”9 III'TGII IIIIIIIIIIIIIIII I 3 1293 10570 6224 This is to certify that the thesis entitled THE INFLUENCE OF COMMUNICATION ON BLACK CHILDREN'S SELF CONCEPT presented by STEVEN T. McDERMOTT has been accepted towards fulfillment of the requirements for Ph D degree in (1}) M. W- Q A} :5 m- Major ‘profecsor U Date CDC";;’F I“; ) 1101.732 0-7639 I CD «\ MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES wiII be charged if book is returned after the date stamped beIow. THE INFLUENCE OF COMMUNICATION ON BLACK CHILDREN'S SELF-CONCEPT by Steven T. McDermott A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirement for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1980 ra in va :19 ii co: the bet and raw ‘- ABSTRACT THE INFLUENCE OF COMMUNICATION ON BLACK CHILDREN'S SELF CONCEPT By Steven T. McDermott The research examined the influence of communication on black children's racial and self-esteem. A review of the literature pointed out deficiencies in past esteem development research, such as a focus on only one or two variables at a time, the failure to account for communication in esteem development, and conception and measurement problems with esteem. To overcome these deficiencies, a survey of fourth and fifth grade black children was conducted to assess the overall and interactive influence of communication on three aspects of esteem development: (1) self-esteem, (2) racial esteem, or that perception of value or worth a child holds for his/her own racial group, and (3) comparative esteem, which was conceptualized as the child's conception of the value or worth of his/her own racial esteem as compared with the esteem held toward another racial group (in this study, the difference rebetween black and white perceptions). A total of 18 hypotheses predicted that (1) communication from parents and peers that emphasized positive characteristics of black people would be related to more positive esteem levels than communication that does not emphasize positive characteristics, (2) frequent viewers of blackrfamily televdsion shows would have more positive attitudes toward black child. and adult television characters than those who are substantially less frequent viewers, (3) children who are heavier viewers of black-family shows and who have highly positive attitudes toward adult and child ti C: C2 to PE: 3U “as Cl» ‘ Steven T. McDermott television characters will have higher self, racial and comparative esteem than those who are less frequent viewers, and (4) communication from.parents and peers which emphasizes positive characteristics of black adult and child television characters will be more highly positively related to positive attitudes toward black adult and child television characters than communi- cation that does not emphasize positive characteristics. Survey data were collected from 82 fourth and fifth grade children. In addition to descriptive results that indicated among other findings that black children have fairly positive esteem levels, and that communi- cation levels are fairly high, 11 of the 18 hypotheses were supported. It appears that communication is an important factorzin socialization and in relation to black children's esteem development. The hypotheses predicting a relationship between peer and parental communication and racial esteem.were supported. Support was also found for the hypotheses relating parental communication to self-esteem but not for the prediction that self-esteem would be related to peer communication. Interpersonal communication was not related to comparative racial esteem. Mere exposure to the black-family television shows was related to self-esteem. Inter- personal communication from.parents and peers appears to heighten positive attitudes toward television characters and high viewership of black—family shows was also related to attitudes toward the characters on the shows. Regression analyses indicated that parental communication was a most important predictor of esteem and attitudes toward television characters was a signficantly larger predictor when the characters were adult and not child. Accepted by the faculty of the Department of Communication, College of Communication Arts. Michigan State University, in partial fulfillment of the require- ments for the Doctor of Philosophy degree. ii. ACKNOWLEDGMENTS To so easily thank people on an acknowledgment page seems to me to be grossly inadequate as each of the‘people mentioned below deserve so much more for what they gave me. Bradley Greenberg, my advisor and dissertation director, was a constant source of inspiration, rigor, and direction. He prodded me when it was necessary and left me alone when that was too. My work with him had a huge impact on this research and it will no doubt influence my future research. Charles Atkin demonstrated the very best teaching and research as I worked for and with him. He continues to be a good model and friend. Gerald Miller has greatly influenced what I think and do about communication research. His insights and skills as a teacher were only added to by the pleasant times I've had learning from him or being with him. Eileen Earhart gave much of her valuable time and expertise in child development to this research. Although not directly connected with this research, I must thank Cal Hylton for his support and friendship during my days of writing and collecting data. Finally, I formally thank Marilynn for her stability, patience, warmth and love. She unfortunately had to bear many of the frustrations of this effort and yet received too few of the rewards. TABLE OF CONTENTS Acknowledgments Table of Contents List of Tables List of Figures CHAPTER I - RATIONALE AND HYPOTHESES Introduction Rationale Socialization Learning Self/Other Concept. Influences on Self/Other Conceptions Communication Influences on Self-esteem Hypotheses la, 1b, and 1c Hypotheses 2a, 2b, and 2c Hypotheses 3a and 3b Hypotheses 4a1, 4a2, and 4a3 Hypotheses 4b1, 4b2, and 4b3 Hypotheses 5a and 5b Hypotheses 6a and 6b CHAPTER II - METHODS AND PROCEDURES Administration Measures Esteem Interpersonal Communication iii. 35. 36. 38. 38. 38. 40. 47. TABLE OF CONTENTS (continued) CHAPTER II - METHODS AND PROCEDURES (continued) Television Exposure Attitudes Toward Television Characters Interpersonal Communication about Television Characters Statistical Analysis CHAPTER III - RESULTS Descriptive Results Test of Hypotheses Regression Analyses Exploratory Analyses CHAPTER IV - DISCUSSION Descriptive Results Hypotheses Theoretical Issues Learning Explanations The Role of Communication Reservations Suggestions for Future Research ‘FOOTNOTES REFERENCES APPENDICES - STUDY QUESTIONNAIRE iv. 49. 51. 52. 54. 58. 58. 64. 76. 84. 89. 89. 93. 97. 97. 101. 105. 108. 112. 113. 120. Table 1. 11. 12. 13. LIST OF TABLES Simple Correlation Coefficients for Selfdesteem Items Simple Correlation Coefficients for Absolute Racial Esteem Items Simple Correlation Coefficients of Interpersonal Communication Items for Parents and Peers Inter-Item Correlations for Exposure to Black-Family Television Shows Inter-Item Correlations for Attitudes Toward Black Television Characters Inter-Item Correlations for Interpersonal Communication from Friends About Television Characters Inter-Item Correlations for Interpersonal Communication from Parents About Television Characters Intercorrelations of Esteem Measures Descriptive Summary of Indices Mean Split Elaboration for Hypotheses 4a, 4b, and 4c: Correlation Between Exposure and Esteem for Two Levels of Attitude Toward Television Characters Hypotheses Results Summary Table Regression of Parental Communication About Black Children, Peer Communication About Black Children and Attitudes Toward Black Child Television Characters on Self-Esteem Regression of Parental Communication About Black Adults, Peer Communication About Black Adults, and Attitudes Toward Black Adult Television Characters on Self-Esteem 46. 48. SO. 53. 55. 56. 60. 65. 71. 74. 78. 79. vi. TABLE OF CONTENTS (continued) Table Page 14. Regression of Parental Communication About Black Children, Peer Communication About Black Children, and Attitudes Toward Black Child Television Characters on Absolute Racial Esteem 80. 15. Regression of Parental Communication About Black Adults, Peer Communication About Black Adults, and Attitudes Toward Black Adult Television Characters on Absolute Racial, Esteem 81. 16. Regression of Parental Communication About Black Children, Peer Communication About Black Children, and Attitudes Toward Black Child Television Characters on Comparative Racial Esteem 82. 1?. Regression of Parental Communication About Black Children, Peer Communication About Black Children, and Attitudes Toward Black Adult Television Characters on Comparative Racial Esteem 83. 18. Regression of Parental Communication About Black Children, Peer Communication About Black Children, and Exposure to Black Family Television Programs on Self-Esteem 85. 19. Regression of Exposure, Parental Communication About Television Characters, and Peer Communication About Television Characters on Attitudes Toward Black Child Television Characters 86. 20. Regression of Exposure, Parental Communication About Television Characters, and Peer Communication About Television Characters on Attitudes Toward Black Adult Television Characters > 87. Figure 3. 4. LIST Research Schematic Results Results Results Results Results for Hypotheses for Hypotheses for Hypotheses for Hypotheses for Hypotheses OF FIGURES la, lb, 1c 2a, 2b, 2c 3a, 3b 5a, 5b 6a, 6b vii. f 73. 73. CHAPTER I RATIONALE AND HYPOTHESES Introduction How others treat us, reward or punish us, is an important determinant of our sense of self. If we are not rewarded or appear to have little control over our environment (Miller & Steinberg, 1975) we develop low self-concept, including a sense of powerlessness and worth. The intrinsic power or worth a person attaches to him/her- self is termed self-esteem (Rosenberg, 1965). People who are high in self-esteem feel they have control over their environ- ment with feelings of self-confidence and self-approval, whereas people who are low in self-esteem have feelings of self— rejections, self-dissatisfaction, and self-contempt. One part of a person's self-esteem is a result of his/her communication with others and how s/he views his/her role in society. Some of the information a person gathers about his/her role is based on such limited characteristics as gender, age, and race. For instance, a black child forms his/her sense of self by comparison with other role groups in his/her environ- ment, such as white peOple. To the extent that self and con- ceptions of others differ, the probability is increased for polarization and the potential for intergroup conflict is enhanced (Ashmore & Del Boca, 1976). Other parts of a person's overall self-concept may not necessarily be based on such group character comparisons, such as race, but is a result of his/her unique reinforcement scheduling. In other words, a person's overall self—concept contains two parts, his/her conception of his/her own role group (race) compared with other role groups, and his/her self-concept which is independent of racial conceptions. The primary emphasis of the present research is that which is race-related, even though the past research on self-esteem has not made clear the distinction between race-related and non race—related self- esteem. Given the importance of self-esteem, especially in terms of its implications for racial interaction, many researchers have examined the determinants of self and conceptions of others (Katz, 1976). A central emphasis of this reserach has centered on the determinants of people's beliefs and attitudes about other's and their own race. Questions of interest have dealt with white people's perceptions and expectations about black people, black people's expectations about white people, and self— images of both groups. Most of the research dealing with self- imege has investigated black children's self-concept compared with white children's self-concept. The current research seeks to overcome certain shortcomings of the past research. The past developmental research in this area has tradition- ally'looked at the independent influence of schools, parents, jpeers, or other people on the self-concept of black children. Most of this research has not adhered to a multi-influence model «sf the process, but instead has focused on one or perhaps two .Lnfluences at a time (Liebert & Poulos, 1976). As a 3. consequence, little is known about the interactive and relative effects of various influences on self-concept. A second shortcoming of the past research is the failure to account for the effect of communication, especially the identification of specific types of communication and their effects on beliefs and attitudes (Maccoby, 1964). Both inter: personal communication and television can provide important sources of information for the development of self-concept in children. I With these concerns in mind, the present research is an attempt to deal with multiple influences by studying the effects of (race-related) interpersonal communication with parents and peers and television experiences on pre-adolescent black children's self-concept. Interpersonal communication shall be studies from two major perSpectives: first, both parents and peers may commu- nicate directly to a child by making evaluative comments about' black and white people, and; second, they may provide evalua- tions of the depictions of certain race roles on television. By affecting children's attitudes toward television depictions, these parental and peer comments about the televised depictions of race roles may act to intervene the direct effects of television exposure on self-concept. A general schematic of the research is shown in Figure l. Arguments for the direction of the relationships and a discussion of the variables shown here, will be presented in the remaining text to follow. Figure 1 Research Schematic Direct Communication From Parents -—- "Se1f—concept Direct Communication From Peers Attitudes Toward Television Characters Television Exposure ' Parent Communication Peer Communication About TV Characters About TV Characters Rationale This section reviews conceptualizations and empirical research findings applicable to an examination of the develop— ment of black children's self/other conceptions. The section begins with a discussion of socialization learning, including the theoretical explanation most suited for examination of com- munication influences on self/other concept. Next, the past research in black self/other concept is reviewed. Following this, the past research examining influences on self/other concept is presented, commencing with interpersonal communica- tion influences and ending in television exposure effects. From this background, past research pertaining to the model is dis- cussed as it pertains to black children's learning from tele- vision. The section finishes with a summary of the relevant hypotheses pr0posed for test. Socialization Learning A person learns what is appropriate for his/her group through informational inputs about what is normative for certain roles. The learning of one's own role is contingent upon the assessment of other's roles in relation to oneself (Brim & Wheeler, 1966). Similarly, the expectation about other's roles is contingent upon the other's relative position with regard to one's own perceived role. Thus, a black child's self-concept is tied to and relative to his/her conception of his/her own ascribed role group (i.e., black people in general) and his/her conception of other ascribed role groups (i.e., white people in general). With this type of conception it is sometimes easy to forget that roles are based on expectations and that certain behaviors do not automatically go with certain groups of people. There is an assumption here, however, that children may expect certain behaviors to be based on a person's role alone, be they stereotypic or not. Children learn about themselves and others through a process of socialization. .The socialization process may be seen as an interplay of varying forces acting on a child which helps define images of self, his/her role group, and other's roles. These expectations are collections of cognitions--beliefs, subjective proba- bilities, and elements of knowledge—-which specify in relation to complementary roles, the rights and duties, the appropriate conduct, for persons occupying a partic- ular position (Sarbin & Allen, 1969, p. 498). 6. Of course, self, self-other, and other expectations are inter- dependent. For instance if one has a low image of his/her social role group, then he/she might be more likely, when iden- tifying with that group, to carry a congruent low self-image. Conversely, if one has a high image of his/her role group, any "other" group might be considered lower, and the congruent high self-image might accrue. As such, any self-image depends on the "other" as a reference anchor. Therefore, it is important that research on self-concept examine both self and other conceptions to assess the relative level of self-concept. Thus, the present research examines black children's self-image, including per- ceptions of themselves compared with Urfir perceptions of others (whites). These roles are thought to be acquired through a process of socialization (Sarbin & Allen 1969). The basic premise of this view is that what is learned through socialization is a series of complex interpersonal relationships (Brim & Wheeler, 1966). The process of learning about roles and self/others is complex, yet certain learning processes may be involved. McLeod and O'Keefe (1972) suggest three types of learning are prominent in socialization processes. These three are (1) reinforcement, (2) modeling, and (3) social interaction. Reinforcement conceptualizations deal with the positive and negative reward mechanisms in the environment which bring about role expectations. A child learns to repeat a given behavior that has been rewarded by a socializing input and/or avoid 7. behaviors which are negatively reinforced. This simplified, paraphrased view of the learning process, as presented by McLeod and O‘Keefe, captures the basic process, yet does not reflect the range of reinforcement conceptualizations of which there are many. There has been an abundant amount of research and theo- retical work exemplifying the reinforcement conceptualization's range for explaining learning with a variety of consequent variables including behavioral, as well as attitudinal, belief, and other responses (see for example, Weiss, 1962). The second type of learning McLeod and O'Keefe referred to is modeling. Modeling is one outcome of learning as conceptual- ized by Bandura (1971; 1977) in social learning theory. According to the theory, social learning may take several forms: (1) observational learning; (2) strengthening or weakening of inhibitions; or (3) response facilitation (Bandura & Walters, 1963). For learning of self/other conceptions, it is useful to look at all three types of social learning, for all assume that learning is a vicarious process whereby the behavior of a person may change as a result of the observation of the reinforced action of others. Unlike reinforcement learning conceptualiza- tions, direct reinforcement of the observer is not a necessary condition for learning to take place. . . . (Mlodeling influences produce learning principally through their informative function. During exposure observers acquire mainly symbolic representations of the modeled activities which serve as guides for appropriate performances (Bandura, 1977, p. 24). 8. Thus, according to social learning theory, the basic process is the same whether the behavior is conveyed symbolically or through live action. This distinction becomes important for explaining the effects of such symbolic representations as are on television. Observational. This type of learning depends on the stimuli of the particular source. The stimuli must be different or novel from previously learned behaviors or attitudes in the child's repertoire of behaviors. The child learns novel response patterns by viewing or hearing a verbal description of the reinforced actions of others. An example of this type of learning would be children who view television which depicts new ways to commit crimes or violence, and thus are influenced by incorporating the new behaviors into their behavioral repertoire of possible behaviors. For the black child, consistently presented new images of black people on television could lead to new or novel ideas or behaviors. For example, children who have seen the TV character, J. J., exclaim "DYN--OOO--MITE" might be more likely to see that as acceptable or probable behavior for similar situations that require exclamation. More broad based behaviors and images might also accrue according to this basic proposition. For instance, if a whole set of verbal behaviors are new to a child, and are reinforced on television, children maylearn these new and varied verbal behaviors. Inhibitory. In addition to novel response acquisitions, 9. previously learned inhibitions may be strengthened or weakened by vicarious experiences that are reinforced. Negatively rein- forced or punished responses inhibit expression of similar behavior, whereas rewarded responses tend to lead to disin- tfibfljon expression of similar behaviors. For example, consider the child who views a model being rewarded for smoking. According to this inhibition proposition, the child is likely to be less inhibited or disinhibited with regard to smoking. Conversely, assume the child sees a model punished for smoking; the child would be subsequently inhibited with regard to smoking. This inhibitory learning, then, differs from the observational learning preposition in that no new responses are acquired and it is assumed here that the child already had the behavior (e.g., smoking) in hiS/her behavioral repertoire. Response Facilitation. This learning type refers to a case where the responses facilitated are not novel, are already a part of the observer's repertoire, and are not socially inhibited responses. In other words, response facilitation occurs when a socially acceptable behavior is cued by some external reminder of some sort, such as television. It is unlike observational learning in that it deals with an existing class of behaviors. In learning about one's self and others, it would seem that television may provide new ideas about one's self, and/or help facilitate the expression of previously learned behaviors. Given that the effects of television should be stronger when 10. there are few cues from the environment about a particular behavior, one might expect a large measure of observational type learning. It is clear that ". . . the mass media play an influential role in shaping behavior and social attitudes (Bandura, 1977)." This learning occurs on the basis of mere observations, whether observation facilitates past learned behaviors or presents new conceptions (Zimmerman & Rosenthal, 1974). Past research (Siegel, 1958; Loviband, 1967; McArthur & Eisen, 1976; Bandura, 1977) has demonstrated the utility of social learning modeling for explanations of learning from media sources. These studies will be discussed below. However, despite all indications that social role learning may be probable, there haven't been any studies to date which demonstrate the learning of children's self—concept specifically, although there have been studies demonstrating racial attitude learning (Atkin, Greenberg, & McDermott, 1979). .As such, it may be useful here to look more closely at the social learning perspective and how it pertains to black children's learning of self-concept. First of all, it is not necessary for a behavior to be emitted for social learning to take place. In addition to verbal responses, ". . . formation of representation can be assessed by measures of recognition and understanding not requiring motor reproduction (Bandura, 1977, p. 36)." There- fore, the process of modeling results from cognitively formed representational structures prior to and/or instead of 11. 'performance criteria. An observer who sees a model reinforced for a particular action may integrate that action into his/her cognitive or representational repertoire. Past empirical investigations have supported this proposition. In an early study by Siegel (1958) second grade children were exposed to a radio program which either displayed taxicab drivers as aggressive or nonaggressive. When questioned later about the taxicab drivers in their town, children exposed to the more aggressive story were more likely to believe that drivers in their own town would act aggressively. Loviband (1967) found that exposure of boys to television crime and violence was correlated with acceptance of the attitudes, values and ideas expressed in the depictions. More recently, McArthur and Eisen (1976) found that the presentation of stereotypic behaviors in storybooks tended to produce an internalized motive to perform behaviors that were displayed (variations in achievement behaviors). But modeling does not itself guarantee that views which have been learned will be articulated. In the case of performance preferences, modeled judgments are learned but not expressed because fifiidiiifimr—X’em’faé’.’ 43f —a—r1§§°:ilined:§§2§‘ifed Accordingly, there is a distinction to be made between learning and performance; a person.may learn but not necessarily perform as implied by the learning. Thus, attitudes and ideas about one's self are within the purview of nonperformance learning. For an observer to behaviorally model a model's behavior they must in a sense "elect" to behave. Thus, modeling 12. differences occur on the basis of model discriminations such as . status, competence and power (Bandura, 1977). Consideration, therefore, may be given to the amount of status, for example the child perceives in the influence source, or in the case of television, the characters involved. Thus a child's attitude toward televised models will influence the extent of modeling effects. The third learning process considered by McLeod and O'Keefe for socialization is social interaction. Through this method of learning, complex interpersonal relations (including social roles) are learned through interpersonal communication. McLeod and O'Keefe argue that social norms involved in a person's interactions with others shape behaviors. This may involve combinations of modeling and direct reinforcement. That is, in social interaction a person may learn by seeing how a person treats another (modeling) and may be rewarded by personally acting a certain way (direct reinforcement). This McLeod and O'Keefe conception is probably not a specification of a learning process at all, but rather a specification of the content of influence and how a relevant other's interaction about social phenomena might be a factor for learning. Social interaction is treated as an important factor of influence for socialization in this research, but is not considered an explanatory mechanism for effect, as McLeod and O'Keefe posit. With this learning theory perspective, research dealing 13. specifically with black children's self-esteem will be reviewed. Self/Other Concept Two major lines of research are relevant to the examina- tion of black children's self-concept. These lines of research concern children's self-esteem. One line of research has been concerned with direct compar- isons between black and white stimuli as indicators of self- concept. This research paradigm is epitomized by the pioneering work of Clark and Clark (1947). They used doll choices as measures of racialpreference and evaluation. The child subject was asked to choose the doll he/she would like to play with, the nice doll, from pairs of brown and near white dolls. If a child chose a doll whose color repreSented his/her own racial group, it was construed as evidence for high self-esteem, and con- versely low self-esteem for a choice from the other color. In the first two decades of this research black children consistently demonstrated a preference for the white doll over the black doll, both as general play choices and evaluatively, i.e., chose the "nice" doll (Clark & Clark, 1947; Goodman, 1952; Morland, 1962; Radke, Sutherland, & Rosenberg, 1950; Radke & Trager, 1950; Stevenson & Stewart, 1958; Armstrong & Gregor, 1966; Greenwald & Oppenheim, 1968; Asher & Allen, 1969). In the past decade, the results of similar studies have been mixed. This has been due to several factors. Greenwald and Oppenheim (1968) suggested that there might be two factors that acted as mediators in this research: (1) denial of self- 14. identity, and "confused" self-image. Regarding their second point, they noted that in the Clark and Clark paradigm white children also made a certain amount of cross-race identification and children seemed to choose dolls closer to their own skin color, not their race. Therefore, they thought darker colored white children and light colored black children might choose a color not typically thought of as their race. To overcome this assumed artifact, Greenwald and Oppenheim expanded the range of doll choices in their study to three, adding a middle range color doll which was light brown. In doing this, they found that when the children were given the opportunity of selecting play-choice dolls more representative of their particular skin colors, they were more likely to choose a doll that was more similar to their own skin color. They also found that white children were slightly more likely to err in matching skin color, apparently perceiving the middle range colored doll to be like themselves. Thus, they found that mis-identification occurred in spite of race. However, the black dolls were still unpopular play choices and were evaluated lower than white dolls. Other stimulus issues have been brought out more recently. Katz and Zalk (1974) found that when they held hair and eye color constant across the dolls, there was no particular preference for doll color. Similarly, Kirchner and Furby (1971) found that hair type could be a significant predictor of doll choice. Katz (1976) also noted that those studies finding black-black preferences often used black testers whereas the 15. others did not, suggesting subtle experimental biases even with young children. - Most‘of these studies have been done at a time when other studies have not found black children preferring White dolls over black ones. Several authors have found in the past few years that black children do not necessarily prefer white dolls (Hraba & Grant, 1970; Fox & Jordon, 1973; Katz & Zalk, 1974). Before the last decade when researchers were finding that black children preferred the white dolls, researchers subscribed to a "deprivation" explanation. For the child who lives in a deprived environment which does not provide rewards appropriate for children to learn to manipulate their environment, his/her self-esteem is affected. The culture of the economically disad- vantaged, usually those in urban ghettos or inner-cities, often includes abundant information about one's shortcomings (Rainwater, 1970), how those around him/her aren't successful (Rosenberg, 1965), and there is high unemployment, restricted opportunities, high levels of instability, family disruption and racial discrimination (Hulbary, 1975). These perceptions may even be amplified by any perceptions of other groups of people who seem to be doing well, seemed to be rewarded, and are successful manipulators of the environment. Thus, social com- parisons with whites or middle class people are likely to further lead to lowering of self-concept (Asher & Allen, 1972). This hypothesis is poignantly demonstrated by the Coleman, Campbell, and Hobson report (1966) which found that black children in 16. integrated schools showed lower academic self-concept than children in segregated schools. . During the late 1960's, however, the inner-city black person's attitude seemed to change, despite small changes in the actual environment. The black liberation movement during this period (Hulbary, 1975), emphasizing black pride, racial aware- ness, and self-acceptance, along with some increased access to opportunities and participation in them (Bringham & Weissback, 1972; Caplan, 1970) may be major factors for the change in black choice and self-esteem seen in the studies done after the late 1960's (Butler, 1976). Katz (1976) notes that there appears to be geographical variations in the children's attitudes, with most of the studies reporting black children's preferences for black dolls having been done in urban centers and the black children's preference for white dolls having been done in the South and smaller towns. This is further evidence that increased opportunities and racial awareness in the 1960's occurred most readily in the inner cities (Caplan, 1970). The second line of research is that exemplified by Rosenberg (1965) and Coopersmith (1967), where black and white children were given carefully designed indices which measured self-esteem. Items in these measures are typified by questions such as "I feel I'm a person of worth. . ." with four responses from strongly agree to strongly disagree (Rosenberg, 1965) or more specific items such as those from Coopersmith (1967) such as ”I'm proud of my school work. . ." with a space the child marks l7. . indicating it is "like me" or "unlike me." Evidence from this type of research, which does not require a child to make com- parisons between themselves and others in explicit fashion, completed up to the late 1960's has demonstrated that black children and adolescents have lower self-evaluations than white children (Duncan, 1968; Rosenberg, 1965; Porter, 1971). Although researchers have concentrated on pre-school children, especially in the doll comparison studies, self-esteem measures have demonstrated low self-esteem for older children and even for adolescents (Rosenberg, 1965). Mere recently, Greenberg (1972) found a change in attitudes as children get older. His research showed that by ninth grade black children are beginning to have higher self-concepts than their white counter parts. ‘Caplan (1970) also found that inner-city black children who are more militant tend to have higher self-esteem. It appears that black self-images when considered alone or in comparison with white images are changing (Katz, 1976). It is still unclear how and when urban children are gaining this self-image; however, the potential information to be gained from examining the communication determinants from a multi—influence perspective (Katz, 1976; Liebert & Polous, 1976) appears to be important. Similarly, identification of critical ages for learning need further exploration, and will be addressed in the next section. Comparative data have not been analyzed from either the doll studies typified by Clark and Clark's research or from the 18. self-esteem studies typified by Rosenberg. The present study attempts to overcome these shortcomings by looking at.a compara- tive self-concept, which includes black children's view of whites. Specifically, children were asked to make three judg- ments: 1. Self-esteem, or how they view themselves; 2. Absolute esteem, or how the black children view other blacks, and; 3. Comparative esteem, or how the black children's view of black people differs from how they view white people. Influences on Self/Other Conceptions As discussed earlier, information about the self and others may come from a variety of sources. The focus of the prOposed study is communication influences. However, the environment and the extent of direct contact the children have with members of other social role groups, i.e., black students ‘with white students, must be considered. The amount of contact la child has with other role groups has certain effects on his/her self-concept and his/her conception of others. There- .foren this section will begin with a discussion of the effects indices of attitudes toward television characters, one assessed attitudes toward black adult and the other black child characters. The child index had a possible range from 6, indicating that each of the six characters in the index were not liked (i.e., a "don't like" response), to 18, indicating that each character was liked "a lot." The aetual mean was more than a standard deviation higher than the theo- retical mean of 12 (i = 14.94, SD = 2.24, and skewness = -0.373). Thus, children appear to have fairly positive attitudes toward child characters. The results of the adult index are also positive but not as much. The theoretical mean of this index with a possible range of 5 to 15 is 10. The actual mean was not quite a standard deviation away at 11.63 with SD of 2.31, and skewed negatively -0.23. A t-test of the differences between the mean attitudes toward black adult characters and black child characters was significant (t = 7.91, p (.001). To do this test, the indices were made comparative by adding a constant of 2 to the adult index. 63. Interpersonal communication about television characters. Here, two communication sources were assessed, parents and peers. For each of these sources, communication about adult and child characters was assessed. Thus, there are four indices of interpersonal communication about television characters. The content of communication from peers could vary from liking black characters "a lot" to "don't like." A low score of 6 for child character preference and a low score of 5 for adult character preference would indicate that a child didn't like each of the characters in the index. These two low scores differ because 6 characters were chosen for the child character index, while only 5 were chosen for the adult. Conversely, a score of 15 and 18 respectively would be an indication that the child's friends preferred all of the characters in the index. The mean for the peer communication about child characters was 13.71 with a SD of 2.77 and skewed negatively at -0.27, indicating that the mean is less than a standard deviation higher than the theoretical mean, and with the negative skew would mean that the majority of children received communication from their peers that was positive about black child television characters. Communication received from their peers about black adult television characters is not so positive. The mean was 10.40, SD = 2.58, and skewed -0.27. This mean is substantially below the theoretical mean. Yet the moderately 64. negative skew indicates that many of the cases are clustered toward or beyond the mean. A t-test of the differences indicates that children receive different communication about adult characters than child characters (t = 4.02, p.<.001). To do this comparison test a constant of 2 was added to the adult index. Similar discrepancies between communication about adult and child characters seem to originate from parents (t = 3.70, p (.001). To calculate the E, a constant of 2 was added to the index. The parental communication about children on television had a mean of 14.06, a SD = 3.10, and was positively skewed at 0.43. The parental communication about adults on television measure had a mean of 11.04, with a SD of 2.93 and skewed at 0.33. Thus, both peer and parental communication, while being positive about black child and adult characters, is significantly more positive toward black child characters. In Table 9 below, a summary of the descriptive statistical results is shown. Test of Hypotheses The section presents the results of the test of each of the eighteen hypothesized relationships. The first set of hypotheses, i.e., Hypotheses la, lb, and 1c, stated that communication from parents that emphasizes positive characteristics of black people is related to higher 65. mm.o.. mm.~ 3.3 o4: ma 0» m muouuouueo panes Home $6.. msé :3: a.ma ma 3 m 3300qu EH5 adobe mambo mm.o+ mm.~ 3.2 o4: ma on m 3.30330 pages pane mme.o+ aim 3.: o.~H ma 0.... m .muouumumno madman poops museums mmmeodmdmo onmH>mAmB 950mm onB¢UHZDZZOU AflzommmmmmBzH 27o: e~.~ 2.: o.~H A ma 0» G 330390 3.30 #003 mad... Hm...” no.3 o.oH ma 0» m anon—opaque ago... 000.8 mmmausmdmo ZOHMH>mqu MUdAm amezoa mmoDBHBB< mmél mwé me...“ c.o 3+ ou m7 mg... our? made 0.0on mm.o+ 34 3.3 Ema on on m msoem mass—om roman mmDmOmxm onmH>mAme mm.on so.e Hm.ma .o.~H on or s Home 36.. SJ a.ma a.ma on on v ucouom onaauHZDzzou 4420mmmmmmazH mm.on ma.m e.c I o.o mH+ 0» man Edmund 0>Huouodeoo ~m.o.+ h~.~ Ema m.~a ow on m sooumo meadows? 3.5+ ms; a.ma m.~.n a on m Eooumoumaom Emmamm coo: oocmm mmocsoxm cm x Hmowuouoone caewmmom xoocH Ammucv mooaocn mo humeezm o>wuofluomoo m manna 66. self-esteem than communication that does not emphasize positive characteristics of black people. This set of three hypotheses, included the criterion variables of self-esteem (i.e., "I think I am . . ."), absolute racial esteem (i.e., "I think black children in real life are. ..") and comparative esteem (i.e., "I think white children in real life are . . ." minus "I think black children in real life are . . ."). For hypotheses la and lb, which hypo- thesized a relationship between communication and absolute esteem and self-esteem, respectively, the data were supportive. The relationship between parental communication and self-esteem was moderate in size (r = .290, p (.006), while the relationship between parental communication and absolute racial esteem was larger (r = .418, p‘(.002). The relation- ship between comparative racial esteem and parent communica- tion was not significant (r= -.155, p (.088), yet in the right direction. See Figure 2. The second set of hypotheses, i.e. Hypotheses 2a, 2b, and 2c, stated that communication from peers which emphasizes positive characteristics of black people will lead to higher esteem among black children than communication which does not emphasize positive characteristics of black people. The relationship between peer communication and self- esteem was not significant (r = -.112, p‘<.l7l). Also, the 67. relationshipcfi?comparative esteem to peer communication was nonsignificant (r=-.168, p<.083). However, the posited relation- ship of peer communication to absolute racial esteem was supported by the data (r= .376, p<.002). See Figure 3 for a summary of results. Given the results of the tests of the first two sets of hypotheses, it is clear that absolute racial esteem is signifi- cantly related to these interpersonal communication patterns. The third set of hypotheses, i.e., Hypotheses 3a and 3b, hypothesized that black children who are high viewers of black television programs will have more positive attitudes toward black television characters than black children who are low viewers of such programs. Hypothesis 3; tested the effects of exposure on attitudes toward black adult characters, while hypothesis g9 tested the effects of exposure on attitudes toward black ghild characters. Both hypotheses were supported. The relationship between viewership and attitudes toward black adult characters was most impressive (r=.604, p<.001); the relationship for child characters was also significant (r=.232, p<.03). See Figure 4 for a summary of results. The fourth set of hypotheses, i.e., Hypotheses 4a1, 4a2, 4a3, 4bl, 4b2, 4b3, predicted that exposure to black family television programs is related to more positive esteem for those children who have more positive attitudes toward black television characters, than for those who are less positive. The simple relationship between exposure to these programs 68. FIGURE 2 Results for Hypotheses la, lb, 1c Parental Communication r = .418* Absolute Racial About Black People =::::::::: ‘\»Esteem (Hla) r = .290*-—+Self-esteem (Hlb) r = -.155 Comparative N Esteem (ch) *p<.05 FIGURE 3 Results for Hypotheses 2a, 2b, 2c .376*\\$Absolute Racial Peer Communication ——’_.——_———r About Black People Esteem (H2a) “-~“-‘~r r .llZ-——+Self-esteem (H2b) -.162 Comparative ‘\“Esteem (H2c) *p<.05 69. FIGURE 4 Results for Hypotheses 3a, 3b Attitudes Toward Black Adult /, Television Characters r = .604* Exposure to Black Family Television Shows ‘ H II .232* ‘\\’Attitudes Toward Black Child Television Characters * p<.05 70. and self-esteem was significant (r==.200, p (.05). In the case of absolute racial esteem the simple relationship between program exposure and absolute racial esteem was nonsignificant (r==.184, p‘(.07). The relationship between exposure and comparative racial esteem was also nonsignificant (r= .027, p (.40) . In order to elaborate the relationship between exposure and esteem, the sample was split at the mean to form groups of those children with high positive attitudes toward black characters versus those with attitudes that were not as positive. Table 10 shows the simple correlation coefficients for both child and adult characters for each of the three hypotheses. A transform from 5's to 5'3 (Blalock, 1972) was done to test the significance of the differences between the correlation coefficients. Two of these were significant: the elaboration for adult characters for absolute esteem (z==3.432, p (.002) and for self-esteem for adult characters (z==4.724, p‘(.001). Three of the four others were in the direction predicted. Given these results, Hypotheses 13; and 32; were supported for adult models. The fifth hypotheses set, i.e., Hypotheses 5a and 5b, predicted that communication from parents which emphasizes the positive characteristics of black characters on television will be related to more positive attitudes toward television characters by fourth and fifth grade black children than Table 10 71. Mean Split Elaboration for Hypotheses 4a, 4b, and 4c: Correlation Between Exposure and Esteem for Two Levels of Attitude Toward Television Characters Attitudes Toward Adult Characters Low High Attitudes Toward Child Characters Low High H: 4a1 Absolute r Esteem H: 4a2 Self- r Esteem H: 4a3 r Gamanije Esteem .030 < .206* .027 4 .263* .091 < .149 4b 4b 4b .115 4 .151 .170 > .152 .040 '< .140 *p<.05 72. communication which does not emphasize positive character- istics. Both of these hypotheses were supported. The relationship between the child's attitudes toward black adult television characters and the type of communica- tion their parents engage in regarding those characters was large (r = .732, p<.001) and significant. Similarly, the relationship between the type of communication they reported their parents gave them regarding child characters and the children's actual attitudes toward those characters was also significant (r = .572, p<.001). See Figure 5 for results. The sixth hypothesis set, i.e., Hypotheses 6a and 6b, predicted that communication from peers which emphasizes the positive characteristics of black characters on television will be related to more positive attitudes toward television characters by fourth and fifth grade black children than communication which does not emphasize positive characteristics. These hypotheses were supported. The relationship between the child's attitudes toward black adult television characters and the type of communica- tion their friends engaged in regarding those characters was large (r = .630, p<.001). The relationship between the type of communication their friends gave them regarding child characters and the children's attitudes toward those charac- ters was also large (r = .777, p<.001). See Figure 6. In summary, of the 18 hypotheses, 11 were supported (see summary Table 11). The support of the hypotheses can Results for Parental Communication About Black Television Characters Parental Communication About Black Television Characters *p‘<.05 Results for Peer Communication About Black Television Characters Perr Communication About Black Television Characters *p (.05 FIGURE 5 Hypotheses r = .732* r = .572* FIGURE 6 Hypotheses r = .630 :8, r = .777* 73. 5a, 5b Attitudes About Black Adult Television Characters (HSa) Attitudes About Black Child Television Characters (HSb) 6a, 6b Attitudes About Black Adult Television Characters (H6a) Attitudes About Black Child Television Characters (H6b) 74. Amman uxoc co omscwucoo. m oz 63mm 865“ 9368950 “ be: . N mumfixfidgu Egon u u oz a m maom fine: >9 odflzu zooam oz Ema H393" 9:50ng u 3: Bozo» moguflud «be: oz ammumm Hmaomm 93mg "mg: a .N mHQHXEHEU mm» 83% fimm ~me 2. pg x83 238E has mm» emoumm Howomm musaomom “ ova pnmzou mmusufluua "new xomam on musmomxm em Hmflbmbwn.>afimagu mm» xuqmmcfiflfib.mflufiAHfimubmm 230398 B. “32 9888s 328m .8» x88 amazon $36.32 ” an: 682 x83 8 8:896 9 oz stHEHHQfiHSIQEDNHREUO 3N2 I . smfizuguufixnm oz .HXEmmmzom mam uaRfiwmfimd mow 880mm Hmflomd 3385 "MN: wcoz scum 23.50% mm oz Ewmumm H393 933.6050 “03 . gmazunuufixfim mm» .fimbflmmamw.ndz onfiwmflfiflbm mm» Emoumm 1.423 3382 23: 0:02 can c3u8§50 Hm «museumsm cowumufluo msflcm>umucH Houowomnm mwmwnuomzz manna zumafism muHSmmm mommsuomzm . Ha manna 75. .mo.v m :ma3 dmuuommsm cmumcwmcoo mmz mammnuomzs was I muoz« muwuomug >9 v.36 mannaksvodunv xumm_u;xfiwmnxm mm» xuqmmuzufiwmmfiflzuam mfiQz EpfiwcofithdfifiiHvunez mHHfiaw>anzafim mumfixuofiu>enzaflc xUQRHuQRfiWMfimm mm» xomam usono moguwufim mcoz Eoum :oUmowSEBUU “we: m: WREUmkfivcddhv mumugumsukavoddnv goommuogawmugmmafl mm» x83 uaonm mogufluc 0:02 8mm 838550 5mm mfiQflH~>BHZB©¢ QMflQNHEUKEvudéa. ufifimmuaxfiwmflmfimA mm» xUQmHuQQfiwmmfifizuam msgz =BQM.HJHQBESEiHv"oau mm «muuommsm coaumufiuo mcflcm>uwucH Houowcmum mwmwnuomzm .omscflucoo. Ha magma 76. be expressed with varying degrees of confidence. For instance, although significant, some correlations ranged from .29 to higher than .70 in support of hypotheses. Thus, regression analyses to identify the relative importance of each input and to note possible interaction effects were done in exploratory fashion in the next section. Regression Analyses Several analyses were completed in order to explore the relative and interactive effect of the variables investigated here on esteem. Regressions were done with each of the criterion measures of esteem, with the three determinants of (1) communication about black children from parents, (2) communication about black children from peers, and (3) children's attitudes toward black television characters. Separate regressions were done with attitudes toward adult and child television characters. A second set of analyses looked at the contribution of exposure to black television programs along with the two interpersonal communication determinants (parent and peer) on self-esteem. A third set of analyses regressed (1) exposure to black television programs, (2) parental communication about television characters, and (3) peer communication about television characters, on the black children's attitudes toward black television characters. The summary of these analyses is shown in the tables as indi- cated below. 77. First, note the regressions on the criterion of self- esteem (Tables 12 and 13). For the regression which included attitudes toward child television characters (Table 12), parental communication had the largest influence on self- esteem (this was the only significant beta). However, note (the results when attitudes toward adult television characters is entered (Table 13): Attitudes toward the adult television characters is the only significant contributor to variations in self-esteem. This demonstrates the importance of adult characterizations and this will be discussed in the next chapter in terms of modeling. Totally, the three predictor variables accounted for 16% of the variance in self-esteem when attitudes toward child characters was entered, but 2l% of the variance when attitudes toward adult characters was entered. Second, note the regressions on the criterion of absolute racial esteem (Tables 14 and 15). In neither the regression where attitudes toward black adult or child characters was entered is there a significant predictor. However, it does appear that the beta for parental communication is the largest, and therefore the most ”important predictor. The three predictor variables accounted for over 20% of the variance when attitudes toward child characters was entered and over 19% when attitudes toward adult characters was entered. Almost no variance was accounted for with the comparative racial esteem variable (Tables 16 and 17), thus once again indicating the lack of accountable variance with this dependent 78. Table 12 Regression of Parental Communication about Black Children Peer Communication about Black Children and Attitudes toward Black Child Television Characters on Self-Esteem Ctihafion RnXHLtor Ikufianlnifized Stmxtudimai Bhfltiphe Baal Beha 2 F R Self-Esteem Parental Ccmmmication .181 .422 7.455* .158 Peer Cmmunication -.142 -.331 3.752 Attitudes Toward.Child .022 .284 4.733 TV<1EuacUus *p<.05 79. Table 13 Regression of Parental Communication about Black Adults, Peer Communication about Black Adults, and Attitudes toward Black Adult Television Characters on Self-Esteem Crflxxion Ikedflflxm‘ lkufiandmxfizai sauflhniumd Muhjrde Beta Beta F R2 Self—Esteem Parental Camnmication .102 .238 2.005 .214 Peer Camunication —.123 -.287 3.121 Attitudes Toward Black .324 .428 8.836* .AifltffllChanxfiers *p(.05 80. Table 14 Regression of Parental Communication about Black Children, Peer Communication about Black Children, and Attitudes toward Black Child Television Characters on Absolute Racial Esteem (Ednadon Ekedkxnr mmnankudiaai Shatbnihmd MuUfiphe Bani Ikma F R? Abaflxte Panamal Racial Caummication .172 .309 4.204 .203 menm Paar Ccmnunication . .050 .090 (1.0 Attflxfles Tawud<3uld TV Characters .138 .137 1.156 *p (.05 81. .Table 15 Regression of Parental Communication about Black Adults, Peer Communication about Black Adults, and Attitudes toward Black Adult Television Characters on Absolute Racial Esteem Cnflxnion Rnxfictdr Uhsbmthmdim31 Stmthudimai F Muhjrfle ' Bea: Beta R2 Absolute Parental Camnmication .144 .258 2.308 .197 Ihcufl.flax£m Peer Carmunication .076 .136 41.0 Attitudes toward .110 .112 (1.0 .AhfltntChanKXEIS *p(.05 82. Table 16 Regression of Parental Communication about Black Children, Peer Communication about Black Children, and Attitudes toward Black Child Television Characters on Comparative Racial Esteem CriUnfion Ihadflflxm' Ethuflanfized Stafibnihmfl. Mquphe 3661 Hana F R? Comparative Parental Ccmmnication -.065 -.085 < 1.0 .035 Rmfiallmneen Peer Ccmmnication -.045 -.059 < 1.0 Attitudes toward Child -.120 -.085 (1.0 TV<3Emadans *p<.05 83. Table 17 Regression of Parental Communication about Black Children, Peer Communication about Black Children, and Attitudes toward Black Adult Television Characters on Comparative Racial Esteem CEMfiIflXl PHXHcUm: mmfiznflufifizai Starbmdnmfl. Muhjtde Ikfla lkfiz. F R2 Carparative Parental Ccmmmication -.135 -.174 2.162 .063 lkmdallkfieem Peer Camunication -.129 -.l67 4 1.0 Attitudes toward Adult .316 .231 2.162 TVCRnrmflxms *p<.os 84. measure. However, it does appear that attitudes toward EQElE television characters are again an important predictor and that the beta is positive and larger than for the other predictors. However, trivial variance is accounted for. In the regression analyzing the contribution of exposure to esteem (see Table 18), together with the contribution of the interpersonal communication predictors, parental communication and peer communication, parental communication is the only significant predictor. Neither peer communication nor exposure are significant contributors. Tables 19 and 20 show the results of regressions of the determinants of the black children's attitudes toward both child and adult television characters. From this, one may note that parental communication is the most important predictor of attitudes toward characters, but that adult television characters' attitudes are significantly influenced by mere exposure to black television programs too. Exploratory Analyses Of interest here was whether the relative amount of viewing of black shows compared with white television shows would con- tribute to attitudes toward characters. This appears to be no different than the pure black show influence. The simple corre- lation between the black show index and attitudes toward black child actors was r = .231, whereas the correlation for the relative amount of black versus white shows was slightly higher 85. Table 18 Regression of Parental Communication about Black Children, Peer Communication about Black Children, and Exposure to Black Family Television Programs on Self-Esteem Grimafion RnXHrtor thsflmtbmdimai Stmxkuufized Muhjrde Bea: Ikma F R2 Self-Esteem Parental Canmmication 0.163 0.381 4.103* .101 Peer Communication -0.071 -0.165 (1.0 Exposure 0.019 0.048 (1.0 *p<.05 86. Table 19 Regression of Exposure, Parental Communication about Television Characters, and Peer Communication about Television Characters on Attitudes toward Black Child Television Characters Uthmthmdimai SUmthdimai F Muhjrfle Beta Beta Test R2 Exposure -.021 -.042 < 1.0 .630 Parental Communication .544 .673 43.375* abmn:BLx$:Ch£kiTV Chanxnzms Peer Camunication .149 .205 3.845 ahmnzBLafl:Ch£kiTV Chanxners *p<.05 87. Table 20 Regression of Exposure, Parental Communication about Television Characters, and Peer Communication about Television Characters on Attitudes toward Black Adult Television Characters Umazmdmfiuzed samflbniumd F Mdhfiphe lkma Bea: Tbst R2 Exposure .164 .313 14.409* .710 Parental Camunication . 364 . 407 13. 571* abmnthxkgAmflt 'fllChnmcUus Peer Carmmication .230 .291 6 .275 abanthxmnkhflt Tvcmaraduns *p<.05 7‘, 88. at r = .284. Thus children who are exposed to black shows are influenced by those shows to a similar degree even if they also are high viewers of white shows. Yet the correlation for the relative measure is slightly higher indicating that if the children watch black shows more than white shows, they may have stronger positive attitudes toward black characters. However, the simple correlation between this relative viewing and self-esteem is near zero (r = .064), indicating that if the children watch black shows more, their self-esteem does not correspondingly become more positive. It appears that sheer watching of black shows, independent of white shows has the greatest relationship to esteem (r = .200). Caution should be taken in this interpretation, however, as the relative viewing measure was constructed only for this preliminary exploration and the white family shows were chosen only because they were at the same time-slots. Even the partial correlation controlling for attitudes towards characters between the relative viewing index and self-esteem is larger (.116) than the partial between mere black viewing and self-esteem (partial = .067), indicating that attitudes toward characters were not a significant mediator when relative viewing was tested. 89. CHAPTER IV DISCUSSION This chapter is divided into five discussion sections: (1) the descriptive results and their implications; (2) the test of the hypotheses; (3) theoretical issues; (4) reserva- tions about the research; and, (5) suggestions for future research. Descriptive Results This section will discuss the implications of the descriptive results of this research. It will begin with a discussion of the criterion variables. First, the study found that both absolute racial and self-esteem of the children were at moderately positive levels. This is a confirmation of the research trend which has indicated a shift from early studies where children had fairly low levels of esteem, to the research in the last decade which has demonstrated a shift to higher levels of esteem for black children (Katz, 1976). This higher level of esteem is further indicated by the results of the comparative measure of esteem used in this study. This measure which subtracted absolute black esteem from absolute white esteem found that black children have a slightly more positive esteem for black children than they do for whites. Second, there has to be consideration given to the type of esteem measure utilized in studies. Correlations among the three measures--absolute, self, and comparative--were 90. rather moderate in size, and different between measures. While absolute and self-esteem were moderately correlated and absolute and comparative were correlated (which could be expected since absolute was part of the comparative measure), there was little relation between a child's report of their self-esteem and their level of comparative esteem.. Self-esteem may be quite different from absolute esteem, and may even have different determinants. The implications of this will be discussed in the third section. The study looked at three classes of predictor variables: (1) interpersonal communication from parents and peers about people, (2) exposure to television programs featuring black families, and (3) interpersonal communication from parents and peers about telvision characters. Interpersonal communication about television characters were predictors of the intervening variable, attitudes toward television characters. Each of these predictors will be discussed below. Exgosure. The amount of television exposure to black fanuly shows indicated that black children are watching, cni the average, most of the shows some weeks. When amount <3f viewership of black family shows was subtracted from *viewership of white family shows, to give a relative measure «of black family show viewing, the results indicated that the black children were watching black family shows slightly less often than they were watching the selected white family shows. Two points must be considered about these findings. 91. First, the comparison white family shows were chosen only because they were in the same time slot as the black family shows and may have been very popular. Second, viewership on an absolute level was fairly high for the black family shows. Future research may want to deal with comparative measures on a more systematic basis. Communication about people. Interpersonal communication from both parents and peers about black children was measured in the study. The results indicate that positive information is often given about black children from parents and peers. The average parent and peer apparently communicates positively about black children sometimes, but not necessarily a lot. These results show that communication about racial groups is fairly common for these fourth-fifth grade children; thus, communication may be influencing both racial attitudes and esteem. This will be discussed in the next section dealing with the hypotheses, but the significance of this finding is rather apparent. Communication about television characters. Communication by peers about television characters (as reported by the children in the sample) was on the average very positive. However, communication from peers about television adult actors was not on the average very positive. Thus it appears that the valence of interpersonal communication about child and adult actors (models) is significantly different. 92. What could account for this finding? The first attribu- tion could be that television programs display child behaviors that are acceptable to children but display adult behaviors which are less acceptable. Unfortunately, the past demographic and behavioral content analyses of black characters has not identified portrayal differences between adult and child characters. It is possible that the child characters engage in different type behaviors than adult characters. A more likely attribution might pertain to what a child viewer deems as acceptable. It could be that similar behaviors from an adult or child, say for instance humor, are more acceptable to a child when it comes from a child actor but is not acceptable from an adult actor who is expected to be more serious. Or, it could be that children are less severe critics of a child actor's behavior than an adult's. Whatever the reason, the difference has important implications for modeling hypotheses and social learning theory to be discussed in the following two sections. Similar response levels were noted regarding parent communication about child vs. adult actors. Parents communicate rather positively about television characters, but do communicate positively more often about child actors than adult. Perhaps this is further indication dun the perceived appropriateness of character behavior is a plausible explanation. However, it must be remembered that the measures 93. consisted of asking the black children themselves what the content of communication from parents and peers was. What the child reports about their communication may not correspond to the actual communication. The child may expect his/her parents to be less positive than his/her friends and report that expectation. Even in spite of this, what the child expects or "thinks" he/she is told may be as important or more important than what the parent intends or actually says. ' The study also investigated a set of variables that were expected to intervene between exposure and esteem. This set of variables deals with the children's attitude toward‘ the television characters. Overall, the children sampled had a fairly high evaluation of the black television characters on black family programs. Thus an important factor for modeling influences has been found--children do see the black television characters in a rather positive way. This finding has important ramifications for television influence and will be discussed at length in the theory section. Hypotheses Out of the 18 hypotheses tested in this research, 11 ‘were supported in whole or in part. Judging from the results of these tests, it appears that communication is an important factor in socialization and in its relation to black children's esteem. The type of esteem and the source of communication are important qualifiers to the above statement, however. 94. The hypotheses predicting a relationship between peer and parental communication and absolute racial esteem were supported. Support was provided for the hypotheses relating parental communication to self-esteem but was not provided. for the prediction that self-esteem would be related to peer communication. Apparently, absolute racial esteem may be more susceptible to communication. This will be discussed in the theory section. Neither the hypothesis relating peer communication nor parental communication to comparative racial esteem were supported. This may be a result of many factors, among which are the nature of the measurement and/or that the children view black and white poeple in similar and in rather positive ways. The results regarding television communication indicated support for the hypotheses that related exposure to self- esteem. This extends the past work that demonstrates that black children identify with black characters (Dates, 1979) especially when they are high viewers of black shows (Greenberg, 1972). Interpersonal communication from parents and peers appears to heighten positive attitudes toward televion characters. The children in this study reported communication valence from parents and peers which were highly related to . the children's own attitudes toward the television characters. Thus, communicationfran,parents and peers may impact esteem 95. directly and may also serve the secondary function of providing evaluational information about televised models. High viewership of black family television shows is also highly related to attitudes toward the characters on the shows, both adult and child characters. In other words, if a child is a high viewer of black family shows s/he is more likely to report that the characters have more positive characteristics. Regression analyses indicated that when interpersonal communication with parents and peers and attitudes toward television characters are entered as predictors of esteem, the type of esteem and whether attitudes toward adult or child television characters was entered was important. Communication from parents appeared to be the most important predictor of esteem as it was significantly related for self-esteem when children's attitudes toward child television characters was entered into the equation and it approached significance for both conditions of absolute racial esteem when attitudes toward child and adult actors were entered. Communication from parents is related to esteem. In addition, attitudes toward television characters was a significantly larger predictor for self-esteem when the characters were adult. This again points out the importance of adult modeling on esteem. Attitudes toward television did not appear as significant predictors for either absolute or comparative esteem, however. It should be noted that 96. almost no variance in comparative esteem was accounted for by the three predictors, but that the conjoint influence of the three predictors on self-esteem and absolute esteem accounted for almost 20% of the variance in those esteem measures. Looking at these regressions in their entirety, one could conclude several things. First, comparative esteem is not influenced greatly by the three predictors. This confirms the simple correlation analysis that found no significant predictors of comparative racial esteem. Apparently there is little variation in comparative esteem. This may be the result of many factors, but it is obvious that black children do not view black children and white children that much different to begin with (as the descriptive statistics have shown). This lack of difference in attitudes toward black and white children is partly a result of the fact that the black children viewed both groups rather positively. Second, parental communication does seem important to these children. The theoretical importance of this finding will be discussed in the next section. Third, the finding that attitudes toward adult television characters is an important predictor of self—esteem has theoretical implications from a social learning theory stand- point to be discussed in the next section. A separate set of regressions analyzed the relative and 97. joint effects of television exposure to black family programs and communication from parents and from peers about television characters on children's attitudes toward both adult and child television characters. This analysis indicated that a large portion of the variance in attitudes toward the characters may be accounted for by interpersonal communication and to an extent by exposure to the programs. Once again, the influence of parental communication was greater than ' the influence from peer. Mere exposure to black family programs did have a relationship to the children's attitudes toward black adult television characters. These findings once again point out the importance of modeling for explaining the influence of television on racial esteem. The next section which looks at the theoretical implications of this research will explore this further. Theoretical Issues This section will discuss the results as they are related to the theoretical rationale. It will contain three major sections: (1) learning explanations, and (2) the role of communication in self-concept formation, and (3) esteem. Learning_explanations. In the rationale section, three types of learning models (McLeod & Chaffee, 1972) were presented: reinforcement, modeling and interactional (or combinatorial learning). It was argued that these three distinctions are useful for explanations of communication influences on socialization processes. Interpersonal 98. communication was classified as interactional. That is, children learn about themselves and others through direct reinforcement of their actions agg_by observation or modeling of their significant others (in the case of the present study, parents and peers). On the other hand, television learning is primarily explained with modeling conceptions. Direct reinforcement does come from peer and parental communi- cation about television, but much of television learning appears to be a result of modeling where television characters act as role models (Bandura, 1977). These learning distinctions may help explain the apparent scattered findings of this research. In line with past research that exposure alone is not highly related to racial attitudes (e.g., Atkin, et al., 1979), exposure to black family programs was not expected to influence esteem except when children had positive attitudes toward black television characters (models). Although past social learning research has identified the importance of model identification in television effects (see for example, Maccoby 8 Wilson, 1957; Tannenbaum & Gaer, 1965; Rosenkrans, 1967; Reeves & Greenberg, 1977), the present research findings indicate that attitudes toward the models may also be an important mediator of television learning. Although it is assumed that a model a child identifies with will also be favorably evaluated, it may be that even though a child has positive attitudes 99. toward a model, they may not wish to be like that model or identify with him/her. At least for adult models, television exposure was related to absolute and self-esteem. This finding extends social learning and modeling theory explanations for the development of self-esteem through television exposure, by accounting for merely attitudes toward the models. However, the modeling explanation is not necessarily straightforward. It should be noted that while attitudes toward adult characters were an important mediator of the exposure effect, attitudes toward child characters were not. Adult modeling was more crucial. The reasons for this may be varied. First, children at this pre-adolescent age may still rely mostly on adults for information about their environment, as the develOpmental research indicates (for example, Saltzstein, 1976). Even adult television models may be more potent sources of influence. Another reason for this difference may be that the adult television models actually may have a more favorable image. Given the portrayals on television, the child actor cannot be viewed in terms of successfully dealing with his/her environment except in terms of how they interact with family members or friends. The adult models, on the other hand, are shown as rather successful on other criteria, such as occupationally. This explanation is rather speculative and certainly would be appropriate for further study. Third, children may be 100. more critical viewers of the appropriateness of child television behaviors than they are of adult behaviors. It is important to note here that these learning explanations may be influenced by developmental factors as suggested in Chapter 1. Age of the child may influence what source of communication may have the greatest impact on esteem and may even influence which television models have the greatest modeling potential. For instance, given the past developmental research which has shown that children at the onset of adolescence rely more on peers than parents for information (Saltzstein, 1976), the same may hold true for television models. It could be that for children a bit older than the ones currently sampled, child or teen actors may be potent models. It should also be noted that the current research did not support the Katz (1976) observation that racial attitudes tend to crystallize at the fourth and fifth grade. The current sample of children was in this age group and yet the results indicated that their attitudes and esteem are in fact related to communication behaviors. Thus, racial attitudes may still be impacted by communication. However, such results are to be taken with some hesitation, since the current research did not assess esteem changes directly, especially from a developmental or maturational perspective. That particular area of research will be appropriate for another research effort. 101. Communication from peers was also related to esteem, < but only absolute and not self-esteem. Perhaps, information and thus attitudes toward one's racial group is easier to impact than attitudes toward oneself. Additionally, the research results indicated that communication from parents and peers was highly related to children's attitudes toward the television characters. This is another example of direct reinforcement. In the case of television characters, the target of attitude is clearly delineated; parents and peers could provide specific forms of reinforcement to the child about any particular character through communication. Apparently parents and peers did communicate specifically about certain characters. This communication or the percept that it occurred does appear to be a very important factor when considering the influence of television on children's learning of esteem, since the research did show that attitudes toward the characters is related to esteem. The role of communication. The present data underline the Liebert and Poulos (1976) recommendation that social- ization research be approached from a multi-influence perspective. Unlike most of the past research, where one determinant was studied, the present research accounted for the influence of several factors. Each of the factors that were hypthesized to influence self-concept were communication variables or variables directly influenced by communication 102. as in the case of attitudes toward television characters. The research demonstrated that racial socialization is indeed a result of many factors in additive and nonadditive ways. Interpersonal communication along with attitudes toward black television characters are related to self—concept. In nonadditive fashion, attitudes toward characters mediate the effect of exposure on self-concept, and attitudes toward characters are highly related to interpersonal communication about the characters. Apparently the influence of television exposure is inextricably tied to interpersonal communication. The effects of television and interpersonal communication 'on self-concept are not, however, simple. Different com- ponents of self-concept seem to be related to some communica- tion but not others. The component distinction appears to have provided some interesting and perhaps fruitful results. The current research found that communication was always related to absolute esteem, or the esteem a child holds for his/her racial group. Communication was not always related to self-esteem. As discussed in Chapter 1, communica- tion may be directed in three ways: (1) it may be directed to how the child functions in a rather direct way and thus influence the child's self-esteem, (2) it may be directed at the child in rather indirect ways by referencing his/her cultural or racial group and thus influence the child's absolute esteem, or (3) it may make comparisons between his/her racial group and other groups, and thus influence 103. comparative esteem. In the present research, communication was more highly related to indirect influences on the child's absolute esteem. Perhaps communication may be easier to accept about others than about oneself directly, even if the others are like oneself in terms of race. Or, perhaps one's attitude toward onself is more deeply entrenched than one's attitude toward others, and thus communication has greater impact on those attitudes with less mass. Traditionally, researchers have lumped together the components of self-concept. The current findings indicate that this may distort the picture, and that reconceptualizing self-concept may be quite heuristic. Esteem. As discussed in Chapter 1, past research into self-concept has taken two research approaches. The self- esteem approach, for example Rosenberg's (1965) merely asks children to indicate their attitudes toward themselves. No comparison is made with their attitudes toward themselves and their attitudes toward others. The second approach epitomized by Clark and Clark (1947) with their "doll choices," asks the child to make preference choices assumed to be on racial grounds. Here, the child is asked to make comparisons between stimuli indicative of racial groups and the individual's choice is assumed to indicate their self-esteem. The current research conceptualized esteem as containing racial components and self components. It was argued that 104. a child's self-concept can be considered to contain two parts, his/her conception of his/her own role group compared with other role groups (racial esteem), and his/her self- concept which is assumed to also be a result of unique reinforcements.. To the extent that a child's racial concept (absolute racial esteem) is related to his/her self-concept (self-esteem) we know that the child's self-esteem is not independent of his/her racial esteem. In addition, the Current research findings suggested that unless the child's attitudes toward his/her racial group were compared with other racial groups, there would be no indication of how the child views him/herself comparatively. The results of the current research demonstrated the fruitfulness of this self-concept breakdown. Self-esteem and absolute racial esteem were partially related, indicating that self-conceptions of these black children were only partly a result of their attitudes toward their race. This is an important finding. The results indicated that communication, while an important determinant of self-esteem when it comes from parents, appears to be a more relevant predictor of absolute racial esteem, or attitudes toward the black child's racial group. With these findings in mind, the value of the esteem conceptualization used in this research should guide esteem research to greater clarification of the role of racial attitudes in esteem formation. 105. Reservations This section will present some of the reservations about the current research effort. It will begin with a discussion of the research design and then explore measurement problems and conclude with sample selection and generalizability. Causation. This field survey study attempted to deter- mine the relationship of communication to black children's esteem. Children were asked to report their levels of esteem and the content of communication they received from others. Since the study was neither an experiment nor even logitudinal in design, no time-order statements can be made. Thus, causation is an issue here. Recent evidence by several authors (see for example, Atkin, Greenberg, Korzenny & McDermott, 1979) have questioned the traditional view that mere exposure to television leads to certain attitudinal or behavior indi- cators, without regard to selective exposure or attitudinal predispositions. From the current research, we do not know if selectivity is operating. It may be that those children with higher esteem do view more black television programs. Similarly, those children who are higher in esteem may have more positive attitudes toward the television characters they do see. And, those children with more positive attitudes toward television characters may expose themselves to those programs featuring those hightly evaluated programs. Specific television diet controls may be necessary to test for these 106. exposure effects, as specific interpersonal control may be necessary in an experimental context to test for direction of effect. Measurement. Some of the problems mentioned above are further confounded by measurement problems. In the study, the children were asked to report not only their own level of esteem (which is the standard way to measure esteem) but also they were asked to assess the amount and content of communication they received from others (parents and peers). This may not in itself be a problem, for we may be interested in not what was actually said by others, but how the child interpreted what others said. There may be, however, two problems with the method employed in this study: (1) social acceptability of answers, and (2) self-image. The first problem, social acceptability, refers to a tendency of people to give answers that appear to be socially acceptable but are not necessarily their "true" attitudes or feelings (Robinson & Shaver, 1971). This may be especially important in sensitive or socially significant areas. Al- though every attempt was made to minimize this potential problem through insistence by the administrators that only the researchers at the university would see the results, and by having both black and white administrators, it may have been a factor. Self-image may also have been a factor. Often in a .107. person's desire to be comfortable in the world, they have a tendency to see others as they see themselves or as they'd like others to see them. This obviously presents a problem when the communication of others is measured. For instance,‘ children who hold negative attitudes toward black television characters may be likely to report that their friends also don't like the characters. Thus, they might be likely to report that the communication they receive from their friends is negative, when in reality it might not be. There is no direct evidence that this may have happened, but further research may wish to validate the child's report by asking others. Sample and generalizability. These data were collected at only one point in time, and they were not cross-sectional, as they did not measure a large age range of children. Developmental conclusions may not therefore be drawn from this research. An additional concern must be aired with regard to the selection of the sample. Although the children did come from an integrated school, the school was selected after a series of schools rejected the solicitation of subjects from researcher. These denials ranged from "these children have been over-studied" to the most common: ". . . the sensitive nature of the questions.“ Thus, the school officials who did permit access to the students had different attitudes toward this questionnaire, and perhaps even racial 108. matters. Although the children in the schools may not have been aware of these perceptions by their officials (e.g., teachers, principals), it may have been a factor. No objec- tions came from the students during administration of the questionnaires, but some teachers reacted negatively to the questionnaire after the administration. Perhaps, the children were influenced as well. Certainly, the results of this study should be replicated to insure generalizability. Suggestions for Future Research It is apparent from the findings of the present research that much awaits and deserves to be done. In addition to problems of measurement and sample generalizability, there are many questions that remain to be answered in determining the causes and correlates of black children's esteem. First, longitudinal studies of the communication process should be a priority. Especially important would be research that follows children through the years when race attitudes are supposed to be crystallizing (Katz, 1976), and which this current research found to appear to be influenced. Addi- tionally, during these years, children are switching from reliance on parents to peers for information about social issues. Longitudinal research should probably look at children from second through at least the seventh grade to assess these changes. Second, a trip to the lab to carry out controlled 109. experiments may shed light on the causative directional links and on relative communication influences. For instance, several input conditions could be manipulated in the lab. In one condition an adult confederate could communicate information about a racial role group to children, or even comment on television. In another condition confederate children could do the same. As a variation of each of these, the valence and amount of communication could be also manipulated. To study television exposure effects, a condition might include either a past program depicting a certain role related image or videotape presentation of role related actions could be produced. Of course, "one-time" exposures would probably have limited effects compared with multiple exposures, but relative effects might still be assessed experimentally. Even program diets might get at such effects. Third, future studies may wish to include some of the well known variables which appaer to mediate television exposure effects. Especially salient would be variables of selective exposure, uses and gratifications and moti- vational factors., Once again, over-time studies would be important for assessing selective exposure, especially since attitudes toward characters may in fact determine exposure to black-family programs. 110. Fourth, distinctions between behavioral portrayals might help specify modeling effects more finely. Both portrayal differences between black and white characters and child versus adult actors would aid in determining model influences more specifically. As discussed earlier in this section, it may be that adult actors are rewarded more often than child actors, thus the adult television model would be a more potent modeling source for children. Two strategies could be used in this research: (1) content analysis of character behavior to determine what the actual portrayal differences are, and (2) children's perception of what the portrayal differences are. The second strategy would require some exploratory work requiring that the children be asked what the characters do to determine salient behavioral categories. In addition, it seems important to develop refined. indices of relative viewing of black family shows versus white family shows. The current research selected the compared shows in a rather unsystematic way. Future research might wish to choose only those shows that are shown at the same hour or those shows in which the characteristics of the families in the black shows are similar to the characteristics in the white family shows. For instance, shows might be chosen that are matched in terms of family size, sex of the head of the family, or are similar in socio-economic levels. 111. Finally, future research may wish to deal with issues of generalizability. Replication of the current research, especially given the selectivity of the sample, would provide more confidence. Additionally, social learning and socialization processes research should be cognizant of the social demands for providing answers that may or may not reflect the behavioral dispositions of the people studied or questioned, especially in sensitive or socially significant areas such as racial esteem. Researchers would move toward developing less obtrusive measures. For example, one could design a story with race "hidden" in the message. Half of the children would be presented with a story containing reference to the character as black, while the other children would receive a story with a white character. Children would then be asked to respond to attitude questions about the character (say, John) and not by race differences, explicitly. Any differences found in the attitudes would then be attributed to race. All in all, given the relationship of communication with esteem as demonstrated in this present field survey, any move back to the lab is grounded in the generalizability of the research findings reported here. 112. FOOTNOTES According to the 1975 Santa Clara County special census, this neighborhood with a population of 55,203 out I of a San Jose total of 551,224, has an annual median household income of $9,977. 1 1 3 . REFERENCES Amir, Y. The role of intergroup contact in change of prejudice and ethnic relations. In P. Katz (ed.), Towards the elimination of racism. New York: Pergamon Press, Inc., 1976, 245-308. Armstrong, C. P., & Gregor, A. J. Integrated schools and Negro character development. Psychiatry, 1964, 21, 69-72. Asher, S. R., & Allen, V. L. Racial preference and social comparison processes. Journal of Social Issues, 1969, 25, 157-165. Ashmore, R. D., & DelBoca, F. Psychological approaches to understanding intergroup conflicts. In P. Katz (ed.), Towards the elimination of racism. New York: Pergamon Press, Inc., 1976. Atkin, C. K., & Gantz, W. The role of television news in the political socialization of children. Paper pre- sented at the International Communication Association, Chicago, 1975. Atkin, C. K., & Greenberg, B. S. Parental mediation of children's social behavior learning from television. CASTLE report #4 to U.S. Office of Child Development, Michigan State University, Department of Communication, 1977. Atkin, C., Greenberg, B., Korzenny, F., & McDermott, S. Selective exposure to televised violence. Journal of Broadcasting, 1979, 23, 5-13. Atkin, C. K., Greenberg, B., & McDermott, S. Race and social role learning from television. In H. S. Dordick (ed.), Proceedings of the Sixth Annual Telecommunications Policy Research Conference. Lexington, Mass.: Lexington—Books, 1979. Bandura, A. Social learning theory. Morriston, New Jersey: General Learning Press, 1971. Bandura, A. Social learning theory. Englewood Cliffs, New Jersey: Prentice-Hall, 1977. Bandura, A., & Walters, R. H. Social learning and personality development. New York: Holt, Rinehart and Winston, 1963. Barcus, F. E. Parental learning on children's television viewing. Television:guarterly, 1969, 8, 63-73. 114. Blalock, H. M. Social statistics. New York: rMcGraw-Hill Book Company, 1972. '« Bogatz, G. A., a Ball, 8. J. The second year of Sesame Street: A continuing evaluation. Princeton: Education Testing Service, 1971. Brim, O. G., & Wheeler, S. Socialization after childhood: Two essays. New York: Wiley, 1966. Brigham, J., & Wéissbach, T. Racial attitudes in America: Analysis and findings of social psychology. New York: Harper & Row, 1972. Butler, R. 0. Black children's racial preference: A selected review of the literature. Journal of Afro- American Issues, 1976, 4, 2, 168-171. Caplan, N. The new ghetto man: A review of recent empirical studies. Journal of Social Issues, 26, Winter, 1970. Clark, K. B., & Clark, M. P. Racial identification and preference in Negro children. In T. M. Newcomb & E. L. Hartley (eds.), Readings in social psychology. New York: Holt, Rinehart and Winston, 1947, 169-178. Coleman, J. S., Campbell, E. 0., Hobson, C. J., McFarland, J., Mood, A. M., Weinfeld, F. D., & York, R. L. Equality of opportunity. Washington, D.C.: United States Government Printing Office, 1966. Collins, A. W. Learning of media content: A developmental study. Child Development, 1970, 41, 1133-1142. Coopersmith, S. The antecedents ofiself-esteem. San Francisco: W. H. Freeman, 1967. . Dates, J. L. The relationship of demographic variables and racial attitudes to adolescent perceptions of black television characters. Unpublished doctoral dissertation, The University of Maryland, May 1979. Dimas, C. The effect of motion pictures portraying black models in the self-concept of black elementary school children. Doctoral Dissertation, Syracuse University (94), in Comstock (ed.), Television and human_§ehavior: the key studies. Santa Monica, CA: Rand, 1975. Duncan, 0. D. Inheritance of poverty or inheritance of race? In D. P. Moynihan (ed.), Understandigg poverty: Perspectives from the Social Sciences. New York: Basic Books, 1968, 85-110. 115. Fox, D. J., & Jordon, U. B. Racial preference and indentification of black, American, Chinese, and white children. Genetic Psychology Monographs, 1973, 88, 229-286. Goodman, M. E. Race awareness in young children. Cambridge, Mass: Addison-Wesley, 1952. Greenberg, B. 3. Children's reaction to TV blacks. Journalism Quarterly, 49, 1972. Greenberg, B. S. Gratification of television viewing and their correlates for British children. In J. Blumler & E. Katz (eds.), Annual reviews of Communication Research, 1974. Greenberg, B. S., & Dervin, B. Use of the mass media by the urban poor. New York: Praeger, 1970. Greenberg, E. S. Black children, self-esteem and the liberation movement. Politics and Society, 1972, 2, 293-307. Greenwald, H. J., & Oppenheim, D. 8. Reported magnitude of self-misidentification among Negro children--Artifact? Journal of Personality and Social Psychology, 1968, 8, 49-52. - Hale, G. A., Miller, L. K., & Stevenson, H. W. Incidental learning of film content: A developmental study. Child DevelOpment, 1968, 32, 69-77. Hawkins, R. P. Learning of peripheral content in films: A developmental study. Child Development, 1973, 44, 214-217. Hicks, D. J. Effects of co-observer's sanctions and adult presence on imitative aggression. Child Development, 1968, 38, 303-309. Hinton, J. L., Seggar, J. P., Northcott, H. C., & Fontes, B. F. Tokenism and improving imagery of blacks in TV drama and comedy: 1973. Journal of Broadcasting, 1969, 13, 331-344. Hraba, J., & Grant, G. Black is beautiful: A reexamination of racial preference and identification. Journal of Personality and Social Psychology, 1968, 8, 49-52. Hulbary, W. E. Race, deprivation, and adolescent self- images. Social Science Quarterly, 1975, 56, 105-114. 116. Katz, P. Towards the Elimination of Racism. New York: Pergamon Press, Inc., 1976. Katz, P. A., & Zalk, S. R. Doll preferences: An index of racial attitudes. Journal of Educational Psychology, 1974, 55, 663-668. Kircher, M., & Furby, C. Racial preferences in young children. Child Developmen§,l97l, 35, 2076-2078. Kohlberg, L. Stage and sequence: The cognitive developmental approach to socialization. In D. Goslen (ed.), 252 Handbook of Socialization Theory and Research, Chicago: Rand-McNally, 1969. ‘ Leifer, A. D., Collins, W. A., Gross, B. M., Taylor, P. H., Andrews, L., & Blackmer, E. R. Developmental aspects of variables relevant to observational learning. Child Development, 1971, 55, 1509-1516. Liebert, R. M., & Poulos, R. W. Television as a moral teacher. In T. Likona (ed.), Moral development and behavior. New York: Holt, Rinehart and Winston, 1976. Loviband, S. H. The effect of media stressing crime and violence upon children's attitudes. Social Problems, 1967' 125—, 91-1000 Lyle, J., & Hoffman, H. Children's use of television and other media. In Rubinstein, Comstock & Murray (eds.), Television and social behavior, Vol. IV, U.S. Department of Health, Education and Welfare, 1972. Maccoby, E. E. Effects of the mass media. In M. L. Hoffman, & L. W. Hoffman (eds.), Review of child develOpment research. New York: Russel Sage Foundation, 1964. Maccoby, E. E., & Wilson, W. C. Identification and observational learning from films. Journal of Abnormal Social Psychology, 1957, 55, 67-87. ‘ McArthur, L. 2., & Eisen, S. V. Achievements of Male and Female storybook characters as determinants of achievement behavior in boys and girls. Journal of Personality and Social Psychology, 1976, 55, 467-473. McLeod, J., Atkin, C., & Chaffee, S. Adolescents, parents, and television use: Adolescent self report measures from Maryland and Wisconsin sample. In Rubinstein, Comstock & Murray (eds.), Telvision and social behavior, Vol. III, U.S. Department of Health, Education and Welfare, 1972. 117. McLeod, J., & O'Keefe, J. R., Jr. The socialization per- spective and communication behavior. In G. Kline & P. Tichenor, Current perspectives in mass communication research. Beverly Hills: Sage Publications, 1972. Miller, G. R., & Steinberg, M. Betweenypeople: A new analysis of interpersonal communication. Chicago: Science Research Associates, 1975. Morland, J. K. Racial acceptance and preference of nursery school children in a southern city. Merril-Palmer Quarterly, 1962, 5, 271-280. Niven, H. Who in the family selects the TV program? Journalism Quarterly, 1960, 51(1), 110-111. Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. The measurement of meaning. Urbana, Illinois: The University of Illinois Press, 1957. Porter, J. D. R. Black child, white child. Cambridge: Harvard University Press, 1971. Proshansky, H., & Newton, P. The nature and meaning of Negro self-identity. In M. Deutsch, I. Katz, 5 A. R. Jensen (eds.), Social classy race and psychological development. New York: Holt, Rinehart and Winston, 1968. Reeves, B., & Greenberg, B. S. Children's perceptions of television characters. Human Communication Research, 1977, 5, 113-127. Radke, M. J., Sutherland, J., & Rosenberg, P. Racial attitudes in children. Sociometry, 13, 154-171. Radke, M. J., Trager, H., & Davis, H. Children's perceptions of the roles of Negroes and Whites. Journal of Psychology, 1950, 55, 3-33. Rainwater, L. Crucible of identity: The negro lower class family. In R. Endo & W. Strawbridge (eds.), Perspectives on Black America. Englewood Cliffs, New Jersey: Prentice-Hall, 1970. Roberts, D. F. Communication and children: A developmental approach. In I. de Sola Pool & W. Schramm (eds.), Handbook of gommunication. Chicago: Rand-McNally, 1973, 174-215. Rosenberg, M. Society and the adolescent self-image. Princeton: Princeton University Press, 1965. 118. Robinson, J. P., & Shaver, P. R. Measures of social psychological attitudes. Ann Arbor, Michigan: Survey Research Center, 1971. Rosenkrans, M. A. Imitation in children as a function of perceived similarities to a social model of vicarious reinforcement. 593rnal_of Personality and Social Psychology, 1967, 1, 307-315. Saltzstein, H. D. Social influence and moral development. In T. Lickona (ed.), Moral development and behavior: Theory, research and social issues. New York: Holt, Rinehart and Winston, 1976, pp. 253-265. Sarbin, T. R., & Allen, V. L. Role theory. In Lindzey & Aronson (eds.), Handbook of Social Psychology, Vol. 1. Reading, Mass.: Addison-Wesley, 1969. Schramm, W., Lyle, J., & Parker, E. B. Television in the lives of our children. Stanford: Stanford University Press, 1961. Schwartz, M., & Tangri, S. S. A note on self-concept as an insulator against delinquency. American Sociological Review, 1965, 55, 922-926. Selman, R. L. Social-cognitive understanding: A guide to educational and clinical practice. In T. Lickona (ed.), Moral Development and Begavior. New York: Holt, Rinehart and Winston, 1976. Siegel, A. E. The influence of violence in the mass media upon children's role expectation. Child Develgpment, 1958, 55, 35-56. Simmons, K., Greenberg, B. S., Atkin, C. K., & Heeter, C. J. The demography of fictional television characters in 1975-76. CASTLE report number 2, Department of Communication, Michigan State University, 1977. Stevenson, H. W., & Stewart, E. C. A developmental study of race awareness in young children. Child Development, 1958, 55, 399-410. Tannenbaum, P. H., & Gaer, E. P. Mood changes as a function of stress of protagonist and degree of identification in a film-viewing situation. Journal of Personality and Social Psychology, 1965, 5, 612-616. Trubowitz, J. Changing the racial attitudes of children. New York: Praeger, 1969. 119. Wand, B. Television viewing and family choice differences. Public Opinion Quarterly, 1968, 55, 84-94. Weiss, R. F. Pursuasion and acquisition of attitudes: Models from conditioning and selective learning. Psychological Reports, 1962, 55, 709-732. Zimmerman, B. J., & Rosenthal, T. L. Observational learning of rule governed behavior by children. Psychological Bulletin, 1974, 55, 29-42. 120. STUDY QUESTIONNAIRE Appendices A. Cover Sheet B. Self Esteem Measure C. Attitudes Toward Television Characters D. Attitudes Toward White Adult E. Parental Communication About Television Characters F. Parental Communication About White Adults and Children G. Absolute Racial Esteem (Adults)/Exposure to Television Programs H. Parental Communication About Black Adults and Children I. Attitudes Toward White Children/Exposure to Television Programs J. Peer Communication About Television Characters K. Peer Communication About White Adults and Children L. Absolute Racial Esteem (Children)/Exposure to Television Programs M. Peer Communication About Black Adults and Children N. Attitudes Toward Black Adult Television Characters Attitudes Toward Black Child Television Characters APPENDIX A TELEVISION‘SURVEY TODAY WE WOULD LIKE TO ASK YOU SOME QUESTIONS ABOUT WHAT YOU WATCH ON TELEVISION. WE ALSO WANT TO FIND OUT ABOUT THINGS YOU DO WITH PARENTS AND YOUR FRIENDS. THIS IS EQT A TEST, SO THERE ARE NO RIGHT 0R WRONG ANSWERS. YOUR ANSWERS WIII.NOT BE SHOWN TO YOUR TEACHER 0R YOUR PARENTS---ONIY THE RESEARCHERS AT THE UNIVERSITY WILL SEE THE SURVEYS. PLEASE BE HONEST WHEN YOU ANSWER THE QUESTIONS, SINCE WE WANT TO KNOW WHAT YOU REALLY THINK. THANK YOU FOR HELPING US. WE APPRECIATE IT VERY MUCH. APPENDIX B PLEASE TELL US WHAT YOU ARE LIKE: I think I an... VERY GOOD VERY IMPORTANT VERY UNSELFISH NOT FRIENDLY NOT KIND NOT SUPERIOR ”ON AN AVERAGE SCHOOL DAY, about how many HOURS PRETTY GOOD_ PRETTY IMPORTANT PRETTY UNSELFISH NOT VERY FRIENDLY NOT'VERY KIND NOT VERY SUPERIOR NOT VERY GOOD NOT VERY IMPORTANT NOT VERY. UNSELFISH PRETTY FRIENDLY PRETTY KIND PRETTY SUPERIOR television after your evening supper? Number of Hours: 0 1% 2 3 3% 4 NOT GOOD NOT IMPORTANT NOT UNSELFISH VERY FRIENDLY VERY KIND VERY . SUPERIOR do you spend watching “i 5 or more APPENDIX C Here is a list of people on TV, please tell us which ones YOU LIKE or DON'T LIKE: LAVERNE...on Laverne & Shirley SHIRLEY...on Laverne .1. Shirley RE-RUN...on.What's Happening ROGAR...on What's Happening DEE...on.What's Happening MRS. WALKER...on.What's Happening THE FONZ...on Happy Days RICHIE...on Happy Days BILL COSBY...on Fat Albert J.J....on.Good Times Thelma...on Good Times Florida Evans...on Good Times Mork. . .on Mark 8: Mindy Mindy. . .on Mark at. Mindy ARNOLD...on Diff'rent Strokes GEORGE JEFFERSON...0n The Jeffersons LOUISE JEFFERSON...on The Jeffersons FREDDY WASHINGTON . . .on Kotter LIKE A LOT LIKE ALOT' LIKE A LOT LIKE A LOT. LIKE A LOT LIKE A LOT DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE AILOT A LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE' LITTLE ' LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE. LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE IITTLE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE LIKE A LOT LIKE A LOT LIKE A LOT .LIKE A LOT LIKE A LOT LIKE _A IOT DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE APPENDIX D I think WHITE adults in real life are: VERY SUPERIOR VERY KIND VERY FRIENDLY NOT UNSELFISH NOT IMPORTANT NOT GOOD PRETTY SUPERIOR PRETTY KIND PRETTY FRIENDLY NOT VERY UNSELFISH NOT VERY .IMPORTANT. NOT VERY GOOD NOT VERY SUPERIOR NOT VERY KIND NOT VERY FRIENDLY PRETTY ‘ UNSELFISH PRETTY IMPORTANT PRETTY GOOD NOT SUPERIOR NOT KIND NOT . FRIENDLY VERY UNSELFISH VERY IMPORTANT VERY GOOD How often do you talk with your PARENTS about BLACerople on TV? A.LOT SOMETIMES NOT MUCH NEVER Do your PARENTS usually say good things or bad things about the BLACK people on TV? SAY GOOD THINGS SAY BAD THINGS APPENDIX E Here is a list of people on TV, please tell us which ones your PARENTS like or dislike: THE FONZ. ..on Happy Days RICHIE. . .on Happy Days BILL COSBY. . .on Fat Albert J .J. . . .on Good Times THEIMA. . .on Good Times FLORIDA EVANS...on Good Times MORK. . .on Mark 8: Mindy MINDY. . .on Mark a. Mindy GEORGE JEFFERSON...on The Jeffersons LOUISE JEFFERSON . . .on The J effersons ARNOLD. . .on Diff'rent Strokes FREDDY WASHINGTON . . .on KO tter LAVERNE...on Laverne 8. Shirley SHIRLEY . . .on Laverne 3. Shirley RE-RUN . . .on What' 8 Happening ROGAR . . .on What ' s Happening DEE. . .on What's Happening MRS‘. WALKER. . .on What's Happening LIKE A LOT LIKE A LOT LIKE A LOT LIKE A IOT LIKE A LOT LIKE A LOT DON'T LIKE DON'T LIKE DON 'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE ' LITTLE LIKE LITTLE LIKE LITTLE LIKE IITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE LIKE. A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE I A LOT DON'T LIE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE APPENDIX F How often do your PARENTS tell you that WHITE adults are... A LOT .....GOOD? A IOT SOMETIMES NOT OFTEN NEVER _ .....IMPORTANT? A IOT SOMETIMES NOT OFTEN NEVER .....UNSELFISH? A IOT SOMETIMES NOT OFTEN NEVER .. ...FRIENDLY? A IOT SOMETIMES NOT OFTEN NEVER .. ...KINO? A LOT SOMETIMES NOT OFTEN NEVER . .SUPERIOR? A IOT SOMETIMES NOT OFTEN NEVER How often do your PARENTS tell you that WHITE mg; are. .. ' .....SUPERIOR? A IOT SOMETIMES NOT OFTEN NEVER .....KIND? A IOT , SOMETIMES NOT OFTEN NEVER . . . ..FRIENOIY? A LOT SOMETIMES NOT OFTEN NEVER . . . . .UNSELFISH? A LOT SOMETIMES NOT OFTEN NEVER . . . . .IMPORTANT? A IOT SOMETIMES NOT OFTEN NEVER . . . . .GOOD? SOMETIMES NOT OFTEN NEVER APPENDIX G I think BLACK gdglts in real life'are: VERY GOOD VERY IMPORTANT VERY UNSELFISH VERY FRIENDLY NOT KIND NOT SUPERIOR PRETTY GOOD PRETTY IMPORTANT PRETTY UNSELFISH PRETTY FRIENDLY NOT VERY KIN D NOT VERY SUPERIOR How often do you watch? LAVERNE & SHIRLEY? MORK & MINDY? WHAT'S HAPPENING? GOOD TIMES?’ ONE DAY AT A TIME? EVERY WEEK E NOT VERY ' NOT GOOD GOOD NOT VERY NOT IMPORTANT IMPORTANT NOT VERY NOT UNSELFISH‘ UNSELFISH NOT VERY NOT FRIENDLY FRIENDLY PRETTY VERY KIND KIND PRETTY 'VERY SUPERIOR SUPERIOR MOST SOME WEEKS WEEKS‘ L., r‘1 r-1 "‘T ..J} __ A... -w ONCE IN AWHILE OR NEVER APPENDIX H How often do your PARENTS tell you that BLACK adulps are” . .....GOOD? A LOT .....IMPORTANT? A LOT .....UNSELFISH? A LOT .....FRIENDLY? A LOT .....KIND? A LOT .....SUPERIOR? A LOT How often do your PARENTS BLACK kigg are... .....SUPERIOR? A .....KIND? A LOT .....FRIENDLY? A LOT .....UNSELFISH? A LOT .....IMPORTANT? A LOT .....GOOD? A LOT LOT SOMETIMES SOMETIMES _ SOMETIMES SOMETIMES SOMETIMES SOMETIMES tell you that SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT OFTEN OFTEN OFTEN OFTEN “OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN NEVER NEVER NEVER NEVER NEVER NEVER 'NEVER NEVER NEVER NEVER NEVER NEVER APPENDIX I I think WHITE kids in real life are... VERY PRETTY GOOD GOOD VERY PRETTY IMPORTANT IMPORTANT VERY PRETTY UNSELFISH UNSELFISH NOT NOT VERY FRIENDLY FRIENDLY NOT NOT VERY KIND KIND NOT ‘ NOT VERY SUPERIOR SUPERIOR How often do you watch EVERY wEEK HAPPY DAYS? '—1 FL DIFF'RENT STROKES? ,_1 EIGHT IS ENOUGH? __1 WELCOME BACK KOTTER? __1 WHITE SHADOW? NOT VERY GOOD NOT VERY IMPORTANT NOT VERY UNSELFISH PRETTY FRIENDLY PRETTY KIND PRETTY SUPERIOR MOST SOME WEEKS WEEKS -T -1 ..1. .—1 ‘F‘T .__ -w ..1 F-w -1' NOT GOOD - NOT IMPORTANT NOT UNSELFISH VERY FRIENDLY VERY KIND VERY SUPERIOR ONCE IN AWHILE OR NEVER L_J P—1 E] APPENDIX J Here is a list of people on TV, please tell us which ones your FRIENDS like or dislike: LAVERNE. . .on Laverne & Shirley SHIRIEY...on Laverne 5. Shirley RE-RUN . . .on What' 8 Happening ROGAR . . .on What‘ 8 Happening DEE. . .on What "8 Happening MRS . WALKER. . .on What' 8 Happening THE FONZ. . .on Happy Days RICHIE. . .on Happy Days BILL COSBY. . .on Fat Albert J.J. . . .on Good Times -THEIMA. . .on Good Times FLORIDA EVANS. . .on Good Times MORK. . .on Mork a: Mindy MINDY . . .on Mork &: Mindy ARNOLD. . .on Diff'rent Strokes George JEFFERSON. . .on The J effersons LOUISE JEFFERSON. . .on The Jeffersons FREDDY WASHINGTON . . .on Kotter LIKE A LOT LIE A LOT LIE A LOT LIKE A LOT LIKE A LOT LIKE A LOT DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE DON'T LIKE LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE A LOT LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE LIKE LITTLE DON'T LIE DON'T LIKE DON'T LIE DON‘T LIKE DON‘T LIKE DON'T LIE LIKE A LOT LIE A LOT LIE A IOT LIKE A LOT LIKE A IOT LIKE A LOT DON'T LIKE DON'T UKE DON'T LIKE DON'T LIE DON'T LIKE DON'T LIKE APPENDIX K How often do your FRIENDS tell you that WHITE adults are... .....GOOD? A LOT .....IMPORTANT? A LOT ...UNSELFISH? A LOT .....FRIENDLY? A IOT .....KIND? A LOT .....SUPERIOR? A IOT How often do your FRIENDS WHITE kids are... .....GOOD? A LOT .....IMPORTANT? A IOT ...UNSELFISH? A LOT .....SUPERIOR? A IOT ...KIND? A IOT .....FRIENDLY? A IOT SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES tell you that SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN NEVER NEVER NEVER NEVER NEVER' NEVER NEVER NEVER NEVER NEVER NEVER NEVER APPENDIX L I think BLACK m in real life arei VERY. GOOD“. VERY' IMPORTANT VERY UNSELFISH NOT FRIENDLY NOT KIND NOT SUPERIOR PRETTY GOOD PRETTY IMPORTANT PRETTY UNSELFISH NOT VERY FRIENDLY' NOT VERY KIND NOT VERY SUPERIOR How often do you watch THREE'S COMPANY? THE JEFFERSONS? THE wALTONS? GOOD TIMES? FAT ALBERT? EVERY WEEK RE 1%! L :1 NOT VERY GOOD NOT VERY IMPORTANT NOT VERY UNSELFISH PRETTY FRIENDLY PRETTY KIND PRETTY SUPERIOR. MOST WEEKS _ SOME WEEKS \Hl NOT GOOD NOT IMPORTANT NOT UNSELFISH VERY FRIENDLY VERY KIND ‘ VERY _ SUPERIOR _ ONCE IN AWHILE 0R NEVER “‘T ‘ APPENDIX M How often do your FRIENDS tell you that BLACK adults are . . . .....GOOD? ....IMPORTANT? .....UNSELFISH? .....FRIENDLY? .....KIND? .....SUPERIOR? A LOT A LOT A LOT A LOT A LOT A,IOT How often do your FRIENDS BLACK m are... .....SUPERIOR? .....KIND? .....FRIENDLY? .....UNSELFISH? .....IMPORTANT? _ .....GOOD? LOT LOT IOT IOT IOT >:>:>:>>:> LOT SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES SOMETIMES tell you that SOMETIMES SOMETIMES SOMETIMES SOMETDMES SOMETIMES SOMETIMES NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN OFTEN NEVER NEVER NEVER NEVER NEVER NEVER NEVER NEVER NEVER NEVER NEVER NEVER APPENDIX N Do you think BLACK adults on TV VERY GOOD VERY IMPORTANT VERY UNSELFISH NOT FRIENDLY NOT KIND NOT SUPERIOR PRETTY GOOD PRETTY IMPORTANT PRETTY UNSELFISH NOT VERY FRIENDLY NOT VERY KIND NOT VERY SUPERIOR are NOT VERY GOOD NOT VERY IMPORTANT NOT VERY UNSELFISH PRETTY FRIENDLY PRETTY KIND PRETTY SUPERIOR NOT GOOD NOT IMPORTANT NOT UNSELFI SH VERY FRIENDLY VERY KIND VERY SUPERIOR How often do you talk with your Friends about BLACK people on TV? A LOT SOMETIMES NOT MUCH ' NEVER Do your FRIENDS usually say good things or bad things about the BLACK people on TV? SAY GOOD THINGS - SAY BAD THINGS APPENDIX 0 VERY GOOD VERY IMPORTANT VERY UNSELFISH NOT FRIENDLY NOT KIND NOT SUPERIOR How old are you? Are you a Do you think BLACK kids on TV What GRADE are you in? are PRETTY NOT VERY GOOD GOOD PRETTY NOT VERY IMPORTANT IMPORTANT PRETTY NOT VERY UNSELFISH UNSELFISH NOT VERY PRETTY FRIENDLY FRIENDLY NOT VERY PRETTY KIND KIND NOT VERY PRETTY SUPERIOR SUPERIOR 9 10 11 12 BOY? or GIRL? 3rd? uth? 5th? THANK YOU VERY MUCH 1 NOT GOOD NOT IMPORTANT NOT UNSELFI SH VERY FRIENDLY VERY KIND VERY SUPERIOR HICH «MIME