THE RACE AND SEX FACTORS IN THE FRIENDSHIP PREFERENCES 0F WHITE PRESCHOOL AGED CHILDREN Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY ALAN I. SUGAWARA 1967 .r,‘ t ‘ I1 11 LIBRA 2:; Y _ Michigan State . _ University INLET: F streams ay 1“ NOAH & SUNS’ MEWEIIJEE- ABSTRACT THE RACE AND SEX FACTORS IN THE FRIENDSHIP PREFERENCES OF WHITE PRESCHOOL AGED CHILDREN by Alan I. Sugawara Studies in the area of race awareness has pointed out that an awareness of race and sex differences does exist at the preschool age level. However, with respect to the correlated variables of age/race awareness and age/sex awareness, conflicting trends have been reported. Some re- searchers conclude that children of preschool ages place a greater emphasis upon sex rather than race in their choice of playmates; others conclude that children of preschool ages place a greater emphasis upon race rather than sex in their choice of playmates; while still others conclude no consistant patterns to be present in the choice of play- mates among preschool aged children. The primary purpose of this thesis, therefore, fo- cuses upon seeking further information regarding the question, "Which factor (race or sex) is more influential in the friend- ship preferences of white preschool aged children?" A total sample of sixty white preschool aged children, thirty boys and thirty girls, ranging in ages from 3.5 to Alan I. Sugawara 5.5 years, were randomly selected from six all-white nursery school classes held at the Michigan State University. All subjects came from homes in which either one or both of their parents were enrolled as students in the university. A Picture-Preference Test designed for this study was used as the research instrument. The test included twelve 5" by 7" black and white facial photographs of chil- dren between the ages of 3.5 to 5.5 years. Of the twelve photographs, three were of dark-skinned Negro boys, three were of dark-skinned Negro girls, three were of white boys, and three were of white girls. These photographs were chosen by a board of five adult judges from a larger sample of twenty-five photographs according to criteria described in this thesis. Statistical measures were used to analyze the fre- quency preferences obtained on the Picture-Preference Test, while a descriptive technique was used to analyze the de- scriptive data (responses subjects made to questions asked about their preferences). Findings resulting from these analyses indicated that: 1) white children of preschool ages preferred friends of their own sex when the variable of race was held con- stant; 2) white children of preschool ages preferred friends of their own race when the variable of sex was held constant; 3) white boys of preschool ages preferred friends of their own sex when the variables of race and sex were varied; and Alan I. Sugawara 4) white girls of preschool ages preferred friends of their own race when the variables of race and sex were varied. These findings add further to the controversy re- garding the race and sex factors in the friendship prefer- ences of white preschool aged children. The conflicting trends occurring in previous research mentioned earlier were not acceptable. For white boys of this sample, sex rather than race occurred as more influential in their friendship preferences, while race rather than sex was more influential for girls. THE RACE AND SEX FACTORS IN THE FRIENDSHIP PREFERENCES OF WHITE PRESCHOOL AGED CHILDREN By Alan I. Sugawara A THESIS Submitted to the Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Home Management and Child Development 1967 ACKNOWLEDGMENTS The writer of this thesis would like to extend his gratefulness to Dr. Martha Dale, Miss Phyllis Lueck and Dr. Charles Mange, members of his guidance committee, for their interest, guidance and support throughout the plan- ning and conducting of this study. Also great appreciation is eXpressed to Dr. Sarah Hervey for her help in statisti- cal advice. Sincere thanks are due to Miss Sue McWilliams, Mrs. Carol Holt, Mrs. Barbara Isenberg, and Miss Mariella Aikman, teachers of young children, for their sensitive in- sights into understanding the preschool aged child. An expression of appreciation is also extended to Miss Jane BOpp, Miss Phyllis Jones, and Mr. Fred Hahn for their helpfulness and encouragement. Finally, appreciation is expressed to the Department of Home Management and Child Development in the College of Home Economics at the Michigan State University, for their permission to undertake this study at their nursery schools. ii TABLE OF CONTENTS Page ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . ii LIST OF TABLES . . . . . . . . . . . . . . . . . . . vi CHAPTER I INTRODUCTION . . . . . . . . . . . . . . . 1 Purpose of the Study 2 Definition of Terms 2 Assumptions 3 Hypotheses 3 II A REVIEW OF LITERATURE 6 Race Awareness and Incipient Attitudes: The Preschool Child . 6 Race Awareness (Phase I) 8 Age Differences 8 Race Differences . . . . . 9 Ingroup Differences: The Negro Child . . . . . . . . . . . . . . 12 Skin Color Differences . . . . . 12 Sex Differences . . . . . . . . . 14 Interracial Contacts . . . . . . 15 North-South Differences . . . . . 16 Socioeconomic Background Differences . . . . . . . . . 17 Ingroup Differences: The White Child . . . . . . . . . . . . . . 17 Sex Differences . . . . . . . . . 17 Socioeconomic Background Differences . . . . . . . . . l8 Incipient Attitudes (Phase II) . . . . 19 Race-Related Words and Value Judgments . . . . . . . . . . . . 19 Racial Preference . . . . . . . . . 20 iii CHAPTER Page The Negro Child . . . . . . . . 20 The White Child . . . . . . . . 23 Awareness of Sex Differences: The Preschool Child . . . . . . . . . . . 24 Physical Attributes . . . . . . 25 Toy Preference and Sex Cleavage . . . 27 Overt Behavior . . . . . . . . . . . 29 Race and Sex Factors in the Friendship Preferences of White Preschool Aged Children . . . . . . . . . . . . . . 32 The Picture- Preference Test Technique . . 33 The Sociometric Test and. Interview Technique . . . . . . . 35 The Observational Technique . . . . . 37 The Matching Technique . . . . . . . 39 III METHOD . . . . . . . . . . . . . . . . . . 41 The Picture-Preference Test . . . . . . 41 The Reliability of the Picture- Preference Test . . . . 43 The Validity of the Picture- Preference Test . . . . . . . . . . . . . . 45 The Pilot Study . . . . . . . . . . . . 46 The Final Study . . . . . . . . . . . . 48 Description of Subjects . . . . . . . 48 The Establishment of Rapport . . . . 50 The Test Room . . . . . . . . 51 Presentation of the Picture- Preference Test . . . . . . . . . 51 Collection of Data . . . . . . . . . 52 IV ANALYSIS OF DATA . . . . . . . . . . . . . 53 Frequency Preferences . . . . . . . 53 Set I (Tasks One and Two) . . . . . . 54 Set II (Tasks Three and Four) . . . . 55 Set III (Tasks Five and Six) . . . . 57 Descriptive Data . . . . . . . . . . . . 6O Criteria For Evaluating the Data. . . 60 Set I (Tasks One and Two) . . . . . . 62 Set II (Tasks Three and Four) . . . . 63 Set III (Tasks Five and Six) . . . . 64 iv CHAPTER Other Findings Sex Awareness and Value Judgments Sex- Related Words Value Judgments . Types of Value Judgments Words of Value Judgments Race Awareness and Value Judgments Race- Related Words . . Kinds of Race- Related Words Value Judgments . Types of Value Judgments Words of Value Judgments . Definite Statements of Dislike Additional Use of the Picture-Preference Test V SUMMARY, IMPLICATIONS, LIMITATIONS AND RECOMMENDATIONS . . . . . . Summary Implications Other Findings Sex Awareness and Value Judgments Race Awareness and Value Judgments Limitations Encountered In The Study Limitations of the Sample . Limitations of the Picture- Preference Test Recommendations For Further Study BIBLIOGRAPHY APPENDIX Page 66 66 67 68 69 71 71 72 73 74 75 76 78 81 81 86 86 88 89 89 90 91 94 103 TABLE II III IV VI VII VIII IX LIST OF TABLES A Reliability Coefficient for Task One and Percentages for Tasks Two, Three, Four, Five and Six Obtained From Two Administra- tions of Each Task in the Picture-Preference Test for Thirty Subjects . . . Age and Sex of Subjects Employed in this StUdy O O O O O O O O O O O O O O O O 0 Frequency Preferences of Sixty Subjects On Set I (Tasks One and Two) of the Picture- Preference Test . . . . . . . . . . Frequency Preferences of Sixty Subjects On Set II (Tasks Three and Four) of the Pic- ture—Preference Test . . . . . . . Frequency Preferences of Boys and Girls On Set III (Tasks Five and Six) of the Picture- Preference Test . . . . . . . . . . . . . . Descriptive Data of Subjects On Set I of the Picture—Preference Test . . . . . . . . . . Descriptive Data of Subjects On Set II of the Picture-Preference Test Descriptive Data of Subjects On Set III of the Picture-Preference Test . . . Subjects Making and Not Making Value Judg- ments of Their Own Sex and Opposite Sex Photographs Used in the Picture-Preference Test . Types of Value Judgments Used by Subjects of Their Own Sex and Opposite Sex Photo- graphs Used in the Picture-Preference Test vi Page 45 49 SS 57 59 63 64 65 68 69 TABLE XI XII XIII XIV XV XVI XVII XVIII XIX Page Words of Value Judgments Used by Subjects in Describing Their Own Sex and Opposite Sex Photographs Used in the Picture- Preference Test . . . . . . . .-. . . . . . . 7O Race-Related Words Used (Correctly) and Not Used by Subjects in Describing Photographs of Their Opposite Race Used in the Picture- Preference Test . . . . . . . . . . . . . . . 72 Kinds of Race-Related Words and Physical Attributes Used by Subjects in Describing Their Own Race and Opposite Race Photo- graphs Used in the Picture-Preference Test . . . . . . . . . . . . . . . . . . . . 73 Subjects Making and Not Making Value Judg- ments of Their Own Race and Opposite Race Photographs Used in the Picture-Preference Test . . . . . . . . . . . . . . . . . . . . 74 Types of Value Judgments Made by Subjects of Their Own Race and Opposite Race Used in the Picture-Preference Test . . . . . . . 75 Words of Value Judgments Used by Subjects in Describing Their Own Race and Opposite Race Photographs Used in the Picture- Preference Test . . . . . . . . . . . . . . . 77 Subjects Making and Not Making Definite Statements of Dislike Concerning Photographs of Their Opposite Race Used in the Picture- Preference Test . . . . . . . . . . . . . . . 78 Choice of Photograph Subjects Made to the Question: "Which Little Girl Do You Think Is My Little Girl?" . . . . . . . . . . . . . 80 Reasons Subjects Gave to the Question: "Why Do You Think That Is My Little Girl?" . 80 vii CHAPTER I INTRODUCTION In the course of the last decade the United States has experienced two historic milestones in the area of race relations. These two milestones include: 1) the Supreme Court's decision of May, 1954, which outlawed segregation in the public schools; and 2) the 1957 enactment by Congress of the first federal Civil Rights Bill since Reconstruction.(36) Following these milestones researchers in the field of early childhood education have recently focused attention upon questions concerned with the development of race awareness, attitudes and prejudices in young children. Of the studies done in the area of race awareness, research points out that an awareness of race differences in young children does exist at the preschool age level. This fact has been well established and the relationship be- (3,13,36,82) How_ tween age and awareness is fairly clear. ever, with respect to the correlated variables of age/ awareness and sex, conflicting evidences have been seen. Three conflicting trends have emerged in research which in- dicates: 1) that children of preschool ages place a greater emphasis upon sex rather than race in their choice of l playmates;(1’26’46’68) 2) that children of preschool ages place a greater emphasis upon race rather than sex in their choice of playmates;(56’65’81) and 3) that no consistant patterns are present in the choice of playmates among pre- school children.(57’83) Purpose of the Study The controversy centering on research mentioned above has continued throughout the first half of the twen- tieth century. No definite solution has yet been found. This thesis, therefore, will focus upon exploring a similar problem. The primary question to be investigated is: "Which factor (race or sex) is more influential in the friendship preferences of white preschool aged children?” Definition of Terms The definition of terms as operationalized in this thesis are: 1) Children of preschool ages--defined as those children between the age range of 3.5 to 5.5 years. 2) Influence--as measured by the results of the Picture-Preference Test designed for this study. It represents the total score derived by summing the choices (preferences) of all subjects on specific tasks in the Picture-Preference Test. 3) Friendship preference-~designated as the choices the subjects make on the Picture-Preference Test designed for this study. 4) Descriptive Data--designated as those responses subjects make to questions asked about their preferences in specific tasks of the Picture- Preference Test. Assumptions 1) Most children of preschool ages are aware of race differences between themselves and other children.(3’13’36’86) 2) Most children of preschool ages are aware of sex differences between themselves and other children.(13’24’50’51) 3) The use of the Picture-Preference Test designed for this study is an effective means of obtain- ing research data from children of preschool ages.(l,46,81) Hypotheses The hypotheses in this thesis to be tested include: 1) White children of preschool ages will prefer friends of their own sex when the variable of race is held constant; that is: 2) a) b) d) white girls will choose white girls when a choice between a white boy and a white girl is given to them. white girls will choose Negro girls when a choice between a Negro boy and a Negro girl is given to them. white boys will choose white boys when a choice between a white boy and a white girl is given to them. white boys will choose Negro boys when a choice between a Negro boy and a Negro girl is given to them. White children of preschool ages will prefer friends of their own race when the variable of sex is held constant; that is: a) white girls will choose white boys when a choice between a white boy and a Negro boy is given to them. b) white girls will choose white girls when a choice between a white girl and a Negro girl is given to them. c) white boys will choose white boys when a choice between a white boy and a Negro boy is given to them. d) white boys will choose white girls when a choice between a white girl and a Negro girl is given to them. 3) White children of preschool ages will prefer friends of their own sex when the variables of race and sex are varied; that is: a) white boys will choose white boys when a choice between a white boy and a Negro girl is given to them. b) white girls will choose white girls when a choice between a white girl and a Negro girl is given to them. c) white boys will choose Negro boys when a choice between a Negro boy and a white girl is given to them. d) white girls will choose Negro girls when a choice between a Negro girl and a white boy is given to them. An analysis of both the frequency preferences and descriptive data (responses of subjects to questions asked about their preferences) will be undertaken to test these hypotheses. Also additional content analysis of the de- scriptive data will be done to determine other findings on sex awareness and value judgments, and race awareness and value judgments. CHAPTER II A REVIEW OF LITERATURE A review of literature related to the problem of this thesis falls into three categories: Section I focuses upon the development of race awareness and attitudes in the preschool aged child; section II deals with the development of sex awareness in the preschool aged child; section III deals with the race and sex factors in the friendship pre- ferences of white preschool aged children. Race Awareness and Incipient Attitudes: The Preséhool Aged Child \ The first six years of life are important for the development of all social attitudes, though it is a mistake to regard early childhood as alone responsible for them. A bigoted personality may be well under way by the age six, but by no means fully fashioned. (3, p. 279) These generalizations were made by Gordon Allport and are supported by many relevant studies concerned with the de- velopment of race awareness in young children. Research findings indicate that the development of race awareness and attitudes is a continuous process through- out childhood.(4’17'22’35’36’38’48’57’58’69’70’78’81'84’87’89) Three essential and overlapping phases are distinguishable. 6 These include: Phase I: Awareness--the consciousness of self and others in terms of race identity; Phase II: Inci- pient Attitude (orientation)--the learning and synthesizing of race-related words, value judgments and preferences; Phase III: True Attitude--the establishment of full-fledged attitudes involving enduring presuppositions about other people, object situations and ideas concerned with race.(35’ 36) Phase I and Phase II of the developmental process are present during the preschool ages.(3’4’17’19'22’34’36’ 38,48,56‘58a65:69:70178'81'84'87’88) True Attitudes (Phase III) among preschool children are rarely found. Only upward of five years, are indications of true attitudes present. (76,77) Radke, Trager and Davis in their studies of 242 Negro and white children from both middle and lower socioeconomic backgrounds found that among some five year olds, Concepts and feelings about race frequently included adult distinctions of status, ability, character, oc- cupations and economic circumstances. Among older children expressions of hostility and aggression were frequent and attitudes more crystalized. (76, p. 476) Kutner's studyc54) SUpported the generalization made (76,77) by Radke, Trager and Davis. Employing seven year old children from the Boston area in his research, he found 12% (sixty children) "clearly prejudiced." Many other studies concerned with the development of race attitudes among chil- dren five years or older are present.(2’3’8’10’28’32’33’41’ 54,79,88) Race Awareness (Phase I) Age Differences Awareness (Phase I) of race differences among pre- school children was first cited in a classical study written .(58) by Lasker. From scattered incidents of anecdotal rec— ords written on children, he infers the possibility of race awareness in a few children one year of age. At the age of two or two-and-a-half, however, more children are likely to begin to give evidence of conscious- ness of their own and others racial characteristics.(4’36’ 48’83) The Horowitz Picture Test,(48) Ammon's interview with children,(4) Goodman's interview with mothers(36) and (83) of both Stevenson and Stevenson's observational records Negro and white children, provide the basis upon which this inference is made. At three years of age the majority of children are (17,19-22,35,38,58,8l-84) Among aware of race differences. three year old Negro children from both Northern and South- ern communities, 75% showed that they were conscious of "(17,19’22) These differences between ”white" and "colored. findings clearly supported the conclusion that race aware- ness is present in Negro children at the age of three. (35,33,57'81'84) show the same to be Other investigators true of white children as well. Dolls, pictures, photographs, interviews, observations, puzzles and clay, were used to collect data from which this conclusion was made. By four years of age all children studied showed (l7,l9—22,36,38, (36) significant awareness of race differences. 46’48’69’70’82’87’89) The extensive study by Goodman of 103 Negro and white children from a Northern city pro- vided substantial evidence for this generalization. Horo- k(17’19'22) concluded from their witzc48) and Clark and Clar studies thatan:four years of age, children ceased to identify themselves with pictures of animals or clowns, but identified themselves with pictures of Negro and white children. Mor- land(69’70) also indicated that the racial recognition ability of children progressed most rapidly between the third and fourth year of a child's life. By five years, race awareness in young children be- comes significantly more definite.(4’17’19'22’35'38’48’58'69’ 70,76,77,8l-84,87-89) (76.77). Radke,Trager and Davis 1n their studies suggests that at this age, beginnings of true atti- tudes are not unlikely. Thus, the assumption that most children of preschool ages are aware of race differences is well documented and established. Race Differences Despite the fact that the relationship between age and race awareness is fairly well established, conflicting evidences are seen with respect to the correlated variables (36,38,48,81,84,87-89) of race and awareness. It is not 10 surely established whether race awareness is likely to dawn earlier in Negro children or in white children. Horowitz(48) in her study of both Negro and white boys, two to five years of age, inferred that white boys seemed to have more difficulty in identifying themselves, then Negro boys on a Picture-Identification Test. On the whole, her white subjects presented a more confused picture than did her Negro subjects. Sixty-six percent of her Negro subjects identified themselves correctly by race, while only 40% of her white subjects made correct identifications. A more recent study directed by Goodman(36)supported (48) Although she found 85% the inference made by Horowitz. of both her Negro and white subjects to be of medium and high awareness, only 24% of her white subjects were of high awareness. Forty percent of her Negro subjects were re- garded as ”highly aware.” Such high awareness among Negro subjects indicated that 1) they had a clearer perception of color and other racial attributes, 2) used race terms more numerously and accurately, and 3) had a strong sense of racial differentiation and separation. However, other studies have found the reverse to be (84) in their study of 225 chil- true. Stevenson and Stewart dren, three to six years of age, from both Negro and white groups, found white children to make more correct identifi- cations on a Picture-Discrimination Test, a D011 Selection Test, and a D011 Assembly Test. Although the difference 11 between the scores of both races were not significant at the .05 level of confidence, a tendency was noted. Also, Morland(69) in his study of 454 Negro and white children found similar results. White children made more correct identifications than Negro children. These conflicting evidences, however, may all be (48) (36) plausible. Both Horowitz's and Goodman's studies involved subjects from Northern communities, while Steven- (84) (69) son and Stewart's and Morland's studies involved subjects from Southern communities. (36) Goodman emphasized the point that minority group children were somewhat more susceptible to the development (74) in reviewing studies done of race awareness. Proshansky on race awareness, supported Goodman's findings. Stevenson and Stewart(84) and Morland,(69) however, indicate that for Southern children, white subjects were more aware of race differences than Negro subjects. The normative character of Negro-white relationships found in the south made these (69,84) More children more sensitive to racial differences. studies are necessary to delineate a clearer understanding of this difference in race awareness. Three studies are present regarding race awareness among preschool children from different cultures. Vaughnc87’88) in his studies of 180 Maori (Negroid) and 180 white New Zea- land children, four to twelve years of age, found his white subjects making more correct race identifications than Maori subjects. This tendency was seen at every age level. 12 Studying 287 preschool children in Hawaii, three to six years of age, Springer(80) also discovered her white subjects to make more correct race identifications than Oriental and part-Oriental subjects. A tendency was seen at all age levels, although all racial groups were aware of race differences at a significant level of confidence (p = .05). Using the Clark Doll Test, Greagor and McPhersonCs8) in their study of thirty-five, three to seven year old Bantu (Negroid) and white children from South Africa, found that subjects from both groups showed marked proficiency in iden- tifying appropriate racial groups and themselves. No sig- nificant difference was discovered between the Bantu and white groups. Both showed almost equal correct identifications. Ingroup Differences: The Negro Child The extensive studies of Clark and Clark(17’19—22) provide the most significant findings for this section. Most of the generalizations made herein are based on their findings. Skin Color Differences: Studying 150 Negro children, three to seven years of age, varying in degrees of skin color (light, medium, dark) Clark and Clark(17’19'22) reported that dark Negro children were consistantly more accurate in their identifications of appropriate racial groups than medium and light Negro children. From such findings the generalization that dark Negro children had a more definite knowledge of 13 race differences than medium or light children was made. Goodman'sCSé) findings in research with similar children supported this generalization. Her study also indicated that Negro children with dark skin colors represented vis- ually a dramatic contrast to persons of white skin color, and thus became aware of race differences at an earlier age. (88) of light, medium and dark Maori children Vaughn's study reported similar findings. An analysis of age differences and race awareness among light, medium and dark Negro children also provides interesting insights. In Clark and Clark's studysz) dark Negro children, three years of age, consistantly identified themselves with pictures and dolls of Negro children, while light Negro children identified themselves with pictures and dolls of white children. The medium Negro children, like the light Negro children identified themselves with pictures and dolls of white children. The generalization that dark Negro children were earlier in their development of race awareness, was thus made. At the age of four, however, an interesting reverse is indicated. While light Negro children as well as dark Negro children remain consistant in their identifications, medium Negro children identified themselves with pictures and dolls of Negro children. Clark and Clarkczz) see this reverse as signifying a deve10pmental stage in race aware- ness among medium Negro children. Negro children of medium 14 color follow dark Negro children in their awareness of race differences. Vaughn'sC89) study of Maori children also re- ported similar findings. At five years of age, light Negro children still re- main consistant in their identification with pictures and dolls of white children. However, an analysis of age trends indicate a slight decrease in identification with pictures and dolls of white children and a slight increase in identi- fication with pictures and dolls of Negro children. Clark and Clarkc17) suggested this to signify develOpmental aspects of race awareness among light Negro children. They see this as a trend toward identification with a socially defined ”race" meaning, based on physical characteristics other than skin color. For five year old dark Negro children, however, a reverse in identification is seen. Dark Negro children now identify themselves with pictures and dolls of white children, indicating another developmental stage in race awareness; Phase II of the developmental process. This will be discussed later under the section "Incipient Attitudes." Negro children of medium color still remain consistant in their identifications with pictures and dolls of Negro chil- dren at this age. Sex Differences: Four studies concerned with sex differences in race awareness among Negro children are avail- (21,36,38,69) able. Conflicting evidences are encountered in these research findings. 15 Clark and Clark,(21) Morland(69) and Greagor and (38) in their studies of Negro children three to McPherson seven years of age, report no significant difference in race awareness between sex groups. Also Clark and Clark's(21) study showed that no significant differences exist between sex groups of Negro children with varying skin colors. Goodman,(36) however, in her recent study encountered re- verse findings. She reported that Negro girls were more aware of race differences than Negro boys. Data collected from administering a Puzzle Test, a D011 Test, a Picture Test, a test with clay of different colors and also record- ing spontaneous reactions of children during these test situations, provided evidence for her generalization. With such evidence she concludes that such may well be plausible, as among Americans, personal appearance is more stressed for girls than for boys. Interracial Contacts: The studies of Clark and Clark(21) and Goodman(36) provide information for generali- zations in this section. However, there is insufficient evidence for delineation of any reliable conclusions. In studying Negro children from racially integrated, semi-segregated and segregated groups, Clark and ClarkCZI) report that at three and four years of age, Negro children from both segregated and semi-segregated groups showed an increase in apprOpriate race identifications. With Negro children in mixed groups, however, confusion was found. 16 Also, whereas Negro children from segregated and semi- segregated groups ceased to identify themselves with pic- tures of animals and clowns at four years of age, some Negro children in mixed groups continued to do so. This suggested that the awareness of race differences in Negro children from mixed groups occurred later than Negro chil- dren from segregated and semi-segregated groups. Clark and Clark(21) also suggested that such confusion and retardation among Negro children from mixed groups may indicate that other factors are present in segregated and semi-segregated groups that lead them to make more clear cut racial identifications. (36) after studying four and five years olds Goodman, from mixed groups and segregated Negro groups, reports simi— lar findings. She indicates high awareness "a little more frequent" among children in her segregated all Negro groups than among children in mixed groups. She continues to state, however, that the frequency of interracial contact is prob- ably less significant than the type of contact with different racial groups a child experiences in his life. Contacts incidental to a generally pleasant situa- tion, like nursery school, are certainly conducive to friendly acceptance across color lines. But this favor- able association can easily be cancelled out by street play conflict, parental cautioning against Negro and a large number of other factors. (36 p. 205-206) North/South Differences: Only one relevant study concerned with the relationship between geographical location 17 and race awareness among Negro children was found. Again, a definite need for more research in this area is indicated. Employing Negro children, three to seven years of age in their study, Clark and Clark(18) discovered no significant difference in race awareness between Northern Negro children and Southern Negro children. However, they indicate in their findings the tendency of Northern Negro children to make more identifications with white dolls then Southern Negro children. Socioeconomic Background Differences: Morland(69) in studying Negro children three to six years of age, from upper and lower socioeconomic backgrounds report no signif- icant differences in race awareness between such groups. Ingroup Differences: The White Child Studies of ingroup difference among white preschool children are sparse, but two divisions regarding these dif- ferences can be made. These divisions include studies dealing with the variables of 1) race awareness and sex (36’57’69’81) and 2) race awareness and socio- (59,69) differences, economic background differences. Sex Differences: Limited research findings indicate the existence of conflicting evidence concerning sex dif- ferences in race awareness among white preschool children. (69) of Southern white preschool children re- (81) Morland's study port no sex difference in race awareness. Springer in 18 her study of white children from Hawaii, also found no sex difference in race awareness. Goodman,(36) however, in comparing sex differences of her white subjects found more girls rated "highly aware" than boys. The generalization that among American children, personal appearance is stressed more for girls than for boys, was made to support this finding. Landreth and Johnson(57) in their study reported still other findings. Studying white children from both Upper and lower socioeconomic backgrounds they found that among upper socioeconomic background children, girls were more aware of race differences than boys, while among lower socioeconomic background children, boys were more aware of race differences than girls. Further study is necessary, if a clearer understanding of this problem is to be acquired. Socioeconomic Background Differences: Two studies present findings that concern this section. Landreth and (57) Johnson in their study of children from upper and lower socioeconomic backgrounds report that children from upper socioeconomic backgrounds were more aware of race differ- ences than children from lower socioeconomic backgrounds. Such findings were obtained by administering a Picture and Inset Test Design which required a matching procedure. Morland,(69) however, in studying similar white chil- dren, reported findings which revealed no significant dif- ference in race awareness among children from both groups. Additional research in this area must still be done. 19 Incipient Attitudes: Phase II Research findings concerned with Phase II of the developmental process in race awareness among children two to five years of age are also present.(4’17’19'22’35'36’57’ 69'70’78'82'84’89) The learning and synthesizing of race- related words, value judgments and preferences are clearly present in research findings. Most of these come from spon- taneous reactions of children in testing situations. Race-Related Words and Value Judgments Children three to five years of age are not only aware of race differences, but also use race-related words and value judgments in expressing their knowledge and feel- ings about these differences.(4’17’19-22’35’36’48’58’69’78’ 82'84’89) Evidence of responses indicate the use and some understanding of such words as "black, brown, colored, Negro "(4,36,57,82,83) and white. Also, valuing of racial groups are sometimes encountered. Whiteness is regarded as posi- tively valued (good, clean, nice, pretty, fine, friendly), while Negro-ness is regarded as negatively valued (bad, ugly, lazy, dirty).(4’36:57s82.83) At three to four years of age, the use of race- related words and value judgments increase.(17’19-22’35’36’ 38:48:58,69’70’82’84’89) It is between four and five years of age, however, that such words and value judgments in- (17,19-22,35,36,38,48,59,69,70,82,84,89) crease rapidly, and becomes more definite at the age of five.(18’76’77) 20 Clark and Clarkcl7’19-22) from their studies con- clude that between the ages of four and five, the associa- tion of awareness of race differences with "rudementary evaluation" of these differences is present. They further state that the child cannot learn what racial group he be- longs to without being involved in a larger pattern of emotions, conflicts and desires which are part of his grow- ing knowledge of what society things about race. Goodmanc36) in her study of Negro and white children, report similar findings. Of her subjects, 85% used race terms from "fairly often" to "often." Also 85% of her subjects thought in terms of color differences, had a sense of racial differentiation and separation and held strongly to values concerning each racial group. Phase II of the developmental process then, is well underway by the time the child enters the public school.(74) Racial Preference Negro and white children, three to five years of age, not only show an increased ability with age to use race terms in making value judgments about different racial groups, but also indicate interesting patterns of racial preference among and between different racial groups.(4’17’19'22’36'38’48'57' 70,81,84,88) The Negro Child: For the Negro child, a strong "out- (48) group orientation" is present. In Horowitz's study some 21 Negro children as young as two years of age indicated a "wish" to be "white." Other studies using subjects be- tween the ages of three to five(4’17’19'22’36’38’84’88) repeatedly report similar findings. Also suggested in these findings were indications that Negro children re- ject children of their own racial groups. This rejection of one's own racial group is likely to be accompanied not only by self-rejection, but also by insecurity, anxiety, and a sense of helplessness.(20’36’57’69’74’84) Concerning age differences, although race awareness and the use of race-related words and judgments increase with age during the preschool period, preference for out- group affinity among Negro children decreases with age.(20, 22’36) This decrease in outgroup affinity is suggested by (20’22) to indicate the young Negro child's Clark and Clark growing realization of a "realistic self-identification." They further state that at seven years of age, the Negro child can no longer escape such realistic self-identifica- tion. It should be mentioned, however, that although a decrease in outgroup affinity is seen between the ages two to five, a majority of these children remain outgroup oriented. Only one study which reports findings on sex dif- ferences in racial preference was found. Greagor and McPherson,(38) reported that among Bantu (Negroid) children of South Africa, more males than females remained outgroup 22 oriented at all age levels. The limited research done in this area, however, prevents any valid generalizations. Clark and Clark's(21) study dealing with the vari- ables of geographical location and racial preference, also indicate the presence of variations among Negro children. They report findings which suggest Northern Negro children to be more definite in their preference for white children than Negro children from Southern communities. Southern Negro children preferred children of their own racial groups. Also, Southern NegrOChildren were more verbal and acted "matter-of—factly" to questions of preference, while Northern Negro children verbalized more escape responses, or indicated emotional strain when questions of preference were asked of them. This tended to support the hypothesis that Northern Negro children have a tendency to repress or escape the painful fact of the "meaning" of color difference in American society. The only study concerned with the variables of socioeconomic backgrounds and racial preference is Vaughn's(88) study of Maori (Negroid) children from New Zealand. His re- sults were inconclusive, but a tendency was reported. There was a tendency for more "lower status” Maori subjects to favor persons or objects signifying their own race, than "high status" Maori subjects. Research in relation to the variables of skin color (light, medium, dark) is also limited. However, studies of 23 Clark and Clark(22) and VaughnC88) are present. Clark and Clark(22) report findings which indicate light Negro chil- dren to be more outgroup oriented than medium and dark Negro children. The conclusion that an earlier development of race awareness based on skin color among dark and medium children was previously mentioned. The findings of Clark and Clark(22) summarized above led them to accept such in- ference with more confidence. (88) in his study on ethnic attitudes among Vaughn Maori children of varying skin colors, also supported the results reported by Clark and Clark.(22) A consistant and significant tendency was observed in Vaughn's study (p = <.02) between Maori children with varying skin colors.(88) The pattern of outgroup orientation in racial pre- ference among Negro children is well supported by many relevant studies. However, with respect to the correlated variables of racial preference and sex, geographical loca- tion, socioeconomic backgrounds and skin color, research is limited and sparse. The White Child: Studies of white children three to five years of age indicate an "ingroup orientation" to be (4’36’38’57'81’84’88) Such ingroup orientation in- present. volves a consistant tendency for white children to prefer persons and objects signifying their own race. Some studies report white children to express negative responses toward Negro children. These include feelings of rejection, 24 (4,36,38,57,76,77,85) There is a hostility and aggression. positive relationship between increasing age, and the ex- tent to which white children express their hostilities to— ward Negro children.(36’48’70’79) Studies concerned with the correlated variables of racial preference and sex, socioeconomic backgrounds and geographical location are sparse. More research is neces- sary for a clearer understanding of racial preference among white children of preschool ages. Awareness of Sex Differences: The Preschool Aged Child "A sex role standard refers to a learned association between selected attributes, behavior and attitudes on one hand, and the concepts of male and female on the other. during the period three to seven years of age, the child gradually realizes that people fall into one of two related language categories--boys and girls, men and women, father and mother."(50: P- 138) These generalizations were recently made by Jerome Kagan and supported by many relevant studies concerned with the development of sex awareness in young children. A review of literature concerned with this aspect of the child's development indicates that the awareness of sex differences among children is a continuous process throughout childhood.(9’11’13'15’23’24’27'29'42'44’47’51’ 67’80’90) Like race awareness, reseachers have attempted 25 to distinguish between essential and overlapping phases in its development; however, difficulty has been encountered.(13’ 15’44’62) For the purpose of this thesis the awareness of sex differences among preschool children will be considered from three aspects of the develOpmental process. These as- pects include understanding the child's awareness of sex differences in terms of l) differences in physical attributes, 2) differences in toy preferences and sex cleavage and 3) differences in overt behaviors. Evidence supports the assumption that young children four years of age have dichotomized the world into male and female pe0ple, and are concerned with boy and girl differences.(5’23’29’30’51’63»75) Physical Attributes The early discrimination of sex differences among preschool children is facilitated by the presence of a wide variety of clearly discernable physical cues. These cues (attributes) include a person's clothing, hair, body form and proportion (genitals and breasts) and urinal posture. Research indicates that some children two to three years of age are already aware of sex differences.(11’12'29’34'52’53) (11) state Though not in the majority, Brieland and Nelson that those two year olds who showed an ability to discrimi- nate between the sexes, did so when presented with paired photographs of clothed and nude children. All paired photo- graphs were identical save the single variable of view, 26 segment of the body, hair and external genitals. Goodenough(34) made similar generalizations through the use of a Picture Drawing Test. Her study indicated that some children two years of age drew figures of male and female with discern- able physical characteristics. Kochfsz’ss) Brim,(12) and Dillon,(29) however, made their generalizations from obser- vations of children in play situations and at toilet time. For two year olds, consciousness of sex differences is understood in terms of clothing and hair, rather than genitals or breasts. Only few children consider the geni- tals and breasts as sex differentiating characteristics at this age. Research points out this to be true of children three to five years of age also.(12’24’29’34’75’80) However, the ages between three and four are most crucial in the development of sex awareness in young chil- (5,24,29,30,51,63,75,80). dren. By the age of four, children have a clear conception of "boy" and "girl."(24’29’30’51’63' 75’80) Dillon(3l) reports that children three-and-a-half years of age have no "misapprehension" of the words "boy" (30) and "girl." Emmerich indicates that at four, children distinguished not only between boys, girls, fathers, and mothers, but also perceived these differences in terms of (30) power relationships. Smith reports from her records of verbal responses among children, questions of sex dif- (63) after studying ferences asked by four year olds. Mamiya four year old Japanese children reported his subjects to show a clear consciousness of their own sex role. 27 Though clothing and hair still remain the dominant sex differentiating characteristics at this age, Conn(24) (51) and Katcher reported an increasing number of children considering breast develOpment, urinal posture and genitals in their differentiations. One-third of Katcher's(51) four year olds and one-half of his five year olds were aware of genital and breast differences. At the age of six, this increased to three-fourths. Thus, research clearly points out that most children of preschool ages are aware of sex differences. The majority of their differentiations are based upon clothing and hair, but an appreciable amount consider the breasts and genitals as sex differentiating characteristics. Toy Preference and Sex Cleavagg The games, toys and play materials chosen by pre- school children have also been utilized as an indicator of sex awareness among young children. A large body of re- search on children's game and toy preference indicates that boys chose objects related to sports, machines, aggression, speed and strength, while girls selected games and objects associated with the kitchen, home, babies and personal ap- pearance.(7’13’15’23’31’34’38’39’43’44’47’50’59’66’67’75’86) These preferences for certain types of toys denoting "mascu- linity” and "femininity” were even seen among children two years of age (7,34,66,67,85) 28 Many of the studies conducted in this area of re- search have used their own test device to assess the toy and game preferences of preschool children. The ITSC (It Scale for Children) created by Brown(13) has become most popular. Other studies have utilized the direct choice (7'24’43’75) or interview with parents(34) to ob- technique tain their data. The major results of these instruments indicate that a majority of boys and girls between the ages of two and five, preferred sex-appropriate toys. Also suggested was the generalization that boys exhibited a much clearer preference for masculine toys than girls did for feminine ones. This was found to be most pronounced at the age of five. A consistant increase in boys' preferences for mas- culine toys is seen with age, while girls show more varia- bility. At the age of five, several girls indicated a strong preference for masculine toys. However, other re- search findings report conflicting evidences. LefkowitzCSQ) reports that girls have as much preference for the feminine sex role, as boys have for the masculine one. Lanskycss) adds to this conflict by indicating that boys were more variable than girls in their choices of sex-appropriate toys. Girls showed more stability in their choices of sex- appropriate toys. Children of preschool ages not only show preference for sex-apprOpriate toys, but also show sex cleavage in 29 play situations. Observation reportsc16’43’44’60’89) Pic- (1,46,80) (26,68) ture-Preference Tests and Sociometric Techniques indicate that sex cleavage does exist at the ages two to five. Boys favored boys in play situations and also chose boys as recipients of their affectional ties, while girls favored girls and chose girls as recipients of their affectional ties.(l,16,21,46,52,60,68,81,90) However, like research findings present on preference for sex-appropriate toys, research on sex cleavage also suggests that boys show a tendency toward stronger sex cleavage than girls. Sex cleavage among girls were more variable.(16’52’53’58’90) Additional research is necessary to clarify the con- flicting evidences present in toy preferences of children three to five years of age, and also differences in sex cleavage among them. Overt Behavior A review of literature on the third aspect of sex awareness also lends support to the assumption made, that most children of preschool ages are aware of sex differences. Children at this age level are not only aware of sex differ- ences based on physical attributes, or have preferences for sex-appropriate toys and show sex cleavage, but also indi- cate some form of "identification"—-molding their behaviors in accordance with a standard appropriate to their biological 58X. 30 (50) Kagan states: The cultures differential standards regarding the behavioral face to be exposed to the public are not as clearly delineated as the standards for physical attri- butes, but these rules are strongly felt nonetheless. (50, p. 138) These involve a realization and identification with behav- ioral characteristics of aggression and inhibition appropriate for boys and girls; men and women. Research points out that this standard requires the inhibition of verbal or physical aggression among girls; but gives boys license and encourage- ment in expressing aggression when attacked, threatened or dominated by another male. Among preschool children, it is rather difficult to find a sound study in which aggressive behavior is not more frequent among boys than among girls.(27’ 45’49’64’72’90) Dawec27) in her study of two and three year olds indicated that boys quarrelled more often than girls. Boys were also more aggressive during their quarrels, and this aggressiveness increased with age. Hattwick(45) also studying children two to four-and-a-half years of age reports boys showing more "aggressive-extroverted behaviors" than girls. Grabbing toys, attacking others, refusing to share, refusing to comply, breaking up toys, rushing into danger and activeness were more present among boys than girls. (64) conducted observations of children Recently, McCandless three to five years of age. His findings indicated that boys initiated more conflicts and were more aggressive in their play than girls. 31 This difference is also present in make believe themes (6.73) children tell to pictures and dolls and in children's differential preceptions of adult males and females. When children were asked which parent was more dangerous or puni- tive (aggressive) both boys and girls agreed that the father was more aggressive than the mother.(30) In reference to the behaviors of inhibition, with- drawal, dependency, passivity and conformity, studies indi- cate girls to have greater license in exhibiting these (45,64) behaviors than boys. Hattwick(45) reports from her observations that girls two to four years of age exemplified "withdrawn and introverted tendencies." They avoided play, expressed jealousy, shrank from notice, sought praise from teachers, stayed near adults, sucked their thumbs, feared people and places, and avoided risks more often than boys. This was true at all age levels studied. Recently, McCand— (64) also reported that girls showed more dependency less upon adults than boys, and asked teachers to solve their problems of conflicts more often. Girls shied away from conflict situations. Thus, most children of preschool ages are not only 1) aware of sex differences based on physical attributes, 2) show preference for sex-apprOpriate toys and 3) indicate sex cleavage, but also 4) show some form of molding their behaviors in accordance with standards appropriate to their biological sex. 32 Race and Sex Factors in the Friendship *Preferences of“White Preschool Aged Children The review of literature on studies done in the area of race awareness and sex awareness has pointed out that the awareness of race differences and sex differences does exist at the preschool age level. This fact has been well estab- lished and the relationship between age and awareness of both race and sex differences is fairly clear. However, with reSpect to the correlated variables of age/race awareness and age/sex awareness, conflicting evidences have been seen. These conflicting evidences have occurred in research done by: l) Moreno,(68) Criswell,(26) Helgerson,(46) and Abel and (1) Sahinkaya, whose studies showed that children of preschool ages placed a greater emphasis upon sex rather than race in their choice of playmates; 2) Lambert and Tachhi,(56) (65) (81) whose studies McCandless and Hoyt, and Springer, showed that children of preschool ages placed a greater em- phasis upon race rather than sex in their choice of playmates, and 3) Landreth and Johnson(57) and Stevenson and Stevenson,(83) whose studies showed no consistant patterns present in the choice of playmates among preschool children. The controversy between the research findings is a continuing one. No definite solution has yet been found. To understand more fully the depth of such a controversy, the researcher of the present study has attempted briefly to critique the techniques used by previous researchers. 33 This is done to provide a basis for the present study. These techniques fall into four basic categories which in- clude: l) a Picture-Preference Test Technique, 2) a Socio- metric Test and Interview Technique, 3) an Observation Technique, and 4) a Picture and Inset Test Design Technique. The Picture-Preference Test Technique Helgerson,(46) Abel and SahinkayaCl) and Springer(8l) all utilized the Picture-Preference Test Technique to obtain their data. Facial photographs of children from varying (46) study, racial backgrounds were employed. In Helgerson's facial photographs of Negro and white children (male and female) with different facial expressions (sober and laugh- ing) were used. These photographs were paired together, creating three different sets. In these sets the variables of race, sex and facial expressions were controlled. Set I held the variables of race and facial expressions as con- stants; Set 11 held the variables of sex and facial expres- sions as constants; Set III held the variables of race and sex as constants. Limits were encountered, however, in (46) Helgerson's neglect of presenting all possible paired choices to his subjects. His Picture-Preference Test allowed subjects to choose between 1) a Negro boy and a Negro girl, 2) a white boy and a white girl, 3) a white boy and a Negro boy, and 4) a white girl and a Negro girl, with varying facial expressions. However, such a test did not allow sub- jects to choose between 1) a Negro girl and a white boy, and 34 2) a white girl and a Negro boy. Generalizations made from data collected by this Picture-Preference Test, therefore, were limited in nature. Abel and Sahinkayacl) also employed paired photo- graphs of Negro and white children (male and female) to (46) their Picture-Pre- obtain their data. Like Helgerson ference Test showed definite limitations in reference to the possible number of paired choices they presented their subjects. Unlike Helgerson,(46) however, Abel and SahinkayaCl) allowed their subjects to make choices between 1) a Negro girl and a white boy, 2) a white girl and a Negro boy, 3) a white boy and a Negro boy, and 4) a white girl and a Negro girl. The omission of the paired choices of l) a white boy and a white girl and 2) a Negro boy and a Negro girl was seen. Generalizations made from their study, therefore, were also limited in nature. Unlike Helgerson(46) and Abel and Schinkaya,(l) (81) in her study of children from Hawaii, utilized Springer another variation of the Picture-Preference Test. This technique included three series of five photographs of chil- dren from different racial groups (Chinese, Japanese, Cau- casion, Filipino.and Part-Hawaiian). In these photograph series the variables of age (each series included photographs of children from different age levels), and race (each series included photographs of children from five racial groups) were controlled. The variables of sex and facial expressions, however, were left uncontrolled. 35 Aside from the neglect of l) presenting all possible paired choices to their subjects, and 2) controlling some significant variables (i.e. facial expression and sex) studies mentioned above made the assumption that choices (preferences) subjects made on the Picture-Preferences Tests indicated race or sex cleavage. Such an assumption was left unsupported by research evidence, thus, the questioning of such an assumption indicated a major limitation in the Picture-Preference Tests employed. The Picture—Preference Test of the present study has attempted to meet all three limitations encountered, so as to increase its effectiveness as the research instrument. It presented all possible paired choices to the subjects, attempted to take into account the variables of race, sex and facial expression, and also recorded descriptive data on all subjects which strengthened the assumption made by previous researchers about the Picture—Preference Test. The Sociometric Test and Interview Techniqge Both Moreno(68) and Criswell(26) in their studies of ethnic cleavage in children employed a standard Socio- metric Technique (a study of proximity) to gather their data. Standard sociometric questions which involved asking their subjects: 1) "Which child in the group would you like to sit next to?" or 2) "Which child would you choose to sit (68) next to?" were used. Since both Moreno's and Criswell's (26) 36 data provided evidence stating that racial cleavage among children was not present until the age of ten, the appro- priateness of their methods were questioned by Lambert and (56) Lambert and TaguchiCSé) felt that the ques- Taguchi. tions employed by Morenoc68) and Criswell(26) did not lead children to discriminate "favored associates." They believed that children of preschool ages "didn't mind" or generally "liked everyone" who sat next to them. Because of this, (56) Lambert and Taguchi incorporated within their choosing procedures, what they believed to be "more significant" questions for preschool children. This involved asking their subjects: 1) to give a piece of candy to an associate, then receive one for themselves, 2) to choose an associate to pose with for picture-taking, and 3) to indicate which picture among individual poses of each school member they liked best and would want to keep. Using such questions in (56) (68) their choosing procedures, Lambert and Taguchi reported findings which countered the studies of Moreno and Criswell.(26) (56) Though the critique Lambert and Taguchi made of (68) (26) methods remain an important Moreno's and Criswell's one, all three studies mentioned above overlooked one major limitation in their techniques. The use of standard socio- metrics involved subjects to make choices of "favored associates” from among children known to them. Because of this, numerous variables, such as the individual child's 37 interests, needs and personality traits, were left uncon- trolled. Generalizations made from their data, therefore, were quite global in nature, and in need of support by other more precise techniques. The critique of past sociometric techniques indi- cate major limitations in: l) the kinds of questions asked of the subjects about their preferences and 2) the numerous extraneous variables left uncontrolled by allowing subjects to make choices of ”favored associates" from among children already known to them. In reference to the first limita- tion, questions in the present Picture-Preference Test were simple in nature. These questions were chosen because of their successful employment in numerous studies dealing with attitudes of young children. Also an attempt at remedying the second limitations was made. In the present Picture- Preference Test photographs of children unknown to the sub— jects (a hypothetical group) were used. This was done in an attempt to control some of the numerous variables left uncontrolled by the previous Sociometric Test and Interview Techniques. The Observational Technique (65) McCandless and Hoyt and Stevenson and Steven- son(83)utilized observational techniques to investigate the interaction patterns of children in integrated nursery school (65) situations. McCandless and Hoyt obtained forty-five, 38 fifteen minute observations of their subjects over a period of three—and-a-half months. Three trained observers were employed to record the type and duration of interactions encountered by their subjects. The use of these trained observers tended to cancel out the individual bias of ob- servers, however, other limitations were present. One major limitation involved the researchers' selection of certain observable social behaviors to indicate race or sex cleavage. Such a selection neglected a total concern for the complex dynamics of interpersonal relationships that exist between children in interaction situations. Also, Proshansky(74) in reviewing studies on racial preference indicates that studies of very young children clearly suggests that there is little relationship between "racial behavior" and "racial orientation"; or between what young children.say and do pub- lically, and what they indicate privately or reveal on doll (36) and picture tests. He cites Goodman's study to support this generalization. (83) Stevenson and Stevenson obtained fourteen, forty-five minute observations of children at the beginning of the nursery school year, and six at its end. No mention of control for observer bias was made in their study. This may indicate a major limitation in their research. Criti- (65) cisms made of McCandless and Hoyt's study are also applicable here. 39 Of the limitations encountered in past Observational Techniques, those mentioned were: 1) observer bias, and 2) the selection of specific observable social behaviors to in- dicate race or sex cleavage. The preSent Picture-Preference Test attempted to meet these limitations by: 1) allowing all subjects to make their own choices on tasks in the Picture- Preference Test, and 2) using direct statements and responses of children to indicate whether their choices were made on the basis of race or on the basis of sex. These methods con- trolled much of the numerous variables left uncontrolled by previous Observational Techniques. The Matching Techniqge Landreth and Johnson(57) employed their own Picture and Inset Test Design to obtain their data. The Picture and Inset Test required subjects to choose one of three in- set pictures designed to reveal their reactions to persons of different skin color. The insets, as well as the pictures were of drawings of a man, woman, boy, girl or baby. The size and shape of each picture and inset were similar. The only difference was in skin color. Regarding this technique, both Landreth and John- son(57) indicated that their data on white subjects showed them choosing an inset picture because of correct racial matching, rather than actual preference. For this reason, the apprOpriateness of their entire Picture and Inset Test 40 Design is questioned. This test may obtain data on race awareness, but whether it reveals information regarding racial preference is questioned. Conflicting evidences in research findings concerned with the race and sex factors in the friendship preferences of white preschool aged children, and the limitations en- countered in previous research techniques employed to under- stand such a problem, indicate a need for more research in the area. The critique on previous techniques employed, provide the basis upon which the research of the present study designed his own Picture-Preference Test. CHAPTER III METHOD The Picture-Preference Test A Picture-Preference Test designed for this study was used as the research instrument. The test was made up of twelve 5" by 7” black and white facial photographs of children between the ages of three-and—a-half to five-and- a-half. The twelve photographs were chosen by a board of five adult judges from a sample of twenty—five photographs of Negro (dark-skinned) and white children (six Negro boys, five Negro girls, seven white boys and seven white girls) taken by the researcher with a Kodak 804 Instamatic Camera. All photographs were taken on a white background to lessen the bias effect of the environmental conditions. The cri- teria for choosing these photographs included: 1) that the photographs chosen be distinguishable by race, 2) that the photographs chosen be distinguishable by sex, 3) that the photographs chosen be of children with smiling faces, and 4) that the photographs chosen consist of three white boys, three white girls, three Negro boys and three Negro girls. After this selection was made, the photographs were paired together, creating six tasks. These tasks were again 41 42 combined to create three sets. These three sets are de- scribed in the following paragraphs. Set I: Set II: Set III: Tasks One and Two comprise Set 1. Task One involves a choice between a white boy and a white girl. Task Two involves a choice be- tween a Negro boy and a Negro girl. This set was designed to test the first hypothesis set forth in this thesis; that is: "White children of preschool ages will prefer friends of their own sex when the variable of race is held constant." Tasks Three and Four comprise Set 11. Task Three involves a choice between a white boy and a Negro boy. Task Four involves a choice between a white girl and a Negro girl. This set was designed to test the second hypothesis set forth in this thesis; that is: "White children of preschool ages will prefer friends of their own race when the variable of sex is held constant." Tasks Five and Six comprise Set 111. Task Five involves a choice between a white boy and a Negro girl for boys, and a white girl and a Negro boy for girls. Task Six involves a choice between a Negro boy and a white girl for boys, and a white boy and a Negro girl 43 for girls. This set was designed to test the third hypothesis set forth in this thesis; that is: "White children of preschool ages will prefer friends of their own sex when the variables of race and sex are varied." Three paired pictures of cats and dogs were used as practice samples. These samples were incorporated into the testing procedures and presented to the subjects prior to the actual administration of the Picture-Preference Test. This was done to familiarize the subjects with the procedures of the test and to establish the needed rapport with the sub- jects. To introduce both practice samples and the Picture- Preference Test to the subjects, the researcher said: (Child's name), I would like to show you some pictures ofianimals and children, and would like to ask you some questions about them. Which animal (child) would you like to choose as a pet (friend)? After each task was administered, the researcher then asked: Can you tell me why you chose this animal (child)? Tell me more. Tell me still more. This was done to obtain descriptive data concerning the child's friendship preferences. The Reliability of the Picture-Preference Test The test-retest. method was used to determine the reliability of all tasks in the Picture-Preference Test. The reliability of a research instrument refers to the 44 extent to which repeated applications of the test to the same population yields the same scores. One-half of the original sample (30 subjects) were randomly selected from three nursery School groups of the original six, for retesting. The retesting took place two weeks after the final testing was done. This decision to allow two weeks to pass before retesting was based upon the factors of memory and learning of the subjects. The length of time between the testings kept memory at a minimum. Since during the final testing no indication was given to the subjects concerning the correctness of their choices, the possibility of learning was also kept at a minimum. A phi-correlation coefficient was obtained for Task One of the Picture—Preference Test. This statistic was used because each of the variables in the task were dichotomous variables. However, due to the nature of the choices pre- sented to the subjects in Tasks Two, Three, Four, Five and Six, the phi-correlation technique could not be utilized. Percentages were obtained for subjects making the same choices on two administrations of these tasks in the Picture-Prefer- ence Test. Table I lists the reliability coefficient for Task One and the percentages obtained for Tasks Two, Three, Four, Five and Six. Both the reliability coefficient and the percentages reveal a high reliability for all tasks in the Picture- Preference Test. 45 TABLE I: A Reliability Coefficient for Task One and Percen- tages for Tasks Two, Three, Four, Five and Six Obtained From Two Administrations of Bash Task in the Picture-Preferences Test for Thirty Subjects Task One Task Two Task Three Task Four Task Five Task Six Correlation Percent- Percent- Percent- Percent- Percent- Coefficient age age age age age .80 96 2/3 96 2/3 90 100 86 2/3 The Validity of the Picture-Preference ' Test During the pilot study an attempt was made to estab- lish an index of validity for the present Picture-Preference Test. Added to the procedures of the test were questions regarding the photographs of children used in three sets of the Picture-Preference Test. These questions were asked of each subject after the administration of the entire test. The questions included: 1) Sex Identification: Which two children are boys? and which two children are girls?; 2) Race Identification: Which two children are Negro (colored or black)? and which two children are white?; and 3) Facial Expression Identification: Are all the children smiling? All subjects(n=ll) were able to identify photographs of both their own and Opposite sex and racial groups used in three sets of the Picture-Preference Test. In reference to race identification, those subjects who made correct identi- fications when the word "Negro" was used (n = 3) made correct 46 identifications when the words "colored" and "black” were used. Those who made correct identifications when the word "colored" was used (n = 4), made correct identifications when the word "black" was used. A Of the responses concerned with the facial expres- sions of children in the photographs, all subjects of the pilot study agreed that these photographs were of children with smiling faces. No negative responses were recorded. Data obtained from subjects of the pilot study for an index of validity, therefore, indicate that the photo- graphs used in the Picture-Preference Test were distinguish- able by sex, race and facial expression. This provided a basis for the assumption that the Picture-Preference Test is effective as a research instrument for understanding the race and sex factors in the friendship preferences of white preschool aged children. The Pilot Study A pilot study was undertaken to test the effective- ness of the proposed Picture-Preference Test. Subjects used in the pilot study were selected from two all white nursery school classes at the Michigan State University cooperative nursery school (Spartan Nursery School). These subjects were randomly chosen from such groups after the subjects for the final study were selected. Eleven children, five boys and five girls, aged 3.9 to 4.10 years, served as subjects for the pilot study. 47 During the pilot study attention was particularly focused upon: 1) the aquaintance of the researcher with the procedures of the Picture-Preference Test, 2) the efficiency and clarity of instructions, 3) the use of a tape recorder to record subjects' responses to questions of preference (descriptive data), 4) the scoring sheet (See appendix), and 5) the establishment of some index of validity for the photographs used in the Picture-Preference Test (previously described). After the results of the pilot study were evaluated, two major modifications in the test procedures were under- taken. The appreciable amount of subjects not responding to the questions of preference indicated a need for modifi- cation in this portion of the test. After experimenting with a variety of questions focused on eliciting responses from subjects, two additional questions were incorporated into the test procedures. This will be discussed later in the section on procedures of the final study. Also a decision not to use the type recorder as a research instrument was made. Statements and responses of subjects were short and simple, thus allowing for the easy recording of such data by a shorthand method. This also cancelled the distractable character of the tape recorder, preserving the initial enthusiasm with which the subjects involved themselves in the test. 48 The Final Study Description of Subjects Sixty white children of preschool ages, thirty boys and thirty girls, ranging in ages from 3.5 to 5.2 years were randomly selected from six all white nursery school classes held at the Michigan State University (Spartan Nur- sery School). All subjects came from homes in which either one or both of their parents were enrolled as students within the university. Of the sixty subjects, fifteen boys and fifteen firls were between the ages of 3.5 to 4.2 years, and fif- teen boys and fifteen girls were between the ages of 4.5 to 5.2 years. The mean age for subjects between the ages 3.5 to 4.2 was 3.9 for both boys and girls. The mean age for subjects between the ages 4.5 to 5.2 was 4.8 for boys and 4.7 for girls. Those subjects between the ages 3.5 to 4.2 years will hereafter be known as the younger group in this study, while those between the ages 4.5 to 5.2 years will be known as the older group (see Table II). Since this study was an attempt to familiarize the re- searcher with the problems present in the field of race aware- ness in young children, the sample has been limited to white children only. Other research will be done in the future with subjects from both integrated and non-integrated Negro groups. The overall permission to conduct this study at the Michigan State University Nursery Schools was granted by the 49 TABLE II: Age and Sex of Subjects Employed in This Study YOUNGER GROUP Subject Boys . Girls No. ' (Age) (Age) H OKDWVO‘U'I-DLNNH I—It—I NH I—‘l—l #04 b-b-b-b-B-RLNLNMLNUIUJCNCNM l—‘ O hpbpbpwmmwwmmmm H O ._a U1 Mean = 3.9 Mean = 3.9 OLDER GROUP Subject Boys Girls No. (Age) (Age) 1 4.5 4.5 2 4.5 4.5 3 4.6 4.5 4 4.6 4.6 5 4.8 4.6 6 4.8 4.6 7 4.9 4.7 8 4.9 4.7 9 4.9 4.8 10 5.0 4.10 11 5.0 5.0 12 5.0 5.0 13 5.1 5.0 14 5.1 5.2 15 5.2 5.2 50 Department of Home Management and Child Development in the College of Home Economics. The Establishment of Rapport In order to facilitate a successful presentation of the Picture-Preference Test, time was Spent in establish- ing rapport with the subjects. Participation in nursery school groups from which the subjects were selected occurred over a period of three weeks. Consequently at the time of testing, most of the subjects were willing to participate in the study. In approaching the subjects for participation in the study, the researcher said: (Child's name), I brought the pictures of animals and children to school today, and would like to show them to you. Would you like to see them now? Although most of the subjects reSponded willingly, several subjects who, because of their involvement in some other activity in the nursery school room, refused to participate. For these subjects the researcher said: Oh, you're busy now. Maybe later on. Only those subjects who were willing to participate were used in the study. The subjects were approached three times before help was sought from the teacher. Two subjects from the original sample refused to participate in the study. A matching procedure with age and sex as variables was used to select two more subjects to fulfill the sample's quota. 51 The Test Room A small room adjoining the nursery proper was used to test the majority of the subjects (53 subjects). Seven subjects were tested in another room Of similar size at the Opposite end of the nursery building. Presentation of the Picture-Preference Test Only one subject and the researcher were present in the room during the interview. During this interview situa- tion the subjects were seated in a chair at a small table with the researcher to the left of him. In the center of the table, in front of both the subjects and the researcher was placed the Picture-Preference Test. The paired photo- graphs in each task of the Picture-Preference Test were mounted on pieces of stiff cardboard with a neutral (white) background. The position of the photographs on the stiff cardboard were randomly arranged. Both the presentation of the practice samples and the actual Picture-Preference Test were the same as pre- viously described. In reference to the questions of pre- ferences, however, modifications were undertaken. When subjects made their preferences, but gave no reason for them, the researcher said to the subjects: What about this one? (Pointing to the photographs the subjects did not choose). If the subjects answered negatively to this question, the researcher proceeded to ask: 52 Can you tell me why you don't like this one? (pointing to the photographs to which the subjects responded nega- tively). Tell me more. Tell me still more. However, if the subjects answered affirmatively to such a question, the researcher then proceeded by saying: Then tell me why you chose this one first (pointing to the subjects' original choice). Tell me more. Tell me still more. These additional questions greatly helped the researcher in Obtaining responses from the subjects. All subjects made some form of response in the final study. Collection of Data The collection of data occurred over a period of eight days, during the nursery school's free play session. One hour in the morning and one hour in the afternoon were utilized for testing. About seven or eight subjects were tested each day. CHAPTER IV ANALYSIS OF DATA Data collected for this study were analyzed to test the three hypotheses set forth in this thesis. Statistical measures were used to analyze the frequency preferences ob- tained on all tasks and sets of the Picture-Preference Test, while a descriptive technique was used to analyze the de- scriptive data. Results of both the frequency preferences and the descriptive data were organized and reported separately. Additional content analyses of the descriptive data were done. Age and sex differences were also determined for all data collected. Only where significant differences oc- curred were they reported. Frequency Preferences The chi-square test of independence was used to test the first hypothesis by analyzing the frequency preferences obtained on Set I (Tasks One and Two) of the Picture-Prefer- ence Test. This statistic was used to test the independence of two variables in the first hypothesis; that of the sex of the subjects and the sex of the photographs chosen by the subjects. In order not to reject this hypothesis a chi-square 53 54 value of 6.64 with one degree of freedom (p = .01) was needed for each task in Set I of the Picture-Preference Test. The binomial method was used to test the second and third hypotheses by analyzing the frequency preferences ob- tained on Set I1 (Tasks Three and Four) and Set III (Tasks Five and Six) of the Picture-Preference Test. This statis- tic was used to determine whether the results obtained differed significantly from chance. In order not to reject these hypotheses a z-score of 2.06 (p = .01) was needed for each task. Set I (Tasks One and Two) The hypothesis to be tested for Set I of the Picture- Preference Test stated that "white children of preschool ages will prefer friends of their own sex when the variable of race is held constant; that is: 1) white girls will choose girls when a choice between a white boy and a white girl is given to them; 2) white girls will choose girls when a choice between a Negro boy and a Negro girl is given to them; 3) white boys will choose boys when a choice between a white boy and a white girl is given to them; and 4) white boys will choose boys when a choice between a Negro boy and a Negro girl is given to them." For Task One and Two of Set I, chi-square values of 39.16 and 33.04 with one degree of freedom were obtained respectively. This indicated that there was a significant 55 difference between the two variables present in the first hypothesis; that is between the sex of the subjects and the sex of the photographs chosen (p = <.01). Boys and girls preferred friends of their own sex when the choices between 1) a white boy and a white girl (Task One), and 2) a Negro boy and a Negro girl (Task Two), were given to them. The hypothesis set forth for Set I of the Picture-Preference Test is therefore not rejected. These results suggest that a significant number of subjects showed signs of sex cleav- age in Tasks One and Two of the Picture-Preference Test. This is in harmony with previous studies of sex cleavage among children of preschool ages. Table III lists the total number of frequency preferences made by all subjects on Set I of the Picture-Preference Test. TABLE III: Frequency Preferences of Sixty Subjects On Set I (Tasks One and Two) of the Picture-Preference Test Task One (p = <.01) Task Two (p = <.01) Sex Boys Girls Total Boys Girls Total Boys 25 5 30 26 4 30 Girls _1 _2_9 _39 __2_ _2_§ fl Total 26 34 60 28 32 60 x2 value = 39.16 x2 value = 33.04 Set II (Tasks Three and Four) The hypothesis to be tested in Set II of the Picture- Preference Test stated that "white children of preschool 56 ages will prefer friends of their own race when the variable of sex is held constant; that is: 1) white girls will choose white boys when a choice between a white boy and a Negro boy is given to them; 2) white girls will choose white girls when a choice between a white girl and a Negro girl is given to them; 3) white boys will choose white boys when a choice between a white boy and a Negro boy is given to them, and 4) white boys will choose white girls when a choice between a white girl and a Negro girl is given to them." For Task Three of the Picture-Preference Test, a z-score of 6.91 was obtained (p = <.01). Both boys and girls chose photographs of their own race in Task Three of the Picture-Preference Test. For Task Four of the Picture-Preference Test, a z-score of 6.30 was obtained. (p = <.01). Both boys and girls chose photographs of their own race in Task Four of the Picture-Preference Test. The hypothesis set forth for Set 11 of the Picture-Preference Test is therefore not re- jected. These results suggest that a significant number of subjects showed signs of racial (ingroup) cleavage in Tasks Three and Four of the Picture-Preference Test. This is in harmony with previous studies of racial preference among white children of preschool ages. Table IV presents the total numbercflifrequency preferences made by all subjects on Set II of the Picture-Preference Test. 57 TABLE IV: Frequency Preferences of Sixty Subjects On Set II (Tasks Three and Four) of the Picture-Preference Test Task Three (p = <.01) Task Four (p = <.01) Sex Own Race Opp. Race Total Own Race Opp. Race Total Choice Choice Choice Choice Boys 28 2 30 27 3 30 Girls 29 l 30 27 3 30 Total 57 3 60 54 6 6O z-score = 6.91 z-score - 6.30 Set III (Tasks Five and Six) The hypothesis set forth for Set III of the Picture- Preference Test stated that "white children of preschool ages will prefer friends of their own sex when the variables of race and sex are varied; that is: 1) white boys will choose white boys when a choice between a white boy and a Negro girl is given to them (own race-own sex choice); 2) white girls will choose white girls when a choice between a white girl and a Negro boy is given to them (own race-own sex choice); 3) white boys will choose Negro boys when a choice between a Negro boy and a white girl is given to them (opposite race- own sex choice); and 4) white girls will choose Negro girls when a choice between a Negro girl and a white boy is given to them (opposite race-own sex choice)." For Task Five of the Picture-Preference Test, all subjects made friendship preferences of their own race-own sex photographs (see Table V); white boys did choose white 58 boys, and white girls did choose white girls as hypothesized for Task Five of the Picture-Preference Test. However, deSpite the fact that all subjects chose photographs of their own race and own sex, their choices tell us nothing about whether these choices were made on the basis of their own race or their own sex. Descriptive data, to be discussed later, is needed to clarify this finding. For Task Six, a z-score of .26 was obtained for sub- jects making choices of their opposite race-own sex photo- graphs. The hypothesis set forth for Set III of the Picture- Preference Test was therefore rejected. However, further analysis of frequency preferences for Task Six indicate interesting differences between sex groups. Twenty-three out of thirty boys chose photographs of their opposite race-own sex (Negro boy).' The use of a binomial test to analyze this data yielded a z-score of 2.90 (p = <.01). For boys, the hypothesis that white preschool children will prefer friends of their own sex when the variables of sex and race are varied, may well be plausible. However, the data reveal that this is not true for girls in this sample. Twenty-two out of thirty girls chose photo- graphs of their own race-opposite sex used in Task Six of the Picture-Preference Test. The use of the binomial method to analyze this data yielded a z-score of 2.55 (p = <.01). For girls, then, the hypothesis stating that white preschool children will prefer friends of their own sex when the 59 variables of sex and race are varied, is rejected. Instead girls seem to indicate choices opposed to the hypothesized direction; that is making their choices on the basis of their own race rather than their own sex. Thus, findings on boys seem to support the conclusion that white preschool children placed a greater emphasis upon sex rather than race in their choice of playmates, while girls seem to support the Opposite conclusion made that white children of pre- school ages placed a greater emphasis upon race rather than sex in their choice of playmates. Table V lists the total number of frequency preferences made by boys and girls in Set III of the Picture-Preference Test. TABLE V: Frequency Preferences of Boys and Girls on Set 111 (Tasks Five and Six) of the Picture-Preference Test. Task Five (p = <.Ol) TaSk Six Sex Own Race- Opp.Race- Total Own Race— Opp.Race- Total Own Sex Opp.Sex Opp.Sex Own Sex Choice Choice Choice Boys 30 0 30 7 23(p=<.01) 30 Girls 30 0 30 22(p=<.01) 8 30 Total 60 0 60 29 31 60 z-score = 7.75 z-score for boys = 2.90 z—score for girls = 2.55 Although statistical measures could be utilized for an analysis of age and sex differences between entire age and sex groups, the size of the sample did not allow for an analysis within sex and between age groups. However, a 60 descriptive technique was used to discover differences that occurred within sex and between age groups. Fre- quency preferences obtained for Task Six of the Picture- Preference Test indicated that more boys in the older group preferred photographs of their Opposite sex and own race (n = 6) than did boys in the younger group (n = 1). This seems to suggest the possibility that a developmental pattern may be involved in the racial preference of white preschool aged boys. For girls, however, data indicated that both younger and older groups were almost equal in their preferences for photographs of their Opposite sex and own race (n - 12 for younger group; n - 10 for older group). Accordingly, if a developmental pattern does in- deed exist, then girls would appear to reach a more mature level at an earlier age than do boys. Descriptive Data A descriptive technique was used to test the three hypotheses in this thesis by analyzing the descrip— tive data obtained from the Picture-Preference Test. Criteria for Evaluating the Data Before an analysis of the descriptive data could be done, some criteria for evaluating this data had to be established and this data quantified. In reference to the establishment of some criteria for evaluating the 61 descriptive data, three categories were established for each task in the Picture-Preference Test. For Tasks One and Two of Set I, the categories of own sex choice, oppo- site sex choice and uncertain were established; for Tasks Three and Four of Set II, the categories of own race choice, opposite race choice and uncertain were estab- lished; and for Tasks Five and Six of Set 11, the cate- gories of own race choice, own sex choice and uncertain were established. The criteria for assigning the descriptive data to each of these categories were as follows: 1) Own Sex or Own Race Choice: This included re- sponses made by subjects which indicated: a) a definite like for one's own sex (I like boys or girls) or race (I like white); b) a definite dislike for one's opposite sex (I don't like girls or boys) or race (I don't like black, brown, dark or colored people); c) a positive value judgment of one's own sex (see Table X1) or race (see Table XVI); and d) a negative value judgment of one's Opposite sex (see Table X1) or race (see Table XVI). 2) Opposite Sex or Opposite Race Choice: This in- cluded responses made by subjects which indi- cated: a) a definite like for one's opposite sex (I like girls or boys) or race (I like 62 black, brown, dark or colored people); b) a definite dislike for one's own sex (I don't like boys or girls) or race (I don't like white); c) a positive value judgment of one's opposite sex (see Table X1) or race (see Table XVI); and d) a negative value judgment of one's own sex (see Table X1) or race (see Table XVI). 3) Uncertain: This included responses made by sub- jects that could not be assigned to any one of the categories listed above. If any question arose in reference to any response being eval- uated, such a response was assigned to this category. With regard to the quantification of this descrip- tive data, a numerical value of one was assigned to each reSponse made by subjects on all tasks of the Picture— Preference Test. These values were then summed, giving a total score for each category which could be descrip— tively analyzed. Set I (Tasks One and Two) A numerical score of fifty-four out of sixty was obtained from subjects making their own sex choice on Tasks One and Two of the Picture-Preference Test. These scores indicate that boys as well as girls preferred friends of their own sex, when the variable of race was 63 held constant. Descriptive data, therefore, indicate a non-rejectance of the first hypothesis set forth in this thesis. The inference made from frequency preferences, that subjects in the sample showed signs of sex cleavage is acceptable. This set did provide the researcher with information regarding the sex factor in the friendship preferences of subjects studied (see Table VI). TABLE VI: Descriptive Data of Subjects On Set I of the Picture-Preference Test Task One Task Two Sex Own Opp. Total Own Opp. Total Sex Sex Uncer- Sex Sex Uncer- Choice Choice tain Choice Choice tain Boys 25 3 2 30 26 3 l 30 Girls 29 l 0 3O 28 l l 30 Total 54 4 2 60 54 4 2 60 Set II (Tasks Three and Four) Numerical scores of fifty and forty-seven out of sixty were obtained for subjects making their own race choice on Tasks Three and Four of the Picture-Preference Test respectively. These two scores indicate that boys as well as girls preferred friends of their own race when the variable Of sex was held constant (see Table VII). Descriptive data, therefore, also indicated a non-rejec- tance of the second hypothesis set forth in this thesis. The inference made from frequency preferences, that sub- jects in the sample showed signs of racial (ingroup) 64 cleavage is acceptable. This set did provide the researcher with information regarding the race factor in the friendship preferences of the subjects studied. TABLE VII: Descriptive Data of Subjects on Set II of the Picture-Preference Test Task Three Task Four Sex Own Opp. Total Own Opp. *Total Race Race Uncer- Race Race Uncer- Choice Choice tain Choice Choice tain Boys 27 0 3 30 26 l 3 30 Girls 23 l 6 30 21 3 6 30 Total 50 l 9 60 47 4 9 6O Set III (Tasks Five and Six) Numerical scores of twenty-three out of thirty were obtained for boys making choices of their own sex photographs on Task Five and Task Six of the Picture- Preference Test, while numerical scores of nineteen out of thirty were obtained for girls making their own race choices on both tasks. These scores suggest that boys seem to prefer friends of their own race when the variables of race and sex were varied (see Table VII). Descriptive data, therefore, indicate a rejection of the third hypothesis set forth in this thesis. The inference made from frequency preferences obtained on Set III of the Picture-Preference Test is acceptable. As previously 65 stated antithetical results concerning the race and sex factors in the friendship preferences of white preschool children was found. For boyssex rather than race was more influential in their friendship preferences, while race was more influential for girls. The frequency pre- ferences obtained on Task Five of the Picture-Preference Test did not indicate this, but descriptive data on Task Five, and both frequency preferences and descriptive data on Task Six, supported this inference. TABLE VIII: Descriptive Data of Subjects On Set III of the Picture-Preference Test Task Five Task Six Sex Own Own Total Own Own 7Tota1 Race Sex Uncer- Race Sex Uncer- Choice Choice tain Choice Choice tain Boys 7 23 0 30 7 23 0 30 Girls 19 8 3 30 19 8 3 30 Total 26 31 3 60 26 31 3 6O As with frequency preferences, an analysis of descriptive data obtained on subjects within sex and be- tween age groups for Tasks Five and Six of the Picture- Preference Test was also undertaken. Despite the limits present due to the sample size (n = 60), descriptive data obtained for both tasks also indicated that more boys in the older group preferred friends of their own race (n = 6) than boys in the younger group (n = 1). This again seems 66 to suggest the possibility that a developmental pattern may be involved in the racial preference of white pre- school aged boys. For girls, however, descriptive data indicated that both younger and older groups were equal (n = 9 for both groups) in their preference for friends of their own race. Accordingly, if a developmental pat- tern does indeed exist, then girls would appear to reach a more mature level at an earlier age than do boys. Other Findings Aside from interest in testing the three hypotheses set forth in this thesis, additional content analyses of the descriptive data were undertaken. The findings deter- mined through this analysis are reported in two sections which include: 1) sex awareness and value judgments, and 2) race awareness and value judgments. Both the binomial method and the chi-square test of independence previously described were used to-analyze the data wherever possible. In binomial tests, a z-score of 2.06 was necessary for significance at the .01 level; in the chi-square test of independence, a chi-square value of 3.84 with one degree of freedom was necessary for significance at the .05 level. Sex Awareness and Value Judgments A content analysis was undertaken of descriptive data concerned with sex awareness and value judgments made by subjects in describing photographs of their own sex and 67 Opposite sex used in the Picture-Preference Test. Find- ings from this analysis are reported under the tOpics: l) sex-related words, 2) value judgments, 3) types of value judgments and 4) words of value judgments. Sex Related-Words In reference to the sex-related words used by sub- jects, data indicate that all subjects (n = 60) used the words "boy, him or he" and "girl, her or she" correctly in describing photographs of their own sex and opposite sex used in the Picture-Preference Test. This suggested that the subjects of the sample were well aware of sex differences between themselves and other children. Value Judgments (see Table IX) The binomial method was also uSed to analyze the data obtained regarding the number of subjects making value judgments (for examples see Table XI) of their own sex and Opposite sex photographs used in the Picture-Pre- ference Test. These value judgments included both positive and negative type value judgments made by subjects. A z-score of 4.11 was obtained (p = <.01). A significant number of subjects, therefore, made value judgments of their own sex and Opposite sex photographs used in the Picture-Preference Test. Also, a chi-square test of independence was used to analyze this data. This statistic was used to test the 68 independence othwo variables; that of age groupings (older and younger) and those subjects making and not making value judgments in describing their own sex and opposite sex photographs used in the Picture-Preference Test. A chi- square value of 5.95 was obtained, indicating a significant difference to be present (p = <.05). Older subjects made more value judgments of their own.sex and opposite sex photographs than younger subjects. TABLE IX: Subjects Making and Not Making Value Judgments of Their Own Sex and Opposite Sex Photographs Used in the Picture-Preference Test Age Making Not Making Total Younger 19 ll 30 Older 27 3 30 Total 46 14 v 60 z-score = 4.11 x2 value = 5.95 Types of Value Judgments In reference to the types of value judgments made by subjects of their own sex and opposite sex photographs used in the Picture-Preference Test, the binomial method yielded a z-score of 2.33 for subjects making positive value judgments of their own sex photographs (p = <.01). Only two subjects of the sixty made positive type value judgments of their Opposite sex photographs, one subject made a negative type value judgment of his own sex photo- graph, and seven subjects (girls) made negative type value 69 judgments of their Opposite sex photographs (see Table X). Since a significant number of boys and girls made positive type value judgments of their own sex photographs, data suggests that subjects of the sample showed signs of sex cleavage. TABLE X: Types of Value Judgments Made by Subjects of Their Own Sex and Opposite Sex Photographs Used in the Picture- Preference Test POSITIVE TYPE Photographs Making Not Making Total Own Sex 38(p=<.01) 22 60 ‘z-score = 2.33 Opposite Sex 2 58 60 NEGATIVE TYPE Photographs Making Not Making Total Own Sex 1 59 60 Opposite Sex 7 53 60 Words of Value Judgments A listing of the words Of value judgments used by subjects in describing their own sex and opposite sex photographs in the Picture-Preference Test are included in Table XI. Value judgments such as best, better, better than, first, looks better, nicer (nice), pretty, and good, were used by subjects to describe photographs of their own sex. Value judgments such as bad, beats me, fights, mean, 70 TABLE XI: Words of Value Judgments Used by Subjects In Describing Their Own Sex and Opposite Sex Photographs Used in the Picture-Preference Test OWN SEX PHOTOGRAPHS OPPOSITE SEX PHOTOGRAPHS Words of Positive WOrds of Negative No. of Value Judgments Subjects Value Judgments Subjects best 4 bad 1 better 20 beats me 1 better than 16 fights 1 first 14 hits me 1 looks better 1 mean 1 nicer (nice) 3 naughty 1 pretty 2 good 1 OWN SEX PHOTOGRAPHS OPPOSITE SEX PHOTOGRAPHS Words of Negative No. of Words of Positive No. of Value Judgments Subjects Value Judgments Subjects fights 1 looks better 1 funny 1 pretty 1 71 naughty, and hits me, however, were used by a few subjects to describe photographs of their opposite sex. Own sex photographs were positively valued, while a few subjects (girls) indicated opposite sex photOgraphs to be negatively valued. Race Awareness and Value Judgments A content analysis of descriptive data concerned with race awareness and value judgments made by subjects in describing photographs of their own race and Opposite race used in the Picture-Preference Test was also under- taken. Findings in this section are reported under the tOpics: l) race-related words, 2) kinds of race—related words, 3) value judgments, 4) types of value judgments, 5) words of value judgments, and 5) definite statements of dislike. Race-Related Words Table XII summarizes the number of subjects using (correctly) and not using race-related words in describ- ing photographs of their opposite race (Negro). The bi- nomial method was used to analyze this data obtained on subjects using race-related words correctly. A z-score of 3.61 was calculated (p = <.01). This finding indicates that a significant number of subjects in this study were able to use race-related words correctly in describing photographs of their opposite race. 72 TABLE XII: Race-Related Words Used (Correctly) and Not Used By Subjects in Describing Photographs of Their Opposite Race Used in the Picture-Preference Test Sex Used (correctly) ,Not Used Total Boys 24 6 30 Girls 20 10 30 Total 44 16 60 z-score = 3.61 Kinds of Race-Related Words The kinds of race-related words and physical attri- butes used by subjects to describe their own race and Opposite race photographs used in the Picture-Preference Test are listed in Table XIII. Physical attributes such as skin color (not white, black, dark, colored, brown) and hair (curly) were used by subjects to describe photo- graphs of their opposite race. Three subjects used the word "Negro," while one subject mentioned eye color dif- ferences and another mentioned speech differences. The physical attributes of skin color (white, not black, light, isn't colored) and hair (not curly) were also used by subjects in describing photographs of their own race (white). Data on race-related words and physical attributes used by subjects in describing photographs of their own race and opposite race, suggest skin color to be the determining factor in race differentiations. 73 TABLE XIII: Kinds of Race-Related Words and Physical Attri- butes Used By Subjects in Describing Their Own Race and Opposite Race Photographs Used in the Picture-Preference Test OWN RACE PHOTOGRAPHS OPPOSITE RACE PHOTOGRAPHS Kinds No. of Kinds No. of Subjects Subjects not curly hair 2 curly hair 5 white (hair, face) 8 not white 3 not black (face, skin) 3 Black (people, chil- dren, face, skin, light (people, lighter) 4 head) 19 isn't colored 2 dark (people, darker) 9 colored (people, skin) 9 brown (people, faces) 7 Negro 3 different (looks, skin, face) 9 different eyes 1 different speech 1 % Value Judgments Table XIV summarizes the data on subjects making and not making value judgments of their own race and oppo- site race photographs used in the Picture-Preference Test. These value judgments included both negative and positive value judgments made by subjects (see Table XVI for exam- ples). The binomial method was used to analyze the data obtained on subjects making value judgments. A z—score of 3.61 was obtained (p = <.01). This indicated that 74 significant number of subjects made value judgments of their own race and opposite race photographs used in the Picture-Preference Test. Also, a chi-square test of independence was used to analyze the data. This statistic was used to test the independence of two variables; that of age groupings, and those subjects making and not making value judgments in describing photographs of their own race and Opposite race used in the Picture-Preference Test. A chi-square value of 5.46 was obtained, indicating a significant difference to be present (p = <.05). Older subjects made more value judgments of their own race and Opposite race photographs than younger subjects (see Table XIV). TABLE XIV: Subjects Making and Not Making Value Judgments of Their Own Race and Opposite Race Photographs Used in the Picture- Preference Test Age Making Not Making Total Younger 18 12 30 Older 26 4 30 Total 44 16 60 z-score = 3.61 x2 value = 5.46 Types of Value Judgments A summary of the amount of subjects making different types of value judgments of their own race and opposite race photographs used in the Picture-Preference Test is presented in Table XV. The use of the binomial method yielded a 75 z-score of 2.58 for those subjects making positive value judgments of their own race photographs (p = <.01). Only five subjects of the sixty made positive value judgments of their opposite race photograph; one subject made a negative value judgment of his own race photograph, while twenty subjects made negative value judgments of their Opposite race photographs. Data again suggests an ingroup (racial) cleavage to be present. A significant number of subjects made positive value judgments of their own race photographs used in the Picture-Preference Test. TABLE XV: Types of Value Judgments Made by Subjects of Their Own Race and Opposite Race Photographs Used in the Picture-Preference Test _ . - POSITIVE TYPE Photographs Making Not Making Total Own Race 40(p=<.01) 20 60 z-score = 2.58 Opposite Race 5 55 60 NEGATIVE TYPE Photographs Making Not Making Total Own Race 1 59 60 Opposite Race 20 40 60 Words of Value Judgments A list of the words of value judgments used by sub- jects in describing their own race and opposite race photo- graphs in the Picture-Preference Test are summarized in 76 in Table XVI. Value judgments such as better (better than, best), good guys (boys), nice (nicer than, nice face), pretty (prettier, prettier than), and others were used by subjects to describe their own race photographs. Value judgments such as bad guys (boys, not good), ugly (uglier, doesn't look good, looks awful, not so cute), not pretty, not nice, and others were used by subjects to describe their Opposite race photographs. Already incipient atti- tudes about racial groups seem to be present among the subjects studied. "Whiteness" seems positively valued, while "Negro-ness'(dark-skinned) negatively valued. Definite Statements of Dislike The number of subjects making and not making defi- nite statements of dislike concerning photographs of their opposite race used in the Picture-Preference Test are summarized in Table XVII. The binomial method was used to analyze the data obtained on subjects making definite statements of dislike. These include statements such as: l) I don't like black peOple, 2) I don't like black chil- dren, 3) I don't like brown peOple, 4) I don't like that (pointing to photograph.of Negro child), and 5) I don't like colored people. A z-score of 2.58 was obtained (p = <.01). A significant number of subjects in the study, therefore, were not only aware of race differences, could use race-related words correctly, and made value judgments regarding their own race and opposite race photographs, 77 TABLE XVI: Words of Value Judgments Used by Subjects in Describing The Own Race and Opposite Race Photographs Used in the Picture-Preference Test OWN RACE PHOTOGRAPHS OPPOSITE RACE PHOTOGRAPHS Words ofTPOSItive No. of‘ Words othegative No. Of Value Judgments Subjects Value Judgments Subjects better, better than, best 22 doesn't scratch I scratches 1 don't fight 1 fights 1 good guys, good boys 4 bad guys, bad boys, not good 5 looks good, handsome, cute 5 doesn't look good, looks awful, ugly, uglier, not so cute, isn't handsome 7 nice, nice face, nicer than 21 not nice, not so nice 10 pretty prettier, prettier than 19 not pretty, not so pretty 6 hits 1 takes bikes and things 1 on‘ -I -'o o -t-' 0". -m -'o o -w-- Words of Negative No. of Words of Positive No. of Value Judgments Subjects Value Judgments Subjects looks funny 1 prettier than 2 nicer (teeth, smile) 3 pretty 2 78 but also indicated a definite dislike for photographs of their opposite race (Negro) at this young an age. TABLE XVII: Subjects Making and Not Making Definite State— ments of Dislike Concerning Photographs of Their Opposite Race Used In the Picture-Preference Test fi ‘ Sex Making Not Making Total Boys 19 11 30 Girls 19 ll 30 Total 28 22 60 z-score = 2.58 Additional Use of the Picture—Preference Test Aside from an interest in understanding the race and sex factors in the friendship preferences of white pre- school aged children, the researcher (Oriental) incorporated into his final testing procedures two additional questions regarding Task Four (which included photographs of a Negro girl and a white girl) of the Picture-Preference Test. The two questions were asked of the subjects after the en- tire test was administered to each subject. They included: 1) Which little girl do you think is my little girl? and 2) Why do you think that is my little girl? The reasons for incorporating these procedures into the final study were highly exploratory in nature. The re- searcher of the present study was interested in: 1) dis- covering with what racial group (Negro or white) subjects of the sample placed persons of Oriental ancestry, 79 2) discovering whether children at preschool ages were aware of differences between Oriental, white and Negro persons, and 3) understanding the reasons subjects gave for their choices. Data collected in reference to the first interest are summarized in Table XVIII. The binomial method was used to analyze the data obtained on subjects who chose the photograph of a white girl as the researcher's "little girl." To be significant at the .01 level of confidence, a z-score of 2.06=was needed. A z-score of 2.58 was ob- tained (p = <.01). A significant number of subjects, therefore, chose the white girl as the researcher's little girl. Although such results remain interesting, no defi- nite conclusions can be made regarding the subjects' awareness of race differences between Oriental and Negro or white persons. Descriptive data including reasons sub- jects gave for their choices were often unrelated to the concern of the researcher or difficult to interpret (see Table XIX). Also more than half of the subjects did not respond to the question asked of them after they made their choices. Further research in this area should be undertaken. 80 TABLE XVIII: Choice of Photograph Subjects Made to the Question: "Which Little Girl Do You Think Is My Little Girl?” Sex White Negro Total Boys 23 -7 30 Girls l7 13 30 Total 40 20 6O z-score = 2.58 TABLE XIX: Reasons Subjects Gave To The Question: "Why Do You Think That Is My Little Girl?” SUBJECTS CHOOSING WHITE GIRL Reason No. of Subjects Because she looks like your little girl (looks like you; like you) 8 Because she's nice and you're nice 1 Because she has a prettier face 2 Because I like her 1 Because her shirt 1 Because she is _l_ Total 14 SUBJECTS CHOOSING NEGRO GIRL Reason No. of Subjects Because she's Negro (black) 2 Black hair looks like you 1 Because she looks like you (it looks like you, just like you) 6 Because she's not like you (pointing to white girl) 1 Because that's your little girl 1 Because I like her 1 Total 12 CHAPTER V SUMMARY, IMPLICATIONS, LIMITATIONS AND RECOMMENDATIONS Summary Of the studies done on race awareness, the re- view of literature has pointed out that an awareness of race and sex differences in young children does exist at the preschool age level. However, with respect to the correlated variables of age/race awareness and age/ sex awareness, conflicting evidences have been reported. Some researchers conclude that children of preschool ages placed a greater emphasis upon sex rather than race in their choice of playmates; others conclude that chil- dren of preschool ages placed a greater emphasis upon race rather than sex in their choice of playmates; while still others conclude no consistant patterns to be pre- sent in the choice of playmates among preschool children. The controversy centering upon these research findings has continued throughout the first half of the twentieth century. No definite solution has yet been found. The primary purpose of this thesis, therefore, was focused upon seeking further information on the question, 81 82 "Which factor (race of sex) is more influential in the friendship preferences of white preschool aged children?" Sixty white children of preschool ages, thirty boys and thirty girls, ranging in ages from 3.5 to 5.5 years were randomly selected from six all white nursery school classes held at the Michigan State University. All subjects came from homes in which either one or both of their parents were enrolled as students within the university. A Picture-Preference Test designed for this study was used as the research instrument. The test was made up of twelve 5" by 7" black and white facial photographs of children between the ages of 3.5 to 5.5 years. Of the twelve photographs, three were of Negro (dark-skinned) boys, three were of Negro (dark-skinned) girls, three were of white boys and three were of white girls. These photographs were chosen by a board of five adult judges from a largersample of twenty—five photographs according to the criteria described earlier in this thesis. After the selection was made, the photographs were paired to create six tasks. These tasks were again combined to create three sets. These three sets and their related hypotheses are summarized in the follow- ing paragraphs: Sgp_l: Tasks One and Two comprise Set 1. Task One involved a choice between a white Set II: Set III: 83 boy and a white girl. Task Two involved a choice between a Negro boy and a Negro girl. This set was used to test the first hypothesis set forth in this thesis; that is: ”White children of preschool ages will prefer friends of their own sex when the variable of race is held constant." Tasks Three and Four comprise Set II. Task Three involved a choice between a white boy and a Negro boy. Task Four in- volved a choice between a white girl and a Negro girl. This set was used to test the second hypothesis set forth in this thesis; that is: "White children of pre- school ages will prefer friends of their own race when the variable of sex is held constant." Tasks Five and Six comprise Set 111. Task Five involved a choice between a white boy and a Negro girl for boys, and a white _ girl and a Negro boy for girls. Task Six involved a choice between a Negro boy and a white girl for boys, and a white boy and a Negro girl for girls. This set was used to test the third hypothesis set forth in this thesis; that is: "White children of 84 preschool ages will prefer friends of their own sex when the variables of race and sex are varied." After each task of the Picture-Preference Test was administered to the subjects, questions regarding the subjects' preferences were asked. This was done to obtain descriptive data concerning the subjects' friendship preferences. Findings resulting from an analysis of frequency preferences and descriptive data indicate that: 1) white children of preschool ages preferred friends of their own sex when the variable of race was held constant (p = <.01). 2) white children of preschool ages preferred friends of their own race when the variable of sex was held constant (p = <.01). 3) white boys of preschool ages preferred friends of their own sex when the variables of race and sex were varied (p = <.01). 4) white girls of preschool ages preferred friends of their own race when the variables of race and sex were varied (p = <.01). Although statistical measures could be utilized for analysis of age and sex differences between entire age and sex groups, the size of the sample did not allow for an analysis within sex and between age groups. However, a 85 descriptive technique was used to discover differences that occurred within sex and between age groups. Both fre- quency preferences and descriptive data obtained on Task Six of the Picture-Preference Test indicated that more boys in the older group preferred friends of their own race than did boys in the younger group. This seems to suggest the possibility that a developmental pattern may be involved in the racial preference of white preschool aged boys. For girls, however, data indicated that both younger and older groups were almost equal in their preferences for friends of their own race. Accordingly, if a developmental pattern does indeed exist, then girls would appear to reach a more mature level at an earlier age than do boys. Implications Implications of these findings add further to the controversy regarding the race and sex factors in the friend- ship preferences of white preschool aged children. The con- clusions made by Moreno, Criswell, Helgerson and Abel and Sahinkaya, that preschool children placed a greater emphasis upon sex rather than race in their choice of playmates, is in harmony with the findings on boys of the present study. The findings on girls, however, reveal conclusions opposed to the hypothesized direction. For girls, the conclusions made by Lambert and Taguchi, McCandless and Hoyt, and Springer, that children of preschool ages placed a greater emphasis upon race rather than sex in their choice of playmates was accept- able. For boys, sex rather than race occured as more influential in their friendship preferences, while race rather than sex 86 was more influential for girls. This sex difference was discovered in data obtained from subjects in Set 111 of the Picture-Preference Test. Implications of findings for Set I and Set II of the Picture-Preference Test are also present. The findings of Set 1, indicate that subjects of the sample showed signs of sex cleavage. Boys preferred photographs of boys, while girls preferred photographs of girls. This finding is in harmony with studies done on sex cleavage among children of preschool ages. The findings on Set II, indicate that subjects of the sample also showed signs of racial cleavage. White sub- jects of the sample preferred photographs of their own race used in the Picture-Preference Test. This finding is in harmony with studies done on racial preference among white children of preschool ages. Other Findings Additional content analysis of descriptive data con- cerned with sex awareness and value judgments and race awareness and value judgments indicate other significant results. Sex Awareness and Value Judgments An analysis of descriptive data concerned with sex awareness and value judgments made by subjects in describing their own and opposite sexed photographs used in the Picture- Preference Test indicated: 1) that all subjects in the study used sex- related words such as "boy, him or he" and 87 "girl, her or she" correctly. This suggested that the subjects of the sample were well aware of sex differences between themselves and other children. 2) that a significant number of subjects made value judgments (both positive and negative) in describing photographs of their own sex and opposite sex (p = <.01). Also older sub- jects made more value judgments of their own sex and opposite sex photographs than younger subjects (p = <.05). 3) that a significant number of subjects made positive type value judgments in describing their own sex photographs used in the Picture- Preference Test (p = <.01). This data suggested that subjects of the sample showed signs of sex cleavage. 4) that a large number of the value judgments used by subjects of their own sex photographs were positively valued (best, better, better than, first,looks better, pretty, nicer, good), while a few subjects (girls) indicated Opposite sex photographs to be negatively valued (bad, beats me, fights, mean, naughty and hits me). These findings on sex awareness and value judg- ments are in harmony with other studies done on sex awareness and cleavage. 88 Race Awareness and Value Judgments An analysis of data concerned with race awareness and value judgments made by subjects in describing their own race and Opposite race photographs used in the Pic- ture-Preference Test indicated: 1) 2) 3) 4) that a signifcant number of subjects used race-related words in describing their oppo- siterace photographs correctly (p = <.01). This finding suggests that subjects in this study were well aware of children of the Opposite race. that the majority of the kinds of race-related words and physical attributes used by subjects in describing photographs of their own and Opposite race, showed skin color to be the determining factor. that a significant number of the subjects made value judgments (both positive and nega- tive) of their own race and opposite race photographs used in the Picture-Preference Test (p = <.01). Also older subjects made more value judgments of their own race and Opposite race photographs than younger sub- jects (p = <.05). that a significant number of subjects made positive value judgments of their own race 89 photographs used in the Picture-Preference Test (p = <.01). This finding suggests an ingroup (racial) preference to be present among subjects of the sample. 5) that a large number of value judgments made by subjects of their own race photographs were posi- tively valued (better, better than, good guys, nice, nicer than, prettier, prettier than), while an appreciable number of subjects indicated oppo- site race photographs to be negatively valued (bad, not nice, awful, uglier, not pretty). Already in- cipient attitudes about racial groups seemed pre- sent among the subjects studied. "Whiteness" seemed positively valued, while "Negro-ness" (dark- skinned) seemed negatively valued. 6) that a significant number of subjects made definite statements of dislike regarding photographs of their opposite race used in the Picture-Preference Test (p = <.01). These findings on race awareness and value judgments are in harmony with studies done previously on race awareness and incipient attitudes. Limitations Encountered In The Study Although attempts at overcoming pertinent limitations in this study were undertaken, still a number of problems were encountered which inhibited successful efforts in this direc- tion. These limitations are briefly discussed in the following sections. 90 Limitations of the Sample_ The major limitations encountered in reference to the sample of this study included: 1) the restriction of subjects to children from student families within the university, 2) the restriction of the sample to white children from non- integrated nursery school groups, 3) the absence of control- ling the variables of extent of interracial contacts and parental attitudes toward race, and 4) the sample size. The restrictions of the sample to subjects 1) from student families within the university, and 2) from all white non-integrated nursery school groups, suggests that genera- lizations of the results to the larger population is limited. Also, the absence of control of variables concern- ing interracial contacts outside of the nursery school and parental attitudes toward race posed problems. Research studies on older children suggest that these variables are related to the development of race awareness and attitudes in young children. Finally, although the size of the present sample (60 subjects) allowed for an analysis of age and sex dif- ferences between entire age and sex groups, this analysis was not possible within and between age and sex groups. Limitations of the Picture-Preference Test The major limitations encountered in the use of the Picture-Preference Test as the research instrument for this study centered around 1) the use of black and white photo- graphs of children from both Negro and white racial groups, 91 and 2) the use of photographs of dark-skinned Negro children to represent the totality of the Negro race. The black and white photographs reduced the effec- tiveness whereby these photographs represented both the "white" and Negro racial groups. Whether these photographs did in fact depict the essential characteristics of "white- ness" or "Negro-ness"is questioned. In addition, the use of photographs of dark-skinned Negro children set further limitations. The factor of vary— ing degrees of skin color found among Negro persons was left uncontrolled. The results obtained from this study, there- fore, provided information which was limited to reactions of white preschool aged children to Negro persons of dark skin color. Recommendations For Further Study As an outgrowth of this present study of the race and sex factors in the friendship preferences of white pre- school aged children, several recommendations for further study have emerged. First, the use of a larger sample in order to understand the developmental aspects of the friendship preferences of white boys and girls of preschool ages is suggested. An analysis of age differences within sex groups previously mentioned, suggested the possibility of a develop- mental pattern in the racial preference of white preschool aged boys. These data also suggested that girls may precede 92 boys in their development of racial preference. A larger sample would allow for a clarification of these tendencies. Second, a study of white children from mixed groups is also suggested. The present study has been limited to white children from non-integrated groups. Differences be— tween white children from mixed groups and from non-inte- grated groups need to be assessed. Likewise, a study of Negro children from both non-integrated and mixed groups would be helpful so that differences within racial groups (non-integrated or mixed) and between racial groups (Negro and white) could be further studied. Studies providing this type of information are lacking. Next, the variable of parental attitudes toward race should be of major concern in future studies, since research with older children suggests that this variable is related to the development of racial attitudes in young children. Fourth, if the Picture-Preference Test Technique is used to assess the racial preference of preschool children, this researcher suggests that colored photographs of chil- dren be used in order to accentuate varying skin colors (dark, medium and light). This would in all likelihood increase the effectiveness of the Picture-Preference Test, and also provide more pertinent information about the re- actions of preschool children to individuals with varying skin tones. 93 Finally, additional questions about the Picture- Preference Test were asked in order to ascertain whether the subjects in the sample were cognizant of Oriental per- sons per se. Although a significant number of subjects chose the white girl to be related to the researcher (Oriental), the reasons given by the subjects showed no in- dication that they were aware of differences between Oriental, Negro and white persons. Investigations for understanding this problem might be undertaken in view of the fact that research in this area is nil. Summarizing briefly, an analysis of the data for this study revealed that the conflicting conclusions occur- ring in previous research; namely, that preschool children placed a greater emphasis upon sex rather than race in their friendship preferences; that preschool children placed a greater emphasis upon race rather than sex in their friend- ship preferences; and that no consistant patterns are present in the friendship preferences among preschool children were not acceptable. White boys of preschool ages in this sample placed a greater emphasis upon sex rather than race in their friendship preferences, while the white girls placed a greater emphasis upon race rather than sex. BIBLIOGRAPHY 10. 11. 12. BIBLIOGRAPHY . . Abel H. and R. Sahinkaya. Emergence of sex and race friendships. Child Development. 33:939-943. 1962. . Adorno, T. W., E. Frenkel-Brunswick, D. J. Levinson and R. N. Sanford. The authoritarian personality. Boston: Beacon Press. 1950. . Allport, G. W. The nature of prejudice. Boston: Bea- con Press. 1954. . Ammons, R. B. Reactions in a projective doll-play in- terview of white males two to six years of age to differences in skin color and facial features. Journal of Genetic Psychology. 45:323-341. 1950. . Angrilli, A. F. The psycho-sexual identification of preschool boys. Journal of Genetic Psychology. 45:323-340. 1960. . Bach, G. R. Young children's play. Psychological Monographs. 59:2. 1954. . Benjamin, H. Age and sex differences in toy preferences in young children. Journal of Genetic Psychology. 41:417-429. 1932. Boyd, G. R. The levels of aspiration of white and Negro children in a non-segregated elementary school. Journal of Social Psychology. 36:191—196. 1952. . Bieliaukas, V. J. Recent advances in the psychology of masculinity and femininity. Journal of Psychology. 60:255-263. 1965. Bird, C., E. D. Monachesi and H. Burdick. Infiltration and attitudes of white and Negro parents and chil- dren. Journal of Abnormal and Social Psychology. 47:695-696. 1952. Brieland, D. and L. Nelson. Age trends in sex identifi- cation as determined by a picture test. American Psychologist. 6:309. 1951. Brim, O. B. {Family structure and sex role learning by children. Sociometry. 21:1-16. 1956. 95 96 13. Brown, D. G. Masculinity and femininity development in young children. Journal of Consulting Psycho- logy. 21:197-202, 1957. 14. . Sex role preference in children. Psychologi— cal Reports. ll:477-478. 1962. 15. . Sex role preference in young children. Psy- cHoIogical Monographs. 70:5. 1960. 16. Callman, R. C. Factors influencing friendship among preschool children. Child Development. 3:146-158 1932. 17. Clark, K. B. Prejudice and your child. Boston: Bea- con Press. 1955. 18. Clark, K. B. and M. P. Clark. Emotional factors in racial identification and preference in Negro chil- dren. Journal of Negro Education. 19:341-351. 1950. 19. . The development of consciousness of the sex and the emergence of racial identification in Negro preschool children. Journal of Social Psychology. 10:591-599. 1939. 20. . Racial identification and preference in Negro children. In T. M. Newcomb and E. L. Hartley, ed. Readings in Social Psychology. New York: Henry H01. 1947. 21. . Segregation as a factor in racial identifi- cat1on in Negro preschool children. Journal of Experimental Education. 8:161-163. 1939. 22. . Skin color as a factor in racial identifica- t1on in Negro preschool children. Journal of Social Psychology. 8:150-169. 1940. 23. Cobb, H. W. Role wishes and general wishes of children and adolescents. Child Development. 25:161-171. 1954. 24. Conn, H. Children's awareness of sex differences. Jour- nal of Child Psychiatry.l:3-57. 1947. 25. .' Children's awareness of sex differences (II). Journal of Child Psychiatry. 2:82-99. 1951. 26. Criswell, J. H. A sociometric study of race cleavage in the classroom. Archives of Psychology. 10:591-599 1939. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 97 Dawe, H. C. An analysis of 200 quarrels of preschool children. Child DevelOpment, 5:139-157, 1934. Deutsch, M. Minority group and class status as related to social and personality factors in scholastic achievement. Monograph No. 2, Society for Applied Anthropology. New York. 1960. Dillon,b4. Attitudes of children toward their own bodies and those of other children. Child Development. 5:165-176. 1934. Emmerich, W. Young child's discrimination of parent and child roles. Child Development. 30:403-419. 1959. Fauls, L. B. and W. D. Smith. Sex role learning of five year olds. Journal of Genetic Psychology. 89:105- ll7. 1956. Frenkel-Brunswick, E. A study of prejudice in children. Human Relations. 1:295-306. 1948. Frenkel-Brunswick, E. and J. Havel. Prejudice in the interviews of children: I, attitudes toward minority groups. Journal of Genetic Psychology. 82:135-140. 1953. Goodenough, E. W. Interest in persons as an aspect of sex difference in the early years.~ Genetic Psycho- logy Monographs. 55:287-323. 1957. Goodman, M. E. Evidence concerning the genesis of in- terracial attitudes. American Anthropology. 48: 624-630. 1946. Race awareness in young children. New York: CoIIier Books. 1964. Greagor, A. J. and D. J. McPherson. Racial attitudes among white and Negro children in a deep-south standard metropolitan area. Journal of Social Psychology. 68:95-106. 1966. . Racial preference and ego identity among wH1te and Bantu children in the Republic of South Africa. Genetic Psychology Monographs. 73:217- 253- 1966. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 98 Hall, M. Sex role preference among children of upper and lower classes. Unpublished Master's thesis. California: Clarmont College. 1960. Handy, G. D. The sex role preference of children. Un- published Master's thesis. Colorado: University of Denver. 1954. ' Hartley, E. L., M. Rosenbaum and S. Schwartz. Children's perceptions of ethnic group membership. Journal of Psychology. 26:387-398. 1948. Hartup, W. W. Some correlates of parental imitation of young children. Child Development, 33:85,96. 1963. Hartup, W. W., S. Moore and G. Sager. Avoidance of in- appropriate sex typing by young children. Journal of Consulting Psychology. 27:467-473. 1963. Hartup, W. W. and E. A. Zook. Sex role preference in three and four year old children. Journal of Con- sulting Psychology. 24:424-426. 1960. Hattwick, L. A. Sex differences in the behavior of nursery school children. Child Development. 8:343- 355. 1937. Helgerson, E. The relative significance of race, sex and facial expression in choice of playmates by preschool children. Journal of Negro Education. 12:616-622. 1943. Hetherington, E. M. A develOpmental study of the effects of the dominant parent on sex role preference, identification and imitation in children. Journal of Personality and Social Psychology. 2:188-194. 1965. Horowitz, R. Racial aspects of self-identification in nursenfschool children. Journal of Psychology. 7: 91-99. 1939. Jersild,A. T. and F. V. Markey. Conflicts between pre- school children. Child Development Monographs.21: 1-139. 1935. Kagan, Jerome. Aquisition and significance of sex typ- ing and sex identity. In L. W. Hoffman and M. L. Hoffman, ed. Review of Child Development Research. pp. 137-167. New York: Russell Sage Foundation. 1964. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 99 Katcher, A. The discrimination of sex differences in young children. Journal of Genetic Psychology- 87:l3l—l4l. 1954. Koch, H. L. A study of some factors conditioning social distance between sexes. Journal of Social Psychology. 21:79-107. 1944. . POpularity in preschool children. Child De- veIopment. 4:164-175. 1933. Kutner, B. J. Patterns of mental functioning associated with prejudice in children. Unpublished Ph. D. dis- sertation. Massachusetts: Harvard College. 1950. Lansky, L. and G. McKay. Sex role preferences of kin- dergarten boys and girls. Psychological Reports- 13:415-421. 1963. Lambert, W. E. and Y. Taguchi. Ethnic cleavage in young children. Journal of Abnormal and Social Psychology. 53:380-382. 1956. Landreth, C. and B. C. Johnson. Young children's re- sponses to a picture and inset test design to re- veal reactions to persons of different skin color. Child Development. 24:63-79. 1953. Lasker, B. Race attitudes in children. New York: Henry Holt Company, Inc. 1929. Lefkowitz, M. M. Some relationships between sex role preference of children and other parent-child variables. Psychological Reports. 24:420-427. 1960. Lippitt, R. Popularity among preschool children. Child Development. 12:305-332. 1941. Lowe, W. P. Children's sex role preference with the It figure. Unpublished Master's thesis. Colorado: University of Denver. 1957. Lynn, D. B. A note on sex differences in the develop- ment of masculine and feminine identification. Psychological Review. 66:126-135. 1967. Mamiya, Takeshi, A study of sex differences; II: its systemization and the study of children's con- sciousness of sex differences. Japanese Journal of Educational Psychology. 6:205-216. 1959. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 100 McCandless, B. P. Peer popularity and dependence on adults in preschool age socialization. Child Development. 32:511-518. 1961. McCandless, B. R. and H. M. Hoyt. Sex, ethnicity and play preference of nursery school children. Journal of Abnormal and Social Psychology. 62:683-685. 1961. McDowell, M. Frequency of choice of play materials by preschool children. Child Development. 8:305-310. 1939. Miller, P. An eXperimental study of the development of sex roles. In G. Seward, ed. Sex and the Social Order. New York: McGraw-Hill. 1946. Moreno, H. L. Who shall survive? Boston: Beacon Press. 1934. Morland, J. Racial preference of nursery school chil- dren in Lynchburg, Virginia. Social Forces. 37: 132-137. 1958. Racial preference of nursery school children 1n a southern city. Merrill-Palmer Quarterly. 8:271-280. 1962. Mussen, P. H. Differences between the TAT responses of Negro and white boys. Journal of Consulting Psychology. 17:373-376. 1953. Muste, M. and D. Sharpe. Some influential factors in the determination of aggressive behavior in pre- school children. Child Development. 13:11-28. 1947. Pinter, M. H. and R. R. Sears. Sex differences in pro- jective doll play of preschool children. Journal of Psychology. 21:73-80. 1946. Proshansky, H. M. The develOpment of intergroup atti- tudes. In L. W. Hoffman and M. L. Hoffman, ed. Review of Child DevelOpment Research. pp. 311-371. New York: Russell Sage Foundation. 1966. Rabban, M. Sex role identification in young children in two diverse social grOUps. Genetic Psychology Monographs. 42:81-185. 1950. Radke, M., H. G. Trager and H. Davis. Children's per- ceptions of social roles in children. Genetic Psychology Monographs. 40:327-477. 1949. 102 89. Vaughn, G. Ethnic awareness in relation to monority group membership. Journal of Genetic Psychology. 105:119-130. 19641 90. Walters, J., D. Pearce and L. Dahms. Affectional and aggressive behavior of preschool children. Child Development” 28:15-26. 1957. APPENDIX 104 ”mam Emma no>flm xmmh ”930m xmmh ”mouse xmme “039 xmme Homo xmme "mhzmzzou swapsoz osmoz muse: :ofiumOMMAuaopH Honoumomom 02 m3 mahfiu mz u: msom xflm Mmme mz u: mango oz m2 msom o>Hm xmme “ammummm oz 03 nsom xmmh mz m3 mouse xmme Ammummm oz m2 039 xmmh 03 m3 Ono xmme «mummm Ammo mausauv humouom "Hanan ”umODOHm ”macho ”Monumoe ”wow "ow< "osmz onbumqqou