Int-15:5 L IE R A I"? 2’ MiChigan Stacy University "1......‘25 COMMUNICATION BEHAVIOR IN POTENTIAL CONFLICT SITUATIONS: .A COMPARISON OF STORY COMPLETIONS BY SCHOOLCHILDREN OF‘GERMANY, BRAZIL AND THE UNITED STATES By Leticia.M. Smith A.THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS EEpartnent of Communication 1966 ABSTRACT COMMUNICATION BEHAVIOR IN POTENTIAL CONFLICT SITUATIONS: A COMPARISON OF STORY COMPLETIONS BY SCHOOLCHILDREN OF GERMANY, BRAZIL AND THE UNITED STATES by Leticia M. Smith The present study explores the empirical significance of the concept of "national character," which is broadly conceived as "a set of Characteristics specific for a given nation as compared with others."* More specifically, it examines whether countries differ in some dimen— sions of national character, such as trust, empathy and respect for ‘ generational status, and.whether such differences coincide with varia— tion in the extent of mass media development. The data are derived from 211 story completions of three Anderson Incomplete Stories by sChooldhildren from.Germany, Brazil and the United States. All stories involve what may be called "violation of the sense of property," where one party is the cause of the damage of the property of another*without any apparent malicious intent. Two of the stories involve siblings (older sibling as offended, younger sibling as offender), and one involve father (as offender) and son (as offended). The protocols were content analyzed. It was assumed that the completions reflect the communication pattern in similar potential tension situations in the countries where the Children are fnmn. The fOllowing traits can be inferred from the pattern and content of communication in the protocols on potential conflict situations: Leticia.M. Smith 1. There is less respect for generational status in Brazil and.Germany than in the united States. 2. Empathy is present to a greater extent in Brazil than in either’the United States and Germany. 3. In Brazil, the united States and Germany alike, trust is more likely to be forthcoming to the offender the higher he is in the generational hierarChy with respect to the offended party. These findings may be explained better empirically rather than theoretically. The towns from.which the protocols were taken are not equal in their position as urban and mass media centers. Rio de Janeiro from which all the Brazilian data were gathered enjoys the position of a cultural center*which is not even equalled by any of the towns from.the United States and Germany studied here. * .. . H.C.J. Duijker and.N.H. Prijda, National Character and National Stereotypes, Amsterdam, NorthéHolland Publishing Co., 1960. 36p. ACKNOWLEDGMENTS This study was possible through the generosity of Dr. Harold H. Anderson of the Department of Psychology in giving me access to his story completion protocols , in providing me an exhaustive background on the process of gathering these protocols and studies that have been done on them, as well as in supplying me a quiet room to work on this paper. I also feel indebted to my husband, Huron M. Smith, Jr. , for relieving me of a lot of important detail work; to Mary Susan Faaborg, for helping me check the reliability of my coding system; and to Dr. Hideya Kumata, for assuming the irmunerable tasks advising me entails . Dr. Bradley S . Greenberg and Dr. Everett M. Rogers have given me very indispensable suggestions, and I am very grateful. To Ruth Langenbacher, I. give special thanks for typing this paper . TABLE OF CONTENTS Chapter I THEORETICAL BACKGROUND II METHODOLOGY Story 1 - Broken Bicyle Story 2 - Broken Axe Story 3 — Soiled Dress Classification of Countries . Coding Procedure Reliability Check Analysis Scheme III FINDINGS Correlation Between Measures of Trust Results of the Tests of Hypotheses IV SUMMARY AND CONCLUSION: A TRAIT PROFILE OF BRAZIL, GERMANY AND THE UNITED STATES Conclusion . . . APPENDIX: Code Book BIBLIOGRAPHY , iii Page 11 13 18 19 25 27 27 '42 nu us 55 LIST OF TABLES Table 10 11 12 13 Mass media statistics fOr Brazil, United States and Federal Republic of Germany, 1964 Correlation between offended party‘s definition of the situation in the two sibling stories Correlation between offended party's affective response to the situation in the two sibling stories Correlation between offended party's behavior response to situation in the two sibling stories Correlation between the offender's behavior response in the two sibling stories Correlation between direction of initial communication in the two sibling stories Relationship between direction of initial communication and generational hierarChy, Stories 1 and 2 Relationship between direction of initial communication and generational hierarchy, Stories 1 and 3 Correlation between age group of respondent and offended party‘s definition of the situation, Story 2 Correlation between age group of respondent and offended party's definition of the situation, Story 3 Correlation between age group of respondent and offended party's affective response to the situation, Story 2 Correlation between age group of respondent and offended party's affective response to the situation, Story 3 Correlation between age group of respondent and offended party‘s behavioral response to the situation, Story 2 iv Page 12 26 26 27 27 28 29 29 30 30 30 Table 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Correlation between age group of respondent and offended party's behavioral response to the situation, Story 3 Relationship between generational hierarChy and behavioral response of offender, Stories 1 and 2 Relationship between generational hierarChy and behavioral response of offender, Stories 1 and 3 Relationship between direction of initial communication and generational hierarchy, Stories 1 and 2 Relationship between direction of initial communication and generational hierarChy, Stories 1 and 3 Correlation between age group of respondent and offended party's definition of the situation, Story 2 Correlation between age group of respondent and offended party‘s definition of the situation, Story 3 Correlation between age group of reSpondent and offended party's affective response to the situation, Story 2 Correlation between age group of respondent and offended party's affective response to the situation, Story 3 Correlation between age group of respondent and offended party's behavioral response to the situation, Story 2 Correlation between age group of respondent and offended party's behavioral response to the situation, Story 3 Relationship between generational hierarChy and behavioral response of offender, Stories 1 and 2 Relationship between generational hierarChy and behavioral response of offender, Stories 1 and 3 Offended party‘s definition of the situation, Story 1 Offended party‘s definition of the situation, Story 2 Offended party's definition of the situation, Story 3 Offended party‘s affective response to the situation, Story 1 31 32 32 34 34 35 36 36 37 37 38 38 39 no 40 H0 H0 Table 31 32 33 34 Offended party's affective response to the situation, Story 2 Offended party's affective response to the situation, Story 3 Offended party‘s behavioral response to the situation, Story 2 Offended party's behavioral response to the situation, Story 3 Pege ”1 41 ”1 1+1 CHAPTER I THEORETICAL BACKGROUND Considerable interest has been registered by social science researChers in setting up a descriptive framework within which to compare human groups across cultural or national boundaries. One area being explored is that of "national character." Essentially, studies on national Character try to determine "a set of Characteristics specific fer a given nation as compared with others."1 Part of the reasoning behind this searCh is the assumption that since human beings have common biological endowments, a culture, impinging upon these unifbrm.consti- tutions will produce relatively homogeneous effects upon these humans. If in fact people within political boundaries are feund more alike than people between political boundaries, then the hypothesis of the exis— tence of "national Character" is deemed supported. Mere specifically, two criteria.must be used to test whether it is valid to assume that "national Character" exists: 1) some degree of homogeneity within the _ group with respect to the Characteristics involved is required, and 2) the Characteristics with respect to WhiCh there is homogeneity should very'from.group to group.2 lH.C.J. Duijker and N.H. Frijda. National Character and National Stereotypes. Amsterdam, NorthéHolland PubliShing, 1960. p.36. 2Bert Kaplan. "Personality and Social Structure." (In J.B. Gittler, ed., Review 9f_Sociology, N.Y., JChn Wiley, c1957, pp. 87- 126). p. 116. l 2 The present study aims to explore the possibility that cultural boundaries and political boundaries do coincide and that pe0p1es of different nations exhibit significantly different response patterns. At the same time, it aims to give meaning to suCh differential responses when they occur on the basis of variation in status criteria and extent of mass media development. If societies are classified according to whether "ascribed" characteristics, suCh as age, sex, or position in a kinship structure, or "aChieved" Characteristics suCh as education, teChnical or athletic or aesthetic skills are the bases for classifying people into the hieru arChical structure of the society, these bases will be reflected in the deference patterns that emerge in the communication process between members of the society in an interpersonal conflict situation. In a society where ascribed Characteristics are the bases fer deference be— havior, these Characteristics are pertinent in the entire life of the individual. We will call this type of society traditional. On the other*hand, in the case of societies where "achieved" Characteristics are the most important bases for deference, these Characteristics are pertinent only in the later life of the individual, at whiCh time he has had ample time to acquire status requirements. During Childhood and adolescence, the individual in this society is responded to and responds solely on his merit as a human being. We will call this type of society modern. Deference patterns that prevail in a society will be reflected in potential conflict situations suCh as one whiCh involves what may be called "violation of the sense of property." Property is that which the individual himself has the right to control. 3 It is presumed that individuals attaCh some degree of positive value to their’property such that when this positively-valued property is either taken away or damaged by an identifiable person, a negative emotional state is aroused in him. The negative emotional state may orImay not induce overt action. If and when suCh emotional state does give rise to overt action, the latter'may take the ferm.of verbal or physical hostility, or*may be sublimated to pleasant, rational interaction with the "offender," depending upon the cultural upbringing of the individual. Tension arises when the individual's hostile or aggressive impulses are contradictory to the values of respect and Charity whiCh.have been inculcated in him. The tension potential is especially high when family members are involved. The consummation of negative impulses may either*be inhibited or facilitated.by the relative positions of the offended (one whose property was taken away or damaged by another; one Whose "sense of property" is violated) and the offender (one who takes away or damages the property of another; one who violates somebody else's "sense of property") in the age, sex, and generational hier— arohy of the family. It is expected that in the traditional type of society, the older family member and those higher in the generational hierarChy will receive more deference than they give when they interact with younger family members and those who are in the lower*kinship positions. The _ greater the age difference, the more the situations where deference is accorded them. In an interpersonal conflict situation where the of- fender is older and higher in the generational structure than the offended.party, the offender's selfeesteem is at stake if the offended younger person perceives him as an offender. The older person is supposed to be the embodiment of wisdom and virtue, whiCh "naturally" u accompanies age and high position in the generational hierarchy. The offender will have to present himself as a person consistent with these Characteristics ascribed to him.in his role as an older person. If the offender is younger than the offended person, self-esteem maintenance is less dependent on the older person's perception of him.as an offender. The offended person in a traditional society will be limited in his overt reaction to the offender depending upon their age and gener- ational position differences. If the offender is younger and lower in . generational position, the offended can confront the fermer with the offense and administer punitive action if he wants. If the offender is older and higher in the generational hierarChy, then the offended person.will be breaking norms if he confronts the offender and especial— ly if he tries administering retributive justice. Tb a family with a secular outlook, that is, a family in.a.mode§n society, suCh variables as age, sex and generational position.may not have any import at all in a similar interpersonal conflict situation. The offender in the.modern_society will not feel the same pressure as the offender in the traditional society does to present himself as one of ”wisdom.and virtue" because errors are expected, though not desired, to be committed by anyone. The primary determinant of action on the part of the offended party in the modern_society is Whether'he perceives his rights as an individual as either*purposely or inadvertently violated. Since the perception of motives does not entirely depend upon the objective characteristics of the external situation because motives cannot be directly observed, it can be expected that perception of the same will be dependent on the amount of trust the perceiver has toward 5 the other person befOre the offense was committed, as well as on the empathic ability of the perceiver. The minimum.requirement fer trust is the belief that the action of the other person does ngt_arise from ill-motives. Empathy, on the other hand, is the ability to imagine oneself in another situation, to role-play and to be "psychically mobile."3 It has a more developmental connotation than trust, that is, empathy develops not only through a long history of varied social interaction, but also along with the intellectual and emotional maturation of the individual. Trust depends on the relative frequency of positive compared to negative results of social interaction (in the sense of fhlfillment or non-fulfillment of expectations) and does not require maturity. Empathy connotes adequacy in reality—testing, while trust does not. Trust involves minimizing cognitively the risks in the prediction of outcomes, while empathy does not. Trust can be an outcome of empathy when a person's understanding of another's roles leads him to conclude that alter's history of acts arising from good intentions, or at least with the absence of bad intentions, is significantly greater than alter's history of acts arising from ill intentions. All the offenses in the three stories used in this study were committed without any apparent malicious intent on the part of the offender. If the offended person is empathic, then he or she will be able to condone the act of the offender. This act of condonation takes the same behavioral manifestations whether it is by 3Daniel Lerner. The Passing oflTraditional Society. Glencoe, Ill., Free Press, 1958. p. H9. 6 virtue of trust alone, or by virtue of both empathy and trust. If we add another variable to our classification of societies into "traditional" or "modern," we will be able to discriminate between these societies better. This variable is the extent of development of mass media in the society. On the basis of Lerner's model,” the society we call "traditional" will have a less developed mass media system.and the society we call "modern" will have a.more completely developed mass media system. .According to Lerner, the development of mass media brings with it the cultivation of the empathic ability of people. People in a "traditional" society will have a lower level of empathy, and people in a "modern" society will have a higher level of empathy. The stages in.mass media development can be placed along a con— tinuunn we will call the "zero point" the stage where absolutely no technological Hess medium.(e.g., radio, television, movies, newspapers, other printed.materials) are available to the people. This is the point where we start labeling societies as "traditional." As we move along to the farthest point in the continuum, we have the stage where people have unlimited access to technological mass media. Deference patterns change along with the level of empathy of a people as one moves from societies at the zero point of'mass media development to societies in the unlimited access stage. The Change is from a stage Where deference is on the basis of ascribed characteristics and where the level of empathy is low to a stage where deference is on the basis of achieved Characteristics and where the level of empathy is high. It is hypothesized here that: 1. In a potential interpersonal conflict situation, the direc— tion of initial communication about the offense is affected Urlerner, op, cit. 7 by the generational status of the offended party with respect to the offender. 2. In a potential interpersonal conflict situation between an older sibling and a younger sibling where the fbrmer is the offended, and the latter the offender, the older sibling will perceive himself as being more empathic than the younger sibling will perceive him to be. 3. In a potential interpersonal conflict situation, the offended party's level of trust is affected by his relative status to the offender in the generational hierarChy. H. There are significant differences between nationality groups with respect to the relationship between generational status and the direction of initial communication about offense be- tween offender and offended in a potential interpersonal conflict situation. 5. There are significant differences between nationality groups with respect to the degree of empathy the older sibling as the offended party will ascribe to himself as against the degree of empathy the younger sibling as offender will ascribe to him.in a potential interpersonal conflict situation. 6. There will be significant differences between nationality 'groups with respect to the relationship between generational status and the level of trust of the offended for the offender in a potential interpersonal conflict situation. 7. The extent of'mass media development in a country affects the level of empathy of its people. Confrontation is defined here as the offended party's act of talking to the offender about the offense. CHAPTER II METHODOLOGY This study attempts to identify patterns in the communication process involved in potential tension situations as perceived by school children from age 9 to age 18 in Germany, Brazil and the United States. These situations all involve "violation of the sense of property" with the family members as exclusive characters in the stories. Both verbal communication and communication through bodily movements are considered as communication behavior in this study. The data used are part of the 6,615 protocols from.the Psychology- Creativity ResearCh conducted by Harold and Gladys Anderson of the Department of PsyChology, Michigan State University, from.l952 to 1962. The Andersons devised eleven incomplete stories as projective tests, and administered them.in two series, Series A and Series B. Series A con- sists of six brief incomplete stories, while Series B consists of five. Three of the.Anderson Incomplete Stories from Series B were Chosen for the present study. These stories appeared to be adequate to test the hypotheses outlined here. The fellowing are the three incomplete stories used in this study: Story 1 - Broken Bicycle JChn worked hard and bought a new bicycle with his money. One Saturday, while JChn is playing with friends, JChn's father decides to ride this bicycle to the store to buy cigarettes. He leaves the bicycle in the street. When.he comes out of the store he finds the front wheel 8 9 bent and some of the paint on the frame badly scraped. No one is around. The father could still ride the bicycle home. What does the father do? What does JChn do? How do they both feel about it? Think about these questions and finish the story quickly with a few sentences. Story 2 - Broken Axe Herbert received fer his thirteenth birthday a.handsome camping axe. It is sharp and has a strong leather Case. While Herbert is at school his foureyear 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 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. Story 3 - Soiled Dress Kate, aged 13 years, has a little fourhyear 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 herself, and looked in the mirror. It was too long and hung to the floor. She gathered.up the skirt in her ains and.went out of the house fer 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? 10 Finish the story in a few sentences. TWO hundred eleven protocols whiCh have all three stories com- pleted, and whiCh constitute five per cent of the total 4,141 proto— cols with Series B completions in the U.S., Germany and Brazil were randomly drawn for this study. It should be noted, however, that the original sample from.whiCh this subsample was drawn.did not involve any randomizing procedure, although the number of schoolrooms of Children were "about equally divided between boys and girls, contain Children frenlhigh, middle and low socioeconomic levels as determined by school authorities."5 This implies that it is hazardous to infer from.the Anderson and Anderson sample to the countries, or even to the entire towns where the respondents were located. With this considera- tion, this study is limited to generating workable hypotheses for future researCh. The Anderson Incomplete Stories were administered in the locations used in this study in the fellowing dates: Germany Hamburg 1954 Braunschweig 1953 Munich ‘ 1954 Brazil (Rio de Janeiro) 1959 United States Benton Harbor, Michigan 1957 Cedar Rapids, Iowa 1960 5HaroldAnderson and Gladys L. Anderson. “A Cross-National Study of Children: A.Study in Creativity and.Mental.Health." Paper presented at the Sixth International Congress on Mental Health, Technical Session, Sorbonne, France, Aug. 31, 1961. 11 In 1960, Theodore and Judyth Anderson administered the incomplete stories to the same classrooms in MuniCh, Germany where Harold and Gladys Anderson obtained their data in 1954. .After a content analysis of three of the story completions, it was concluded that "there was high stability in the fantasy productions of Children in the same school grade and same neighborhoods in MUnich in 1954 and in 1960."6 It will be noted that the date of administration of the incomplete stories differ~within and between the countries involved in this study. It is assumed here, however, that the finding in the MuniCh replication study is also true fer Hamburg and BraunsChweig, as well as fer the U.S. and Brazil. The Anderson Incomplete Stories were completed by the Children in their local language. These completions were translated into English, which translations the author used. Geierhaas7 studied the reliability of translation. He coded the German protocols independently from another coder'who worked on the English translations of the same protocols. The inter-coder index of reliability was 97.2 per cent. Classification of Countries On the basis of the data in the fellowing page (Table l), we shall arbitrarily call Brazi1.from.the mass media development point of 6HaroldH. Anderson and Gladys L. Anderson. "Creativity and Cross—National ResearCh: A.Comparison of Children's Values in MuniCh, Germany in 1954 and 1960." Paper~presented to the Society fer ResearCh in Child Development, Pennsylvania State University, March 17, 1961. 7F.G. Geierhaas. Problems 9f_Reliability‘in_Evaluating Story Completions About Social Conflict by;German.Adolescent Children. M.A. thesis. East Lansing, Michigan State University, 1955. 12 Table 1. Mass media statistics for Brazil, United States and Federal Republic of Germany, 1964 Brazil U.S. Germany Pepulation ' 70,799,000 ' 180,670,000 ' 55,577,000 Illiteracy , 51% , 2.2% ' l-2%a Press Dail newsp. . v 291 ' 1,763d . 073 Tot daily Ciro. 3,837,000 59,211,464 17,388,135 Copies/100 ' ' ' people ' 5.4 32.6 30.7 I I Radio Transmitters . 870.AM. 3,624.AM . 83 AM 54 FM 931 FM 155 FM Receivers ' 0,700,000 . 183,800,000 v 16,332 w1f 100,000 wf Receivers/100 ' ' ' people ' 6.6 ' 100 ' 29.6 Television Transmitters ' 28 main' 607 ' 45 main l4 aux. 280 aux. I I I Receivers 1,030,000 60,000,000b 7,800,000f Receivers/100 I ' ' people 2.0 33.2 13.9 Filfll ' ' ' Cinemas , 3,284—35MM 8 ' 12,300-35MM FC , 6,667-35MM FC 16MM FCC 4,700—35MM 403—35MM.MUg . . [tie . l-35MM DrI Total seating capacity ' 1,899,100 ' 10,000,000 ' 2,760,600 Seats/100 , , , people 2.7 5.6 4.9 aAround 1950 eDrI = Drive-In bEstimated no. in use fBased on the no. of licenses CFC = Fixed Cinemas . gMU = Mobile Units dEnglish language dailies only Source: UNESCO. World Communications: Press, Radio, Television, Filnn N}Y. and Paris, 1964. 13 view alone as "traditional,” and U.S. and Germany as "modern." This diChotomy is perhaps more applicable during the time when the data fer this study were gathered than it is today, although even today, Brazil still lags behind the U.S. on a per capita availability of the various media. We do not have mass media statistics for the period when our protocols were obtained, nor for the specific towns involved in the present study for‘Whatever1period. Since Rio de Janeiro, the only place in Brazil in this study, is the "cultural capital" of the country, the mass media data for Brazil surely underrepresent the availability of mass media facilities in Rio. The German and American statistics in Table 1 can be assumed to represent the extent of mass media develop- ment in the towns included in this study. Coding Procedure A coding manual was set up after scanning the original set of protocols from.whiCh our sample was drawn. The three story completions were content analyzed in terms of the following variables: I. Deference pattern Category 1: Direction of communication regarding the offense, i.e., who initiates the communication, the offender, the offended party, or a third party. Category 4: Content of offender's communication to offended party after the latter learns of property damage.— Coded here are the responses indicating whether the offender 1) asks forgive- ness or expresses regret, 2) promises repair or replacement of the damaged property, 3) attempts to evade l4 responsibility by accusing offended party of guilt, or 4) merely tells reason for doing the act that led to the dame page of the property. This category, however, cannot be used for hypothesis testing because of very small or no frequencies in many cells of the matrix. II. Trust Category 3: Offender's behavioral response to the situation (damage of somebody else's property).- This category overlaps with Category 1 in that the behavioral response involved in this Category 3 is whether the offender tells the offended of the offense with or*without the questioning from.the offended party. However, it allows for coding lying and attempts at deception on the part of the offender with or without ques— tioning from the offended party, which kind of coding is not provided fer by Category 1. III. Empathy Category 2: Definition of the situation by the offended party or by the respondent himself.- In this category, subcategories ranged from.responses attributing no illemotive to the offender, to responses attributing illamotive to the offender. Category 5: Offended party‘s affective response toward the offender after'the fermerilearns of property damage.- Positive responses toward the situation, suCh as "happiness" and "joy," and negative responses suCh as "depression," "shock" and "anger" are coded here. 15 Category 6: Offended party's behavioral response to.the situation after learning of the damage of'his/her~property.- The responses involved here ranged from constructive acts on the part of the offended.party toward.the offender, suCh as 1) laughing it off, 2) attempting to placate offender's feelings by promises of material or companionship reward, 3) advises offended party about the inappropriateness of the act, and/or future desirable action, or 4) repairing or replacing the damaged property; to punitive acts suCh as 1) seeking third party to help in punishing offender and 2) punishing offender’by himself/herself. Story 1 was content analyzed for Categories 1—5, while Stories 2 and 3, for Categories 1—6. It was not possible to make a parallel between the actions of the offended party in Story 1 (father-son story) with the ones in Stories 2 and 3 (sibling stories) fer Category 6 be- cause of the difference in relative status and probably physical size between the offender and the offended parties in the three stories. To illustrate, the offended parties (older siblings) punished the offenders (younger'siblings), or asked their parents to administer punisnment in most cases in Stories 2 and 3. Another*mode of response was to attempt to please the offender by verbal or material reward. In the case of Story 1, the offended party (the son) was not in a position to ad- minister punishment to the offender (the father), nor was there any logical need to please the offender’based on the plot of the story. The level of empathy of the respondent is to be inferred from Categories 2, 5, and 6, while the level of trust, from Category 3. As we defined empathy earlier, and as we explained that the "offenses" 16 involved were presented in the incomplete stories without any apparent malicious intent on the part of the offenderg we expect that 1) the respondent or the offended party will define the situation as to acknow- ledge lack of ill—intent on the part of the offender (Category 2) and that the offended party will respond.with positive affect toward the offender since the "offense" was perceived as without ill—intent (Category 5) and 3) the offender will act constructively (e. g., replace property, advise offender on future desirable action, etc.) toward the offender since he/she acknowledged that the "offense" was of no ill- intent. All these conditions should obtain if the respondent is empathic, or is able to grasp the situation of the offender. Trust will be inferred from.Category 3 because it codes the action initiated by the offender'himself. The alternatives open to the offender in this category is 1) tell the truth to the offended party without being questioned by the latter 2) tell the truth after being questioned by the latter 3) lie or deceive offended.party without being questioned by the same 4) lie or deceive offended party after being questioned by the latter. If the respondent made the offender choose either Alternative 1 or Alternative 2, he is deemed to be indi— cating trust toward that character in the story. If on the other'hand the respondent made the offender Choose.Alternative 3 or Alternative 4, he is deemed to be indicating distrust toward that character in the story. The coding manual fer all three story completions contains twenty-two categories: l- 10- 11/17— 12/18- 13/19- 14/20- 15/21- 16/22- 17 .All Protocols Home Country of Respondent Year of Administration (of incomplete stories) Age Sex Grade Level of Respondent Story 1 Protocols (The Bicycle Story) [Erection of Communication (regarding property damage) Son's or ReSpondent's Definition of Situation (re: damage of bike) Father's Behavioral Response to the Situation (re: damage of bike) Content of Father's Communication to Son (after son learns of property damage) Son's Affective Response Toward Father (after learning of bike damage) Story 4 (The Broken Knife) and Story 5 (The Soiled Dress) Protocols Direction of Communication (regarding property damage) Older Sibling's (or Respondent's) Definition of the Situation (re: damage of property) YOunger Sibling's Behavioral Response to the Situation (re: damage of older Sibling's property) Content of Younger Sibling's Communication (after older sibling learns of property damage) Older Sibling's Behavioral Response to the Situation (after learning of damage of property) Older Sibling's Affective Response to the Situation (after learning of damage of property) 18 It is in terms of these categories that the hypotheses of this study will be tested. Reliability Check Two coders, an.M.A. candidate in PsyChology and the author, inde— pendently coded 15 protocols from eaCh country (a total of 45) to CheCk on inter—coder~reliability. This resulted in 92 per cent agreement. The fermula used to determine per cent of agreement was the fellowing: Number of Agreements Total No. othategoriesCBded x 100 A proportionate stratified random sample of 70 protocols eaCh were drawn randomly for the U.S. and Brazil, and 71 fer Germany. Proportion- ate drawing was deemed necessary because of the several towns involved in the German and U.S. samples. Thus, a total of 211 protocols were used in this study, distributed as fellows: Gernany 71 Hamburg Braunschweig MuniCh Brazil (Rio de Janeiro) 70 United States 70 Benton Harbor, MiChigan 45 Cedar Rapids, Iowa 25 Total 11 .All of these 211 protocols were coded by the author. A.frequency count of all the codebook subcategories by country served as the guide fer collapsing some of these subcategories. The reason fer collapsing is that some of the subcategories did not occur frequently enough (less than five per cent) to justify using them for 19 cross—tabulation. Analysis Scheme The operational statement of our theoretic hypotheses and the manner in WhiCh these operational hypotheses will be tested are as fellows: Hypothesis 1: The respondent's perception of confrontation between offended party and offender is affected by the generational status of the former with respect to the latter in the situation (story); that is, the elder Child is more likely to be made to confront the younger sibling (Stories 2 and 3) than the son to confront the father (Story 1). This hypothesis will be tested by obtaining a correlation measure (phi) fer Categories 11 and 17 which are the responses to the sibling stories on the direction of communication, and correlating Category 6 whiCh is the response to the Broken Bicycle story on direction of com- munication to Category 11 and also to Category 17. It is expected that responses to Categories 11 and 17 will be highly consistent; in other words, positively correlated. If the hypothesized relationship is equally true to all three countries in this study, then the deference pattern is no basis for classifying them into "modern" and "traditional" societies. If, however, there are significant differences in the proportion of responses to the direction of initial communication regarding the offense between the sibling stories and the fatherbson story, then it might be usefu1 to make the classification. 20 Hypothesis 2: The age of the respondent affects the perception of the degree of empathy of the older sibling for the younger sibling; that is, the older respondents will have a greater tendency to ascribe empathy to the older sibling than will the younger respondents. The ages of the respondents range from age 9 to age 18. The "Older? group consists of ages 12 to 18, while the "YOunger" group con— sists of ages 9 to 11. It is expected that this age grouping represents significant differences in maturational development, though no more than an "educated" guess was used to distribute the respondents into suCh . groups. The "Older“ group is expected to be more mature than the "Younger" group. There are three measures of empathy in this study which are not mutually independent. One measure is the offended party's or the re- spondent's definition of the situation (damage of property and younger sibling), which is Category 12 fer Story 2, and Category 18 fer Story 3. The second measure is the offended.party's affective response toward the offender'whiCh is Category 16 for Story 2, Category 22 fer Story 3. The third measure is the offended party's behavioral response to the situa— tion (after learning of damage of property), whiCh is Category 15 for Story 2 and Category 21 for Story 3. A measure of correlation (phi) will be taken fer all three measures, and responses to Categories 12, 15, 16, 18, 21 and 22 will be correlated with Category 3 (age group of the respondent). It is ex- pected that if in fact the three empathy measures are positively cor- related with eaCh other, then there will be a.high correlation between results of correlating Category 3 with 12, 15, 16, 18 and 22. In other words, eaCh of the three.measures serve as a Check on the relationship 21 between empathy and age. Hypothesis 3: The respondent's level of trust (empathy) is affected by the generational status of the Characters in the situation (story); that is, respondents are more likely to make the father lie or deceive his son in the "Broken Bicycle" than make the younger sibling lie or deceive the older sibling in the "Broken Knife" and "Soiled Dress" stories. The reason fer this expectation is that even families in “modern" societies carry over a certain amount of the traits in the "traditional" society with respect to deference. Category 8, the father's behavioral response to the situation (damage of bike) for Story 1 will be correlated with Categories 13 and 19, which are the behavioral responses of the younger sibling to the damage of property in Story 2 and Story 3 respectively. A measure of correlation between Categories 13 and 19 will be taken, and it is expected that the correlation will be high, and a positive one. Hypothesis 4: There are significant differences between nation— ality groups with.respect to the respondent's perception of the effect of generational status on the direction of initial communication about the offense between the offender and the offended in all three stories illustrating potential tension situations. In Brazil, the father (offender) will be perceived as being more likely to initiate communication about the offense than the son (offend- ed) is. In Gernany and the U.S., the son will be perceived as more likely to initiate communication about the offense than the father will. In Brazil, the older sibling (offended) will be perceived as more likely to initiate the communication than the younger sibling 22 (offender) will be. In Germany and the U.S., the younger sibling (offender) will be perceived to be just as likely to initiate the com- munication as the older sibling. The correlations (phi) obtained for each country for variables 6 and 11, and fer variables 6 and 17 will be compared.with eaCh other. Hypothesis 5: There are significant differences between nation— ality groups with respect to the relation between the respondent's age and their perception of the degree of empathy of the older sibling fer the younger sibling in the stories illustrating potential tension situations. In Brazil, more older'respondents than younger respondents will ascribe empathy to the older sibling in both sibling stories. The prOportion of the older respondents ascribing empathy to those not ascribing empathy will be smaller for Brazil than fer the U.S. and Germany. FUrthermore, there will be a greater proportion of younger respondents who will ascribe empathy to the older sibling as against those WhO‘Will not in the U.S. and Germany than in Brazil. This hypothesis involves only Stories 2 and 3, and not Story 1. Again, the correlations (phi) obtained fer eaCh country between vari— . able 3 and variables 12, 15, 16, 18, 21 and 22 will be compared with eaCh other. Hypothesis 6: There are significant differences between nation- ality groups with respect to the effect of generational status of Characters in the stories and the level of trust respondents show through these Characters. In Brazil, more respondents will show distrust toward the father (offender) than toward the younger sibling (offender). In the U.S. and 23 Germany, there will be an equal proportion of respondents who show dis- trust toward either father or younger sibling, to respondents who show trust toward either Characters. This hypothesis involves all three stories, and the correlation between variables 8 and 13 will be compared with the correlation between variables 8 and 19 for eaCh country. Hypothesis 7: The extent of'mass media development of a country affects the level of empathy of its people; that is, the higher the level of development of mass media, the more empathic people are, and vice versa. It is expected that there will be a significant difference be- tween Brazil on the one hand, and Germany—U.S. combined on the other. More precisely, it is expected that there will be more empathic re— sponses in Germany—U.S. combined which we have called "modern" on the basis of the extent of mass media development as contrasted with Brazil whiCh we have called "traditional." This hypothesis will be tested by pooling the data for the U.S. and Germany and comparing them with those of Brazil, on the three measures of empathy whiCh are contained in Categories 7, 10, 12, 15, 16, 18, 21 and 22. In using a projective measure to test the above hypotheses, the following assumptions were made: 1. The respondents, as individual personalities, are controlled to a large extent by the culture in which they live. Con— versely, the individual personalities, idiomatically using the prescribed social and cultural patterns, themselves create the social and cultural.milieu. 20 The respondent shares meanings with the society to which he belongs. The story completions represent what the situations presented in the stories mean to the individual. The story completions reflect the pattern of interpersonal relations arising in potential tension situations in the culture where the respondent has been reared. YOunger respondents will identify with the younger sibling in the story, while older respondents will identify with the older sibling in the story. CHAPTER III FINDINGS Not all the story completions had responses fOr eaCh category in the code book that was set up fOr this study. The net result of this is that the frequencies fOr a number of subcategories were very low, and.made it necessary that the subcategories be collapsed so that the data could be analyzed by appropriate statistical means. Most of the variables were diChotomized into "positive" and "negative" values, indicating the presence (positive) or absence (negative) of the particular trait being measured. This is particularly true of the three measures of empathy and the measure fOr trust. As to the category "Direction of Communication," the diChotomy consists of l) offender initiates communication and 2) offended party initiates com- munication. Responses indicating that either offender or offended party approaChed a.third person about the offense were eliminated in the correlational analysis as they are not necessary in the testing of the hypotheses. For Hypotheses l, 3, 4, 6 and 7, the X2 test is used here. The fourfold point correlation coefficient, or the phi_co- efficient is the statistic used here to test Hypotheses 2 and 5. Correlation Between the Measures of Empathy_ The three measures of empathy fOr the two sibling stories were correlated, as shown in Tables 2, 3 and 4. 25 26 TWO of the three ("Definition of Situation" and "Behavioral Response") measures of empathy are positively correlated across the two stories. "Affective Response," though indicating possible negative correlation, did not attain significance. It should be remembered that the correlation is between the responses of the same individual to the two incomplete stories. As mentioned earlier, not all of the story completions could be coded on all the categories in the code book. Some of the protocols were coded on a measure of empathy in Story 2, but could not be coded on the same measure in Story 3, which accounts for the zero frequency and conse- quently fOr the negative correlation obtained in Table 3. Table 2. Correlation between offended party's definition of the situation in the two sibling stories (o = .400, p < .05) Story 2 Non-empathic Empathic Empathic 2 13 Story 1 Non-empathic 10 9 Table 3. Correlation between offended party's affective response to the situation in the two sibling stories (0 = -.036, N.S.) Story 2 Negative Positive (Non-empathic) (Empathic) Positive (Empathic) 2 0 Story 1 Negative (Non-empathic) 53 2 27 Table 4. Correlation between offended party‘s behavior response to the situation in the two sibling stories (0 = .332, p <:.01) Story 2 Negative Positive (Non—empathic) (Empathic) Positive (Empathic) 13 21 Negative (Non—empathic) 44 17 Correlation Between Measures of Trust The same problem.encountered in correlating the three measures of empathy across the two sibling stories was present in correlating the one measure of trust ("Offender's Behavior Response") across the two sibling stories. As shown in Table 5. There were only six protocols which were coded fOr trust on both Stories 2 and 3. It is not surprising therefore that a negative correlation is indicated for trust responses across the two stories. Table 5. Correlation between the offender's behavior response in the two sibling stories (0 = .500, N.S.) Story 2 lie/deception Truth Truth 2 2 Story 1 Lie/deception 0 2 Results of the Tests of Hypotheses Hypothesis 1 is a prediction that generational status will make a difference in the direction of initial communication about the offense. More specifically, it was expected that the Character*who is higher in 28 the generational hierarChy will initiate the communication, whether*he or she is the offender or the offended party. Table 6 shows an indication that if there is a correlation be- tween the direction of communication in the two sibling stories, it may be a negative one. The obtained correlation, however, is very small (0 = -.064) and is not significant. The negative direction it took can be explained by the fact that in Story 3, the incomplete story specifi- cally states that the younger sibling (offender) initiated the communi- cation. It was deemed important to code Story 3 on the "Direction of Communication" because a number of respondents altered the direction of communication as if they could not believe that the younger sibling could indeed initiate the communication about the offense. Table 6. Correlation between direction of initial communication in the two sibling stories (0 = —.064, N.S.) Story 2 Younger sibling- Older sibling—to— to—older sibling younger sibling Older sibling—to— 6 52 younger sibling Story 1 Younger sibling— 0 2 to-older sibling Hypothesis 1 is confirmed by the data, as shown in Tables 7 and 8. In Story 1, the offender (father) initiated the communication in 74 our of 141 or 53 per cent of the protocols coded fer this category. In Story 2, 91 out of 95, or 92 per cent of the protocols coded for this category had the offended (older sibling) initiate the communication. In Story 3, 93 out of 104, or 90 per cent of the proto- cols coded fOr this category had the offended (older sibling) initiate 29 the communication too. Table 7. Relationship between direction of initial communication and V generational hierarChy, Stories 1 and 2 Offenderhto— Offended-to- Offender'higher in offended offender X2 generational hierarchy (Story 1) 74 67 57.38943 Offender lower in generational hierarChy (Story 2) 4 91 df = l, p.< .001 Table 8. Relationship between direction of initial communication and generational hierarchy, Stories 1 and 3 Offenderhto— Offended-to— 2 Offender*higher in offended offender X generational ' hierarchy (Story 1) 74 67 4.4440 Offender lower in 7 generational hierarChy (Story 3) ll 93 df = l, p‘<’.05 Hypothesis 2 is a statement of expectation that the older re- spondents will tend to ascribe a.higher degree of empathy to the older sibling than will the younger respondents. This is based on the assump- tion that the older respondents will identify with the older sibling in Stories 2 and 3. Since empathy is developmental, a self—insight that they do have a capacity to understand the role of the younger sibling in the stories, will make the older respondents attribute more empathy to the older sibling than the younger respondents will. The data does not support Hypothesis 2, as shown in Tables 9, 10, ll, 12, 13 and 14. Moreover, there is a slight indication that the relationship between age and empathy may even be negative, though the negative correlations obtained for both stories are very small and non— significant. It is more likely that in suCh potential conflict 30 situations, the offended party will perceive the offender as having no ill motive regardless of the age of the offended parry. Table 9. Correlation between age group of respondent and offended party's definition of the situation, Story 2 Non—empathic Empathic 0 Total Older 11 22 -.279 Younger 0 7 N.S. Table 10. Correlation between age group of respondent and offended party's definition of the situation, Story 3 Non-empathic Empathic 0 Total Older 13 16 -.027 Younger 5 7 N.S. Table 11. Correlation between age group of respondent and offended party's affective response to the situation, Story 2 Negative Positive (Non—empathic) (Empathic) 0 Total Older 63 l -.033 Younger 23 l N.S. Table 12. Correlation between age group of respondent and offended party's affective response to the situation, Story 3 Negative Positive (NCn—empathic) (Empathic) 0 Total Older 53 3 —.024 Younger 24 2 N.S. 30 situations, the offended party will perceive the o " ill motive regardless of the age of the offended ”Ht Table 9. Correlation between age group of 5;..i; f party's definition of the situation, ....‘_..; Non—empathic Total Older 11 Younger 0 ,— - . Table 10. Correlation between age group of. j? ’ party's definition Of the si. .- a ' 0 Non—erpathic N031 , Total ‘ Older 13 Younger 5 . ‘ l of trust 15 x in the potential Table 11. Correlation between age 7.." party's affective res-{.5 ' ‘ likely ”C0 make -icycle" than make ’ riled Dress" stories. .1? Hypothesis 3, as shown Total Older . Younger 2 (Table 15), it can be w. a the father than toward Table .12. Correlation < .05) by making the father party's - af' .ced) more oftenthanthe to the older sibling (offended). able 16) did not yield any 31 Table 13. Correlation between age group of respondent and offended party's behavioral response to the situation, Story 2 Negative Positive (Non-empathic) (Empathic) 0 Total Older 58 16 —.l43 Younger 29 5 N.S. Table 14. Correlation between age group of respondent and offended party's behavioral response to the situation, Story 3 Negative Positive (Non—empathic) (Empathic) 0 Total Older 53 34 .031 Younger 32 18 N.S. Hypothesis 3 states that the respondent's level of trust is affected by the generational status of the Characters in the potential conflict situation; that is, the respondents are more likely to make the father lie or deceive his son in the "Broken Bicycle" than make the younger sibling in the "Broken Knife" and "Soiled Dress" stories. The data.partly supports the opposite of Hypothesis 3, as shown in Tables 15 and 16. Comparing Stories 1 and 2 (Table 15), it can be seen that the respondents show more trust toward the father than toward the younger sibling (X2 = 4.2509, df = 1, pt( .05) by making the father (offender) tell the truth to his son (offended) more often than the younger sibling (offender) tell the truth to the older sibling (offended). The comparison between Stories 1 and 3 (Table 16) did not yield any significant difference. 32 Table 15. Relationship between generational hierarChy and behavioral response of offender, Stories 1 and 2 Truth Lie/deception X2 (Trust) (Distrust) Total Father 91 17 4.2509 Younger sibling 35 16 df = l,p <3 .05 Table 16. Relationship between generational hierarChy and behavioral response of offender, Stories 1 and 3 Truth Lie/deception X2 (Trust) (Distrust) Total Father 91 17 .1281 Younger sibling 12 1 df = l, N.S. Hypothesis 4 is a prediction that in Brazil, which we have arbitrarily called "traditional," the character involved in a dyadic potential conflict situation.who is higher in the generational hier- archy will initiate the communication about the offense, regardless of his role in the conflict (i.e., whether he is the offender or the offended). Conversely, in Germany, as well as in the United States, both of which we have called "modern," it is predicted that the offended party will be the one to initiate the communication, regardless of his generational position relative to the offender. The data partly supports the opposite of the prediction fOr Brazil and the united States as shown in Tables 17 and 18. In Brazil, the offended.party is more likely to initiate the communication, regard— less of his relative generational position to the offender. The likeli— hood that the offended party will initiate the communication is increased 33 by the decrease in generational distance between offender and offended, however. In Story 1, the offender (father) initiated the communication in 57 per cent of the cases, while in Story 2, the offended (older sibling) initiated the communication in 100 per cent of the cases. (x = 10.9333, df = 1, p < .001) In the United States, the Character'higher in the generational hierarchy is more likely to initiate the communication about the offense, regardless of his role in the conflict. The prediction fOr Germany is confirmed by the data, for Stories 1 and 2 (Table 17), but the data for Stories 1 and 3 (Table 18) did not reach significance. Comparing Stories 1, it can be seen that the offended party is more likely to initiate communication, regardless of his rela- tive generational position to the offender. As in Brazil, the likelihood that this will happen is increased by a decrease in generational distance between the two parties in the conflict situation. In Story 1, the offended party initiated communication in 59 per cent of the cases, while in Story 2, the offended party initiated communication in 93 per cent of the cases. Hypothesis 5 is an expectation that in Brazil, more older re- spondents than younger respondents will ascribe empathy to the older sibling in both sibling stories. However, it is expected that this proportion of the older respondents ascribing empathy will be smaller for Brazil than for the U.S. and Germany. As shown in Tables 19, 20, 21, 22, 23 and 24, no conclusive statement can be said about Hypothesis 5 no significant direction of re— lationship between the age group of the respondent and the offended party's empathy toward the offender is indicated. Two exceptions may 30 Table 17. Relationship between direction of initial communication and generational hierarchy, Stories 1 and 2 Offenderbto- Offended-to- Germany: Offender’higher in offended offender X2 ‘ generational hierarchy (Story 1) 22 29 16.2068 Offender lower in generational hierarChy (Story 2) 2 41 df = l,p<<.001 Brazil: Offender higher in generational hierarChy (Story 1) 18 24 14.9333 'Offender lower in generational hierarChy (Story 2) 0 30 df = l,p<<.001 U.S: Offender higher in generational hierarChy (Story 1) 34 14 20.6169 Offender lower in generational hierarChy (Story 2) 2 20 df = l,p