- I - -.x — '_ _— N » I-v .__I “A , . - > - - . . .. ..., W ‘ "" “’3 ‘r; . . . ’R,‘ -‘ ‘f .’ III" “” ° '3‘ '.. (3 " 7 {'7‘ \""' . 'l' . I ‘I -. '- .U . - . I V ' b& 1' ' ‘ .". 7".c I.“ . III. I '. I -I_ ‘ . ' "Q; 'I III‘:'II LI. ‘ ...I,II- .: “II... V'IIII ':-II:'I- . .. I. . . -» . I.. I . . I II II I I‘ I-. III ‘ I ‘ ’ .‘I ' ' Ix' . .i “MI-f." I“I III“ .'-.' ".' . . I. ' I ‘ . 3“" . w“ . .II'IA" v‘ 'I': " l ' I " ‘ 't‘ I l I '5; It“. ' I -“ ‘ II “I u . I‘ \\ ' ' "I", 1' ““‘IIn “h”. ‘v‘ ‘ I II':I' I'll” ’QHI I, II I II I‘ I‘:‘ II I. ‘ ' ' r“ -‘ I - IQ'~.' ' I I I\.I I II I ‘I II. A“. I II W; I .’ I I II * 'I'I' I ‘ .M I‘II.“.I~ I- IHI‘.‘ I'” I I'WI “ I“ “I“ I "- IIvi‘ “ "" ' w ‘ "' ‘Nl ‘ I;.,"I'."- .I'AII ' ITI1II' II“ .1 '~‘. w W"? "' «'I' " 3' M 3'“ n . .‘ ‘,..g;.;l I'l'nu‘ I‘. ' IIIII‘IIIII "‘ ’IIg. ' w “ . . "I" E735" 'I "I"'I ‘ ‘ ' . "H. "“ L.I;-:‘ I -JI.‘;; .. 1.? .I. . .-.""." . " ‘ . “' "J“:I'm " "‘1‘- ‘7 ,IIMIIII. '. II .IIMI .LI .-_. q I .I III.I«-_ .I 7 .."~' ;I I ..I I. J ; II“ ,'.\ ' I‘HI "I. . f I". QI !'. Ca. "I I I 1'" h V. I "' I.“ o . ‘ ‘ 'I I ‘V‘I .. I" '| ' I' «'I I".« “Vl'v‘ll ". . l_ .| "...'.'II'.I‘1!I""“ 0‘". '. " J. """w'|nL'II' '1'”! W h" '4‘“ ""1“ . I; 573'?" “‘- “1’ . "‘Z 1 “H “II ' ' ‘1 ‘ J IQIIII (; II‘ 'I .II I II . If.” :I‘In II IoIIklIIa: II:‘ I '4‘ 'Ivb‘ 'In:III'.fl I”: II 'II'II‘ ,‘II. .‘ I" “I. l",'“‘~"U’\.“. . ', :I“ II I“ "'I“"’ N . .III IIIIIIII I V“. II‘..‘ IIIKII-Q‘ I. IliuI'QI III I": Is, '“:--a'.-“I .IQ‘IQII,I ."J-U' I'II-I: Ir‘. I‘H“ ..'.' I . .III 1’ in. I I ‘. I.» r sun-II; III M m . """Ww'tI' '1’.'"I‘IP3-7"5‘4~'I-‘I"-"' I . . I‘I“. “WP IIIIII III.” IJII I‘. .QI‘ !' “'1' I I, Inn. .\v “I; 'IIL' 1:. . 33‘ {I . "(ii-""31 '_; I- . I. '. LII-I ‘ I_I - U. - I a' A IIQII I W .Q' “III '\ ‘Q “Q. 1:1: .' VI- ;‘ ' . ' I. III; I. .. II. mi NM" " . " I I "~ 7'3? 'I ‘ '“I‘I' " “ , ”I I ‘II . I I “I - .' I'I II. . 'I.II:,I I 'II‘ IIx'I'Q;)I‘II ['1' II.II- I .1". AI . I_'I- . I§t alj <._II ' I *’ II’lIIIIIQII-IIII;}IIQIIII‘IIII:II tn“ I III‘ III p. “i.“ N?“ If?“ “av-'79.“! : {Is . "0;. . . I’I . :‘lI‘ - '3. :I r .‘ | II‘IbI' !I W“ “3,“;“15 Ida»: '“ It" . '. "z ‘ ‘J-J II' 'I - I‘ll.‘ Q'IIII :II ') II “WI“ M7 "II: “I“ 5“"? :‘l: ~ '2 . I1 MI I... III: III III“ 1.. Iv ,. I: I; . I W I III “Half-”kl "I ‘l‘n 6" ‘ I I I '“I'QJ 'IHI' II M“ “II"II‘I This is to certify that the dissertation entitled Common Trai ts of Emotionall y Heal thy Children : The Merging of Self-Esteem, Sex-Role Orientation, and Internal-External Locus of Control presented by Edward C. Cook has been accepted towards fulfillment of the requirements for degree in Urban Counseliaiig Ph.D. November 5, 1984 I)ate MSU is an Affirmuriw Action/Equal Opportunity Institution 0-12771 _._l AA~ “.ni MSU BEIURNING MATERIALS: Place in book drop to LJBRAfiJES remove this checkout from 1—9—- your record. FINES will be charged if book is returned after the date stamped below. .5: w «l- ‘ Re? i‘I-J 115' M P631 - ”if‘t} i fififiifrtmflil.’r ‘ l)", 'J 1,2 / 94am " 1‘ O I $“¢‘ ‘V y l‘. -*:~ 14-92,:- 1 2 .. ‘ f if 8:" J I " "‘V‘hmi-“S'TI: m: 1 6 wt! ,‘_lhi OCT J o {990‘ ” l MB A 1:; COMMON TRAITS OF EMOTIONALLY HEALTHY CHILDREN: THE MERGING OF SELF-ESTEEM, SEX-ROLE ORIENTATION, AND INTERNAL-EXTERNAL LOCUS OF CONTROL BY Edward C. Cook A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Counseling, Educational Psychology and Special Education 1985 ©1986 EDWARD C . COOK» ‘ All Rights Reserved ABSTRACT COMMON TRAITS OF EMOTIONALLY HEALTHY CHILDREN: THE MERGING OF SELF-ESTEEM, SEX-ROLE ORIENTATION, AND INTERNAL-EXTERNAL LOCUS OF CONTROL BY Edward C. Cook Research on self-esteem, sex role orientation, and internal-external locus of control has typically been analyzed among adult subjects with each trait being scrutinized separately or paired td’one other trait. This study combined the analysis of these three areas and investigated their relationship among a middle school population since this has not yet been attempted empiri- cally. Two hundred children, 50 each of male and female students, from sixth and eighth grade, completed the Who Am I and children's versions of the Ben Sex-Role Inventory and Rotter Internal-External Locus of Control Scale. Results showed that children with androgynous and internal locus of control perceptions were also those with greater self- esteem than other children as demonstrated by them writing a greater number of positive statements about themselves Edward C. Cook themselves, fewer statements regarding their appearance and possessions, and more statements referencing their ties to others such as friends and classmates. Implications are discussed, e.g., those for parents and several professional groups; as well as suggestions for further explorations of the utility of the Who Am I. V: 31 ACKNOWLEDGEMENTS For those of you who are reading or skimming this section because of your compulsivity or desire to gain an ounce of wisdom, please take a hint and skip to the last sections of the thesis - where the fluff is most evident - it will do you some good to practice a bit of impulsivity or callousness. After all, this section is written largely for the sake (and the humor) of individuals who would be pleased if this volume contained cookie recipes or travelogue scripts. Actually it does not contain these, but you probably guessed that. Back to the point then, the majority of those people (supporting this effort) are proud of the fact that this volume has at long last made its appearance. Not that its completion was ever in doubt understand, but it has simply taken a few years and a few dollars to bring it to fruition. And, since you have read this far down the page, you may as well know that the tediousness of completing this work has rendered especially valuable the encouragement and support of what has become a sizable cast. ii The principal actors in this scenario are (as you likely guessed) the family of the author. These are: William and Mary Ellen Cook (parents); Pollianne Cook: Bill, Pam, Paula, and Andrea Cook: Gary, Leah (Say Eah?), Jared, and Aaron Johnson; and Virgil P. Hogwash (modest tailwagger that he is). It has always been a point of amazement that these family members have provided a vocal and continual cheering section during these many years of study. Certainly God will continue to bless such loving and giving people. There are many others to pay tribute to before you reach the sleep-inducing text, so do not hurry away if you have gone this far. There is plenty of time for sleep 1:33:- On with it then. A heartfelt thanks to Belinda Castiglione, who provided a free and urgent typing service when deadlines were live monsters that gobbled sleep in huge chunks. Also recalled is her disbelief, expressed late in the summer of 1984 that the thesis actually existed. What nerve! A vital need was filled by Carl and Pat Wren - who opened their home and refrigerator to a near-starving (looked the part anyway) grad student. Again, Carl, what better humanitarian thought can be had beside the one that iii a man is hardworking and is concerned for his fellow man? And your son fits that description. In human terms, that smells like honor and esteem; if not success. May your family have peace and security always. To the CATS (Correctional Assessment and Treatment Services) staff a cheery THANKS(!) for the patience (and tolerance of mood swings) and understanding you provided. It was a blessing to me to be able to practice my increasing skills on your clients. My wish is that you and they benefited. For you and the probation departments (that includes 299, Eldine, Mike, Kathy, Dave, Pam, Wilbur, Gene, Cheryl, Roxane, Sue, Paul, and John) for furnishing referral questions which challenged my skills. Your professional manners and avid interest in my assessments (and recommendations) made working with you easy and fun. May I become even more name conscious? Before you protest at the length of this section I exercise my right to write anyway or is that to write right? Or write to wright? They are just words anyway. Symbols that forever chase but never capture the entire thought. We had best continue. Three note-worthy CATS staffers are special to this effort: Bonnie Copher (I should have asked you to help. Not doing so lost more time and money than I care to think about). iv The Pat Relleys with Gale and girls are to be revered. I always felt sheepish that I seemed to forever be on the receiving (versus giving) end of our interactions. I hope to one day give back, with interest, your overabundance of kindness. I respect you immensely and cherish the knowledge and surety of our friendship. Still, you Pat, are one of the most conscientious people I have ever met. On the lighter side, bless you Mike, Theresa, Katie, and David Steely. Your company, friendship, and food (I) were savored long past their immediate utilization. Oral fixation seems to get its urge expressed from time to time. Freud might have had a valid point there. . . I almost forgot but thanks Shelly Villeral for your generous time in finding things I only hoped existed. I have not forgotten that you expressed an interest in getting together for lunch sometime. If only this degree would put money in this old, thin (and leaking) wallet. A wish goes to Mike Pierce, with a thanks, that he find someone who would be willing to follow him constantly and write down his wonderful creative ideas as they leap from his lips. You have lost more workable ideas than I am likely to have if I live to see 75, Mike. We really should write a book, or two. Or maybe a 3 - volume set. The topic is immaterial. A sigh of relief escapes me whenever I think of you, Dave Harley. It was your brainstorming that initiated and kept rolling my attempts to analyze the mass of data that threatened to win a stare down with me. May your ambitions be fulfilled. A very true thank you for Dr. Gloria Smith who guided me through the maze labelled academia and yet also provided realistic appraisals of my words. Idealism and pragmatism: what strongly wonderful characteristics you have Dr. Smith. Further thanks is due to the Guidance Committee members: Dr. John Schweitzer, Dr. Lee June; and Dr. Gary Stollak. My association with Dr. Stollak has been particularly enlightening and challenging. You caused me to think many important thoughts that may have escaped my notice. And for Dr. Schweitzer how can I thank you for the incredibly long hours you spent in personally rethinking and critiquing the mess I had, transforming it into coherent data and finally expressable results? I only wish that Dr. Beverly Parker could have remained officially on the Guidance Committee. Her genuine personal interest in students and their growth is, to my thinking, the very essence of education. My answer for the educational woes of any age is to have (recruit and maintain on staffs) educators like you. vi Bob Sherman was, and always will be, one of those whose greatness is only known in times of great need. Bob, your service met my great need. May God increase your blessings for your amazing desire to serve others. Thanks also to your wife Debbie, for letting you develop those great computer programs and instruct me in their use. Thanks to the families of Christ's Church at Ballard's Corners and University Christian Church, who live the desire of Christ and His kingdom in their fervent loving and sharing of the life. What can be said of one who is not a brother but lives the part? Dave Arend (aka SLID bother), if you ever need life please take mine, as it has long been pledged to you anyway. The motivation and desire to become a psychologist came from you - Dr. Shelly and this work is a tribute to a fond memory of you. ' Also to a family - Lloyd, Mark and Cyndi and kids - I love you deeply and pray often that God would smile on you and hold you close to Him, and I see the results of His life in you already. For all of you, I cringe at being so wordy. I am keenly aware that when all is said (and especially when all is done) the verbal skills are the least effective. May I be, as you have been to me, your servant. vii TABLE OF Acknowledgements. Contents. . . . . List of Tables. . List of Figures . CHAMER I O O O O O 0 Introduction and Overview . Need for the Study. . CONTENTS Historical Research Antecedents. The Nature of Research with Children. Measurement Techniques. Generalization of Adult Results . Summary of Problems and Issues. Purpose of the Study. Technical Rationale. . Practical Rationale and Theoret Orientation. . Glossary of Terms . . Use of Terms and Assumptions. Hypotheses. . . . . . CHAPTER II. . . . . . . Survey of the Literature. Locus of Control Studies. Overview. . . . . . . Developmental Trends. Locus of Control and Self-Esteem. Locus of Control and Sex-Role Orientat Sex-Role Orientation Studies. The Development of Sex-Role Orientation Sex-Role and Self-Esteem or Positive Adjustment. . . . i Developmental Trends of Self-Esteem . Self-Esteem and Sex-Role Orientation. Summary of the Literature Review. CHAPTER III . . . . . . . Methodology and Procedures. Introduction. . . . Restatement of the Purpose. Restatement of Rationale and Ant Hypotheses. . . . . Analytical Procedures The Population . . . The Sample. . . . . . The Measures. . . . Limitations of the Study. viii ceden eeeeueeeeeeeeee io e e e e e e "e e e e on Page viii XX p. F‘H Home mmmeaeHHHH 13 13 13 13 14 15 17 18 21 22 27 27 28 3O 30 30 30 30 31 32 33 33 34 40 Page CHAPTER IV. . . . . . . . . . . . . . . . . 41 Results of the Study. . . . . . . . . . . 41 Classifications and Groupings. . . . . . 41 Hypotheses . . . . . . 44 Additional Analysis and Significant Findings . . . . . . . . . . . . . . . . 48 CHAPTER V . . . . . . . . . . . . . . . . . . . . 58 Introduction. . . . . . . . . . . . . . . . . 58 The Findings . . . . . . . . . . . . . 59 Additional Findings . . . . . . . . . . 63 Summary of Findings. . . . . . . . . . 67 Implications and Suggestions . . . . . . 69 REFERENCES. . . . . . . . . . . . . . . . . . . . 76 APPENDICES. . . . . . . . . . . . . . . . . . . . 84 ix LIST OF TABLES Table Page 3.1 Means, Medians, and Standard Deviations of Dichotomized Variables. . . . . . . . . , 36 4.1 Sex-Role Self-Concept Group. . . . . . . . 42 4.2 Sex-Role Self-Concept Group, Masculinity and Femininity Only. . . . . . . . . . . . 43 4.3 Mean Scores of Androgynous and Other Sex- _ Role Orientations on Locus of Control. . . 45 4.4 Means of Dependent Variables for Androgynous and Internal Versus Other Children 0 C O O I O O I O O O O O O O O O 46 Figure 4.1 4.2 LIST OF FIGURES The Interaction of Sex and Femininity The Interaction of Masculinity and Femininity O O O O O O O O O O O O O Anova Analysis Design: Dependent and Independent Variables. . . . . . . . Interactions of Sex-Role and Locus of Control on Mean Number of Responses. xi Page 50 51 53 54 CHAPTER I not a ew Although there is a voluminous literature dealing with the separate topics of internal-external locus of control, sex-role orientation, and self-esteem, very little work has demonstrated a correlation among these areas. This study explores a region even fewer researchers have entered: the correlations of these areas for children. e tu I. Historical Research Antecedents The three theoretical models which are the focus of this dissertation, internal-external locus of control (hereafter "locus of control"), sex-role orientation, and self-esteem, were initially studied among adult test subjects. The theories' historical backgrounds will now be examined by discussing them in the order of locus of control, sex-role orientation, and self-esteem. Rotter's (1966) social learning theory, of which locus of control was a component, was conceptualized as a model dealing with adults' tensions and anxieties. This is also 2 the focus of more recent treatises. A very extensive and detailed example is Novaco's (1979) article "The Cognitive Regulation of Anger and Stress," in which he reviewed locus of control studies. It was during the 1970s that researchers used this concept to design experiments teaching children that they were in control (an internal locus of control perception) of their circumstances (Kendall and Finch, 1979). Since children's research has a newer history than that of adults in regard to self-control theories, fewer data have been generated and results are thus less conclusive than for adults. In fact, a major portion of the self-control studies of children have dealt only with academic and social adjustment (Meichenbaum and Asarnow, 1979: Kendall and Finch, 1979). Sex-role orientation has perhaps an even more entrenched history as a model exclusively for adults. Several factors appear to account for this occurrence. Bem (1972: 1974: 1975: 1976), a major pioneer of sex-role research, choose adult subjects who were typically college students. Since sex-role orientations and behaviors are so very intriguing, Bem's followers sought to refute or replicate her results with adult subjects also. The con- cept of sex-role orientation, being new and different, was not universally accepted, and not surprisingly the research 3 after Bem's studies has given conflicting conclusions (Gilbert, 1981: Gonzalez and Williams, 1981: Mennigerode, 1976: Whitley, 1983). Since the initial research focused on adult subjects and the results were contested, researchers did not begin to address questions of children's sex-role orientations for some time. Although work has been conducted specifically on sex-role orientation, it seemed to follow social learning theories such as Rotter's (1966) or Bandura's (1977), that is, it frequently examined the sex role or influence of a parent or other role model as it affected the child (see, for example, Lamke, Bell, and Murphy, 1980). These experiments less often examined sex-role orientations in their component parts, as did Ban (1972: 1974: 1975: 1976). The research history of self-esteem is not dissimilar to that of locus of control and sex roles. Given the interest of social scientists in self-esteem as a general concept, more has been published on it than on the other models with which this paper is concerned. Even so, work has focused more on adults than on children. Given the great difficulty of quantifying what an individual believes of him/herself, it is equally difficult, if not more so, to determine whether the view of self is more positive than negative or vice versa (Gordon, 1968). Furthermore, researchers have tended to confine their interest to adult populations, since research with children is in itself difficult and nebulous. II. The Nature of Research with Children As the preceding comments suggest, much research focusing on children is conflicting and inconclusive. One explanation is that as children grow they change, and so do their test scores and other research data they produce. Perhaps the only consistent characteristic of children is continual change. Researchers have this problem to contend with, above and beyond any hypothesis to be investigated. III. Measurement Techniques Although more will be said in later chapters about the measurement techniques employed in this study, method- logical issues are central to the history of the theories and consequent results of interest here. Just as the theoretical models have applied to adults, so, too, have the measures used to investigate the models. Researchers who have wanted to apply the three theories to children have altered the adult scales and forms so that children could follow directions and items without diffi- culty. Since these children's versions have been in use for some time, it is to be inferred that the modifications were successful. IV. Generalization of Adult Results As was noted earlier, research on locus of control, sex-role orientation, and self-esteem has focused on 5 adults. Furthermore, application of these rubrics to children is comparatively less developed than the work on adult subjects. Understandably, the conclusions drawn from adult populations have, with few modifications, been generalized to children. This is not without problems. The changing nature of a child's personality makes aay generalizations difficult, especially those based on older persons. There is considerable uncertainty about what tendencies develop in children, much less how they develop (see, for example, Gecas, Calonico, and Thomas, 1974). In addition, the result of research with children (especially when compared to that of adults) is itself less conclusive, which compounds the generalization problem. Another important point, examined in Chapter II, is that most studies to date which examine these models simultaneously investigate only two of them at once. Researchers have been hesitant to examine all three theories simultaneously. .§EEméI!_Q£_EI991§E§_QDQ_I§§E§§ The three areas of research investigated here have a history and concomitant literature focusing on adults. Consequently, measures and methodology have been tailored to that population. Results and conclusions have been generalized to children, but at least two problems emerge: (1) Children continually grow and change, whereas adults are much more static: and (2) children require measures, 6 techniques, and scales suited to their level of functioning, comprehension, and interest. Research is not abundant which combines these models, especially which synthesizes them into a single paradigm. As will now be discussed, such an approach may have merit. W I. Technical Rationale Despite the rather bleak prospects of drawing conclusive evidence from a study conducted with children using the three models, that is the objective of the present research. Although numerous experiments have shown correlations between two of the three models, this study seeks to combine all three and demonstrate positive correlations among a middle-school population. II. Practical Rationale and Theoretical Orientation The linking of these models has far-reaching consequences at home and at school. It relates to the manner in which children are reared and what they are expected to become and hence has ramifications for child-rearing techniques and children's mental health. For example, although parents may desire a particular sex-role orientation or internal locus of control for their child- ren, these become even more important and powerful when they are conceputalized in terms of positive self-esteem. Expressed another way, most parents strive to guide 7 children to healthy self-esteem, yet the concept is as vague as are the means to achieve it. Discovering what methods of teaching (training) children will be an aid along these lines is extremely important. Defining a workable construct of self-esteem in terms of sex-role orientation and locus of control yields at least two benefits: the potential to increase a child's self-esteem and concepts that are more easily understood and thus more easily manipulated. For example, extant research suggests that those with androgynous sex-role orientations exist (Bem, Martyna, and watson, 1976): that they have more adaptable personalities than do those with other sex-role orientations (Bem, 1975): that androgynous children are relatively.more apt to succeed in social and academic situations (Sprafkin, Serbin, and Elman, 1982): that androgynous individuals have a high degree of positive self-esteem (Flaherty and Dusek, 1980). Such evidence will be discussed later, but here it is sufficient to note that positive correlations among the three constructs in total represents positive adjustment of children. A finding of this sort means an increased likelihood that parents, teachers, and others concerned with the welfare of children would want to develop and achieve means of rearing them with these positive characteristics. As conceptualized here, rather than attempt to increase positive self-esteem, it would be easier for parents, 8 teachers, and child-care-givers to know the correlates which can positively increase self-esteem, namely, locus of control and sex-role orientation. As scientists have tended to do in recent years, this research applies psychological (measurement) techniques to the classroom setting. This study also looks at a younger age than would most paradigms, given the variables involved. Yet, in so doing, it goes beyond mere academic and social achievement or adjustment and aims at drawing conclusions about how to orient children to school, society, self, and life. It is widely accepted that a healthy self-esteem means more initiative, greater achievement, and greater satisfaction with life. In the author's major area of interest, urban ocounseling, it is all too plain that children from low socioeconomic status groups, the minorities, the under- achievers, those with external locus of control, and those with rigid roles and orientations suffer considerably. In drawing a tighter parameter around self-esteem, this research may indicate how to guide the development of children most in need of improvements in how they see themselves. Modifying sex-role orientation or locus of control may be a more effective means of bolstering a child's self-esteem than has been recognized to date. This study will draw some conclusions in this regard. ELQ§§§I1_QI_I§IEE Specific words and terms are used in this paper. They are defined here in order of theoretical overlay, that is, each builds on the understanding of the previous term. Examples will be offered when clarification is necessary. Self-esteem - The relative level or amount of regard one holds of oneself, whether positive or negative. Note that some researchers use self-concept and self-esteem interchangeably. Except when noted, that practice is followed here. Global self-esteem - An individual's overall, comprehensive self-esteem. Component or aspect of self-esteem - The positive or nega- tive valence one holds of self in regard to particular activities, situations, or tasks. Example: academic self-esteem is the self-valence one has for academic situations and tasks. Masculine - The traditional behaviors and attitudes associated with the male gender. These include (but are not limited to): achievement, leadership, dominance, authoritativeness, aggressiveness, and risk taking. Feminine - The traditional behaviors and attitudes associated with the female gender. These include (but not limited to): congeniality, nurturance, expressive- ness, responsiveness, passiveness, and inhibitedness. 10 Androgynous - A combination of both masculine and feminine characteristics. Undifferentiated - A lack of orientation to (or valuing of) either masculine or feminine characteristics. Sex-role orientation - A person's behaviors and attitudes in terms of available masculine/feminine combinations, that is, (1) masculine: (2) feminine: (3) androgynous: (4) undifferentiated. Internal-external locus of control (or locus of control) - A person's perception of his or her interactions with the surroundings, that is, whether a person believes he/she controls (internal locus) or is controlled by (external locus) the environment. WW Based on a general consensus among available studies, the following assumptions are considered positive and are valued, whereas their opposites are less desirable. These theoretical assumptions involve androgyny, internal locus of control, and global self-esteem. Androgyny has been posited as a beneficial sex-role orientation. Given the preponderance of evidence, it is accepted in this study as being desirable and as contributing to positive self-esteem. Similarly, the majority of studies cited here view internal locus of control as related or contributing to positive self-esteem. It is so valued in this research. 11 There is less evidence supporting global self-esteem as a beneficial personality characteristic, but it has been described by several researchers in such terms as the sum of all positive views a person holds of him/herself. It is also an intuitive construct, possessing face validity within its very definition. Consequently, for our purposes it is assumed valuable. Three other assumptions relate to the procedures and methods employed in gathering and categorizing data. The first is that all children in this study understood all instructions and the purpose of the measures they completed. The second is that the children completed the measures and scales to the best of their ability. The third is that all fourteen undergraduate volunteer coders attempted consistently and accurately to rate and classify subject's responses on the measures according to the trainer's directions. nastiness H1: Those children who score above the mean on both masculinity and femininity will be those who are androgynous by definition and will also be those who have a greater degree of internal locus of control. H2: The children with androgyny and internal locus of control will be those with the greatest self-esteem as measured by the Who Am I 12 instrument. Specifically, these children will have higher scores in terms of: a) b) C) d) e) f) More positive statements: More mention of Abstract Identifications: More statements of Interests and Activities: Fewer statements of Material References: More mention of Personal Characteristics: Fewer negative statements. Points a through f were found by those who studied the Who Am,I instrument to be the greatest indicators of positive self-esteem (see Gordon, 1968, 1969: Kuhn, 1960: Kuhn and McPartland, 1954: McLaughlin, 1966). CHAPTER II Sazyay of tag Litagature The purpose of this chapter is to examine research on the three theoretical models with which this study is concerned: locus of control, sex-role orientation, and self-esteem. The discussion format is as follows: an overview and explanation of the theory: developmental trends: and the relationship of each theory to the others. A summary details the significant facets of each model and its salience to the others. Wise Overview Locus of control (LOC) refers to an extensively researched personality variable posited by Rotter's (1966) social learning theory. It describes the relationship between the individual and the environment and has two dimensions, internal and external. Those with internal LOC believe they manipulate their surroundings: those with external LOC perceive the environment (or luck, circum- stances, and so forth) as determining their situation. In a very complete review, Novaco (1979) termed LOC a trait of 14 "expectancy for personal efficacy." That is, control over an event can be understood as the subjective probability of exerting or obtaining reinforcement in the face of an environmental demand. In developing this theory, Rotter (1954) proposed the intermediate concept of behavioral potential as a function of expectancy and of reinforcement value. Rotter's version of LOC is a form of expectation, specifically, a generalized expectation for control over reinforcement. Internal control refers to the perception or expectation that positive and/or negative events are a consequence of one's own behavior and are thus under personal control. External control refers to the perception or expectation that events are independent of one's own behavior and are therefore beyond personal control (Lefcourt, 1966). - DQ!§122E§D§QI.II§DQ§ Researchers have not been particularly interested in the conceptual development of LOC theory as related to either children or older populations. In his review of "control" theories, Novaco (1979) lists 209 references, none of them being a study of the development of LOC. Thus, while much is made of the components of LOC, its characteristics, and correlated tendencies, very little is reported on stages or trends in the unfolding of an individuals' LOC perceptions. 15 A study by wu (1982) was designed to compare age and sex groups and did suggest that in their school years males seemed to become more internally oriented, where as females appeared to become more externally oriented. He also noted that junior high females were more internally oriented than were their senior high counterparts. L92Qfl_9I_QQDEIQL_ED§_§QLI:E§§§£E Since internality is assumed to be a desired personality characteristic it should be linked to positive or high self-esteem in both a practical and theoretical sense. A number of studies have linked LOC and positive judgment. Nowicki and Schneewind (1982) found evidence that internals show greater family cohesion and expressiveness, coupled with lower conflict, than do externals. In a similar study, Sadowski and Wenzel (1982) claimed that externals express more hostility than do internals. It is expected that an internally oriented person with high self-esteem would be more in harmony with self than are others. Adjustment and LOC studies have not been confined to interpersonal and family relationships. In research clearly related to self-control and efficacy, Rost, Neuhaus, and Florin (1982) discovered that bulimorexics had significantly higher external and fatalistic self- perception than did comparison subjects. In a more general study, Burke (1982) concluded that individuals with 16 internal LOC possessed greater self-esteem, were more active, and tended to be more satisfied in a helping role. In an educational setting, Bhagat and Chassie (1978) also found that undergraduates with an internal orientation performed better than others. In 1983 Maqsud discovered correlations showing that Nigerian students with internal LOC were those with higher 10s and self-esteem. In a very interesting study by Hanes, Prawat, and Grissom (1979), individuals were asked to rate their own and the other sex's LOC. It was revealed that each sex perceived itself as more internally oriented than the other. Research by Topol and Reznikoff (1982) examining the LOC of a younger population concluded that adolescents who attempted suicide were significantly more external than were controls. Among a related population, Cole and Kumchy (1981) found that juvenile delinquents with low self-esteem were external. A few studies linking LOC to the self-esteem of children have been published. Stipek (1981) found that first graders, on the average, had a more internal LOC when they were optimistic and motivated. Kanoy and Johnson (1980) concluded that fourth-grade achievers had greater self-concept and more internal LOC compared to external LOC and lower self-concept for underachievers. Using even more ‘variables, Kuo (1982) matched teacher expectations, children's motivation, self-esteem, and LOC. He found that 17 among elementary students for whom teachers held high expectations, internally controlled children had greater self-esteem and achievement motivation than did externally controlled students. In the teachers' low expectation group, the internal LOCs scored higher on self-esteem than did external LOCs, but there was no significant difference between them on achievement motivation. Walden and Ramey (1983) studied preschool intervention programs and stated that high-risk intervention and low-risk children had stronger beliefs in personal control over academic success, and that those beliefs were good predictors of achievements and task-related classroom behavior. For high-risk nonintervention children, only IQ was related to achievement, while IQ was not related to achievement in intervention children. W A few researchers have found evidence that particular sex-role orientation configurations are aligned with internal or external LOC perceptions. Research conducted by Mennigerode (1976) and re- analyzed by Gonzalez and Williams (1981) concluded that highly feminine women (the female sex-role stereotype) had an external LOC, while an internal LOC was congruent with the male sex-role stereotype (highly masculine). Substantiating that finding, Brehony and Geller (1981) found androgynous females to be reliably more internal than 18 stereotypic females. Burke (1982) found that nursing staff with greater self-esteem were also those of internal LOC, and that they were more active and satisfied. Individuals who described themselves as emotionally cold, abrasive, ineffectual, timid, not perceptive, and lacking self- confidence were less active, less satisfied, and external in LOC. §2x:BQlfi.QIi§nEAEiQD_§£Q§i§§ A person's sex-role orientation (SRO) has assumed greater importance in recent years than perhaps at any other time in modern culture, due largely to changes in societal values and the individual's functions in employment, home, and society at large. While it was once thought that highly masculine men and highly feminine women were the most well adjusted individuals, evidence now points to androgynous people as setting the standard for mental health. Androgyny is generally defined as the "coexistence of masculinity and femininity in the same person" (Gilbert, 1981: Ginn, 1975). Androgyny has been used to discuss and explain such diverse behaviors as male-female interpersonal interaction (Orlofsky, 1982), and children's frequency of TV watching (Stericker and Kurdek, 1982). Androgyny was conceptualized by Bem (1972: 1974: 1976) as one of four SROs: masculine (high positive masculine, low feminine): feminine (low masculine, high positive 19 feminine): androgynous (high positive masculine, high positive feminine): and undifferentiated (low masculine, low feminine). Whitley (1983) views androgyny as a particular sex-role model (theory). He expands on Bem's work by characterizing androgyny as greater self-esteem and low depression resulting from the additive effects of a high degree of both masculinity and femininity in the personality. To delineate the composition of these four SROs, how they are derived, and with what they are associated, the following paragraphs present a brief discussion of each. Masculinity is defined as the sex role of individuals who score above the sample mean on the masculinity scale and below the sample mean on the femininity scale. This SRO is associated with assertion and aggression, achievement, dominance, and other traditional masculine descriptors and traits. Femininity is defined as the sex role of individuals who score above the sample mean on the femininity scale and below the sample mean on the masculinity scale. The femininity SRO is associated with passivity, sensitivity, empathy, and the various other stereotypic feminine jpersonality traits. Androgyny is defined as the sex role of individuals who score above the sample mean on both the masculinity and femininity scales. Androgyny is associated with both the 20 masculine and feminine traditional personality traits, descriptions, and labels. An undifferentiated sex role is defined by those individuals who score below the sample mean on both the masculinity and femininity scales. This SRO is associated with a lack of any particular polarity a; combination of the sex role behaviors and attitudes. Several studies have supported this general view of SRO with evidence that various components comprise the sex role. For example, Orlofsky (1981) wrote that sex-specific behaviors tend to be mutually exclusive, while male and female behaviors that are more positive form a dualism that is actually positively correlated. In addition, those with nontraditional sex-role attitudes or personality trait organization (specific sex type) were somewhat less conventionally sex typed in behaviors and interests than were those with traditional attitudes or sex-typed personality traits. Overall, a general independence of sex-role traits, attitudes, and behaviors was supported. Other researchers (such as Bernard, 1981: Waters, 'Waters, and Pincus, 1977) even suggested that the masculinity or femininity orientations may be broken down into singular rudimentary factors. Although they are beyond the scope of this paper, a brief mention will be ‘made of these ideas. Bernard (1981) and Antill (1980) stated that masculinity and femininity might each be regarded as consisting of several components. For Bernard 21 (1981), these fundamental aspects involved aesthetic interests, manual and physical interests, hypersensitivity, timidity and sentimentality, and temerity. In other studies, the links are so significant between androgyny, sex role in general, and various personality factors that sex roles could be viewed as particular social skills or social competencies that individuals can use to obtain reinforcement or otherwise interact with their environment (Kelly and WOrell, 1977). WWW Given the interest in sex roles and particularly in androgyny, one would expect many studies examining the development of children's SRO, but thisis not the case. Aside from articles such as Mischel's (1966), which claimed sex-role behavior is learned from an adult who is perceived to be powerful, or the claim by Lamke, Bell, and Murphy (1980) that androgyny mostly develops in females who have an older brother, very little has been written about sex-role development. Hyde and Phillis (1979) examined 289 people aged 13 to 85 and concluded that, with age, androgyny increased in men and decreased in women. Williams, Bennett, and Best (1975) attempted to measure the awareness of sex stereotypes among kindergarten children. It was found that their knowledge, ‘which developed similarly in boys and girls, increased ‘until the second grade and then leveled off. Similar 22 tendencies were revealed among college students who rated themselves on their current and earlier sex-role identification. That is, males shifted from androgyny on the ”early" measure to masculine on the "current” form: females shifted from feminine to androgynous. In comparing the development of sex-trait stereotypes among children of several English-speaking nations, Best and colleagues (1977) found a clear progression in sex stereotype learning from ages five through eight. Knowledge of female traits increased more rapidly during these ages. Many studies have demonstrated correlations between SRO and self-esteem. This review highlights the issue by discussing diverse and more conclusive articles. One study by Orlofsky (1982) instructed adults to describe ideal partners and themselves. All subjects, whether androgynous or sex typed, tended to describe complements, that is, the male was somewhat more masculine and less feminine than his partner, the female somewhat more feminine and less masculine than hers. In a follow-up study reviewed in the same article, Orlofsky (1982) found males had equal preference for sex-typed and androgynous females. All females described ideal partners as being androgynous. 23 Sex-role orientation figures in more than attractive- ness, however. Several studies have shown that adolescents and adults have higher self-esteem when they are androgynous than when they have other sex-role orientations. Studies of Spence, Helmreich, and Stapp (1975), O'Conner, Mann, and Bardwick (1978), Puglisi and Jackson (1980-1981), and Lamke (1982a) present evidence that androgynous people have the greatest self-esteem, with masculine, feminine, and undifferentiated SRO individuals following in decreasing order. Two similar studies by Moore and Rosenthal (1980) and Lamke (1982b) claimed that both androgynous and masculine people possess higher levels of self-esteem compared to feminine and undifferentiated SRO individuals. Taking these results as a whole, especially those by Spence, Hemreich, and Stapp (1975), Puglisi and Jackson (1980-1981), Moore and Rosenthal (1980), and Lamke (1982b), it could be concluded that androgyny is a more significant contributor to a female's self-esteem than to a male's, since masculinity is a greater predictor of a female's self-esteem than femininity is for a male's. Findings contrary to the articles reviewed above have been published. Most notably, Logan and Kaschak (1980), Richter, and colleagues (1982), and Watkins (1982) concluded that self-esteem was not greater for androgynous individuals or any other particular SRO group. Lubinski, Tellegen, and Butcher (1983) found only partial support for 24 masculinity and femininity as predictors of self-esteem. Baumrind (1982) investigated androgynous parents, expecting them to have well-adjusted children showing assertion, responsibility, and competence, yet she found that the children of sex-typed parents were the most well adjusted and competent. Baumrind also stated that androgynous women do not differ behaviorally from other women in child- rearing practices. Androgynous men are more like androgynous women than like masculine men in their child-rearing methods. These men are unconventional and autonomous in their personal lives. Sex-typed mothers were found to be responsive, and sex-typed fathers were firm. The androgynous parents were not likely to be authoritative but were child-centered. In conclusion, Baumrind said androgyny may not be a model of ideal adjustment or child rearing. Other researchers have concluded that androgyny is a desirable trait. In fact, as noted earlier (Orlofsky, 1982) androgynous individuals may be chosen as a companion or partner over a sex—typed individual. In a humorous study, Bridges (1981) found that while individuals rated a sex-typed person as more attractive than an androgynous one, both androgynous and sex-typed females liked an androgynous male significantly more than a masculine one. 1Males liked androgynous and sex-typed females equally well. Many studies support the claim that androgyny is integral or contributes to an individual's self-esteem. 25 Halpin and Whiddon (1980) discovered correlations between supposed feminine characteristics (instrumental companionship, nurturance, and so forth) and high self-esteem for both sexes. Other indicators of positive adjustment have been found within androgynous people. Heilbrun (1981) found androgynous females to be more competent in judging than were other females, while androgynous males demonstrated the best social cognition of those tested. DeGregorio and Carver (1980) found that females lacking a masculine orientation (males had less of this tendency) were especially prone to low self-esteem, high anxiety, and depression. This result also emerged from Whitley's (1983) meta-analysis of 34 studies which examined the relationship between sex-role orientation and self-esteem. The nature of the evidence is such that we concur with Hoffman and Fiddell (1979): A review of the sex-typing literature led Bem (1972) to conclude that masculine and feminine sex-typing, in men and women respectively, is generally associated with poor adjustment. They also found that masculine adult men tend to have more ego control, less dominance, less capacity for status, less self-acceptance, more need for abasement, less self- assurance, less sociability, and less capacity for intro- spection than do men reporting masculine traits less strongly. Hoffman and Fiddell also believed that more evidence would eventually be found to correlate masculinity 26 in men with anxiety, guilt proneness, neuroticism, and other indicators of poor mental health: as well as a negative correlation with warmth, emotional stability, sensitivity, and other indicators of positive mental health. Hoffman and Fiddell (1979) made a strong case for the relationship between androgyny and positive adjustment in women by stating its antithesis: "Feminine women have significantly lower self-esteem than do androgynous and masculine women." They went on to describe feminine women as anxious, low in social acceptance, and low in self- concept. In their lengthy review, Hoffman and Fiddell (1979) concluded: In addition, strong and biologically consistent sex-typing seems to have a negative effect on intellectual development in children of both sexes. In sum, the evidence suggests that masculinity in men and femininity in women are related to poor adjustment. Recent research has suggested that masculinity in men and femininity in women do not in themselves indicate poor adjustment. As Whitley (1983) said, the additive effects of both sex-role orientations within each person are the real markers for positive adjustment. Hoffman and Fiddell (1979) suggested that self-concept be viewed as the individual's set of self-descriptive behaviors and that self-esteem be the subset of those behaviors relating to self-valuation of a trait or quality. They then added: 27 While these definitions deviate from the perceptual tradition, they are consistent with the implied operational definitions underlying much of self- concept research. Although more will be said later regarding self-concept measurement, it is helpful to note at this point that when self-esteem or self-concept is quantified, it is usually broken down into components of global self-esteem. v - t When Bohan (1973) attempted to determine the manner in which global self-esteem develops, he concluded that only longitudinal studies seemed helpful, indicating an increase in self-esteem over the time from pre-adolescence to adolescence. In contrast, cross-sectional research showed age differences but left the course of developmental changes unclear. Research on self-esteem usually is quite different from Bohan's study, focusing on a more narrowed aspect, such as academic self-concept. - e - t Apart from the literature review discussed earlier, one study will be examined here to draw the link between self- esteem and sex-role orientation closer. It has been noted that self-esteem is usually broken down into components and then measured in relation to other personality variables. 28 In contrast, Spiegel, Jackson, and Stollak, in an unpublished 1982 study, examined global self-esteem in relation to other personality characteristics and found it more consistently and positively correlated to androgyny. This finding is partially supported by Whitley's (1983) conclusion that social self-esteem is not necessarily positively correlated with androgyny, but this was dependent on the methods used to measure self-esteem. Sammazy 9f tha Literature Review Locus of control, derived from Rotter's (1966) social learning theory, has two dimensions, internal and external. The former is associated with competence, high achievement, harmony with others, and positive self- esteem. Internally oriented people also tend to be androgynous. Externality is negatively correlated to these traits. Little evidence exists regarding the development of locus of control in children. It has been suggested that males become more internally oriented and females more externally oriented with age. Sex-role orientation refers to the behaviors and/or attitudes one manifests regarding various sex-specific traits. SRO was given a boost in research popularity by Ram (1974), who hypothesized four orientations: masculine, feminine, androgynous, and undifferentiated. Androgyny in particular (and sometimes masculinity) is the SRO 29 correlated with positive mental health and other desired personality traits such as achievement, competence, and emotional stability. With age, androgyny apparently increases in men and decreases in women. Self-esteem is usually studied in relation to such issues as achievement or social adjustment (Larned and Muller, 1979). Nonetheless, the value individuals attribute to themselves is quite important and worthy of close scrutiny. The overall or global self-esteem an 'individual possesses has been positively linked to androgyny. Self-esteem seems to increase from pre-adolescence to adolescence when viewed longitudinally. Cross-sectional. studies have discerned no clear developmental trends. In general, locus of control, sex-role orientation, and self-esteem are not well represented in the literature about children since many more studies have been completed using adult populations. Relevant research with children typically does not yield results as conclusive as for adults, although it has been widely accepted that generalizations to younger populations are appropriate. Studies such as Baumrind's (1982), in yielding as much inconclusive data as conclusive, make this point more apparent. CHAPTER III Wm IEEIQQBELIQD This chapter will relate the information presented in chapters I and II to the methodology adopted in this study. The population and sample chosen will be described, as well as the research instruments and procedures. Limitations of the study also will be noted. 89W _ This study seeks positive relationships among data provided by middle school students on forms measuring sex-role orientation, perceived locus of control, and self-esteem. An harmonious linkage is expected among these three areas: that is, children who are well adjusted in relation to self, others, and their environment are expected to be found androgynous and to have an internal locus of control perception. Since past research has dealt primarily with adults, 'measures, techniques, results, and conclusions are not 31 likely to be very applicable to children. For these and other reasons stated previously, studies of these theories among childhood populations are less than conclusive. The available data on children does, however, provide some support for the hypotheses set forth here. More data, both in quantity and conclusiveness, are needed to confirm tentative statements regarding the self- esteem of middle school children. If internal locus of control, androgynous sex-role orientation, and positive self-esteem are found in this study to be positively related, this could have a powerful effect on how children are taught, guided, and trained in the home, at school, and in other areas where adults are concerned about children becoming wholly well adjusted. In short, it may be that children should be reared with an androgynous sex-role orientation and internal locus of control {9; aha pagpaaa of causing them to be better adjusted to relationships and circumstances they will face in life. EIEQSBQEQE H1: Those children who score above the mean on both masculinity and femininity will be those who are androgynous by definition and will also be those who have a greater degree of internal locus of control. 32 H2: The children with androgyny and internal locus of control will be those with the greatest self-esteem as measured by the Who Am I instrument. Specifically, these children will have higher scores in terms of: a) Mere positive statements: b) More mention of Abstract Identifications: c) More statements of Interests and Activities: d) Fewer statements of Material References: e) More mention of Personal Characteristics: f) Fewer negative statements. Points a through f were found by those who studied the Who Am I instrument to be the greatest indicators of positive self-esteem (see Gordon, 1968, 1969: Kuhn, 1960: Kuhn and McPartland, 1954: McLaughlin, 1966). W The hypotheses were evaluated by means of t-tests because the variables were dichotomized and lent themselves to this statistical procedure (Glass and Stanley, 1970). After the hypotheses were tested, additional analyses ‘were completed to determine whether relationships existed among the independent and dependent variables which were not hypothesized at the study's outset. These additional tests included ANOVA, and Pearson correlations. 33 11112923133191: Unfortunately, demographic and other detailed information was not available for the study. Since the students live near a major university and many are from academic families, they would be considered advantaged (Finlayson, 1984: Mitchell, 1984: Schneider, 1984). Even students from families of below average socioeconomic status would probably be influenced by parents, teachers, and others having ties to the university. W Middle school students in the sixth and eighth grade from two buildings within a single school district completed the research measures. All were enrolled in an elective course, Reproductive Health and Family Life, which was taken according to parental preference. Each student was instructed to complete the forms during one class session by the course instructor. Of the 531 students who completed the measures 200 were selected randomly, yielding groups of 50 each from males and females in the sixth and eighth grades. Since among the initial 200 there were measures with missing data, random selections continued until 200 subjects were obtained with at least representative data on all three measures (one-half the items of each measure completed). 34 W ' v e e - v . (See Appendix A.) Based on Bem's (1974: 1981) Sex-Role Inventory for adults, the children's version was developed by Stericker and Kurdek (1982). The many criticisms of Bem's theory and her sex-role orientation measure as a device used in self-esteem assessment are well taken, but it is important to note that Whitley's (1983) meta-analytic study revealed that androgyny is a more salient characteristic to measure if the study is investigating global self-esteem, as does this one. Stericker and Kurdek (1982) endeavored for two years to develop the children's version of the sex-role orientation inventory and believed it to be adequate for its intended purpose. It has been well received by researchers. Since this measure necessitated much administration time due to its four-choice format, it was shortened for use in this study. From the original 60 items (20 each of masculine, feminine, and filler items) used by Rotter (1966) and modified by Stericker and Kurdek (1982), 33 were selected randomly. From this process a mix of 12 masculine, ll feminine, and 10 filler items were chosen. The form and scoring procedure developed by Stericker and Kurdek (1982) features a four-choice response continuum 35 (0-Never, 1-Sometimes, 2=Often, and 3-Always). Scores were tabulated for masculinity on the masculine items and femininity on the feminine items for each child. A total score range of 0-36 was possible for the masculine scale and 0-33 for the feminine scale. As recommended by Spence, Helmreich, and Stapp (1975) and as done by Stericker and Kurdek (1982), children were classified into one of four sex-role self-concept groups by means of a double mean split of their masculine and feminine scores. As noted earlier, the four groups were masculine, feminine, androgynous, and undifferentiated. A tabulation of these classifications for the present sample is presented in Table 3.1. The mean score for the masculinity scale was 20.39, with a standard deviation of 5.74. The femininity scale yielded a mean score of 20.57, with a standard deviation of 5.84. To estimate the internal consistency of the sex-role inventory, coefficient alpha was computed separately for the masculinity and femininity scale. The resulting values were r-.72 for the former and r=.74 for the latter. To compare these findings more directly to Bem's (1974) original reliability data, a Spearman-Brown prophecy formula was employed. It predicted r-.81 for the masculinity scale, which compares closely to Bem's (1974) finding of an alpha coefficient of .86. The counterparts for the femininity scale, are r-.84 and an alpha coefficient of .80. 36 TABLE 3.1 MEANS, MEDIANS, AND STANDARD DEVIATIONS OF DICHOTOMIZED VARIABLES Mean Median Standard Deviation Sex-role orientation Masculinity 20.39 19.79 5.74 Femininity 20.57 19.91 5.84 Locus of control 14.55 14.12 4.45 37 IDLQIDQI'EEEQEDQL Locus of Control Scale fog Children. (See Appendix B.) This device was born of Rotter's (1966) interest in social learning (Rotter, 1954). In 1973 Nowicki and Strickland modified the scale for children and gave it the name listed here. This scale was scrutinized further by Nowicki, (1976) and by Wolf, Skolv, Mac D. Hunter, and Berenson (1982), who found it reliable and valid. It is important to note that Nowicki and Strickland (1973) and Nowicki (1976) examined children in grades 3-12, while Wolf et a1. (1982) studied children in grades 5-12. The number of subjects for these studies were 102, 1,226, and 368, respectively. The general practice for scoring the Children's Locus of Control Scale was followed here: a child's score is the number of external items s/he endorsed. This yielded an external score for each subject. Mean splits were then taken, and subjects with scores above the mean were classified as those with an external locus of control orientation. Internally oriented children had scores lower than the mean. Among the researchers listed above, only Nowicki and Strickland (1973) published actual normative data. They found a cumulative mean score of 14.36 for the sixth and eighth grades, without reporting how many subjects they tested in those grades. The present study resulted in a mean score of 14.55 for the 200 subjects. 38 The reliability of the locus of control scale was estimated in the same fashion as employed by Nowicki and Strickland in 1973 when they developed the scale. This was a Spearman-Brown split-half coefficient. The present study yielded r-.63, similar to Nowicki and Strickland's (1973) r-.68. Wha_Am_1. (See Appendix C.) The use of the Who Am I as a measure of global self-esteem for the present study is based primarily on the work, arguments, and resulting conclusions of Kuhn and McPartland (1954) and Gordon (1969). Although these scientists use differing words, both studies concluded, after examining myriad forms of ”self” assessments, that the researcher must assess the subjective self to know how the individual values him/herself. They decided that tapping the individual's own statements without providing ready-made responses or dimensions was the surest method for determining the nature of self-concept and global self-esteem. The Who Am I was used here precisely as it was by the aforementioned researchers. Totals for each category were tabulated for positive, negative, and neutral statements. Tabulating the results of the Who Am I necessitated training 14 undergraduate volunteer students to perform coding tasks. To ensure that they were blind to the study's purpose, the students were told that the research investigated metalinguistics and metacognition--how 39 children reflect on their thoughts and how they report them. The coders were not shown the other measures the children completed. The coders were given the Training Manual for the Who Am I Questionnaire (See Appendix D). This was devised by the author purely to ensure that the training would be consistent and in agreement with Gordon's (1968: 1969) coding procedures, and to maintain coders' ignorance of the study's true objectives. They were asked to read the manual before beginning coding tasks. Any questions were answered in a nondirective fashion. As can be discerned from the manual, the coders followed the categories and format of Gordon (1969). These students were trained on bogus Who Am I questionnaires according to the manual's examples and directions. They coded five questionnaires or more until they agreed with at least 90% of all responses of master (coded) Who Am I questionnaires drawn up by the author. For the 14 students, the number of instruments coded in training ranged from five to seven. A random pairing of 40 Who Am I forms coded during training resulted in an intercoder agreement of 90.5% of the total possible coded responses. The results of the Who Am I responses will be addressed specifically in the Results section. 40 MW The "weak link" in this study is the transition from the theoretical to practical. It has been stated at various points that the three theories and related research in regard to children are quite nebulous. That is, while this study seeks to make their application to children more definitive, it is very obvious that this has been a problem for other researchers as well. Not least of which is the pitfall of basing children's research on that begun with adults. These issues make the task formidable. It also has been noted that the subjects resided within a major university setting. That may make the population sufficiently distinct to prevent valid comparisons with previous findings and/or generalization to other middle school populations. CHAPTER IV 39521£§_21_£h§_§£2§1 The format followed in this Chapter is to discuss the categorization of the sample into various groups and describe them. Each hypothesis, its analysis and results, and associated (not hypothesized) results is then listed. classificationemsreumioss The percentages of students by sex is presented in Table 4.1. Among males, the greatest number fall in the masculine category (n-39) and the next largest in the androgynous category (n-31), followed by undifferentiated (n-23) and feminine males (n-7). The comparable ranking for females is feminine (n=4l), androgynous (n-28), undifferentiated (n-25), and masculine females (n-6). These data may be broken down by masculinity and femininity only, as shown in Table 4.2. versus the percentage given in Table 4.1. In general, the males are more masculine than feminine, and vice versa for the females. When the subjects were divided by their locus of control scores, 106 were classified as internal, the remaining 92 as external. 42 TABLE 4.1 SEX-ROLE SELF-CONCEPT GROUP % % % % N Masculine Feminine Androgynous Undiffereniated Males 100 39 7 31 23 Females 100 6 41 28 25 43 a>mrm s.m mmxuxorm mmrminoznmva mzoco. 3>mncr~zuq< >2: mmzmz~2H4< ozr< z>mocr_z~a< mmzmzmzwq< :mme «mammmm Zmme “mammmm zcacms om yin: No we wm mo zcscms om Lo: mo mm mm as 44 H1: Those children who score above the mean on both masculinity and femininity will be those who are androgynous by definition and will also be those who have a greater degree of internal locus of control. Evidence was found to support this hypothesis. As shown in Table 4.3, the mean of the androgynous children was 1.07, and the mean of all other children was 1.14, resulting in a t value of 1.69 (p < .05) It is to be noted that, for the remaining hypotheses to be tested, androgyny and internality are given status as independent variables so that the remaining variables are the focus of the analysis. That is, the status of androgyny and internality among the subjects is no longer an issue, but is accepted. 32a: Children who are both androgynous and internal will have a greater number of positive self-statements on the Who Am I. Evidence was found to support this hypothesis. The children with androgyny and internality had a mean of 2.15, other children a mean of 1.86 with regard to positive statements. As shown in Table 4.4 this resulted in a t value of 1.92 (p < .05). 32b: Children with androgyny and internality will have a greater number of statements referenc- ing Abstract Identifications on the Who Am I. 45 TABLE 4.3 MEAN SCORES OF ANDROGYNOUS AND OTHER SEX-ROLE ORIENTATIONS ON LOCUS OF CONTROL N Mean Stand. Dev. Androgynous 59 1.07 .25 Other sex-role orientations 140 1.14 .35 MEANS OF DEPENDENT VARIABLES FOR ANDROGYNOUS 46 TABLE 4.4 AND INTERNAL VERSUS OTHER CHILDREN Internal and androgynous Other children Dependent Mean Standard Mean Standard variables deviation deviation Positive statements* 2.15 .834 1.86 .783 Abstract identification .455 .869 .493 1.08 Interests and activities 1.91 2.29 1.59 2.59 Material references* .697 .732 1.08 1.29 Personal characteristics* 3.55 2.59 2.65 2.33 Negative statements .886 1.39 1.14 1.69 *Significant at p=.05 47 Using a t-test, no significant values were found between the two groups. H2c: Children with androgyny and internality will have a greater number of Interests and Activities references on the Who Am I. Using a t-test, no significant difference was found between the two groups. sz: Children with androgyny and internality will have fewer Material Reference statements on the Who Am I. This hypothesis was supported. Children with both androgyny and internality had a mean on Material Reference statements of .69, while other children had a mean of 1.08. This result is shown in Table 4.4. The t value was 2.36 (p < .05). H2.: Children with both androgyny and internality will have a greater number of Personal Characteristics references on the Who Am I. Significant values were found between the two groups, with a mean of 3.55 for children with both androgynous and internal characteristics, a mean of 2.65 for the others. The t value given by this finding was 1.99 (p < .05). This result is also shown in Table 4.4. 32:: Children possessing androgyny and internality will have fewer negative Who Am I statements. 48 Using a t-test, no significant difference was found between the two groups. To investigate further the relationship of these variables with the measures of self-esteem, two- and three-way ANOVAs were performed. The independent variables were sex, high or low scores on masculinity, and high or low scores on femininity. The design for this analysis was as follows: High (N-70) aim (IL-31) Male Maagaliaity Is Fem tit-39) (N-lOO) Low (-30) Hi r... (N-7) Wt L9_F_em (rt-231 High (N-34) We) Female Maaaaliaitv ILa_Eam:tN-6) (N-100) Lag (11-66) Hi Fan (Iv-411 Jaw (II-'25) The three-way ANOVA was performed using each of the six measures of self-esteem on the Who Am I as a dependent variable. These analyses yielded several significant results. 49 One main effect was found for sex, which showed that the number of negative statements was greater for females than for males. The mean for females was 1.32 and for males .88. This resulted in an F value of 5.18 (p < .05). The interaction of sex and femininity was significantly related to the number of negative statements given by subjects on the Who Am I. Males of low femininity demonstrated the lowest mean number, .73. Females of high femininity were next to lowest, with 1.09, and males of high femininity were next, with 1.13. Females with low femininity had the highest mean number of negative statements, 1.84. The resulting F value was 20.12 (p < .01). This finding is graphed in Figure 4.1 and will be discussed in Chapter V. The interaction of masculinity and femininity was significant in terms of Material References on the Who Am I. Subjects with both high masculinity and high femininity showed the lowest mean, .79, while those with low masculinity and low femininity had the second lowest mean, .89. Subjects with high masculinity and low femininity had a mean of 1.13: while those with low masculinity and high femininity had the greatest mean number, 1.29. This interaction is graphed in Figure 4.2 and is discussed in Chapter V. The resulting F value was 4.18 (p < .05). AVERAGE NUMBER OF NEGATIVE STATEMENTS 50 FIGURE 4.1 THE INTERACTION OF SEX AND FEMININITY FEMALES .8 “ MALES LON HIGH FEMININITY FEMININITY AVERAGE NUMBER OF MATERIAL REFERENCES 51 FIGURE 4.2 THE INTERACTION OF MASCULINITY AND FEMININITY \ LON FEMININITY HIGH FEMININITY LON MASCULINITY HIGH MASCULINITY 52 An ANOVA two-by-two design was then constructed to detrmine whether androgyny and internality had interactive effects. The rationale and implications of this phase of the analysis will be treated more extensively in Chapter V, but it is sufficient to state at this point that the results had so far been inconclusive in the determination of the interactive effects of androgyny and internality. The design of the analysis is given in Figure 4.3. The results of this analysis were not significant: no interactive effects for androgyny and internality were revealed. When this particular analysis is taken as a whole, it is found that of the sum of the differences for androgyny and internality, taken by the six dependent variables, eight of the twelve resulting variations are in the hypothesized direction, even though they are not all significant. This result is shown in Figure 4.4 and will be discussed in Chapter V. A series of Pearson product moment correlations also was calculated. Actual scores on the masculininity, femininity, and locus of control scales were used, without dichotomizing them. Findings will be given by discussing, first, relationships among the dependent variables: second, relationships among the independent variables: and, third, relationships between the dependent and independent variables. Seven relationships among dependent variables are noteworthy. 53 Figure 4.3 Anova Analysis Design: Dependent and Independent Variables Independent Variables Androgyny Other Sex-role orientations (n=59) (n=141) Internal External Internal External (n=106) (n=92) (n-106) (n=92) Dependent Variables Positive Statements Abstract Identification Interests and Activities Material References Personal Characteristics Negative Statements .15 .10 .00 .90 o—eo-‘o-INNNNN .20 - 05 A .95 . .854 POSITIVE STATEMENTS \\ ANDROGYNOUS OTHER SROS A_F_ INTERNAL EXTERNAL ABSTRACT IDENTIFICATIONS OTHER SROS ANDROGYNOUS INTERNAL EXTERNAL INTERACTIONS OF SEX-ROLE AND 511 FIGURE 4.4 LOCUS OF CONTROL ON MEAN NUMBER OF RESPONSES HHHHHHfi-‘HHHHH IOU-OOOIII'. PERSONAL CHARACTERISTICS 3.6 , 3.51. 3 4 i ANDROGYNOUS 3.3 » 3.2 - 3.1 - 3.0 - 2.9 » 2.8 » 2.7 - 2.6 _ OTHER SROS 2.5 2.4 ‘ A4, 0 INTERNAL EXTERNAL NEGATIVE STATEMENTS . OTHER SROS l» o ANDROGYNOUS INTERNAL EXTERNAL HH H H H H 0-0 9—. H H H N N N U I O I I C . . I D 1 C) N w 5 UI 0’1 V m ‘0 O H N .70 .90 - .00 .10 .20 MATERIAL REFERENCES I \\\\\\:NOROGYNOUS r OTHER SROS .: - INTERNAL EXTERNAL INTERESTS & ACTIVITIES . ANDROGYIV OTHER SROS Y INTERNAL EXTERNAL 55 (l) The number of positive self-statements and Interests and Activities were related significantly (p < .01), with a Pearson correlational coefficient of .43. (2) Positive self-statements were correlated to Personal Characteristics (p < .01), with a Pearson coefficient of r-.31. (3) Positive self-statements were linked to Abstract Identifications (p < .01), with a Pearson correlation of .25. (4) Abstract Identifications were correlated to Interests and Activities (p < .01), with a Pearson coefficient of r-.21. (5) Negative statements were correlated with Material References (p < .05), with a Pearson coefficient of r-.15. (6) Abstract Identifications correlated negatively with Material References (p < .05), with a Pearson coefficient of r--.13. (7) Interests and Activities were negatively related to Material References, with a resulting Pearson coefficient of -.16 (p < .05). As will be discussed in detail in Chapter V, these correlations demonstrate general support for the notion that the dependent variables measuring positive self-esteem correlate positively with one another and negatively with indicators of negative self-esteem. One relationship among dependent variables was found to be significant, that of masculinity and femininity, which resulted in a Pearson coefficient of .23 (p < .01). This result will be treated more completely in Chapter V. In 56 general, this population was similar to those in other research in that the characteristics of masculinity and femininity have points in common (for example, are androgynous) and are not mutually exclusive. One significant finding was revealed between dependent and independent variables: Masculinity was correlated to Interests and Activities, with a Pearson coefficient of .18 (p < .01). This finding will be explored in the next chapter. The final analysis of the data was accomplished by using t tests to compare the means of the data for each variable being measured. Only one result was significant, the relationship of masculinity to Interests and Activities. Highly masculine children produced more statements of Interests and Activities on the Who Am I than all other children (a mean of 1.33). This resulted in a t value of 1.69 for n-104 (p < .05). This finding will be given more extensive treatment in Chapter V. In sum, four of the seven hypotheses were supported: Androgyny and internality were positively related: children with androgyny and internality wrote a larger number of positive statements on the Who Am I than did other children: subjects with androgyny and internality wrote fewer Material Reference statements than did other children: and children with androgyny and internality wrote a greater number of Personal Characteristic statements than did other children. 57 In addition, the combination of high masculinity and high femininity was related to a low number of Material Reference statements. It was also found in general that the dependent variables were correlated in a positive direction. These major findings and supporting results will be explored in greater detail in Chapter V. CHAPTER V 11133332911911 In general, the findings provided support for the hypotheses proposing that positive adjustment of middle school children can be viewed in terms other than "global self-esteem." Precisely, four hypotheses received direct, affirmative support. Secondary evidence also appears to support the hypotheses. Not every expected relationship among variables measured was supported, but sufficient support was found to draw favorable conclusions about the majority of the hypotheses. In presenting these conclusions, the discussion will follow the format of Chapter IV: a section on general findings, categorization of the sample, and treatment of each hypothesis. A separate section will examine additional findings, followed by a summary of specific results. The chapter concludes with a section on implications and suggestions for future research, and a discussion of the author's view of the results and their relevance. 59 Minding; General Findings The sample was divided into sex-role groups as hypothe- sized. As Table 4.1 showed, the sexes favored their stereotyped sex role above the other three sex-role orient- ations. Interestingly, this sample did not demonstrate the high percentages (51.5 males and 37.0 females) of androgy- nous subjects that Stericker and Kurdek (1982) found. Since the reliability data for the sex-role inventory was acceptable (male-.66, fem-.78) (Glass and Stanley, 1970), the rationale for this curious fact may lie in the rather small sample size of the previous study, and also in the possibility that the samples are from differing popula- tions. (Table 4.2 reflected Table 4.1, with subjects categorized into only the two basic sex-role orientations.) Subjects were classified as internally and externally oriented, as per Nowicki and Strickland (1973). The means of the two samples (14.36 in 1973 and 14.55 here) are surprisingly close, and the reliability data were acceptable, although not outstanding (.63) for the present sample. These facts lead to the conclusion that the present sample was indeed comparable in regard to the internal-external locus of control variable to those of other researchers who classified their subjects via the Children's Locus of Control Scale. 60 Hypotheses H1: As anticipated, androgyny and internality were correlated. This is interpreted as indicating that a child with a high degree of masculinity and femininity typically believes s/he is in control of his/her circumstances. These children presumbly would be well adjusted and positive in their outlook, as was hypothesized at the outset of this study, but the other findings must be taken into account before such a conclusion can be drawn. H23: Evidence was found that androgyny and internality are related to the positive self-statements a child provides, and it can be concluded that androgyny and internality are correlated with the positive statements. The inference is that androgyny and internality are also related to positive adjustment, since pOsitive statements are held to be indicators of positive self-esteem. On the strength of the additional evidence gathered, it also may be properly inferred that the component of androgyny, femininity, interacts with the value a child holds of him/herself. As was shown in Figure 4.1, males of low femininity had very few negative self-statements, while those with high femininity had more negative things to say of themselves. Conversely, low-feminine females had a relatively high number of negative views of self, while those of high femininity had nearly half as many. 61 The obvious conclusion is that a stereotypical view of self leads to fewer negative self-statements. That does not allow for androgyny being a predictor of the valence one holds of oneself, however. This conclusion is actually in opposition to the hypothesis, which held that androgyny was related to a greater number of positive statements. A traditional view of self is not linked to a higher number of positive statements, but is significantly linked to fewer negative statements, still an indicator of positive self-esteem. Since the hypothesis was directly supported and since other results of this study also provide evidence that both androgyny and internality are indicators of positive adjustment, undue emphasis need not be placed on this particular result. 32b: No result supported the expectation that androgyny and internality are related to an esoteric view of self, as shown by a high number of statements in the Abstract Identification category of the Who Am I. The inference that androgyny and internality contribute to a view of self not entangled in temporal characteristics and hence connected to a secure sense of self as an independent entity, was not borne out. It cannot be said, then, that positive self-esteem as evidenced by androgyny and internality is reflected in a child's identity being tied to perceptions of him/herself as a unique, nonmaterial, and irreducible being. 62 32c: No direct evidence spported the claim that androgyny and internality were linked to Interests and Activities. The notion that androgynous and internally oriented children are more active and interested in their world and the lives of those around them was not confirmed. Nevertheless, a component of androgyny, masculinity, was a predictor in this area. High masculinity alone was related to a larger number of statements in the Interests and Activities category on the Who Am I. A masculine child was more likely to report a larger number of hobbies, and the like. Rather than androgynous and internally oriented children being active and involved in the environment and surroundings, it can be said that masculine children are more "involved" and seek to be more highly stimulated and challenged. 32d: As hypothesized, children with androgynous and internal orientations were less likely to be concerned with their appearance and possessions. A parallel finding was shown in Figure 4.2, where the combination of masculinity and femininity in high amounts is the lowest point on the graph. Other sex-role orientations are at much higher points in terms of physical concerns and references. Due to fewer references in this category it appears that children with high masculinity and high femininity as well as both androgyny and internality are more secure in their appearance and with their possessions than are others. 63 32¢: Evidence was provided that children with both androgynous and internal locus of control characteristics had a greater number of Personal Characteristics referenced on the Who Am I. This indicates children with high self-esteem, as measured by androgyny and internality, also are more interested in and in touch with those around them and are able to report this involvement, as opposed to being concerned with whether they are attractive to others and have nice possessions. This result supports the view that an androgynous and internally oriented child's relationships are more satisfying and come more readily to mind to be reported than are those of others. H21: The final hypothesis stated that androgynous and internally oriented children would have fewer negative statements on the Who Am I. Since this was not supported by the study's results, it is concluded that no link exists between positive self-esteem, as defined by androgyny and internality, and a lack of unfavorable views and thoughts of self. Winding: Analyses performed in addition to the tests of the hypotheses revealed several findings. One was that the number of negative statements expressed by females was greater than for males. This is interesting but not surprising, since other researchers have found that 64 masculinity is a greater predictor of high self-esteem in females than is high femininity for males, hence females would probably have the greater number of negative statements. Another finding, shown in Figure 4.1, was that males of low femininity and females of high femininity had the fewest negative statements on the Who Am I. Conversely, males with high femininity and females with low femininity had the most negative statements. This finding, again supporting the notion that a traditional sex-role orientation promotes high self-esteem, goes against the bulk of the evidence showing androgyny (and internality) to be the more salient personality characteristics associated with high self-esteem. A third finding is not so difficult to explain because it was very much as hypothesized: Subjects with a high degree of both masculinity and femininity had the fewest number of Material References on the Who Am I. This finding is very similar to that which originally supported hypothesis sz, namely, the association of androgyny and internality with fewer Material References. The difference here is that the components of androgyny are linked to the dependent variable, Material References, without internal- ity being a factor. Regardless of how this finding is explained, it lends more credibility to 32d! as well as to the others. 65 A fourth finding is most curious in that it is not technically significant but offers compelling evidence in partial support of the hypotheses. As reported in Chapter IV and shown in Figure 4.4, when the twelve possible combinations of androgyny and internality were compared on the six dependent variables, the result was eight expected directions and only four not hypothesized. Although not significant by itself, this preponderance of agreement in the hypothesized direction may be meaningful. It seems at least to bolster the results supporting the hypotheses whose tests were significant. It may also give increased importance to the hypotheses whose results were not significant. Several correlational links were demonstrated among the possible combinations of positive and negative indicators of the Who Am I. Overall, it was expected that all positive variables and areas would be positively correlated and all negative areas would be positively correlated. Although that did not happen, the following links were supported by correlational findings: POSITIVE TO POSITIVE RELATIONSHIPS: -Positive statements and Interests and Activities -Positive statements and Personal Characteristics -Positive statements and Abstract Identifications -Masculinity and Interests and Activities 66 NEGATIVE TO NEGATIVE RELATIONSHIPS: -Negative Statements and Material References POSITIVE TO NEGATIVE (Negative Relationships): -Abstract Identifications and Material References -Interests and Activities and Material References These findings suggest that the positive areas of the Who Am I are (generally) those that were taken to be indicators of positive self-esteem. To wit, a child with a healthy self-esteem as indicated and measured by a greater number of positive Who Am I statements has more interests and hobbies: has more to say about his/her relationships and personality characteristics, being more interested and invested in those around him/her and reporting same: and has a more esoteric (Abstract) opinion of self. A high degree of masculinity, already shown to be a predictor of a greater number of Interests and Activities, is also posi- tively correlated to Interests and Activities, showing that masculinity is related to an active and more well-rounded personality. Also as expected, negative statements and Material References are linked. The inference is that children with a negative view of self are much more concerned with appearance and trappings of success and status. 67 Finally, Material References are negatively correlated to Abstract Identifications and to Interests and Activities. This is not surprising since the possibility was just raised of how a negative or positive view of self is expressed via these categories of the Who Am I. It could be said, then, that children concerned with trappings and appearances of success and to whom these come easily to mind are not those with an enduring and stable view of one's self as a unique and irreducible being, nor those with interests and investments outside of self. It was shown that masculinity and femininity were positively correlated. This has been found by various researchers (such as Orlofsky, 1981), who believe this is the reason positive adjustment is linked to androgyny. It is held that those with a high degree of masculinity or femininity can assimilate the opposite tendency into their personality to benefit their view of self and emotional balance (See Orlofsky, 1981). Windlass The subjects of this study can be classified in the fashion of other researchers measuring these same variables, and many of the suspected indicators of positive adjustment were in harmony. This suggests that a child with many references in these areas would have positive self-esteem: the description of an emotionally healthy child as one with androgyny and internality was also 68 supported by the data. The conclusion to which the data lend credit is the relationship posited at the outset. The personality characteristics in children of androgyny, internal locus of control, and positive self-esteem are in harmony. What have been largely tentative statements in these areas of personality with regard to this age group are now on firmer ground, since this research found evidence that androgyny, internal locus of control, and positive self-esteem are correlated for children. The design of this study raised the question of whether the make-up of the population - being derived from a university setting - had a direct bearing on the findings. Though no evidence delivered conclusive data on this issue, more will be said of it later. The children described favorably by the weight of the evidence of this study are androgynous and view themselves as having a blend of many masculine and many feminine behaviors. They also perceive that they are in control of their destiny or circumstances, and they recall and believe a larger number of positive things to say of themselves than do other children. Those shown to be well adjusted are also less concerned with their appearance and their possessions: they list more things about their relationships and their investments in other's lives than do the majority of the other children. 69 W Since the setting from which this study drew differed from that of previous research and yet the results are similar, an obvious conclusion is that the generalizability of the previous studies would now be accepted. This similarity is quite surprising since few other studies have dealt with all three personality factors for children, and none have examined children this age from a similar population. The fact that this research found androgyny and internality to be associated with the factors which other authors asserted they would he leads to the statement that their claims have the breadth of meaning for all children which they ascribed to their findings. Although none of the studies cited selected subjects from a university setting, they implied that their findings would be applicable to all children. Furthermore, since the studies of Nowicki and Strickland (1973), Nowicki (1976), and Wolf et a1. (1982) drew from a rather large number of subjects in many grades (including sixth and eighth), it was also important to note that the present study's conclusions support their findings of internality being related to positive adjustment and self-esteem.' Similarly, although this study does not settle the debate as to whether androgyny is an indicator of positive self-esteem, internality has been shown as linked to androgyny and to those characteristics which signify positive adjustment. 70 In addition, while replications have successfully supported the relationship between androgyny and internality and positive self-esteem in several populations, this conclusion is now borne out for subjects from a university setting. Since very few of the studies cited analyzed the data of their sample by grade, it may still be true that androgyny and internality are not related to positive self-esteem in some grades, while they may be in others. Meet research has lumped the grades together in performing correlative calculations to determine whether androgyny and internality are linked to self-esteem. If the children of some grades do possess a sex-role orientation that is related to self-esteem while others do not, then a study's findings could still be conclusive in an overall sense. Since no other work has examined these characteristics for middle school students exclusively, additional effort may be needed to assess this possiblity, particularly by analyzing the results grade by grade. A study of this sort would yield very interesting data (to compare to the present study), especially if samples were drawn from a university and nonuniversity setting for comparative purposes. Then, with analyses performed by grade for a range of grades encompassing the puberty years, conclusions might be forthcoming as to whether sex-role orientation and internality are related to self-esteem and, if so, for what particular age ranges and populations. To 71 elucidate, a study that would examine self-esteem, sex-role orientation, and locus of control perceptions in grades 4-12 (each examined separately) across the populations of urban, rural, and advantaged students should be able to state which grades demonstrate the positive correlation of the characteristics as well as which populations. The Who Am I deserves close scrutiny as a device to be used in future experiments. This is due to the fact that its variables appear to be interrelated, and they are claimed to be indicators of self-esteem (Gordon, 1968, 1969), yet they have not (in other studies) been linked to independent variables. Thus, although the Who Am I presumably is related to other self-esteem measures (such as the Rosenberg self-esteem scale), its relationship to measures of other personality characteristics remains a mystery. The Who Am I is a valuable tool in self-esteem assessment. One has only to read a few of the forms from any population to be fascinated by them, but this only yields descriptive statistics (a major use of the Who Am I by Gordon, 1968: 1969). Correlational studies are needed to place this device in its proper place among mesures of personality, as will be delineated below. Given the use of the Who Am I among high school and adult groups, validating studies with younger populations are sorely needed. Fortunately, the present research made some contributions in this regard, but more work among 72 younger populations could be very valuable to determine whether age and sex differences exist among differing populations as measured by the various Who Am I categories. Also useful would be further correlational calculations on the variables derived from the Who Am I categories since they have been investigated more in relation to themselves than to other (that is, independent) variables. Although the Who Am I has traditionally been tabulated by category designations, future researchers may find that it actually measures more of a centralized and constricted trait or characteristic. This possiblity needs to be investigated with additional data showing trends and tendencies within Who Am I results, but it may be found that a single score derived from the instrument would be sufficient for determining the significance of its results when compared to those of other measures and scales. This possibility is raised with the evidence presented in Figure 4.4, where the Who Am I variables show results in the hypothesized direction, even though not of a significant level. Regardless of specific technical considerations, the claim made here indeed seems valid: that if it is desirable to teach, guide, and treat children in such a manner as to build their self-esteem, than parents, teachers, and child- care-givers will want to encourage an androgynous and internal locus of control perception, as well as reinforce the concomitant behaviors. The ultimate goal would be to 73 prepare children to be more well adjusted and thus better able to meet the conditions of life they will face. WW Given the author's interest in the underadvantaged and urban dwellers, a personal word might be added as to what this study's results mean. The author began this research with the idea that androgyny and internality are stable and solid indicators of healthy adjustment. This study's findings support the advisability of fostering an androgynous and internal locus of control perception in middle school children, and perhaps to others. In this author's opinion, the fact that the Who Am I is a free-response test is of vital importance. That is, of all the possible statements the children could have provided on this measure, those with greater self-esteem gave more positive statements, fewer Material References, and more Personal Characteristics. It is sad to note that current parenting and educational trends stress gadgetry and programs. Perhaps just as easily, an emphasis could be given to promoting stable relationships and even lasting, meaningful friendships. When a child can recall on a Who Am I that s/he has three friends with whom s/he enjoys playing at lunchtime, versus whether s/he learned the latest computer math program, then truly that child has learned a great lesson. 74 The study's findings also suggest that children be reared with less emphasis on their appearance and possessions. This may go against the grain of an "us" and "me” society, but the results certainly point in this direction. It is perhaps especially vital tht underadvantaged children be reared with less regard for their physical attributes, which is no small task. This study and those of Gordon (1968: 1969) imply that young and underadvantaged individuals should be helped to feel more comfortable with their role and themselves in an overall sense, rather than attempting to give them a different view of self. That would entail, of course, the challenge of helping others believe and have confidence in themselves, whether they are from a family of low socio- economic standing, are underachievers, and so forth. This is much easier to say than to carry out, but if the effort is successful, all of us would benefit. Following are specific examples of what various influential adults could do to assist children in the development of healthy self-esteem. The present culture, as much as at any time in our history, places enormous emphasis on physical attributes (appearance) in defining the individual. Those in the field of education who come in contact with children would do a great service by teaching him/her to emphasize interpersonal ties and other positive personality characteristics already developed as a means to bolster self-esteem. A welfare worker could offer 75 suggestions to a child in a family of little monetary means to be mindful of (and hold dearly) those circumstances in which the child has acknowledged authority (an internal locus of control perception) as a way to bolster that child's self-esteem. Even if the area in which this child has authority is only a sandbox, that fact may become the foundation for the child learning his/her value as an individual. Likewise, a psychologist in any setting would impart a benefit to a child by seeking to attenuate attitudes that are prevalent in that child's surrounding's regarding appearance and possessions. One may initially perceive that this is the undisputed goal of any psychologist, but, again more than lip service is needed. Furthermore, the parents or guardians of disadvantaged children usually do not have the means to pay for the services of a psychologist, and responsibilty for much of any child's development resides at home. Parents may not perceive that they or their child could benefit from reordering their life's priorities to deemphasize possessions, yet this is what would have lasting value for the child. Indeed, this need is great, the challenge immense, but the potential unfathomable. REFERENCES Anastasi, A. £§Y§h912§1§§1.1§§§1fl9. Fifth Edition. New Yerk: MacMillan Publishing, 1982. Antill, T. & Cunning, T. D. Comparative factor analysis of the Personal Attributes Questionnaire and the Bem Sex-Role Inventory. v o s t 1980, 19, 163-172. I Bandura, A. fig;ia1_Laa:ning_1haazy. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. 1977. Baumrind, D. Are androgynous individuals more effective persons and parents? thld_Dexelssmsat. 1982. 51. 44-75. Bem, S.L. Psychology looks at the sex-roles: where have all the androgynous people gone? Paper presented at the University of California at Los Angles, Symposium on Women, May 1972. In D.M. Hoffman and L.S. Fiddell, Characteristics of androgynous, undiffereniated, masculine, and feminine middle-class women. Sex Rales, 1979, 5, 765-781. Ben, S. L. The measurement of psychological androgyny. IQHIDQJ Q: QQBEHJEIDQ QDQ gljnjgal Egygnglogx, 1974, 52, 155-162. Bem, S. L. Sex-role adaptivity: One consequence of psychological androgan- 12uroal_gf_£ers2nalitz_aad_ Social_£sxsholesx. 1975. 11. 634-643- Bem, S. L. Probing the promise of androgyny. In A.G. Kaplan & J- P. Bean (Ed8-) 8sI2nd_§er:8212_§tsrsstxees: Bsadinss_T9sard_a.2sxsbglssr_sf_Andrssxnx . Boston: Little Brown, 1976. Bem, S. L. - o v s an . Palo Alto: CA. Consulting Psychologists Press, 1981. Sam, S.L., Martyna, W., & Watson, C. Sextyping and androgyny: Further explorations of the expressive domain. u s a o s cho o , 1976, 11, 1016-1023. Bernard, L. C. The multidimensional aspects of masculinity-femininity. o o erso a t n §2§1§1.2§Y§h2129¥: 1981. $1. 797-802- Best, D. L., Williams, T. E. Cloud, T. M., Davis, S. W., Robertson, L. S., Edwards, T. R., Giles, H., & Fowles, T. Development of sex-trait stereotypes among young children in the U.S., England, & Ireland. Shilg DQEEIQBEEDEI 1977: $5: 1375'1334- 77 Dover, T. G. The cognition basis for linguistic structures. In J. R. Hayes (Ed.), Dexelonment_2f_Lansuase- New York: Wiley. 1970- Bhagat, R. S., 8 Chassie, M. B. The role of self-esteem and locus of control in the differential prediction of performance, program satisfaction, and life satisfaction in an educational organization. Jaagaal of_Yecationa1_nshaxior. 1978. 11. 317-326- Block, T., VonDerLippe, A., 8 Block, T. H. Sex-role and socialization patterns: Some personality concomitants and environmental antecedents. C and_£linical_rsxsholosx. 1973. 41. 321-341- Bohan, T. S. Age and sex differences in self-concept. AQQLQEEQDQQI 1973: Q: 379-3840 Brehony, K. A., & Geller, E. S. Relationships between psychological androgyny, social conformity, and perceived locus of control. Bayahalagy_afi_flaman QHRIIQILY. 1981. 5. 204-217- Bridges, J. S. Sex-typed may be beautiful but androgynous is good- Persholosical_8222rts. 1981. 18. 267-272. Burke, R. J. Personality, self-image, and informal helping processes in work settings. Psxcholosical_ae22rts. 1982, 59, 1295-1302. Cole, E., 8 Kumchy, C. I. The CIP battery: Identification of depression in a juvenile deliquent population. Jeurna1_ef_slinical_zaxcholesx. 1981. 31. 880-884- DeGregorio, E., & Carver, C. S. Type A behavior pattern, sex-role orientation, and psychological adjustment. IQHIDHJ Q: Egzggngljgx gag gagial ESXQRQLQQXI 1980: Q2: 236-293. Finlayson, L. Conversation between author and Ms. Finlayson, a graduate student at Michigan State University, who also is manipulating this data set, on 3-2-84 at 1:30 p.m. Flaherty, T. F., & Dusek, J. B. An investigation of the relationship between psychological androgyny and components of self-concept. 12_rn_l_9f_£ersooalitx_and Social_£sxcbologx.1980.15. 934-992. Flavell, J. H. Cognitive monitoring. In E. Dickson (Ed.), 9hildrsols_Qral.§2mmunication_§kill_. New York: Academic Press, 1981. 78 Gecas, V., Calonico, T. M., 8 Thomas, D. L. Development of self-concept in children - mirror theory vs. model thsorY- 1ournai.of.§ocial.£sxcholosz. 1974. 22. 67-76- Gilbert, L. A. Toward mental health: The benefits of psychological sndrogyny- Professional_zsxchologr. 1931, 12, 29-36. Ginn, R. O. Psychological androgyny and self-actualiz- ation. Ps22holosisal.8eeorts. 1975. 11. 886- Glass, G., and Stanley, T. C. §tatistical Mathods in o o o . Englewood Cliffs, New Educati_n_and_2sxcb_l_sx Jersey: Prentice Hall, 1970. Gonzalez, C. T., 8 Williams, K. E. Relationship between locus of control and sex-role stereotyping. Esygholo- Mm. 1981. :12. so. Gordon, C. Self-conceptions: Configurations in content. In C. Gordon 8 K. J. Gergen (Eds.), Ina_§alfi_ia_§agial Ingagag§1_n. New York: John Wiley, 1968. Gordon, C. Self-conceptions methodologies. Jaagaal_a§ Esrxous_and_nsntal_nisease. 1969. 158. 328-364. Hakes, D. T. v in_§hildrsu 0 es . New York: Springer-Verlag, 1980. Halpin, G., 8 Whiddon, T. The relationship of perceived parental behaviors to locus of control and self-esteem among American Indian and White children. Journal a; Social_PsYcholosY. 1980. 111. 189-195- Hanes, B., Prawat, R. S., 8 Grissom, S. Sex role percep- tions during adolescence. IQEIDAI_21_EQQEALIQDAI WI 1979: 11.: 850-855- Heilbrun, A. B. Exploration of antecedents and attributes of androgynous and undifferentiated sex-roles. Jgarnal of_§snstic_2sxchologx. 1978. 112. 97-107- Heilbrun, A. B. Gender differences in the functional link- age between androgyny, social cognition, and competence. so oc Psychology. 1931. Al. 1105-1118- Helmreich, R., 8 Stapp, J. Short forms of the Texas Behavior Inventory, an objective measure of self- sstssm. 8ullsIin_2f_ths_2sxch2n2mic_§ocietx. 1974. s. 473-475. Hoffman, D. M. 8 Fiddell, L.S. Characteristics of andro- gynous, undiffereniated, masculine, and feminine middle- class women. §a3_galaa, 1979, a, 765-781. 79 Hogan, H. W. Measurement of psychological androgyny- sxtcnd replication. J2uroal_of_slioical_£sxchologx 1977, 11, 1009-1013. I Hyde, J. S., 8 Phillis, D. E. Androgyny across the life span. DE!§192E§DLA1.E§¥QDQlQSY. 1979. lé. 334-335- Kanoy, R. C., Johnson, B. W., 8 Kanoy, K. W. Locus of con- trol, and self-concept in achieving and underachieving bright elementary students. Psxchologx_in_the_§chools. 1980, 11, 395-399. Kelly, T. A., 8 Worell, T. New formulations of sex-role and androgyny-Critical review. 0 and_§linical_fsxchologr. 1977. 45. 1101-1115. Kendall, P. C., 8 Finch, A. J. Jr. Developing nonimpulsive behavior in children. In P.C. Kendall 8 S. D. Hollon (Eds.), ' ve- ° , BQEQAIQDI_ADQ_EIQQQQDI2§o New York: Academic Press, 1979. Kuhn, M. H. Self-attitudes by age, sex, and professional training- Sociological_euartsrlx. 1960. 2. 39-55- Kuhn, M. H., 8 McPartland, T. S. An empirical investi- gation of self-attitudes. Ana:19an_§agiglagiga1 332131. 1954, 12, 68-76. , Kuo, 8. Relationship between teacher expectations and internal-external control in elementary school children. 8ullatin.ef_zducational_£srshologx. 1982. 15. 139-147. Lamke, L. K. (a) Adjustment and sex-role orientation in adolescence. l9urnal_of_Youth_and_Adolescsnce. 1982. 11, 247-259. Lamke, L. K. (b) The impact of sex-role orientation on self-esteem in early adolescence. §h11a_nayalapaaa§, 1982, :1, 1530-1535. Lamke, L. K., Bell, N. J., 8 Murphy, C. Sibling constel- lation and androgynous sex-role development. Jaagnal_ 91 EEXSDQLQEYI 1930. 125. 139-144- Larned, D.T., and Miller, D. Development of self-concept in grades one through nine. Ths_Journal_of_£sxcholosxl_ 1212, 1. 143-155. Lefcourt, H. M. Internal versus external control of rein- forcement: A review. Psychological_nulletin. 1966. as, 206-220. 80 Logan, D. D., 8 Kaschak, E. The relationship of sex, sex- role, and mental health. Psychology of Woaaa QQBILRIIY. 1980. 1. 573-580- Lubinski, D., Tellegen, A., 8 Butcher, J. N. Masculinity, femininity, and androgyny viewed and assessed as distinct concopts- J2uraal_of_£srsooalitx_aod_§ocial ESYEAQIQSY. 1983. Al. 428-439- Maqsud, M. Relationships of locus of control to self- esteem, academic achievement and prediction of per- formance among Nigerian secondary school pupils. a O , 1983, 5;, 215-221. McLaughlin, B. S. The Who Am I Dictionary and self- perceived identity in college students. In P. J. Stone, D. Dunphy, M. S. Smith, 8 D. M. Ogilvie (Eds.), e ° 0 e Analyaia. Cambridge: MIT Press, 1966. Meichenbaum, D., 8 Asarnow, J. Cognitive-behavioral modif- ication and metacognitive development: Implications for the classroom. In P.C. Kendall 8 S. D. Hollon (Eds-1. Q2gnitixs:8ehaxi2ral_Intsrxentionsi__Thsorx. . New York: Academic Press, 1979. ‘ Mennigerode, F. A. Attitudes toward women, sex-role stereotyping, and locus of control. anghalagiaal nggrgg, 1976, 18, 1301-1302. Mischel, W. A social-learning view of sex differences in behavior. In E. E. Maccoby (Ed.), Ina_gayalgpaaaa_afi fiax_nif1a:anga§, Stanford: Stanford University Press, 1966. Mitchell, K. Conversation between author and Ms. Mitchell, teacher at schools participating in the study and one of the individuals who formed the study's design and implementation, on 2-7-84, at 10:30 a.m. Moore, 8. M., 8 Rosenthal, D. A. Sex-roles: Gender, gen- eration. and self-esteem. Australian_2sxcholesist. 1980, 15, 467-477. Myers, A. M., 8 Gonda, G. Empirical validation of the Ben Sex-Role Inventory. EEXQDQlQQY. 1932. $2. 304-313- Myers, A. M., 8 Gonda, G. Utility of the masculinity- femininity construct: Comparison of traditional and androgyny approaches. on o , 1982, A}, 514-522. 81 Novaco, R. W. The cognitive regulation of anger and stress. In P. C. Kendall 8 S. D. Hollon (Eds.), - v : o c ang_2:aaaaa;aa. New York: Academic Press, 1979. ' Nowicki, S. Factor structure of locus of control on children. Iournal_of_§enstis_£sxshologx. 1976. 122. 13-17. Nowicki, S., and Schneewind, K. Relation of family climate variables to locus of control in German and American studsnts. Journal_of_§snetic_£srsholosx. 1982. 151,(2), 277-286. Nowicki, S. 8 Strickland, B. R. A locus of control scale for children. EEIEDQIQQY. 1973. 22. 148-154- O'Conner, K., Mann, D. W., 8 Bardwick, J. M. Androgyny and self-esteem in the upper-middle class: A replication of Spence. u EEYEBQIQQY. 1973. 22. 1158-1159- Orlofsky, J. L. Relationship between sex role attitudes and personality traits and the Sex Role Behavior Scale-1: A new measure of masculine and feminine sex role behaviors and interests. and_Sosial_£sYshologY. 1981. 42. 927-940- Orlofsky, J. L. Psychological androgyny, sex-typing, and sex-role ideology as predicators of male-female interpersonal attraction. fiax_gglaa, 1982, a, 1057-1073. Piaget. J. 2usesss_and_nndsrstandins- Csmhridgs. Mass: Harvard University Press, 1978. Puglisi, J. T., 8 Jackson, D. W. Sex Role identity and self esteem in adulthood. International_lournal_of Asins_and_numan_nsxeloemsnt. 1980-81. 12. 129-138- Pyke, S. W., 8 Graham, J. M. Gender schema theory and androgyny: A critique and elaboration. Iaaaznagiaaa; 12urnal_of__2men_s_§tudies. 1983. s. 3-17. Richter, A. L., Reaves, M. G., Deaver, H.D., 8 Lacy, S. G. Social stereotypes as a variable in egocentrism. Journal_of_zarlY_Adolsscease. 1982. 2. 173-133. Rost, W., Neuhaus, M., 8 Florin, I. Bulimia nervosa: Sex- role attitudes, sex role behavior, and sex role related locus of control in bulimorexic women. aaagna1_ag Psychosomatis_asssarch. 1982. 22. 403-408. 82 Rotter, J. B. a New York: Prentice-Hall, 1954. Rotter, J. B. Generalized expectancies for internal versus external control of reinforcement. anahalagigal Manggrapha, 1966, Whole, No. 609, 59, 1-28. Sadowski, C. J., 8 Wenzel, D. M. The relationship of locus of control dimensions to reported hostility and aggression. Journal_of_zsrchologx. 1982. 112. 227-230- Schneider, T. Conversation between author and Ms. Schneider, a teacher at schools which participated in the study, on 3-7-84. Spence, J. T., Helmreich, R., 8 Stapp, J. Ratings of self and peers on sex role attributes and their relation to self-esteem and conceptions of masculinity and fonininity- J22rnal_9f_PsrsonalitY_and_§osial EEYSDQIQQY. 1975. 22. 29-39. Spiegel, P., Jackson, L. A., 8 Stollak, G. E. Masculinity, femininity, and components of self esteem. Unpublished dissertation, Michigan State University, 1982. Sprafkin, C., Serbin, L. A., 8 Elman, M. Sex-typing of play and psychological adjustment in young children: An empirical investigation. Isurnal_of_Ahnormal_thld WI 1982: 1.9! 559-5680 Stericker, A. B., 8 Kurdek, L. A. Dimensions and correlates of third through eighth grader's sex-role self-concepts. fiax;gala§, 1982, 8. 915-929. Stipek, D. Social-motivational development in first grade. C2ntsma9raIY_Educational_zsxchologx. 1981..§. 33-45. Topol, P., 8 Reznikoff, M. Perceived peer and family relationships, helplessness, and locus of control as factors in adolescent suicide attempts. L11§_Ih£§s§22183_§§h22121. 1982. 12. 141-150- Walden, T. A., 8 Ramey, C. T. Locus of control and academic achievement: Results from a preschool intervention program. Journal_gf_zdusational Esxshologx. 1983. 12. 347-358- Waters, C. W., Waters, L. K. Pincus, S. Factor analysis of masculine and feminine sex-typed items from Ben Sex-Role Inventory. Psychological_8222rts. 1977. 42. 567-570. ‘ Watkins, D. Sex role perceptions of Filipino adolescents. 1ntsrnati2aal_Josrnal_of_BsYshologz. 1982. 11. 359-368- 83 Whitley, B. E. Sex role orientation and self-esteem: A critical meta-analytic review. Jougaal a; Rexsgaaligy and_Social_PsYshologY. 1983. .4. 765-778- Williams, J. E., Bennett, S. M., 8 Best, D. L. Awareness and expression of sex stereotypes in young children. Dexeloumsotal_£sxshologx. 1975. 11. 635-642. Wolf, T. M., Sklov, M.C., Mac D. Hunter, S., 8 Berenson, G.s. Factor analytic study of the Children's Nowicki- Strickland locus of control scale. Eaaaaaiaaal_aag 2sxcholosisal_Measurement. 1982. 42. 333-337. Wo, T. The relationship between moral judgement and attribution for male and female adolescents. 22112212 9f_Edusational_2§Yshologx. 1982. 12. 149-165- 84 APPENDICES 85 APPENDIX A CHILDREN'S VERSION OF THE BEM SEX-ROLE INVENTORY 86 Code No.& 2:54: 23’ ZZZ. The sentences below ask about what kind of person you think you are. After each sentence there are four choices: Never Sometimes Often Always 0 0 0 0 If the sentence was "I like to smile and laugh" and you thought you were a person who was often this way you would blacken in the circle under the word "Often". If you thought you were a person who was sometimes this way you would blacken in the circle under "Sometimes". Read each sentence and blacken in the circle under the word that best describes the way you think you are. 1. I stand up for what I think is right. Never Sometimes Often Always O O 0 O 2. I feel real happy one day and sad the next. Never Sometimes Often Always O O O 0 3. I like to hug and kiss. Never Sometimes Often Always' 0 0 0 0 4. I depend on myself for what I think and do. Never Sometimes Often AlwaY8 O O O O 5. I am fair and do right towards others. Never Sometimes Often Always O O 0 O 6. I act like girls do. Never Sometimes Often Always O O O O 7. I say what I like or don't like. Never Sometimes Often ' Always 0 0 0 0 8. I can be counted on to do things. Never Sometimes Often Always 0 0 O O ll. 12. 13. 14. 15. 16. 17. 18. 19. 87 I care about how other people and animals feel. Never Sometimes Often O O O I make a big impression on other people. Never Sometimes Often O O O I am jealous. Never Sometimes Often O O O I am sensitive to the needs of others. Never Sometimes Often O O O I push to get my own way. Never Sometimes Often O O O I am truthful. Never Sometimes Often O O O I understand how other peOple think and Never Sometimes Often O O O I am able to lead people. Never Sometimes Often O O O I keep a lot of secrets. Never Sometimes Often O O O I want to help other people in trouble. Never Sometimes Often O O O I am willing to take risks. Never Sometimes Often 0 O 0 feel. Always Always Always Always Always Always Always Always Always Always Always (2) 88 20. I brag a lot about myself. Never Sometimes Often Always O O O O 21. I am eager to soothe hurt feelings. Never Sometimes Often Always O O O O 22. I try to make other people do what I want. Never Sometimes Often Always O O O O 23. I can quickly learn to like something new. Never Sometimes Often Always O O O O 24. I show people that I like them. Never Sometimes Often Always O O O O 25. I act like boys do. Never Sometimes Often Always O O O O 26. I am.careful to say unpleasant things in a way that the other person won't be hurt. Never Sometimes Often Always O O O O 27. I don't treat others roughly. Never Sometimes Often Always O O O O 28. I am.willing to take a stand. Never Sometimes Often Always O O 0 0 29. I act and think like everyone else. Never Sometimes Often Always O O O O 30. 31. 32. 33. I show people gently that I care about them. Never 0 I am always ready to fight or yell at someone. Never 0 Sometimes 0 Sometimes O I love children. Never 0 I like to play games to win. Never 0 Sometimes 0 Sometimes O 89 Often O Often O Often Often Always Always Always Always (4) 90 APPENDIX B THE INTERNAL-EXTERNAL LOCUS OF CONTROL SCALE FOR CHILDREN 10. 11. 12. 13. 14. 15. 16. 91 We are trying to find out what boys and girls your age think about certain things. We want you to answer the following questions the way vou feel. There are no right or wrong answers. Don't take too much time answering any one question. but do try to answer them all. Do you believe that most problems will solve themselves if you just let them? . C o o C O O I O I o o o o o I O O 0 Do you believe that you can stop yourself from catching a cold? 0 o o O O o o O o o O O I O O O 0 o O 0 O O O O 0 Are some kids just born lucky? . . . . . . . . . . . . . . . Most of the time do you feel that getting good grades means a great deal to you? . . . . . . . . . . . . . . . . Are you often blamed for things that just aren't your fault? Do you believe that if somebody studies hard enough he or she can pass any subject? . . . . . . . . . . . . . . . . Do you feel that most of the time it doesn't pay to try hard because things never turn out right anyway? . . . . . . Do you feel that if things start out well in the morning that it's going to be a good day no matter what you do? . . Do you feel that most of the time parents listen to what their children have to say? . . . . . . . . . . . . . . Do you believe that wishing can make good things happen? . When you get punished does it usually seem it's for no good reason at all? . . . . . . . . . . . . . . . . . . . . Most of the time do you find it hard to change a friend's mind or opinion? . . . . . . . . . . . . . . . . . . . . . Do you think that cheering more than luck helps a team to win? 3 . . . . . . . . . . . . . . . . . . . . . . . Do you feel that it's nearly impossible to change your parent's mind about anything? . . . . . . . . . . . . . . . Do you believe that your parents should allow you to make most of your own decisions? . . . . . . . . . . . . . . . . Do you feel that when you do something wrong there's very little you can do to make it right? . . . . . . . . . . yes . yes yes yes . yes . yes yes 'es '4' “1 “I (1 no no no no no no no no no no no no SC no b 17s 18. 19c 20. 5321-" 22. 23. 2“. '26 . 27 e 23. 29. 30. 36. 3?- 38. 39- b0. 92 Page 2 Do you believe that most kids are Just born good at sports? . yes no Are most of the other kids your age stronger than you are? . . yes no Do you feel that one of the best ways to handle most problems - is Just not to think about them? . . . . . . . . . . . . . . . yes no Do you feel that you have a lot of choice in deciding who your .fl'londfl ”.7 e,'e 0:.e e e e e e e .e 0.2. e e e e e e e e e yes no If you find a four leaf clover do you.believe thatit might bring you good luck? .r. .t. . . . .H. . ... . . . . . . . . . yes no Do you often feel that whether you do your homework has much to do with what kind of grades you get? . .'. . . . . . . . . yes no Do you feel that when a kid your age decides to.hit you. there's little you can do to stop him or her? . . . . . . . . yes no Have you ever had a good luck charm? . . . . . . .'. . . . . . yes no Do you believe that whether or not people like you depends on how you act? . . . . . . . . . . . . . . . . . . . . . . . yes no Will your pirents usually help you if you ask them to? . . . . yes no Have ou felt that when people were mean to you it was usual y for no reason at all? . . . . . . . . . . . . . . . . yes no Most or the tine.do you feel that you can change what might happen tomorrow by what you do today? . . . . . . . . . . . . yes no Do you believe that when bad things are going to happen they Just are going to happen.no matter what you try to do to stop them? . . . . . . . . . .1. . . . . . . . . . . . . . . . yes no Do you think that kids can get their own way if they just keoptrylng?....c......o.............,...ycs no Most of the time do you find it useless to try to get your own way at home? . . . . . . . '.- . . . . . . . . . . . . . . yes no Do you feel that when goodithings happen they happen because ‘. thlrdwork?..-.........'o............yesno Do you feel that when somebody your age wants to be your enemy there's little you can do to change matters? . . . . . . yes no Do you feel that it's easy to get friends to do what you‘ Ntht“thcutb?'...3o.o..o_..o............yesno Do you usually feel that you have little to say about what YMb’gOt to '8‘ ‘thouC? e '..' ,e- e e 010 e.e~ e eye‘s... e e e o yes no Do you feel that when someone doesn't like you there's little ‘you can do..b°ut it? e .0 O I ‘ O I O O ' O O O O ' O O O 0 O yea no Do you usually feel that it's almost useless to try in school because most other children are Just plain smarter tmyOU?eeeeebeeeeeeeeeeeeeeeeeeocyanno Are you the kind of person who believes that planning ahead ”“38 thing! turn out bflttOr? e e e e e e e e e e e e e e e e 3'30 no Most of the time. do you feel that you have little to say about what your family decides to do? . . . . . . . . . . . . yes no Do you think it's better to be smart than to be lucky? . . . . yes no 93 APPENDIX C THE WHO AM I 94 Code No. There are fifteen numbered blanks on the page below. Please write up to fifteen different answers to the simple ' question. "Who Am I?" in the blanks, answering as if you were giving the answers to yourself, not to somebody else. Write the answers in the order that they occur to you. Don't worry about “logic" or “importance." H e m m \l 05 0"! ¢ 00 N e H O 12. 13. l4. 15. 95 APPENDIX D THE TRAINING MANUAL FOR CODING THE WHO AM I 96 TRAINIi‘G MANUAL FOR THE NHO AW I QLESTIONAIRE 'This packet contains the necessary information for classifying subjects responses onto categorization sheets (example attached), as performed by Gordon in his l968 study of the use of the Who Am I form. The purpose of the manual and the research that spawned it is to empirically delineate developmental trends in the metacognitive and metalinguistic ability of children and adolescents. 97 INTRODUCTION: Recent years have seen rapid expansion in research investigating the (cognitive development of children. Two "hot" topics among the many under this rubric are metacognition and metalinguistics. Metacognition is usually described as knowledge about one's own thinking, i.e., cognitions about cognitions. Likewise, metalinguistics can be defined as cognitions about one's own use of language. ‘Within this body of research one particular area of snafu has occurred in that researchers are not at all convinced that a child knows only what s/he says, s/he knows. (Hakes, l980) Thus, there is considerable debate generated around the issue of whether a child is held back by his/her in- ability to report on what s/he is thinking. In theoretical terms, the linguistic and metalinguistic ability of a child may lag behind his/her metacognitive development. (See Bever, 1970; Flavell, l981; Hakes, l980; and Piaget, l978 for in-depth discussions on these phenomena) DESIGN: Regardless of how this matter is theoretically stated, an empirical base must be laid which demonstrates the child’s metacognitive and meta- linguistic level of sophistication. To accomplish this goal a data bank of approximately 450 middleschool students (sixth and eighth graders) will be utilized by analyzing the statements they provide regarding themselves on the Who Am I questionaire (which will be introduced and explained in the training sessions). It is obvious and taken for granted that these children (and nearly everyone) have thoughts about themselves which are cognitions. They may reflect on who they think they are , and these are metacognitions. Then, asking them to tell about themselves would be metalanguage (also meta- communications), since they must write and think about themselves and reflect on their self-statements as they fill out the questionaire. It is hypothesized that developmental differences will be demonstrated between the sixth and eighth graders. (This data may be compared to high school students, but only in a later study). The differences are expect- ed not just in vocabulary, but in concepts, categories, and other diverse self-referencing statements and criteria. The data of the Who Am Is will be coded according to the following descriptive categories, as designed by Gordon (1968). (Refer also to attached grid system) - Detailed Categories With Typical Examples: A. Ascribed Characteristics. FUndamental to the idea of social identity used in this section are the ascribed role and category designation conferred on the individual at birth, which typically remain with him/her throughout life. l. §gx; aman, a boy, a son, clear name that is a reference to one's sex, etc. (almost always categorical). 98 2. Age; 15 years old, a boy, young, a teenager, in sixth grade, etc. (at least public if not exactly categorical). 3. Name: John Jones, Clarie, etc. (at least public and partly categorical). 4. Racial 95_National Heritage: a Negro, white, a chinese (meaning ancestry or race, not current citizenship); of Italian, Irish ancestry; an immigrant, etc. (usually categorical). 5. Religious Categorization:A Catholic, Protestant, Metho- dist, Jewish, etc. ( not just "Christian", "atheist", etc. , must be definite religious group) (predominantly categorical). B. Roles andlflemberships. The other basic set of social identity elements is comprised of roles and categorical designations which are to an appreciable degree under the control of the in- dividual. 6. Kinship Role. a son, mother, sister, aunt, daughter, etc. (also coded’as sex references), engaged, going steady, etc. (generallyccategorical). 7. Occupational Role: specific occupation, employed, working part-time, Baby-sitter, hoping to become a doctor, etc. (mainly categorical). 8. Student Role. a student, at South Boston High, getting bad grades, going tb Harvard, taking four courses, trygng to get into a good college, etc. (mainly categori- ca 9. Political Affiliation. a Democrat, an Independent, other clear party*(hot liberal, conservative, etc.) (almost always categor1calL l0. Social Status: from a poor family, an elite neighborhood, middle class an aristocrat, of an old- line family, etc. (somewhat more frequently attributive). ll. Territoriality, Citizenship: now a Michigan resident, liv1ng on oak’St. , a Bostonian, from Alabama, an American, a German (current citizenship, not "heritage"), a foreign student, etc. (usually categorical). 12. Membership in Actual InteractingiSroup: on the football team, in the science club, at Hannahliiddle School, a friend, in a clique, member of a certain family, etc. (almost always categorical). C. Abstract Identifications. A very interesting set of three dimensions contains those that (although often categorical rather than attributive) are usually too abstract or too private to serve as distinct social identity elements. The first of these portrays the individual as a unique, irreducible particle of Being, not definable by reference to anything outside him/herself. The second type places the person in some universal or very large and abstract category, without implication of interaction among members. The third form associates the person with some relative- ly comprehensive idea system, whether theoretical, philosophical, idealogical, religious, or more narrowly political. 99 13. Existential, Individualing: me, an individual, an existing being, myself, nothing, unique, undefinable, etc. (generally the denial of categories). 14. Membership in gg_Abstract Category: a person,a human, a teenager (disc an age reference) a speck in the cosmos, etc. (almost exclusively categorical by definition). l5. Ideological and Belief References: a liberal, a conser- vative, a Christian, very religious, a pacifist, against the middle eaSt conflictggnot prejudiced, etc. (somewhat more frequently categorical). D. Interests and Activities. It is often very difficult to dis- tinguish among the various forms of personal connection to ob- jects outside the self, and the cognitive, cathectic, and active modes are frequently blurred. .Nevertheless, there are good grounds for distinguishing references to judgements of quality, concern over intellectual questions of meaning and substance, and actual participation in activities. 16. Judgements, Tastes, Likes: one who likes acid rock, abstract art, hates jazz, loves Bach, rock'n'roll, etc. (usually attributive). 17. Intellectual Concerns: getting an education, a reader, aithinker, an'intellectual, interested in literature, trying to understand modern theater, etc. (generally the verb-form of an attribute). l8. Artistic Activities: a painter, dancer, poet, musician, singer, céllo player, etc. (usually a category refer- ence the noun form of a verb). 19. Other Activities: a football player, athlete, hiker, stamp callector, movie goer, one who dates a lot, a good swimmer, etc. (categorical on the whole). ‘ E. Material References.. There have always been references to the body as a primary object of self-conscious awareness, but William James was among the first to point out the importance of other material objects as elements of identification. 20. Possessions, Resources: one who has pretty clothes, hoping for a secure future, one who never has enough money, goin to own a car someday, etc. (usually categorical]. 1 k 21. Ph sica Se f, Bod Ima e: good- 00 ing, pretty, strong, tall, 5'5", too t 1n, Blonde, healthy, ugly, llZ lbs., etc. (preponderantly attributive). F; Major Senses of Self. Categories 1 to 21 of this coding scheme are designed to encapsulate the meaning of relatively specific self-representations. 22. Sense of Com etence: intelligent, smart, talented, creatiVE} skillful, low in ability, good at many things, always making mistakes, etc. (primarily attributive). 23. Sense gf_Self-Determination: trying to get ahead, deciding things for myself, ambitious, hardworking, a self-starter, etc. (almost always attributive). 100 24. Ihg_Sense gf_Uhity: in harmony, mixed up, confused, ambivalent, a whole person, straightened out now, etc. (predominantly attributive). 25. Ihg_$ense 2f Moral Worth: self-respecting, a sinner, bad, good, honest, reliable, trustworthy, responsible, evil, a thief, etc. (preponderantly attributive). G. Personal Characteristics. As we move from the categories and roles of social identity through the interests, activities, material references, and the senses of self, there is a large and interesting set of more general self-descriptions which refer to the individual's typical manner of acting and his typical style of psychic functioning. 26. Interpersonal Style (how I_typically act): friendly, fair, nice, shy, introverte , ar to get along with, quiet, demanding, good with children, affectionate, cool, etc. (almost exclusively attributes). 27. Psychic Style, Personality (how I typically think and feel): ludky, happy, dadjimoody, a daydreamer, in love, depress- ed, confident, "crazy", lonely, curious, calm, searching for love, mature, objective, optimistic, etc. (pre- dominantly attributes). H. External References. Two remaining categories of relatively infrequent Who Am I elements refer not to the individual him/ herself in any typical or continuing manner, but rather to the impression s/he feels that s/he makes on others, or to the immediate testing situation itself. References to the impressions or attitudes of others toward the respondant are actually repre: sentations of them notahim/her. r ‘ , '"28. Judgements Im uted tg_0thers: popular, respected, well- liked, well t oug t of, loved, etc. (preponderantly attributive). 29. Situational References: tired, hungry, bored, filling out this questionaire, oing out tonight, late for dinner, finished, etc. usually attributive). 3D. Uncodable Responses: superman, President of the U.S., a flower, the sea, a shell on the beach, etc. (usually categorical, if decipherable at all). In addition to this coding, the statements of each individual will be jud ed as being positive (+ in the gridwork), negative(-), or neutral 0). To ensure uniformity among coders, the following criterion will be used for this procedure: Positives will be given precedence over negative and neutral statements. Thus, the following list is considered all "positives": loves sports; a friend; wants a good education; likes basketball; helping; kind; friendly; caring; not shy; funny; etc. In sum, any descriptor that could be positive is so labeled. In using the grid system (see attached example), each statement is represented by its category designation with the appropriate symbol (+, -, or O ). Each statement may be represented several times because of the various meanings attributable to any phrase or word. 101 Note, for example, that "boy" is both an indicant of youth (age) 'and gender. Wiale" would be only gender indicative. Practice sessions of coding will be arranged and carried out with surplus Who Am Is until coders achieve .90 agreement in coded re- sponses with the trainer. At that time coding will begin on the middle-school students' questionaires. 102 REFERENCES Bever, T.G. The cognition basis for linguistic structures. In J.R. Hayes (Ed.), Cognition and the Development gf_Language. New York: Wiley, l970. Flavell, J.H. Cognitive monitoring. In E. Dickson (Ed.), Children's Oral Communication Skills. New York: Academic Press, 1981. Gordon, C. Self-conceptions: Configurations in content. In C. Gordon and K.J.(3ergen (Eds.), Ihg_Self jg_Social Interaction. New York: Wiley, 1968. Hakes, D.T. The Development gflietalinguistic Abilities jn_Children. New York: Springer-verlag, l980. Piaget, J. Success and Lhderstanding. Cambridge, Mass: Harvard University Press, 1978. Ascrihed Characteristics ' 1. Sex l()3 10 11 12 13 14 15 ll TOTALS 4» - own Age Name éUN Racial or national heritage (.7! . Religious categorization Roles and Memberships 6. Kinship roles 7. Occupational role 8. Student role 9. Political affiliation 10. Social status 11. Territoriality, citizenship 12. Actual group nnnbership Abstract Identifications 13. Existential. individuating 14. Abstract category 15. Ideological and belief references Interests and Activities 16. Judgments, tastes, likes 17. Intellectual concerns 18. Artistic activities 19. Other activities Material References 20.- Possessions. resources 21. Physical, body image Major Senses of Self 22. Competence 23. Self-determination 24. Unity 2'. Moral worth Personal Characteristics 26. Interpersonal style '..' 27. Psychic style. personality External References 28. Judgments imputed to others 29.. immediate situation references 30. Uncodable responses___ CODE NO. _1 2 3 a 5 6. 7 9 SCORER INITIALS & NO. 10 11 12 13 16 15 GRAND TOTALS