned limits. Although psychoanalytic theorists (e.g., Fenichel, 1945) described an additional function of conscience, i.e., the establishing of goals and aspirations toward which one strives, the ego ideal or ideal self, this function was not investigated in the present research. -Although the growth of conscience should be viewed as an ongoing process, appropriately viewed as a develop- mental continuum, the present study investigated only the two extremes of this continuum. A primitive or immature level of conscience was inferred when the response to trans- gression was mainly a fearful anticipation of punishment. This response indicated that the child's self—control was still dependent on the immediate environmental consequences. Guilt or remorse, which involved a loss of self-esteem, relatively independent of the actual or anticipated reaction of others, evidenced an allegedly more mature level of conscience. Although a number of alternative theoretical explan— ations have been offered for the process by which conscience develops, e.g., the introjection of parental values, or the learning of behaviors rewarded by parental figures, there has been general agreement that conscience functions as the mediating agent in the task of individual self-control. The activity of conscience has been typically measured in two kinds of experimental situations. In a temptation situation, the individual's behavioral self-control has been tested. In a transgression situation, the individual's response to a failure in self-control has been investigated. The Anderson Stories. .Anderson and Anderson (1954; 1959; 1961; 1965) have developed a set of eleven incomplete stories which depict hypothetical transgression situations. In each of the Anderson Incomplete Stories, one character (the transgressor) has committed or has been involved in an act of carelessness and/or disobedience, a loss of or damage to the property of another (the property owner). Using this instrument, Anderson and Anderson have found that children from several cities in Germany wrote more stories in which the transgressor expressed anxiety, guilt, deception, and submission than did children from United States cities. More German children also wrote stories where the transgressor was punished than did United States children. The Andersons concluded that these differences reflected culturally de- termined modes of parent-child interaction and/or inter— personal communication. Because a mature level of conscience, as reflected in feelings of guilt and remorse, was designated to be re- lated to the extent of one's independence from the actual or anticipated reaction of others, parent—child relations which foster such independence should also foster the development of maturity in conscience. That there is a relationship between parent—child relations and conscience, as variously defined, has already been postulated in the literature. However, the nature of this relationship is still vague. Context Effects. In a recent review of the litera- ture on conscience development, Kohlberg (1963) suggested that much research in this area has been either inconclusive or it has presented results inconsistent with the results of previous studies. The studies reviewed by Kohlberg, however, consistently neglected the relationship between behaviors allegedly manifesting conscience and the interpersonal con- text in which these behaviors occurred. Rather than assessing the effect of interpersonal context, i.e., context effects, upon the behaviors being measured, they focused on one aspect of the parent—child interaction at a time, e.g., self-criticism and resistance to temptation; dependency and confession, etc. Previous content analyses of the cross cultural story completion data collected by the Andersons (Erikson, 1961; Levitov, 1960; Robinson, 1955) have thus far not investi- gated the comparative effect of interpersonal context. These investigators tabulated the frequency with which a particular kind of response appeared in a given culture sample and then, related the frequency of occurrence to the cultural origin of the sample. Terdal (1961) investigated situational differences in his study on the consistency of themas across several of the Anderson stories. Further in- vestigation of context effects, however, is necessary in order to detect subtle cultural differences in parent-child relations and the relationship between conscience and parent-child interaction. Aims of the Present Study. The present study ana- lyzed completions to the Anderson Incomplete Stories in order to make inferences regarding both the level at which conscience was operating and the type of parent—child re— lationship in which conscience was manifested. The major purpose of this research was to investigate cultural differ- ences in conscience development and its relationship to parent-child interaction, using a Coding Manual (see Ap- pendix A) designed to assess context effects. Materials. The Anderson Incomplete Stories (see Ap- pendix C) is a projective technique designed to elicit fantasy material from children. Each of the eleven incom— plete stories comprising the instrument presents a social conflict situation. Although the Andersons administered the entire set of stories to each subject, the present study analyzed completions to only four of the stories. The first two stories, taken from Series-A of the Anderson instrument, describe a conflict between parent and child following transgression by the child. Ar2, the Lost Meat story, reads as follows: Michael's mother sends him to the store to get one pound of wieners. On the way home he puts the package of meat on the edge of the sidewalk and plays for a short time with his friends. A shepherd dog darts forward quickly, claws half the wieners out of the package and rushes away with them. Michael wraps up the remaining wieners and takes them home. What does Michael say to his mother? What does his mother do? How does Michael then feel about it? Think about these questions, then finish this story quickly with a few sentences. A—4, the Ink Spot story, reads as follows: Susan is sitting in the living room working on her homework. Suddenly she remembers her mother's new coat. She wonders how she would look in it. When taking the coat off she notices that she had got some ink spots on her mother's new coat. Just as Susan is trying to wipe out the ink spots her mother enters the room. What does her mother say? What does Susan say? How do they both feel about it? What does each one do? Think about these questions and finish this story quickly with a few sentences. The second two stories, from Series B of the Anderson instrument, each describe a conflict between siblings, following transgression by the main character's younger sibling. B-4, the Damaged_Axe story, reads as follows: Herbert received for his thirteenth birthday a'hand— some camping axe. It is sharp and has a strong leather case. While Herbert is at school his four- year—old brother, Billy, sees the axe, looks at it a long time, picks it up, puts it back, and finally takes it outdoors with him to play. Billy does not take the case off. He sings to himself as he walks about the garden, tapping the axe gently against a tree, a post, and the pavement. Herbert comes home from school, finds the axe in its leather case with some other toys. But he sees that the axe has cut through the leather case and the blade is chipped and blunted. What does Herbert Do? How does Herbert feel about it? Think about these questions and finish the story quickly with a few sentences. B-5, the Soiled Dress story, reads as follows: Kate, aged 13 years, has a little four-year—old sister, Clara. When Kate comes home from school Clara often wants to play with her and follows her and her older girl friends around. One day Clara took a very pretty new dress from Kate's wardrobe, put it on her- self, and looked in the mirror. ~It was too long and hung to the floor. She gathered up the skirt in her arms and went out of the house for a walk. Kate came home from school, found her new dress crumpled on a chair. The skirt had been stepped on and dragged in the dirt. Clara said, "I wore your dress." ' What does Kate do? What do Kate and Clara think about it and how do they feel? Finish the story in a few sentences. The four incomplete stories described above were de- signed as two pairs, one pair from Series A (parent-child stories) and onegmfiu: from Series B (sibling stories). This fact enabled the author to obtain some measure of the relia- bility of the story instrument using parallel or alternate forms. (See p.14,.) All four stories are alike in that in each case a child character has committed a transgression. In the first pair of stories, the property involved belongs to the mother. In the second pair of stories, the property involved belongs to an elder sibling. One major difference between the two stories in each pair respectively, is that one story depicts a female transgressor and the other story depicts a male transgressor. In the stories, as designed, it was possible to study the relationship between the sex of the author and the sex of the story's protagonist. Such an investigation did not occur in the present research. In the construction of their Coding Manuals, the Andersons discovered a further difference between the ngt Meat and the Ink Spot stories, which does not occur in the pair of sibling stories. In the Lost Meat story, Michael loses the meat before he comes home, and there is an interval between the time of transgression and the time Michael confronts his mother. In the Ink Spot story, Susan's mother catches her in the act of transgression, and there is no interval between the time of transgression and confronting the mother. This situational difference may alter the psycho— logical situation enough to elicit different kinds of com— pletions. A discussion of this difference in the parent— child stories is included in Section II of this paper. Administration. The Anderson Incomplete Stories were administered to school children in their classrooms. Each series required one school period to complete. In both Germany and the U.S.A. the stories were administered under the direct supervision of the Andersons and in the native language of the children. Standardized instructions for ad- ministration were followed in all locations. (See Appendix B.) Subjects were assured that their stories would remain anonymous, that neither teacher nor principal would read their completions. Translation. All story completions were translated into English for purposes of coding. Translators encountered some difficulty in accurately translating moods and feelings. Geierhaas (1955), however, reported that when comparisons were made between coding directly from the German and coding from translations, a reliability of 90 per-cent agreement was obtained. The Andersons' Sample. The Andersons administered their stories to more than ten thousand school-age children in eight countries. Subjects were obtained from the fourth, seventh, and tenth grades in local schools. ~Although the subjects were not randomly selected, there were, in most locations, samples of upper, middle-class and lower socio— economic levels, in approximately equal numbers of boys and girls. Protocols Used in the Present Stugy. The sample for the present study consisted of 187 randomly selected proto- cols of seventh graders from Germany and the U.S.A. Fifty male and fifty female protocols were drawn from the Andersons' 1954 sample of 350 school children from Munich, Germany. Forty male and forty-seven female protocols were drawn from the Andersons' 1957 sample of 299 school children from Benton Harbor, Michigan. The present study required protocols of subjects who had completed each of the four stories investigated. When a protocol failed to meet this requirement, it was discarded from the sample and replaced by another, randomly selected protocol,by a child of the same sex. The subjects ranged in age from 11 to 13 years. Information on religion, age, and socio—economic level was available on each subject but was not analyzed in the present study. 10 Content Analysis of Story,Completions. Each sub— ject's set of four stories was content analyzed according to a Coding Manual (see Appendix A) organized specifically for the assessment of context effects. Certain assumptions about the nature of inter— personal communication were implicit in this organization of the Coding Manual. It was assumed that communication is not a linkage of stimuli, responses, response-produced stimuli, etc. Rather, it is a dynamic, continually evolving process. Thus, the author's description of the transgressor's communi— cation was considered in the context of: l) the immediate, observable responses of the property owner; 2) the set of expectancies, accured through past communications, regarding the probable response of the property owner; 3) the trans- gressor's feelings and valuations of himself and the property owner; and 4) the property owner's feelings and valuations of himself and the transgressor. In order to use the author's description of communi- cation as a basis for inferences about conscience develOp- ment, the assumed processive, contextual nature of communi- cation must somehow be preserved in the process of content analysis. The Coding Manual attempted to do just this by focusing on three major phases of the communication, se- quentially: l) the transgressor's initial feelings and com- munication regarding the transgression; 2) the prOperty owner's initial feelings and action/communication after ll learning of the transgression; and 3) the transgressor's response to the property owner's feelings and action/ communication. These phases were analyzed in relation to one another in Section III of this paper. These three phases were subdivided into six gate- gories to assess specific aspects of the interaction between transgressor and property owner. Three of these categories focused on the transgressor and the level of conscience sug— gested by his response. The remaining three categories focused on the property owner and the amount of domination suggested by his response. The six categories were as follows: (See Appendix A for definition of categories.) Category I Transgressor's initial feelings Category II Transgressor's communication regarding transgression Category III Interpersonal context of communication Category IV Property owner's initial feelings Category V PrOperty owner's action/communication Category VI Transgressor's affective/behavioral response Each parent-child story was coded once and only once in each of the above six categories. Each sibling story was coded once and only once in Category IV and Category V. The above categories were comprised of from seven to eight mutually exclusive and exhaustive items. While a gate- ggry was descriptively defined as some behavioral aspect of the interaction between transgressor and property owner, a 12 category was operationally defined by the series of mutually exclusive and exhaustive items which comprised it. These items were ordered from 1 to 5 or 6, with those items most indicative of a dominance—submission relationship between transgressor and property owner receiving a rank of 1. Items most indicative of a relatively egalitarian, inte— grative relationship between transgressor and property owner received a rank of 5 or 6, i.e., the highest number in the series of items. Items between 1 and 5 or 6 were ranked ac- cording to the relative amount of dominance-submission sug- gested by the item. The definitions of domination-submission and inte- gration used in the Coding Manual were taken from Anderson and Anderson (1954). The authors defined domination- submission as follows: Stories having enforced obedience and conformity by the child are classified here. The parent makes the decisions, has the answers, uses threats, shame or punishment. There is no evidence that the child has, or is allowed, ideas of his own; there is low ego involvement. The outcome is usually an expression of a specific detailed act, vow or intent. An integrative relationship was defined as follows: rFor the story to be classified with this outcome there must be evidence of high spontaneity in both the child and his mother; evidence of a high degree of harmony and of working together; active communication and awareness of mutual confidence. There should be evi— dence of "action" in these respects in preference to "thinking" or fantasy about them. .Reliability. Two types of reliability were con— sidered in the present study. The reliability of coding was l3 computed to establish the extent to which two independent orders agreed on their choice of items in each category for a given sample of stories. The reliability of the story instrument was computed to establish the consistency with which the same subjects responded to similar but not identi— cal stories. Reliability of Coding. On random samples of Munich (N = 30) and Benton Harbor (N = 30) stories, independent coders showed per-cent agreements of 96. The following formula was used to compute the per-cent agreement scores: two times the number of items on which Coders l & 2 agree »X 100 total number of items coded by both Coders l & 2 The computation of Pearson product moment corre- lations allowed an additional measure of inter-coder relia- bility. However, the frequency of zeros (0 = no information) in some categories necessitated the computation of corre- lations in two different ways. Method I involved deleting every pair of scores where one or both numbers was zero, and decreasing N accordingly. Method II involved translating each zero to a 4 and increasing every item coded as >14 by a factor of l. The following rationale seemed to justify Method II. The items comprising each category were ranked from 1 to 5 14 or 6 in terms of the relative presence of domination- submission or the relative maturity of conscience suggested by the item. Coding no information as a zero would unjustly classify this item as even more indicative of dominance— submission than items coded as l. The no information re- sponse would be more apprOpriately classified as a neutral point on the 4'to 5 or 6 scale. Therefore, the score of 4 was assigned to each no information response, the number 4 chosen because each category contained 3 items sug- gesting a predominance of dominance—submission. Making the 0 to 4 translation allowed the no information items to be retained in the sample from which correlations were computed. No decrease in N was necessary. The following correlations were obtained by Method I and Method II. .According to both methods, Category III, Interperson- al context of communication, was coded with the lowest relia— bility. However, this category required the greatest amount of inference by the coder and, hence, the lower correlations for this category are not surprising. These results indi- cated that coders were able to use the Coding Manual on all four stories with reliability sufficient for research purposes. Reliability of the Story-Instrument. Although pro- jective techniques are frequently low in reliability, Anderson (1961) found his story instrument to be a reliable Table l. Inter-coder reliability computed by Method I and Method II Method I Method II Category r .E r _N I 'Transgressor's initial feelings .78 21 .84 60 II Transgressor's communication regarding transgression .64 59 .76 60 III Interpersonal context of communication .51 56 .48 60 IV 'Property owner's initial feelings .93 28 .81 60 V Property owner's action/ communication .94 47 .68 60 VI Transgressor's affective/ behavioral response .82 33 .72 60 Note: p less than .0005. all correlation coefficients are significant, l6 indicator of cultural differences and similarities when each cultural sample was taken as a group. Comparing the fantasy productions of Munich children in the same school grade and same neighborhoods in 1954 and again in 1960, he obtained only a very few significant differences between the two groups. Anderson reported high consistency in the story completions of the two groups. The present study attempted to establish the relia- bility of the story instrument for the same subjects across ostensibly similar but not identical stories. Since the two parent-child stories used in this study had similar plots, reliability coefficients were computed using these stories as alternate forms. Since the two child—sibling stories were also similar in plot, reliability coefficients using these stories as alternate forms were also computed. IX subject's response in Category I of one story of a given story pair was compared to this same subject's response in Cate- gory I of the other story of the same story pair. This pro- cedure was followed for each of the six coding categories used on the parent-child stories and for each of the two categories used on the child-sibling stories. The frequency of zeros (0 = no information) on some categories necessitated the computation of product moment correlations in the same two ways used to compute inter- coder reliabilities, i.e., Method I and Method II. Accord- ing to Method I, the first set of correlations was computed 17 by dropping every pair of scores containing one or more zeros and decreasing N accordingly. The second set of corre- lations was computed according to Method II, by translating each 0 to 4 and increasing all items coded as z,4 by a factor of l. The correlations in Table 2 were obtained. Table 2 suggests that for the parent-child stories, Categories II, III, and V contained items to which subjects responded with some degree of consistency across stories, i.e., we can reject the hypothesis that r = 0 with the probability of error less than .05. Further significance tests regarding the size of r were not computed. In general, the correlations for the parent-child stories were low, sub-_ stantially lower than the correlations for the child—sibling stories. Since neither of the story pairs contained identi- cal stories, they can only be considered as alternate forms and the obtained correlations are probably conservative. To account for the difference in the size of the correlations obtained from parentvchild stories and those obtained from child-sibling stories, we should return to a discussion of the plots of each story pair. .As was mentioned previously, the parent—child stories contained a difference in their respective plots which the child-sibling stories did not contain. In the Lost Meat story, the property owner was absent at the time of the transgression, while in the Ink Spot story, she was immediately present. This plot l8 Table 2. Correlations of each category on one parent-child story with the same category on the other parent- child story and correlations of each category on one child-sibling story with the same category on the other child-sibling story by Method I and Method II. Stories Category Method I Method II r _N r _N Parent- I Transgressor's initial Child feelings .197 38 .03 187 Stories II Transgressor's communi- cation regarding * * transgression .33 186 .33 187 III Interpersonal context of * * communication .15 170 .15 187 IV Property owner's initial * feelings .16 38 .10 187 V PrOperty owner's action/ * communication .16 130 .12 187 VI Transgressor's affective/ behavioral response .06 76 .04 187 Child— IV Property owner's initial * * Sibling feelings .62 75 .33 187 Stories V Property owner's action/ * * communication .43 138 .33 187 *p less than .05. 19 difference could account for the relatively lower corre- lations obtained from parent-child stories. The results according to Methods I and II were generally the same. Although the correlations computed ac- cording to Method II were lower, the same categories proved to be reliable according to both methods with only one ex- ception. 0n Category V, Property owner's actioanommunication, responses to the parent-child stories were significantly correlated according to Method I, but the correlation did not reach the .05 level of significance according to Method II. II. COMPARISON OF GROUPS ACCORDING TO CATEGORY ITEM FREQUENCIES Previous analyses of the Anderson data revealed cultural differences in terms of the frequency with which certain types of human interactions occurred in the subject's story completions. The Andersons (1954) reported that com— pletions by children in Karlsruhe, Germany were characterized by domination—submission parent-child themas. Hoffman (1963) in his review of the literature on conscience de- velOpment, concluded: An internalized moral orientation is fostered by an affectionate relationship between parent and child, in combination with the use of discipline tech- niques which utilize this relationship by appealing to the child's personal and social motives. . . . The use of coercive measures that openly confront the child with the parent's power . . . apparently con- tributes to a moral orientation based on the fear of authority. In accord with the Andersons' data on parent—child relations in Karlsruhe, Germany story completions, we might expect Munich story completions to also be characterized by dominance-submission and a moral orientation based on fear of authority (primitive or immature conscience). we would expect U.S.A. story completions to be characterized by egali- tarian parent-child relations and a corresponding moral orientation based on internalized values (mature conscience). 20 21 The newly develOped Coding Manual first assessed cultural differences according to the previously used method of comparing category item frequencies. The investigation of context effects was presented later in Section III. Procedure. Four stories were used in the present analysis, the Lost Meat, the Ink Spot, the Damanged Axe and the Soiled Dress. The first two stories were coded into all six coding categories and the second two were coded into only two of the categories, Category IV, PrOperty owner's initial feelin s; and Category V, Property owner's action/ communication. Following the coding of stories for the total sample, the coded protocols were divided according to cultural location, i.e., Munich (N = 100) and Benton Harbor (N = 87), and the frequency with which each item occurred in each category of each story was recorded for the two groups. When item frequencies were large enough, chi—square analyses tested the possible relationship between culture group and category items for a given story. When frequencies were too small, items were combined into psychologically meaningful groups and chi-square analyses were made. The total sample was then combined and redivided according to sex, i.e., males (N = 90) and females (N = 97). .As above, chi-square analyses tested the possible relationship between sex group and category items for a given story. 22 Hypotheses. The null hypothesis would predict no relationship between culture group and the frequency with which certain category items appear, or between sex group and the frequency with which certain category items appear. Generally, it was hypothesized that Munich stories would contain more dominance—submission parent—child themas and more instances of a primitive level of conscience develop— ment based on fear of authority. Benton Harbor stories, on the other hand, should contain more egalitarian parent-child themas and more instances of a mature level of conscience development. Hypothesis A I. On Category I, Transgressor's initial feelings, Munich stories will show more instances of fear, physiological symptoms of anxiety and evasion of confronting, while Benton Harbor stories will show more instances of guilt, shame, and sadness. Hypothesis A II. On Category II, Transgpessor's com- munication regarding transgression, Benton Harbor stories will show more instances of truth than will Munich stories. Munich stories will show more instances of deception or give no information. Girls will write stories with more instances of truth than boys, who will write stories with a predominance of deception or pp information. Hypothesis A III. On Category III, Interpersonal context of communication, Munich stories will show more instances of external pressure than Benton Harbor stories which will show more instances of ig- ternal pressure and spontaneity. .Ac- cording to the literature suggesting the establishment of earlier and stronger conscience in girls, girls should show more instances of internal Hypothesis A IV. Hypothesis A V. Hypothesis A VI. 23 pressure and spontaneity than boys, which should show more instances of external pressure. On Category IV, Property owner's ini— tial feelings, Munich stories will show more instances of anger and saddness than Benton Harbor stories which will show more instances of understanding and acceptance. No spe- cific difference between the sexes was hypothesized. On Category V, Propertypowner's action/ communication, Munich will show more instances of punishment than will Benton Harbor. Benton Harbor stories will show more instances of forgive— ness and consolation than will Munich stories. No difference between sex groups was hypothesized. On Category VI, Transgressor's af- fective/behavioral response, Munich stories will show more instances of anger and submission while Benton Harbor stories will show more instances of shame and voluntary promise to reform 24 Findings: Hypothesis A I. The first comparison in- volved a test to determine whether the two culture groups wrote stories which differed on Category I, Transgressor's initial feelings. Ar2, the Lost Meat story and.A-4, the Ink Spot story were used to test this hypothesis. Both stories portray parent—child conflict following a transgres— sion by the child. However, the Ink Spot story includes the mother in the incomplete story, while the Lost Meat story does not. Hence, some differences between the completions to the two stories may be expected as differences on Cate— gory I. Completions to both stories were coded according to Category I of the Coding Manual. This category was designed to record the child's feelings about 1) his transgression, 2) confronting his mother, or 3) both. All feelings which occurred prior to the mother's response after learning about the transgression were coded in this category. In general, the category attempted to assess the presence and content of the child's evaluation of his own behavior, insofar as it could be separated from the mother's evaluation after learn— ing of the transgression. For further description of this category, see Appendix A. The transgressor's initial feel- ings were coded as one of the following mutually exclusive and exhaustive category items: 1. Anxiety, fear 2. Physiological symptoms of anxiety 3. Evades confronting 25 4. Guilt 5. Shame 6. Sadness 7. No information The first three category items were devised to assess the various manifestations of fear. They were thought to re— flect an immature level of conscience develOpment, i.e., the extent to which the transgressor judged his behavior in terms of what he expected would happen to him as a result of it. The second three category items were devised to assess the extent to which the transgressor judged his behavior by an internalized value system, wherein his conscience dealt out the rewards and punishments rather than his mother. The analysis of the data to test Hypothesis A I counted responses to the first three category items as indi— cations of Egg; and counted responses to the second three category items as manifestations of conscience. Subjects who wrote stories coded as giving no information on Category I were excluded from this analysis. The following results were obtained when sex groups and culture groups were com- pared on the Lost Meat and the Ink Spot’story° Table 3 gives the frequencies of items 1-3 combined and items 4-6 combined on Category I for stories-A—Z and A—4, comparing males with females and Munich subjects with Benton Harbor subjects. The table indicates that the Lost Meat story revealed no significant differences between either the sex groups or the culture groups on Category I. 26 Table 3. Sex group and culture groups differences on Category I, Transgressor's initial feelingg- Anxiety, GuiltJ Fear, ShameJ Story Group N Evasion Sadness -Chi—square Ar2 Male 12 9 3 -.20 Female 40 25 15 Lost B. Harb. 22 13 9 Meat Munich 30 21 9 .25 Ar4 Male 24 13 11 2.64 Female 48 15 33 Ink Munich 31 20 11 13.21* Spot B. Harb. 41 8 33 *p (.001. 27 However, the table reveals significant differences between the Munich and Benton Harbor protocols on the ka fiixfl:Storyw Consistent with Hypothesis I, more.Munich proto— cols were coded in_apxiety, fear, and evgsion, while more Benton Harbor protocols were coded in guilt, shame, and Sad- pppp. In other words, Munich subjects completed the ka .gpgp story such that the transgressor expressed fggp, anxiety, or physiolggical symptoms following his transgres- sion or else, he evaded confronting his mother altogether. The majority of Benton Harbor subjects completed the ng pppp story such that the transgressor expressed guilt, shame, or sadness over his transgression. The differences between sex groups on the Ink Spot story were not statistically sig— nificant. However, inspection of Table 3 reveals that while males wrote stories revealing fggp and guilt or shame with ap— proximately equal frequency, about two-thirds of the female subjects wrote stories revealing guilt or shame and only one- third wrote stories revealing fggp. Although the hypothesis regarding differences between the sex groups was not sta- tistically supported, results were in the predicted direction. Findings: Hypothesis A II. This comparison involved a test to determine whether the two culture groups and the two sex groups differed with respect to Category II, Trans- gressorLg communication regarding transgression. Compari— sons were made for both the Lost Meat story and the Ink Spot story. This category was designed to record the 28 transgressor's communication to his mother regarding the fact of transgression before, during, and after any response by the mother. Hence, the entire content of the communi- cation regarding the fact of transgression was coded. Six mutually exclusive and exhaustive items comprised the category: Consistent lie Evades confronting and/or verbal communication Non-essential truth Inconsistent, truth after lie Consistent truth No information ampules- For further specification of this category and the criteria for truth in each of the stories, see the Coding Manual, Appendix A. .Low frequencies on some of the category items necessitated the combining of items into a truth—non—truth dichotomy for statistical analysis. Items 1-4 and item 6, no~information, were combined and considered as non-truth responses. Item 5 was considered the only type of response acceptable as ppyph. The following results were Obtained. Table 4 gives the frequencies of non-truth and ppgph, on Category II for the Lost Meat and the Ink Spot stories. The table indicates that although the obtained differences between sex groups were not statistically significant on either story, the differences between culture groups were significant on both stories. ~On the Lost Meat story, 72% of the Benton Harbor subjects wrote stories coded as 53332, while only 40% of the Munich subjects wrote such stories. 29 Table 4. Sex and culture group differences on Category II, Transgressor's communication regarding transgression. Story Group .N Truth NonLTruth Chi-square, df Ar2 Males 90 47 43 l The Females 97 56 41 n.s Lost “eat Munich 100 40 60 18 . 39* 1 B. Harb. 87 63 24 A—4 Males 90' 65 25 1 The Females 97 75 22 n.s Ink m Munich 100 65 35 10.00* 1 B. Harb. 87 75 12 n.s. p >.05 *p (.01.. 30 On the Ink Spot story, 87% of the Benton Harbor subjects wrote stories coded as truth, while only 65% of the Munich subjects wrote such stories. Findings: Hypothesis A III. This comparison in- volved a test to determine whether the two culture groups and the two sex groups differed with respect to Category III, Interpersonal context of communication. It was assumed that truth or lie is a process of relating, a verbal behavior oc— curring between two people in a situational context. This category focused on the interpersonal context of the trans- gressor's communication regarding the fact of transgression, i.e., the extent and nature of the mother's participation in the transgressor's communication. The following six types of situational contexts are listed in order of the amount of domination expressed by the prOperty owner during the child's communication. The Egg; ‘Mggp and the I k §pgp stories were each coded once and only once in one of these mutually exclusive and exhaustive items: (For further definition of these items, see Coding Manual, .Appendix A.) External pressure 'Internal and external pressure Internal pressure Information-seeking, fact finding Spontaneity No information. O‘U'I-bOJNI-J In analyzing this category, protocols revealingjyg information were not included. The infrequent occurrence of 31 this item rendered it of less interest, and not amenable to analysis by a chi-square statistic. All protocols revealing some information about the context of communication were coded into one of the forementioned items on Category III. The following results were obtained for the Lost Meat and the Ink Spot stories. Table 5 gives the frequencies of items 1 through 5 on Cagetory III, Interpersonal context of communication, for the Lost Meat and the Ink Spot stories. The table indicates that the Lost Meat story revealed significant differences between sex groups, while it did not reveal significant differences between culture groups. However, the signifi— cance of sex-group i differences is consistent with Hypothesis A III. Although more than k of both males and females wrote stories revealing spontaneity as the interpersonal con- text of communication, twice as many females wrote stories revealing internal pressure as did males. These results are consistent with the prediction of Hypothesis A III. On the Ink Spot story, the obtained frequencies were too small for the necessary chi-square analyses. Findings: Hypothesis A IV. This comparison involved a test to determine whether the two culture groups and the two sex groups differed with respect to Category IV, PrOperty owner's initial feelings. category IV recorded the property owner's initial feelings after communication with the child regarding the fact of transgression, or after learning of .Ho. cmsu wmma mass 32 HHmEm 00p VN «m v N ma on .Qnmm .m .183 canommxm mm Mm m 5 mm mm coacsz mafia. MGH HHmEm oou mN mm m 0 ON Nm mHMEmm .gmum Umpommxm wN vm m m ma Nw mam: vlfl w Hman.v co. m o v m No .Qumm .m we Ha w Ha m mm scans: .pmmz m e assmmmm.©a 00 NH . OH OH N em wHMEmm u oq mm c v m m hm mam: NIG mp mumsgwlaso muawcmugomm mgaxmwm madmmwnm ngnwuxm musmmoum msouw muoum .omcH HmcumugH nguwDCH Hmcnmuxm z I‘ iil.‘ .coapmoacgeeoo mo uxmucoo HmCOMHmQHoch .HHH muommumo so moogmummmap msoum mnsuaso paw xmm .m magma 33 the transgression from some other source. Such feelings could concern either a) the fact of transgression; b) the nature of the transgressor's communication, i.e., truth or lie; or c) both. The feelings need not have been overt or verbally communicated to the transgressor. The author only needed to state that such feelings existed, or state that the physiological symptoms of such feelings existed. The Lost Meat, the Ink Spot, the Damaged Axe, and the Soiled Dress stories were each coded once and only once in this category in terms of the following items: .Angry Sad Surprised, shocked Happy, glad Understanding, accepting No information O‘U'lnwal-J Item frequencies for the Lost Meat story were too small for chi-square analysis. In the analysis of results from the three remaining stories, frequencies in Items 3, surprised, and 4, hgppy, were combined for each group. No information items were deleted from this analysis and the following re- sults were obtained. Table 6 gives the frequencies of items 1 through 6 on Category IV for the Ink Spot, the Damaged Axe, and the Soiled Dress stories. The table indicates that on the Ink Spot story, the obtained difference between sex groups waS‘ not statistically significant while the difference between culture groups was. More Munich stories were coded in the 34 .mO. cmsu MMMH m .1 m O OH mm mm .Qumm .m m OO.m OH O m Om we sngsz .MMMflM 6 Ho m .mc.m 6H H OH mm mm mHmsmn c H. w m m m Om sq mHmz mum OH O OH Om OO .Qumm .m m Nm.¢ OH O NH mN mv SUHGSE mmfl. gmfim mm m ma cm as mHmsmm c a m *OH.hH m H m cm he me2 film m m mN OH hm .Quwm .m m L¢O.¢N OH MH H Om Ow soHcsz H mm . xgH m m 5H Hm OO meEwm m m¢.m OH m O NH He mHmS cud mp mHMDOMIHQO OCHumwoom wmmmm pmm mumgm m. msouu mnoum OCHpcmumHmOCD UmMHHmHsm .mmCHHmmm HmHDHcH m.umc3o >uawmoum .>H huommumo co mwocmuwmme msoum musuHso paw xwm O wHQMB 35 anger item than any other single item. rSimilarly, more Benton Harbor stories were codedzna sadness than any other single item. The expectation that more Benton Harbor stories than Munich stories would be coded in the under— standing, accepting item was not borne out in the obtained reSults. The table indicates that on the sibling stories, the Damaged Axe and the Soiled Dress, the obtained differences between sex groups and between culture groups were sta- tistically significant. The fact that the sex-group differences wmnxasignificant for the sibling stories and not significant for the parent-child stories could be a function of the nature of the incomplete stories in each series. The Series A stories involved parent-child conflict, while the Series B stories involved child-sibling conflict. The ob- tained results suggest that the sex groups did not differ with respect to their perception of parental feelings follow— ing transgression by a child, but fihat they did differ with respect to their perception of child feelings following transgression by a sibling. Table 6 indicates that on the Damaged Axe and the Soiled Dress, girls revealed higher fre- quencies of understanding, accepting initial feelings on the part of the property owner than did boys. There was little difference between boys and girls in the frequency with which initial feelings of anger were coded. 36 Findings: Hypothesis A V. This comparison involved a test to determine whether the two culture groups and the two sex groups differed with respect to Category V, Property owner's action/communication. Comparisons were made for all four stories. This category was designed to record the prOperty owner's behavior, i.e., action and/or communication toward the transgressor, after learning of the transgression, but before the transgressor's attempt to reconcile himself to the property owner. Only overt actions or communications which actually occurred or were indicated as being highly probable were coded in this category. The following mutually exclusive and exhaustive items, ranked in terms of the degree of domination suggested by the items, were included in this category: (FTn:further specifi- cation of category items, see Coding Manual, Appendix A.) Punishment by property owner Punishment, outside source Punishment, source unspecified Punishment and forgiveness by property owner No punishment Restitution by property owner without punishment Forgiveness without punishment Consolation, explanation, advice NO information 0 LomqoxmwaI-a A chi-square analysis of the data necessitated the combining of items 1 through 4 and calling them punishment. The re— maining items were kept separate and the following results were obtained. Table 7 gives the frequencies of items 1-4 combined punishment and compared with item 5, 6, 7, 8, and 9 on 37 .mO.Vnm¥ nH HH a O m mO 5O .Qnmm .m m «OO.mN O m H MH OH mO OOH QUHCSS mmmmm. O HO HH O O O h hm OO OHMEmm O H. O m OO.N OH O N O O mm OO me2 mum on m m m a as am .nan .m mmmf m st.NH OH O H OH O OO OOH sngsS mmmm. law! mH m O m 5 mm am mHmsmm a m O0.0 OH H O O O NO OO me2 Olm OH O O OH O mm NO .Qumm .m m O0.0 OH O O NN O OO OOH £0Hg52 mmmw. 5H HH 6 ma m am so mHMEmn McH m OOHm OH m O ON O NO OO mHmz O14 HH OH O NH OH Om hO .thm .m m O0.0 ON HH N O OH OO OOH QUHGSS ummz mH NH m a OH as am chamm anon m OO.H OH O N 5 MH mO OO mHmz NLm mp mumsvm COHumE COHumH mmmc COHusu Dame ucwE m. -msouw huoum IHQU nuomcH, I0mgoo Iw>HOHom IHumwm ISMHgsm uschsm oz oz .COHumoHcsEEOU \coHuom m.umc30 >DHOQOHO .> whomwumo co chMHHmQEOU QSOHO OHSUHDU Ogm xmw .5 mHQMB 38 on Category V for all the stories. The obtained differences between sex groups were not statistically significant on any of the stories. However, differences between culture groups were statistically significant for the sibling stories. On both the Damange Axe and the Soiled Dress stories, a higher proportion of Munich subjects wrote stories coded as punish— ment than Benton Harbor subjects, and a higher proportion of Benton Harbor subjects wrote stories coded as consolation than Munich subjects. While the results from the sibling stories did support Hypothesis A V, the completions of parent—child stories did not. Because of the plot differ- ences between the two pairs of stories, it would not be ex— pected that cultural differences manifested on one series would necessarily be manifested on the other series. -The frequency method did not reveal cultural differences on this category when the conflict depicted in the story was between parent and child. Hypothesis A VI: Findings. This comparison in— volved a test to determine whether the two culture groups and the two sex groups differed with respect to Category VI, Transgressor's affective and/or behavioral response. This category was designed to record the transgressor's feelings regarding: l) the fact of his mistake; 2) the manner or content of communication with his mother; or 3) the thought of punishment after communication or action by the mother. 39 The following mutually exclusive and exhaustive items ranked in terms of the amount of submission suggested by the items, were included in this Category: (For further specification of category items see the Coding Manual, Ap- pendix A.) 1. Angry 2. Submissive 3. Shameful, self-critical 4. Voluntary promise to reform, spontaneous restitution without pressure Relief, tension reduction, hapiness No information mm The results obtained were as follows. Table 8 gives the frequencies of Item 1 through 6 on Category VI for the Lost Meat and the Ink Spot stories. The obtained differences between sex groups and between culture groups were not statistically significant on either story. The results of this frequency analysis do not support Hy- pothesis A VI. 40 .moswum u .OHM .c m .OHm.c Om N NH OH ON H NO .snmm .O NO O OH s OH O OOH OOHcss mmmw. Ham. HO O OH OH ON N am anmsmm m .OHm.c OO O NH s OH O OO mmHmz Ola m .mHu.c ON O ON ON O O NO .numm .O OO a ON ON O O OOH HOHOOS Haws wO Om a ON NN OH O NO anmamm H H m ON.N ON O OH NN O HH OO memz NLO HO mumswm coHme OOOCHmmmm EHOOOO H5O m>HmmHE MHOG< m. macaw OHODO IHso JHOOGH humucsHo> IwEmsO IQOO oz .mmgommmn HOH0H>O£OQ \M>Huommmm m.H0wmeOmcmHB .H> OHOOODOU co mmocmummmHO ODOHO mudesu Ocm Xmm .O mHQMB 41 Discussion: Hypothesis A l. The prediction that Munich completions would show more instances of fear, physio- logical symptoms of anxiety and evasion of confronting on Category I, Transgressor's initial feelings, while Benton Harbor completions would show more instances of guilt, shame, and sadness was not confirmed for A—2, the Lost Meat story. However, this hypothesis was confirmed for A-4, the Ink Spot story. In order to understand why Story A—2 did not reflect cultural differences while Story A—4 did, it is profitable to examine both stories more closely. In the Lost Meat story, Michael puts down his package of wieners, plays and a dog eats the contents of his package. In other words, Michael is only indirectly responsible for the transgression. However, in the Ink Spot story, Susan is directly responsible for getting ink spots on her mother's coat. This difference between the two stories, i.e., that the Lost Meat story in— volves indirect responsibility and the Ink Spot story in- volves direct responsibility for the transgression, may ac- count for the presence of cultural differences on the ka .§pgp and the absence of differences on the Lost Meat:story. Both culture groups expressed a predominance of anxiety, fggp, and evasioncni the Lost Meat story. In other words, it may be that neither Munich nor Benton Harbor authors judged their own behavior as "bad" when the responsi- bility for the transgression was only indirectly theirs. 42 Hence, both groups responded to the mother's probable valu— ation of the transgression and expressed feelings about the fact of confronting rather than about the fact of trans- gressing. However, shown the responsibility for transgres— sion was direct, as in A—4, the Ink Spot story, cultural differences were reflected and significantly more Benton Harbor children than Munich children wrote stories where the transgressor engaged in self—evaluative behavior and ex- pressed feelings about the fact of transgression rather than the fact of confronting. On the other hand, more Munich children wrote stories where the transgressor responded to his mother's evaluation of his behavior and expressed feel- ings about the fact of confronting rather than the fact of transgressing. In terms of conscience development, the obtained cultural differences suggested that in situations where the transgressor was directly responsible for his transgression, Benton Harbor stories reflected more self-evaluative feelings, indicating the activity of conscience. »Munich stories re- flected feelings where the evaluation of the parent was pre- dominant. Such feelings indicated the activity of a primi- tive conscience, based on the fear of authority. Discussion: Hypothesis A II. Hypothesis A II pre— dicted that on Category II, Transgressor's communication regarding transgressions, Benton.Harbor~Stosiesawenld show more instances of truth than Munich Stories, while 43 0 Munich stories would show more instances of deception or no information. The results confirmed this hypothesis on both the Lost Meat and the Ink Spot stories. These results suggest that honesty in communication, i.e., the whole truth and nothing but the truth, as defined in the Coding Manual, was more characteristic of parent-child relations in Benton Harbor than in Munich. That this differ- ence was reflected regardless of whether the story depicted a transgressor who was directly or indirectly responsible for his transgression suggests, further, that the honest communication between parent and child was more highly valued in Benton Harbor than in Munich. In addition, the results suggest that this difference was not just a reflection of the way children responded to their parents when they felt responsible for having erred and wished to display their subsequent good intentions through honesty. When these results are considered in relation to the interpretation of Hypothesis A 1 results, it seems to follow that the Munich child, who was more concerned about his mother's valuation of his behavior, would attempt to "juggle the facts" so as to avoid her disapproval. The Benton Harbor child, on the other hand, who responded to his own value system rather than to his mother's, would, as explained in the Coding Manual, tell the truth about his transgression rather than engage in deception. He would thus avoid his mother's disapproval but heighten his disapproval of himself. 44 Although Hypothesis A II also predicted that girls would show more instances of truth than would boys, the results did not confirm this hypothesis in either the Lost Meat or the Ink Spot stories. Discussion: Hypothesis A III. Hypothesis-A III pre— dicted that both the Munich group and the male group would show more instances of external pressure than the Benton Harbor and female groups which would show more instances of internal pressure and spontaneity. Only the comparison of sex groups on the Lost Meat story confirmed the hypothesis. Females showed more instances of reflective communication indicated by delay, thinking about what to say, etc., than males. However, since the results of Hypothesis A I and A II revealed no significant differences between the sex groups, this reflective communication cannot necessarily be in- terpreted as indicative of conscience and/or self-evaluative behavior. The interpretation of these results is difficult since this category separates the content of communication from the context. Therefore, an investigation of the re- lationship between Category II, Transgressor's communication ,regardingjtransgression,'and Category III, Interpersonal context of communication, for sex and culture groups will be reported in a later section of this paper. Discussion: Hypothesis A IV. Hypothesis A IV pre- dicted that on Category IV, Property owner's initial feelings, 45 Munich stories would show more instances of anger and sadness than Benton Harbor stories which would show more instances of understanding and acceptance. The results confirmed this hypothesis on the Ink Spot story, but not on either of the sibling stories. rAs predicted, anger was the predominately coded feeling in Munich completions of the Ink Spot story. However, Benton Harbor story completions showed a predominance of sadness on this story rather than understanding and ss- ceptance as Hypothesis A IV predicted. On the Ink Spot story, Munich children described a mother who typically expressed her anger toward the trans— gressing child. The Munich transgressor was punished by his mother's anger for his misdeeds. The Benton Harbor trans- gressor, on the other hand, punished his mother by making her sad over his misdeeds. The Benton Harbor transgressor experienced the full impact of his behavior not through his mother's angry devaluation of that behavior, but through his own devaluation of the behavior which led to her sadness. The predominance of 22325 in Munich stories suggests that the evaluation of the child's behavior was dominated by the mother. The predominance of sadness in Benton Harbor stories suggests that a good portion of the evaluation of the child's behavior was left up to the child himself, the mother merely reflecting how this behavior affected her. Thus, the amount of dominance attributed to property owners in each culture group seem consistently related to the amount of 46 self—evaluative behavior attributed to the transgressors in each culture group. In fact, the behaviors attributed to property owners can be interpreted as facilitating or in— hibiting the develOpment of mature conscience, i.e., a moral orientation based on internalized values. Although no specific hypothesis regarding sex group differences on Category IV was proposed, statistically sig— nificant differences between sex groups were manifested on each of the child—sibling stories. Females consistently ex- pressed more understanding and acceptance toward their sib— lings than did males. That no significant differences be— tween the sex groups were manifested on the Ink Spot, the parent-child story, is not surprising._ Girls of twelve, the age of most of the authors, are already beginning to adopt "maternal behavior" toward their siblings. The fact that significant differences between sex groups were found on the sibling stories but not on the parent-child story, suggests that there was no significant difference in the way males and females described the feel- ings of their mothers although there was a difference in the way males and females treated their siblings. Discussion: Hypothesis A V. Hypothesis A V pre- dicted that Munich stories would show more instances of ppnishment than Benton Harbor stories which would show more instances of forgiveness and consolation. The obtained re— sults supported this hypothesis on the two sibling stories. 47 Analysis of the parent-child stories did not confirm this prediction. No differences between sex groups were hypothe- sized and no statistically significant differences were found. These results suggest that more Munich stories de— scribed punitive sibling relations when the younger sibling has transgressed than did Benton Harbor stories. However, there was no support for the hypothesis that parent—child relations as described by Munich authors, were more punitive than those described by Benton Harbor authors. Although cultural differences revealed in the area of sibling re- lations need not necessarily be present in the area of parent— child relations, previous analyses of the Anderson data (Anderson, 1959; Terdal, 1961) have found that munich stories contained more instances of punishment than did U.S.rstories. That the present study encountered no such differences which were statistically significant between culture groups, sug- gests that the frequency method may be insensitive to the nature of the cultural differences in this area. That is, Munich mothers may not necessarily have punished more fre- quently but they may have punished for different reasons than did Benton Harbor mothers. Further investigation of the context in which punishment occurred may reveal cultural differences. Such an investigation will be reported in Section III of this paper. 48 .Discussion: Hypothesis A VI. Hypothesis A VI pre- dicted that Munich completions and male's completions would show more instances of anger and submission on Category VI, Transgressor's affective behavioral resppnse, than Benton Harbor and female completions. The latter two groups would show more instances of shame and voluntarygpromise to reform. Analysis of the Lost Meat and the Ink Spot stories did not support these hypotheses. However, this category excludes the Property owner's action/communication, i.e., Category V, to which Category VI is, by definition a response. It is not surprising, therefore, that no culture or sex group differences were revealed when only the Category VI item frequencies were considered. .A comparison of sex and culture groups on the relationship between Categories V and VI would be a more appropriate test of this hypothesis. Such a comparison will be carried out in Section III of this paper. III. COMPARISON OF GROUPS ACCORDING TO THE CONTEXT IN WHICH CATEGORY ITEMS OCCUR The preceding results were generally consistent with the hypothesized differences between sex and culture groups. Three exceptions occurred on Categories III, Interpersonal context of communication, V, Property Owner's action/ communication, and VI, Transgressor's affective/behavioral response. However, the frequency method, whereby each cate- gory is considered in isolation from other categories is in- herently insensitive to context effects, i.e., the relation- ship between a particular response and the interpersonal context in which it is embedded. This section of the present study proposes that important cultural and sex group differences in conscience development which are obscured by the frequency method are revealed when category items are analyzed in terms of their relation;to another relevant cate- gory. More specifically, it is prOposed that each of the content categories is psychologically related to the others. That is, what the child tells the property owner (Category II) is related to how he feels about what he has done (Cate- gory I). Similarly, the interpersonal relationship occurring between child and property owner (Category III) is related to what the child communicated about his behavior (Category 49 50 II). The property owner's response, affective and behavioral (Categories IV and V) is related to the child's communi- cation (Category II).This, in.turn, affects the child's final response (Category VI). -Although numerous inter— relationships could be examined in this section of the present study, the hypotheses formulated here follow from the preceding section. Procedure. The method described in Section I of this paper was followed in the present analyses, except that in this case, sex and culture groups were compared in terms of the relationship between two categories. Chi-square an— alyses tested the hypothesis of the independence of the two categories in question. Hypotheses. The following hypotheses are an out- growth of the inconclusive results in Section I of this paper: B I. Culture and sex group differences will be re- vealed when responses coded in Category III, Interpersonal context of communication are re— Iated to Category II, Transgressor's initial communication. B II. Culture and sex group differences will be re— vealed when responses coded in Category V, Property owner's actigp/communication are re- lated to Category II, Transgressor's initial communication. B III. Culture and sex group differences will be re- vealed when responses coded in Category VI, Transgressor's affective/behavioral response are related to Category V, Property owner's action/communication. 51 Findings: Hypothesis B I. Stories by each sex and culture group were dichotomized according to Category II, Transgressor's communication regsrding transgression. Stories coded in items 1—4 were called deception while stories coded in item 5 were called truth. Each of these groups was further subdivided into the following four sub- groups, based on Category III, Interpersonal context of com— munication, stories coded in item 6, spontaneity; items 4.+ 5 information seekipg; item 3, internal pressure; and items 1 + 2, internal and external pressure plus external pressure. Stories coded as no information in Category III were ex- cluded from this analysis. An eight celled chi-square table was set up to test the relationship between ppgph and deception on Category II and the four groups on Category III. Such an analysis was set up for each sex group and for each culture group for the Lost Meat story. An identical procedure was followed in setting up the analyses of the Ink Spot story. The follow- ing results were obtained. Table 9 reveals that the null hypothesis, i.e., that Category II and III are independent, can be rejected in the case of each sex and culture group for the Lost Meat and the Ink Spot stories. Further inspection of Table 9, however, suggests that there was only one significant difference be— tween sex or culture groups. On the Lost Meat story, females showed more internal pressure following deception than males, Hmcumuxm I .uxm Hmcumch I .ucH coHummowQ I .an GOHHOEOOGH I .oanH OOOOI... he #22. O O O H OH H HO ON OO .fimm .m an Ear... NH O H N OH O ON ON OO €252 OIOOO. C mo «moi. O O HH H NH O ON ON OO mHms x H flOO OOH. a O O N OH H HO ON NO mHmsmm One OOOOOOH O N H OH H O O OO NO .anmm .m mg #3.: OH O O OO O H O OO OO H0352 Hmoz, m , O... 16...: HH N H ON O N O OO O OHmz O OH nu O. O.NN HH . O O OH H! O O OO OO 33% Num MO mumswm OHSOOOHO madmmmnm OOHMme mpHmcmp wusmmwum muswwwnm.Ocwa®O mpHmcmu m. msouw muoum -Hso .uxm .ucH .oOcH Icomm .uxm .ucH .omcH Icomm, O .cog O .omn O .omn O .osg. O Hague O Hanna O cusua O spans .COHDMUHGOEEOU mo uxmucoo ngoOHmmwwucH .HHH Osm .conmeOwcmHu Imcmua .HH mmHHOprmo cwwzpwn mHsmcoHumHmH man go OCHOHOOOH coHHmoHcsEEoo m.H0mm0HO mmongOOMHO QSOHO OHSDHOU Ocm wa .O anma 53 X2 = 9.00, p less than .05, df=3. Although the data sup- ported the hypothesis that these two categories were not independent, the data did not support the hypothesis that sex and culture group differences would be manifested in terms of this relationship. Findings: Hypothesis B II. The results of previous research led to the conclusion that Munich children's com— pletions contained fewer instances of truth in Category II, Transgressor's communication regardinggtransgression, and more instances of punishment as in Category V, Property owner's action/communication, than American children's stories. What is the relation between truth and punishment in the Munich stories? The following chi-square analyses investigated the relationship between these two categories on the parent-child stories. On Category II, Transgpessor's communication regarding transgression, the combination of items 1-4 and item 6 were considered deception, while item 5 was considered truth. On Category V, Property owner's actiop/ communication, items 1-4 were considered punishment, while items 5-8 were considered deception. The following results were obtained. Table 10 suggests that there was some dependency be- tween truth and punishment on both parent-child stories for the Munich sample. No such statistically significant re— lationship is suggested by the Benton Harbor sample. Further inspection of Table 10 reveals that on the Lost Meat story 54 OO.AQ ma NO.vnH.. H mamm. O O OO Nm .Qnmm .m pomm xcH H «HO 5 ON O ON hm £0HC52 OLfl H msOO. HH OH NO ON .Qnmm .m ummz #mOH H OON.O Om HN OH ON QUHCDE NLN pcwfiancsm OO OHODOO oz ucmEschsm acmeancsm pcmESOHgsm msauw muoum HQU O COHpmmomQ O COHummowQ oz O Lynne O SDDHB .coHumoHOOEEOU\COHDUm m.Hm230 Nuummoum .> OHOOODOO Ocm qOOHmmeUmcmHDIOCHOHmOOH coHumoHcsEEoo m.HowmeO. ImcmHB .HH muommuwu cmm3umn QHAOOOHHOHOH map so mmugmuwmmHO QUOHO OHSDHSO .OH OHQOB 55 62% of the Munich completions coded as truth also contained themas of punishment, while only 35% of those coded as_gs- ception also contained themas of punishment. The same pattern was also revealed on the Ink Spot story with 57% of those stories coded as truth also containing punishment themas and only 26% of those stories coded as deception con— taining punishment themas. Thus, there appear to be differ- ences between the culture groups not in the frequency with which punishment themas appear but in the conditions under which punishment occurs. Table 10 suggests that Munich children were more likely to be punished if the told the truth and less likely to be punished if they did not. This is consistent with an earlier report by the Andersons (1954) for Karlsruhe children. Findings: Hypothesis B III. Completions by each sex and culture group were dichotomized according to Cate— gory V, Property owner's action/communication. Stories coded in items 1-4 were called punishment while stories coded in items 4—8 were called no punishment. Each of these groups was further subdivided into the following six sub— groups according to Category VI, Transgpessor's affective/ behavioral responseO stories coded in items 1 + 2, anger and submission; item 3, shame and guilt; items 4 + 5, volun- tary promise to reform and happiness. Stories coded as pg information on Category VI were deleted from this analysis. 56 A six-celled chi—square table was set up to test the relationship between_punishment and no punishment on Category V and the three item groups on Category VI. Such an analysis was set up for each sex group and for each culture group on the Lost Meat story. .An identical procedure was followed in setting up the analyses of the Ink Spot story. The follow- ing results were obtained. Table 11 reveals that on the Lost Meat story, the null hypothesis, i.e., that Category V and VI are inde— pendent, can be rejected for males. On the Ink Spot story, the null hypothesis can be rejected for males and Munich authors. The male group expressed a predominance of spgss and submission following punishment and a predominance of shame and voluntary promise to reform following pg punishment. The Munich group, as well as the male group, mani- fested a similar trend on the Ink Spot story with one interesting variant. -Although voluntary promise to reform was more frequent following no punishment, shame was much less frequent under this condition. The female and Benton Harbor groups showed more frequent instances of voluntary promise to reform following no punishment, but there was no difference in the shame response under the two punishment conditions. 57 ucmfischsm I .schsm humucsHO> u pc5H0> .OO. cmsu mmmH ms O N xonmmm OH O HH O O OH NO .nnmm .m m «O0.0H OH H OH O O NH OO soHcsz uomm MGH m *O0.0H OH H O O O NH OO me2 O O.H xonmmm NH O OH O O OH NO mHmsmO OIO O O xoummm OH OH O HH O O HO .nnmm .O O O xonmam OH OH O OH OH OH OO HOHOOZ Hams OO O OOO.O OH HH O O OH OH OO mHms n H m H cmsu mme OH OH O OH O O OO mHmEmm NLm mmmchIuHst I. OO OHOOOOIQOO Ho Ho conmHE OOOCHmmmm z QOOHO muoum IHSUOEHOMOM OEmflm lflsm HO HO #HHSO COHmmHE .Ho> O st Hmmcd O Euowwm Ho IQDO Ho O .smH Icsm .schsm p25H0> O mamsm O mecd O Icsm 02 oz .ancsm .ancsm .ancsm OZ .mmcommmu HOH0H>mst \m>Hpomme OHHOOOOHOOCOHB .H> Odm .OOHumUHCSEEOU\moHuom O.Hmckm.>uumaonm .> mmHuommumU cmmzuwn OHSOCOHHOHOH map CH mmocmummmHO OSOHO musuHso Ocm xmm .HH mHQmB 58 Discussion: Hypothesis B I. This hypothesis pre- dicted that sex and culture group differences would be re— vealed when responses coded in Category III, Interpersonal context of communication, were related to Category II, Trans- '_gpessor's communication regarding transgpession. Although a chi-square analysis of the data revealed that these two categories were related, i.e., they were not independent, the analysis did not reveal any differences between sex and culture groups in the nature of the relationship. In every case, there was a predominance of spontaneity in interpersonal relations under the truth condition. However, there was a noticeable drop in spontaneipy under the deception condition and an increase in the external pressure type of relating. That this trend appeared in every group for each of the parent-child stories suggests that there is no sex or culture related difference in the kinds of changes that oc- curred in the mother—child relationship following trgth or deception on the part of the child. These data suggest that in all groups, mothers took a more active, dominant role in interacting with their children following deception than they did following honest communication from the child. Discussion: Hypothesis B II. This hypothesis pre— dicted that sex and culture group differences would be re- vealed when responses coded in Category V, Property owner's action/communication, were related to Category II, Trans- gressor's communication regarding transgression. A 59 chi-square analysis of the data supported this hypothesis, and revealed that there was some dependency between truth and punishment in the parent—child stories by Munich sub- jects, while such a dependency could not be inferred in the stories by Benton Harbor subjects. These data suggested that Munich stories showed a predominance of punishment following truth about the trans— gression, while they showed a predominance of no_punishment following deceppion by the child. Table 10 suggests that Munich children were more likely to be punished if they told the truth and less likely to be punished if they deceived their mothers. The fact that Munich stories showed a predominance of deception in communication between mother and child follow— ing transgression was not at all surprising in the light of the preceding results. It appears that Munich children were rewarded for deception and punished for honesty. This con— clusion further substantiated the discussion of Hypothesis A II in Section I of this paper which concluded that there were cultural differences in attitudes toward honesty in com— munication. It was previously concluded that these attitudes were reflected in the behavior of children, i.e., that more Munich children attempted to deceive their mothers than did Benton Harbor children. The present results suggest that these attitudes were also reflected in the behavior of the mother toward her child, i.e., that Munich mothers rewarded 60 their children for deception and punished them for honesty. No such relationship was discovered in the Benton Harbor story completions. Discussion: Hypothesis B III. Hypothesis B III pre- dicted that sex and culture group differences would be re- vealed when responses coded in Category VI, Transgressor's affectivelbehavioral response, were related to Category V, Property owner's action/communication. These data supported this hypothesis for sex groups on both parent—child stories. In each case, males showed an increase in voluntary promise to reform following the nogppnishment situation. The interpretation of this result can be facilitated by further inspection of the definition of voluntary promise to reform presented in the Coding Manual, Appendix A. This behavior is characterized either by its spontaneity, i.e., that it was offered by the child, himself,without pressure from the mother, or by its silence. In either case, the behavior is interpreted as a genuine wish on the part of the child to change his behavior following some "insight" gained through interacting with the mother or on his own. The item is not to be checked if the child's promise can in any way be con— strued as an attempt to placate his mother. Hence, this be— havior can be interpreted as resulting from the child's self— evaluation of his own behavior and his wish to change that behavior. ,Since this study has already presented self- evaluative behavior as indicating the activity of mature 61 conscience, the present results can be interpreted such that males showed a greater proportion of responses indicating the activity of conscience following no punishment than they did following punishment. Females did not show this differ— ential kind of response to the punishment condition. -In fact, females showed a predominance of voluntary promise to reform under both punishment conditions. Differences between the culture groups were revealed only on the Ink Spot story. Munich stories suggest a re- lationship between the punishment condition and Category VI, Transgressor's affective/behavioral response, similar to the male group. However, in addition to the increase in volun- tary promise to reform following no punishment there was a drOp in shame and guilt responses under the no punishment situation. The Benton Harbor stories did not show this drop in shame and guilt under the no punishment situation. Al- though shame and guilt can be considered as conscience re- sponses, they are of a less constructive nature than volun— psry promise to reform. The former are self destructive and have no socially valuable outcome while the latter are con— structive to the child and his behavior. Thus, although the Munich stories revealed that conscience was active under both the ppnishment and_pp punishment conditions, the latter condition appears to be more conducive to constructive, mature conscience responses which have socially valuable outcomes. 62 Further discussion of the nature of the differences between culture groups must of necessity be conservative. .Since protocols coded as no information on Category VI,~Trans— _gressor's affective/behavioral response, were excluded from the present analysis, there is no assurance that the number of male and female protocols in each culture sample were equal. Since there were differential sex effects on these two variables, the obtained culture effects may be*a re- flection of a different porportion of males in each sample rather than a true reflection of cultural differences. However, one additional comment should be made. Since more Munich authors wrote stories where the child de- ceived his mother and escaped punishment, the increase in porportion of voluntary promise to reform responses under p9 punishment must be explained further. It appears that the Munich author's transgressor lied out of the fear of con- fronting his mother with the truth, since he was more likely to be punished in the latter case. Having escaped punish- ment and his mother's devaluation of his behavior, he was left to consider his behavior in terms of his own value system. Hence, the pp ppnishment condition can be seen to facilitate self-evaluative behavior on the part of the child. However, if the Munich transgressor confronted his mother with the truth and she punished him, he experienced devalu— ation in his behavior from his mother, i.e., he was not left to his own self-evaluative activities. Hence, the greater 63 proportion of guilt and shame reflected self—devaluation following devaluation by the mother and was a more submissive, unconstructive type of response than voluntarygpromise to reform. General Discussion. .The results of this series of analyses suggest that valuable information about cultural differences is reflected when groups of categories are con- sidered in a contextual relationship. The general hypothe— sis that the frequency count method, wherein each category is considered in isolation from the others, obscures many subtle cultural differences, is borne out in this section of the present study. IV. SUMMARY AND CONCLUSIONS The Anderson Incomplete Stories, a projective tech- nique, were developed to elicit children's perceptions of social conflict situations. These stories were administered under the supervision of the Andersons to over 10,000 children in eight countries. The present study randomly selected and analyzed 100 of the 350 protocols obtained by the Andersons in their 1954 sample of Munich, Germany school children and 87 of the 299 protocols obtained in the 1957 sample of Benton Harbor, Michigan school children. Completions to four of the eleven Anderson Incomplete Stories were utilized in this report: A—2, the Lost Meat -Story; Ar4, the Ink Spot story; B-4, the Damaged Axe story; and B-5, the Soiled-Dress story. Each of these stories de- picts a hypothetical transgression situation wherein one character (the transgressor) has committed or been involved in an act of carelessness and/or disobedience, a loss of or damage to the property of another (the prOperty owner). The purpose of the present study was to utilize con- tent analyses of the completions to these four stories in order to investigate cultural differences in the following areés: 64 65 l. the level of conscience development revealed by the actions and thoughts attributed to the transgressor 2. the relative amount of domination or egalitarianism depicted in the interaction between transgressor and property owner 3. the relationship between the level of conscience manifested by the transgressor and the amount of domination or egalitarianism depicted in the trans— gressor-property owner interaction. Although conscience development can be placed on a developmental continuum, the present study considered only the relative extremes of this continuum, i.e., the primitive, immature type of conscience based on the fear of authority and dependent upon the immediate environmental consequences cxf behavior versus the more mature type of conscience based on internalized values and independent of the im- mediate environmental consequences of behavior. To facili- tate the content analysis of story completions in terms of these two levels of conscience development, the Coding Manual contained three content categories which contained items de- scribing various behaviors from which the relative maturity of conscience could be inferred. The three content cate- gories designed to assess the level of conscience development are as follows: I. Transgressor's initial feelings II. Transgressor's communication regarding transgression VI. Transgressor's affective/behavioral response 66 Three additional content categories focused upon the relative amount of domination or egalitarianism present in the transgressor-property owner interaction: III. Interpersonal context of communication IV. Property owner's initial feelings V. Property owner's action/communication The mutually exclusive and exhaustive items comprising these three categories were ranked in terms of the relative presence of domination or egalitarianism. Only the first two of the four stories analyzed in this study, the Lost Meat and the Ink Spot stories, were coded in each of the six categories above. The sibling stories, the Damaged Axe and the Soiled Dress,were coded in only Category IV, Property owner's initial feelings9and Cate- gory V, Property owner's action/communication. Hypotheses. The present study hypothesized that the culture group containing a predominance of dominance— submission themas would also show a relative immaturity of conscience, i.e., a moral orientation based on fear of authority. Similarly, the culture group containing a pre- dominance of egalitarian themas would also show a relative maturity of conscience, i.e., a moral orientation based on internalized values. Furthermore, it was hypothesized that cultural differences in conscience development would be most clearly revealed when context effects were considered. 67 In other words, the transgressor's response must be con- sidered in relation to the interpersonal context in which it occurred before subtle differences between the two culture groups can be assessed. Findings. Chi-square tests assessed the relationship between cultural origin and level of conscience. Of the three content categories designed to assess the level of conscience, two of the categories (Category I, TransgressorLs initial feelingsjand Category II, Transgressor's communication regardisg transgression) revealed cultural differences on at least one of the two parent—child stories considered.‘ In both cases, Munich completions reflected a preponderance of responses indicating a primitive or immature level of con- science, based on a fear of authority. The majority of Benton Harbor completions, on the other hand, reflected a mature level of conscience,based on internalized values. Cultural differences in the amount of domination or egalitarianism expressed in the transgressor-property owner interaction were also assessed by chi-square tests. Of the three content categories designed to assess domination or egalitarianism, two of the categories (Category IV, Property owner's initial feelings, and Category V, Pr0perty owner's action/communication) revealed cultural differences on at least one of the four stories considered. In each case where significant differences were obtained, Munich com- pletions reflected significantly more domination than did Benton Harbor completions. 68 Cultural differences in the relationship between level of conscience (primitive versus mature) and transgressor—prOperty owner interaction (domination versus egalitarianimn)were also assessed by chi-square tests of significance. The relationship between three specific cate- gory pairs was tested: Category II, Transgressor's communi- cationgand Category III, Interpersonal Context of communi- cation; Category II and Category V, Property owner's actiop/ communication; and Category VI, Transgressor's affective/ behavioral response. Only two category pairs, II and V, and\7 and VI, proved to be differentially related in the two culture groups. In each case, however, Munich completions showed a dependence between the transgressor-property owner inter— action and the level of conscience development, while Benton Harbor completions showed no such dependence. Maturity of conscience was originally defined as a moral orientation based on internalized values, independent of immediate environmental consequences. That Munich com- pletions revealed a dependence between transgressor's re— sponse and the amount of environmental domination supports the previous conclusion that Munich completions revealed greater immaturity of conscience, i.e., their responses were dependent upon the immediate environmental consequences. Be- cause Benton Harbor completions did not reveal subh a de- pendence, i.e., because their responses were not dependent upon the immediate environmental consequences, we conclude that they revealed greater maturity of conscience. 69 Furthermore, the nature of the inter—category de- pendence revealed in the Munich completions gave additional support and clarification to this study's major hypothesis. 0f the two category pairs which proved to be differentially related in the two culture groups, both pairs supported the hypothesis that parent-child relations characterized by domination and submission, foster a dependency relationship which hinders the development of independent, internalized value systems, i.e., maturity of conscience. REFERENCES Anderson, H. H. and Anderson, Gladys L. Children's per- ceptions of social conflict situations: a study of adolescent children in Germany. Am. J. Ortho., 1954, .23 (2), 246-257. .Anderson, H. H. and Anderson, Gladys L., Cohen, I. H. and Nutt, F. D. Image of the teacher by adolescent children in four countries: Germany, England, Mexico, United States. J. Social Psychol., 1959 s9, 47-55. Anderson, H. H. and Anderson, Gladys L. Creativity and cross—national research: a comparison of children's values in Munich, Germany in 1954 and 1960. Pre- sented at the Biennial Meeting, Society for Research in Child Devpm., Penn. State Univer., 1961. Anderson, H. H. and Anderson, Gladys L. .A cross-national study of children: a study in creativity and mental health. In Gordon, I. H. (Ed.), Human Development Readings in Research. N.Y.: Scott, Foresman and Co., 1965, 307—315. Erickson, D. J. .A comparison of children's social values in story completions by culture groups. M.A. thesis, Michigan State Univ. Library, East Lansing, Michigan. 1961. Fenichel, O. The psyphoanalytic theory of neurosis. N.Y.: W. W. Morton and Co., 1945. Geierhaas, F. G. Problems of reliability in evaluating story completions about social conflict by German adolescent children. M.A. thesis, Michigan State Univ. Library, East Lansing, Michigan. 1955. Hoffman, L. L. Child rearing practices and moral develOp— ment: generalizations from empirical research. Child Develpm., 1963, s3 (2) 295-318. Kohlberg, L. H. Moral development and identification. In H. W. Stevenson (Ed.), Child psychology. Yearb. nat. Soc. Stud. Educ., 1963, Q2, Part II. 70 Levitov, -Terdal, 71 Edith S. The broken bicycle, a comparative study of story completions by adolescents in United States and German cities. M.A. thesis, George Washington Univ. Library, washington, D.C., 1960. L. G. Consistency in use of themas in story com- pletions: a cross national study in seven countries. M.A. thesis, Michigan State Univ..Library, East .Lansing, Michigan, 1961. APPENDIX A CODING PMNUAL Anderson Incomplete Stories A-2, A-h, 5-4, and B-5 Introduction and Purpose of the Manual. This manual has been designed for the content analysis of four of the eleven Anderson Incomplete Stories, A-Z, the Lost Meat story; A-b, the ng'Spot story; 8-6, the Damaged Axe story; and 8-5, the Soiled Dress story. It is assumed that the stories are projective in nature, i.e., that the completion reveals, to some extent, the thoughts, feelings, values, and expectations of the child author. Each of the four stories considered in this manual depicts a hypothetical tranSgression situation. In each case, a child character ( the transgressor ) has committed or been involved in a transgression, an act of carelessness and/or disobedience, a loss of or damage to the property of another ( the property owner ). In the first two stories, the property involved belongs to the mother. Hence, the Lost Meat and the Ip§_Spot stories are referred to as the parent-child stories. In the second two stories, the Damaged Axe and the Soiled Dress, the property involved belongs to an older sibling. Thus, these two stories are called the sibling stories. This Coding bsnual is designed to assess the author's re- sponse to a hypothetical transgression situation in terms of conscience development, and in terms of patterns of paront-child~relating following transgression. Given this focus, the CodingAPhnual analyzes three molar phases of the interaction between property owner and transgressor: l) the transgressor's initial feelings and communication regarding the transgression; 2) the pro- perty owner's initial feelings and action/communication after learning of the transgression; and 3) the transgressor's response to the property owner's feelings and action/communication. These three phases are then sub-divided into six content categories: I. Transgressor's initial feelings II. Transgressor's communication regarding transgression III. Interpersonal context of communication IV. Property owner's initial feelings V. Property owner's action/communication VI. Transgressor's affective/behavioral response Categories 1, II, and VI are designed to assess the level of conscience attributed to the transgressor by the author. In general, a response to transgression based on the fear of authority suggests an immature or primi- tive level of conscience. 0n the other hand, a response to transgression based on internalized values, independent of the immediate, environmental con- sequences, suggests a mature level of conscience. Categories III, IV, and V were designed to assess the relative amount of domination-submission or egali- tarianism present in the relationship between proporty owner and transgressor. A-Z Certain assumptions about the nature of interpersonal communication are implicit in the organization of the Coding_}mnual. It is assumed that communication is not a linkage of stimuli, reSponses, response-produced stimuli, etc. Rather, it is a dynamic, continually evolving process. Thus, the author's description of the transgressor's communication must be con- sidered in the context of 1) the immediate, observable responses of the pro- perty owner, 21 the set of expectancies, accrued through pash communications, regarding the probable reSponse of the property owner, 3) the transgressor's feelings and valuations of himself and the property owner, and 4) the pro. perty owner's feelings and valuations of himself and the transgressor. In order to use the aunnanls description of communication as a basis for infer- ences about conscience development, the assumed, processive-contextual nature of communication must somehow be preserved in the process of content analysis. The present manual attempts to do just this by considering the-content cate- gories sequentially, and in relation to one another. General Instructions. The Coding'Mnnuar consists of six content categories designed to assess the relative amount of domination-submission and egalitarianiam in the inter-I action between property owner and transgressor, as well as assessing the level of conscience development reflected in the behavior of the transgressor._ A category is, thus, descriptively defined as some aspect of-tho behavioral communication between the property owner and transgressor. A category is oper- ationally defined by the series of mutually exclusive and exhaustive items which comprise it. For each_parent-child story, the coder is to select from each of the six categories,-the single item which most accurately describel the completion being coded.- Thus, both the Lost_Meat and the IQEISpot stories are to be coded once and only once_in each of the six content categories. The sibling stories, the Damaged Axe and the Soiled Dress, are to be coded in only two of the six-categories,,Category IV, Property owner's initial feelings, and Category V, Property owner's action[communication. A descriptive.statement of the purpose of each category precedes a list of the definitions of items comprising that category.. Specific instructions . for coding are statedlin the introduction to each category. In general, coding is to betdone on the basis of explicit statements or descriptive adjec- tives used by the author. Inferences by the coder are permitted only when so stated in the definitions of category items. . Column Number 1 cows.» 10 ll 12 A-3 IBM CODING KEY Information Coded Location: NMnich Benton Harbor Sex: Pmle Female Serial number of protocol starting with 0001 to 9999 A-2, the Ink §E__ story Category I, _Tranggressor's initial feelings Anxiety, fear Physiological symptoms Evades confronting Guilt Shame Sadness No information Category II, Transgressor's communication regarding transgression Consistent lie Evades confronting Non-essential truth Inconsistent, truth after lie Consistent truth No information Category III, Interpgrsonal context of communication External pressure External and internal pressure Internal pressure Fact-finding, information seeking Spontaneity' No information Category IV, Property owner's initial feeling}; Angry , , Sad .‘ Surprised Happy Understanding No information - Category V, Property owner's actionlcommunication Punishment by property owner Punishment, outside source Punishment, source unspecified Punishment and forgiveness No punishment Restitution by p-o without punishment Forgiveness without punishment 'Consolation, explanation, advice No information I Code H omwao—I ouwaI-f ouwat-t ooubwwr— u—a omuompwww 13 14 15 16 17 18 I9 20 21 22 23 24 25 26 Category VI, Transgressor's affective/behavioral response Angry Submissive Shameful voluntary promise to reform A-é Relief, tension reduction No information 12§_Spot story I, (same as above) A-é ’ the Category 'ICategory Category Category Category Category 5-4, the Category Category B-S, the Category Category II, (same as III, (same as above) IV, (same as above) above) V, (same as above) VI, (same as above) Damaged Axe story IV, (same as V, (same as Soiled Dress IV, (same as V, (same as above) above) story above) above) NOU‘IDUNH Category I Transgressor's initial feelings A-2, the Lost Meat story, and A-4, the I3§_Spot story, are to be coded once and only once in this category according to the appropriate item. This category is designed to record the child's feelings about 1) his transgression, 2) confronting his mother, or 3) both. All feelings which are initial, i.e., which occur prior to the mother's response after learning about the transgression, should be coded in this category. Hence, feelings which occur at the following points in the story should be coded in this category: 1) before the child con- fronts his mother; 2) while telling the mother what happened; 3) during ques- tioning by the mother re. transgression; or 4) during discovery by the mother ( e.g.,.the mother comes upon the child in the act of transgression as in A-é.) In general, this category attempts to assess the presence and content of the child's evaluation of his own behavior, insofar as it can be separated from the mother's evaluation after learning of the transgression. Does the child affoctively respond to an internalized value system of his own oy means of which he judges his behavior? Or, does he respond to the value system‘of his mother? '- It is assumed that the child who is capable of evaluating his own behavior and whose own system of self-evaluation is most salient will feel guilt, shame, or sadness, and the fact of transgression will be the source of these feelings. The child whose mother's value system is most salient will feel fear, anxiety, or evade confronting, and the fact of confronting will be the source of his feelings. It is further assumed that the type of initial feeling expressed, i.e., guilt, shame, sorrow, fear, is sufficient to differentiate children whose feelings arise from internalized evaluative forces (conscience) from children whose feelings arise from external evaluative forces (parents, authority figures). The units to be coded are either descriptive adjectives used by the author and corresponding to the items defined below, or verbs describing physiological symptoms of affect. For stories in which an author uses more than one adjec- tive,check the item appropriate to the first feeling described. The rationale for this decision is two-fold: l) Combinations of feelings which reflect dif- ferent category items are almost non-existent; 2) the general focus of this category is on the child's feelings separate from those of the mother and hence, first feelings are given priority. CHECK ONE: 1. Anxiegy. fear: Explicit statement that child is afraid, or statement of vague, non-specific agitation, eg.g., nervous, excited, bothered, upset. 2. Physiological symptoms: Disturbances of speech, e.g., stutters, stammers, is speechless; or disturbances of coordination,e.g., trembles, shakes; or changes in metabolic rate, e.g., heart pounds, perspires, breathes hard. If initial feeling is directly specified, give priority to specified feeling over any expressed physiological symptoms. Evades confronting: Avoids, attempts to avoid, or wishes to avoid con- fronting his mother, e.g., sneaks into house, hides evidence, may con- template action or communication, but says nothing or waits to be questioned by the mother. When explicit statement of initial feelings accompanies an evasion of confronting, give priority to the specific feeling. 'Guilt: Specific statement that child feels guilty, has a bad conscience, is regretful, conscience-stricken, etc. If child feels blameworthy or that he deserves punishment before action by the mother, check this item. E.g., "the mother of Elizabeth thinks that her child is lying, that the child is bothered by her conscience? Check guilt, since mother's obser- vation is noted by the child-author and it is thus inferred that the child- character does have a.bad conscience. Shame: Specific statement that child is ashamed, embarassed, disappointed ‘in himself, or mad at himself. If child blushes,.check this item. Sadness: Child is sad, unhappy, or sorrowful. If child weeps, cries, sobs, or has tears in his eyes, check this item. No information: No information re. initial feelings. A-7 Category II TranSgressor's communication regardingitransgression. A-2, the Lost Meat story, and A-4, the I§§_Soot story, are to be coded once and only once in this category according to the appropriate item. This category is designed to record the transgressor's communication to his mother regarding the fact of transgression, before, during, and after any response by the mother. Hence, the unit to be coded is the entire content of the com- muncation regarding the fact of transgression. ' Three aspects of the communication are to be recorded: 1) the occurrence or non-occurrence of confrontation and some communication about the_fact of transgression; 2) the content of communication, i.e., truth or lie; and 3) the consistency of the communication in the course of the story. The first aspect can be coded by asking whether the child told his mother anything about the transgression. The second aspect, content of communication, necessitates an operational definition of truth for coding purposes. In A-2, the Lost Meat story, §£_least gg£_of the following facts must be communicated in order for the communication to be coded as true: 1) bhchael put the meat down. 2) Michael played with friends. In A-b, the Ink Spot story, Susan must communicate at least one of the following: 1) Susan got ink spots on the coat. 2) Susan played with her mother's coat. In either story, if the author simply.states that the truth was told but does not indicate the content of communication, score this as truth. In A-a, the child-author is asked to complete the story as the mother enters the room where the transgression occureed, and hence, it might.be assumed that the fact of transgression is obvious. In this case, unless there is a basis for inferring that the transgressor is attempting to conceal the truth, check truth as the appropriate item. The third aspect, consistency of communication, pertains to the truth- lie dimension. Most children elaborate and/or change the content of their communication during the course of interactbngwith the mother. Here, we are concerned with the extent to which the total content of communication is consistently true, consistently lie, or modified such that initially, com- munication could be coded as lie and later as truth. Nbdifications whereby an initial truth is changed to a lie are so infrequent that such items should be coded as £12. The coder must be careful not to code a halting, communi— cation, completed only after hesitations and delays as truth after lie. CHECK ONE: 1. Consistent lie: Failure to meet either of the criteria for truth, or is a misstatement of”the facts as given in the incomplete story, or is a de- liberate attempt to conceal such facts from the mother, or the child- author simply states that the transgressor lied. Child must not modify his communication to meet the criteria for truth throughout the story. 2. Evades confronting and/or verbal communication: Child avoids the mother, says nothing, may contemplate action, but says nothing. Check this item only if evasion is, in fact, successful. When the child is pressured to respond, or is caught in his evasion, check the content of his communi- cation according to truth or lie. I A-8 Non-essential truth: Child relates some aspect of the transgression but omits the relevant criteria for truth as defined above. On A-a, the Ink Spot story, check this item only when there is some indication that Susan is trying to omit the essence of the transgression from her communi- ,cation. InConsistent, truth after lie: Child's communication meets criteria for truth only after initially telling a lie. Child amends his communication such that a previously deliberate misstatement of the facts of the story are changed to become a truth in the course of the Story. Consistent truth: Child's communication meets criteria for tguth as defined above. The entire content of the child's communication during the course of the story'is under consideratoon in this item. Hence, truth following non-essential truth is coded as truth since there was no need for the child to reverse the facts of his story in order to incorporate truth in his communication. No information. No information regarding child's communication with the mother regarding the fact of transgression and no indication that the child evaded confronting his mother. A-9 Category III Interpersonal context of communication. A-2, the Lost Neat story, and A-4, the Ink Spot story, are to be coded ence and only once in this category according to the appropriate item. It is assumed that truth or lie is a process of relating, a verbal behavior occurring between two peeple in a situational context. This category con- siders the interpersonal context of the transgressor's communication regarding the fact of transgression. For coding purposes, interpersonal context is defined as the nature of the mother's participation in the transgressor's com- munication. Such participation may be verbal or non-verbal, expressed by the mother or anticipated by the transgressor and revealed as internal pressure. This category focuses on the nature of the mother's participation during the truth process, i.e, while the transgressor is relating the facts of transgression. The mother's communication and/or action after learning of the transgression should be coded in Category V. CHECK ONE: 1. External_pressure: Presence of a) intimidating questions, questioning with evidence of negative emotion by motherfiyexpressions-of disbelief or mis- trust, threats, onJany communication or behavior by the mother which can be construed as attempting to influence the transgressor's communication. 2. Externalgplus internal pressure: Presence of external pressure as defined above plus delay, evasion, weighing of circumstances by the transgressor, either before or after he begins his communication. Spontaneous change from lie to truth would be coded as internal pressure alone. However, change from lie to truth after delay, evasion, etc., and in the context of external pressure, would be coded as external plus internalAprcssure. Stories coded as evades confronting in Category II, and in which the mother provides external pressure to confront, should be coded in this category. 3. Internal pressure: Internal pressure, as defined above, wdthout the presence of external ressure, as defined above. 4. Information-seeking. fact-finding: Non-judgemental attempts on the part of the mother to find out what happened. Behavior must not.be chastising or suspicious or Judgements., but merely aimed at providing the child a neutral atmosphere in which to communicate. Transgressor may be puni- shed after communicating, but the communication itselm must occur in a neutral context. When information seeking occurs in the context of internal_pressure, code as internal ressure, above. 5. Spontaneity: Communication is immediate, consistent, and occurs in the absence of external pressure. Communication must meet the criteria of truth or lie before the mother participates. ' ' ‘ O. No information: Story provides no basis for inferring the nature of the interpersonal context of communication. A-IO Category IV Property owner's initial feelings. All four stories are to be coded once and only once in this category accordir" to theappropriate item. This category is designed to record the property owner's initial feelings after communication with the child re. the fact of transgression, and/or after learning of the transgression from another source. This category is independent of the truth of communication. If the property owner never learns of the communication, but expresses feelings to- ward the child after the child's communication, code.these feelings in this category. ~ Such feelings may concern 1) the fact of transgression, 2) the nature of the transgressor's communication, i.e., truth or lie, or 3) both. They may be specifically directed at the transgressor, or they may be more gen- eral, undirected feelings. These feelings need not be overt or verbally communiated to the transgressor. If the child-author simply states that such feelings existed, check the appropriate item. Inferences should be kept so a minimum, and the unit of coding should be specified by overt physiological manifestations of feelings or by descrip- tive adjectives, nouns, adverbs, or idioms. CHECK ONE: 1. Angrv: P-o was annoyed, cross, mad, angry, excited, or irritated. If P-o has unfulfilled intentions re. punishment, i.e,, feels like puni- shing but does not and no additional feelings are Specified, check this item. When anger occurs in combination with other feelings, e.g., sad- ness, check this item. 2. Sad, unhpppy: P-o was sad, unhappy, disappointed, including vague nega- tive emotions such as upset, regretful, etc. Physiological symptoms indicative of sadness, eg, crying, sobbing, could have cried, etc., should be checked in this item. 3. Surprised, shocked: P-o was surprised, bewildered, dismayed, amazed, "couldn' t believe her eyes" etc. If more specific feeling is indicated, give priority to that feeling and check the appropriate item. 4. Happy,_giad: P-o was blad that transgressor was not hurt, told the truth, etc. 5. Understanding,_accepting: Lack of emotional expression in the context of intellectual acceptance of transgression by P-o. bay include realization of transgressor's lack of ill intent, lack of maturity and in general, making excuses for transgressor. E.g. "‘bther wasn' t angry because Susan told the truth"--infer acceptance and code in this item. 0. No information: P-o learns of transgression but no indication of initial feelings as specified by the author. A-ll Category V Propergy owner's action/communication. All four stories are to be coded once and only once in this category according to the appropriate item. This category records the property- wner's behavior, i.e, action and/or communication, toward the transgressor after learning of the transgression, regardless or the source of that know4 ledge, but before the transgressor's attempts no reconcile himself to the property owner. The items described below concern only these behaviors directly related to the fact of transgression and/or the nature of the transgressor's communi- cation. Only those overt actions or communications which 1) actually occurred or 2) are indicated by the child-author as highly probable are to be coded in this category. Unfulfilled intentions regarding punishment are to be coded in Category IV. ‘ ‘ The unit of coding is the entire content of the property owner's action and/or communication following and consequent to the content of the trans- gressor's communication regarding the fact of the transgression. CHECK ONE: 1. Punishment by property owner: Specific statement of the fact of punishment, material deprivation, vengeful destruction of the transgressor's property, isolation, or activity deprivation,or forced restitution. Threats of punishment, scolding, chastising, berating, should also be coded in this item. Note: quarreling should be coded as item 1 in category IV, since quarreling-is not considered punishment. 2. Punishment,_outside source: Punishment, as defined above, by source other than property owner. Only include if property owner "tells on" trans- gressor and is,.in fact, reSponsible for the factof punishment. ' 3. Punishment. source unapecified: Punishment as defined above but no infor- mation as to source. If it can be inferred that punishment was administered by the P-o, check item 1. J 4. Punishment and forgivenesszP-o punishes transgressor but communicates for- giveness by verbal statement, friendly act, etc. It must be clear that forgiveness is, in fact, communicated to the transgressor. S. No punishment: Statement that the transgressor was not punished along with the absence of punishment as defined above. I 6. Restitution by_P-o without punishment: P-o restores, replaces, or recovers property without punishing the transgressor. Restoration by both P-o and transgressor to be included in this item. ‘ ' 7. Forgiveness without punishm~ntz Communicated forgiveness without punishment. 8. Consolation._explanation, advice: P-o advises, explains, attempts to correct without punishment or scolding. Constructive concern with transgressor's learning something from the transgression. O. No information: Story included no information on punishment. A-12 Category VI Transgressor's affective an_/or behavioral response. A-2, the Lost Meat sto y, and A-4, the Ink Spot story, are to be coded once and only once in tiis ca toggory according to the approprh e item. this cate3.ory records the transgressor' s feelings re. the fact of his mistake, the manner or content of communication With the mother, or the ohought or fact of punishment after communication or action by the mother. If there'is no information regarding mother's communication, i.e., if no information is checked in Category IV, code the child‘s feelings and behavior after his initial com- muncation regarding transgression. CHECK ONE: 1. Angry, resentful, hostile: Transgressor is angry with the mother for not believing him, for punishing him, etc. Include indications of externally projected anger following action/communication by the mother. E.g., Michael becomes angry at_the dog. Do not check statements that the child is mad at or angry with himself. 2.. Submissive,_fearful.mplncating: Child begs, requests forgiveness, apologizes, submits passively to mother’s action or communication. Child communicates promise to reform or voluntarily restores property after punishment, threat, scolding. If author simply states that child is afraid, check this item. 3. Shame, regret,_self-criticism: Child is ashamed, regrets his behavior, is disappointed with himself after mother's communication/action. Physiolog- ical symptoms of shame or regret, e. g.,blus hing, crying, should be checked here. If child feels that he deserves punishment, regardless of the fact of punishment, check this item. 4. ihluntary promise to reform,_snontaneous restitution without_pressure: Communicated promise to reform or spontaneous restitution after confrone ting the mother but without pressure, scolding, or opposition by the mother. Uncommunicated promise to reform , regardless of external pressure, should also be checked in this item. Indications that transgressor has learne d his lesson, has gained insight following communication with the mother, should be chhcked here. If transgressor' s promises to reform can be con- strued as an attempt to placate the mother, or regain her affection, check item 2. . 5. Relief, tension reduction, happiness: Child is pleased that he confronted his mother, feels happy or positively about the mother's communication] action toward him, etc. If child feels happy after submission, placation, reform, etc., check the appropriate item describing the child's initial feeling or behavior after mother's communication/action. 0. No information: IV ' whv‘lr" f’ 221).):‘4LMJ465 .-fi Dirretiorc for ~dninisterirq the Anderson Ineorelete Stories C. 0 « 0'4 re are six short stories. Boys and girls in other inerican schhols as ‘n Germany, Mexico, and England have also "mitten th;.;e stories. .13: hszpens in the story, then write w.at, in your opinion, happens . Use your ira~ination. write clearly, but do not try to write beauti- y. ILy to write as quickly as you ca.. If you make a mistake, strike the trork and continue item diately. J Hxn firm :3 f) r: U: n he ." U H n' Cs :1 O and .I H C) O C You will not be graded for these stories. Your teaches, the principal, an d other teachers in this school will not even get to see these stories. he do not want you to put your name on tr e paper. Write honestly what you really think. If your name is not on the paper, you may even write something which you uéuld not tell to a teacher or to anyone else. I will now hand out the forms. Yen need not look at cacheothor's papers as there are no right or wrong answers. YOu each have enough ircgination to write for yourselfies. (Hand out the papers. The teacher does not par- ticipate). On the tOp of the first page please write the date of your birth. Write the :ontn, tr .e day, and the ye at, like this (write on the board) JUne 4, 19h3. Then write low old you were on your last birthday (write 13). New make a cross after boy or girl. Write the name of your schhol and givd your school grade. trite what clurch you atter d, or if you do not attend church, write {:5}, Kh-;n you wiite the occupation of your father, do not write the name a: ;.e ce.. any, but the kind of work he does, like carpenter, laborer, book- keeper. If your mother works a say from home, write what she does. Other- wise write 1.ome. If you read the first story you will see that there are some questions t t;-e end Please think about these questions, but do not try to answer hcn one by one. These questions are put there to help you finidh the story. then you finish the first story, continue with the next story without delay. If l.(test administrator) go through the aisles I am not trying to look at your paper and read qhat you have written. I will only see how far you havd proceeded. ' I"??? Try to finish all six (five) stories. You will have the entire period. Do you have any questions? If you cannot read a word, or if you have a question, or need a pencil, please raise your hand and I will then come to your seat. And now are you ready? If so, ready, start. Appendix C arson Incomplete Stories: Series A e. I ,. {L (1 Age at last Birthdatc: Birthday: Boy: Girl: School: Grade: Church: Occupation of father: Occupation of mother: ._A and Peter are walking to school. Suddenly Frank grabs Peter's cap and c : high up into the nearest tree so that Peter could not reach it with his hand. F dnk had never done anything like this before. Neither had Frank and Peter Why did Frank do this? What does Peter do? What does Frank think? How does Peter feel about it? Think about these questions, then finish the storie quickly with a few sentences. 2. bachacl's mother sends him to the store to get one pound of wieners. On the way home he puts the package of most on the edge of the sidewalu and plays for a short time with his friends. A shepherd dog darts forward quickly, claws half the wieners out of the package and rushes away with them. Michael wraps up the remaining wieners and takes them home. ' What does Rachael say to his mother? What does his mother do? How does Michael then feel about it? Think about these questions, then finish this story quickly with a few sentences. 3. The teecher suddenly discovers that fifty cents has disappeared from her des‘. She looks up and sees that all the class are working on their arithemetic. She wonders what happened to the money and what she should do. What does the teacher do? "‘ ,1 -_ 3 M! 3_;A;;3 this story with a few sentences. Tess what happened to the money and also exactly how the teacher feels and what hhe does? a. Susan is sittdng in the living room working on her homework. ke Suddenly she . r mother's new cost. She wonders how she would look in it. When taking the coat off she notices that she had got some ink Spots on her mother's new cost. Just as Susan is trying to wipe out the ink spots her mother enters the room. what does her mother sty? so How do they each feel about it? What does Susan say? What does each one do? Think about these questions and finish this story quickly with a few sentences. C-3 5. George and Tom are playing with a football. They know that they shouldnot play fr %3311 on the narrow space in front of the house. George gives the ball a strong . ad the ball hits a wind w and makes a big crack in it. Tom thought that some- on. “rd ajpesrod behind the window. No one could have seen who had kicked the ball A. -. ." .- h ‘ .9 Finish this story with a few sentences. Toll how the two toys felt about it and what they did. 6. Betty often handed in her homozork composition late to the teacher. This tine it wss an especially important compositiongsnd she had, moreover, finished it. ‘On the way to school she lost her composition book and could not find it anywhere. What does Betty say to'the teacher what does the teacher day? Think about these questions and finish this story with a few sentences. Anderson Incomplete Stories: Series 3 Age at last Birthdsts: Birthday: Boy: Girl: rchsol: Grade: Church: Occdpaticn of Esther: Occupation of mother: 1. Eary's grandzothor gave her a besttiful vase for her thirteenth birthday. One dc} L;ry;s mother came in with flowers which she puts in the vase. She places the vase in the window, though she knows the window ledge is too narrow for this vase and that it might fall off. As her mother is dusting, she bumps the vase and it crashes to the floor and breaks. While Mary's mother is in the kitchen getting e cloth, bury returns from shcool and finds the vase in pieces on the floor. What does bury do? What does the mother say? How do they both feel about it? 2. John worked hard and bought a new bicycle with his money. One Saturday, while John is playing with friends, Hahn's father decides to ride this bicycle to the store go buy cigarettes. He leaves the bicycle in the street. When he comes out of the store he finds the tron t wheel bend and some of the paint on the frame badly scsaped. Co one is around. The father could still ride the bicycle home. What does the father do? What does John do? HOW'dO they both feel about it? 3. The teacher has promised the class that if they work hard during the last month of the term they will have one day off to make a special trip. Several times during the last month the teacher talks of this trip. The children want to make this trip and they work very hard. Now it’is the last week of the term and there is no time for a trip. What does the teacher do? What do the children think, and how do they feel about it? Think about these questions and finish this story quickly with a few sentences. a. Htrbcr: rcco vcd for his tLi rtcc nth birthday a handsoraa enjping axe. It is ‘ .3 lidfih'? :130. while Eczvcz t is at school, hi 5 four-y: ar-old tie and, locks at it a long tlr e, pick: it up, puts it bo.ck, and doors with him to pla y. Bi 11y does not take the case off. Lo filo walks about the 3a mdcn,,tapoing the axe gcntly against a tree, a r ' ‘“4 t~e p .vcznont. harbort cones b.ome from school, finds his axe in its loathe: as- with as“: other toys. But he sees that the axe has cut through the leather case and tho blade is chipped and blunted. that does Herbert do? How does Herbert feel about 1t? 5-} U Think about these qeustlons and finid h the story quickly with a few sentences. 5. Kate, aged 13 years, has a little four-year-old sister, Clra. Khan Kate comes hare f:om school Clara often wants to play with her and follows her and her older rirl frinnds around. One day, Clara took a very pretty dress from Kate's ward- roba,.putit on herself, and looked in the mirror. It was too long and hung to the floor. She gathered up the skirt in her bras and went out of the house for a walk. Kate cc”: home from school, found her new dress crumpled in a chair. The skirt had been stopped on.and dr~wved in the dirt. Clara said, "I wore your dress.".. What does Kate do? What do Kate and Clara think about it and how co they feel? Finish the story in a few sentences. o‘]. O" 5' MICHIGAN STATE UNIVERSITY LIBRARIES In; |l||||I|l| 93 03056 2924 31