WNWWIIIIHWHHWIIHHIH’WNWHHIIIHWI ICC—x _mooo '—1_‘_\ llllllllllflllllllgllgllllljlglgllll This is to certify that the thesis entitled THE EFFECTS OF THE VERBAL ENVIRONMENT 0N SELF-ESTEEM IN YOUNG‘=CHILDREN presented by Kimberly Ruth Michaud has been accepted towards fulfillment of the requirements for MA Child Development degree in Date fl/él/ yfl 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return 'on or before date due. DATE DUE DATE DUE DATE DUE II p... . n() , ,1 ; \J :r‘illli l' » n 9 if? gamma 5;; gjglfikz '5 AW 200% ‘ __ ll rfl l J ——l| MSU Is An Affirmative Action/Equal Opportunity institution ammo-9.1 .._ _.._______._ __ THE EFFECTS OF THE VERBAL ENVIRONMENT 0N SELF-ESTEEM IN YOUNG CHILDREN By Kimberly Ruth Michaud A THESIS , Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Family and Child Ecology 1990 ,, ‘ n} ,4 ‘7 (*Ci Q47“ ABSTRACT THE EFFECTS OF THE VERBAL ENVIRONMENT ON SELF-ESTEEM IN YOUNG CHILDREN By Kimberly Ruth Michaud Researchers have found several components of the human behavioral environment have an effect on self-esteem development. The purpose of this research is to examine the relationship between the verbal environment within the child care classroom and the self-esteem of the children in the class. The sample consisted of 50 children aged 4-7 from six child care classrooms in the Lansing area of Michigan. Self- esteem was measured on 50 children and observations were made to rate the verbal and physical environments of six classrooms selected by administrator ratings of verbal environment to locate extremes within centers. Analysis of the data revealed a significant difference in classrooms with a positive verbal environment and those with a negative verbal environment with regards to self-esteem when other factors were taken into account. Age was found to be a significant factor in this difference. The physical environment was not found to be significantly related to the verbal environment. DEDICATION This thesis is dedicated to my family: to my brothers, sisters and their families for providing encouragement and support throughout my education and especially for serving as examples to me and teaching me that any education is of value, and mostly to my parents Terry and Doris Michaud for their continual financial and emotional support of my educational pursuits, for providing me with an optimal home environment and especially for their love. iii ACKNOWLEDGMENTS I would like to thank Dr. Alice Whiren for her continued guidance throughout the writing of this thesis. Also, I wish to express my appreciation to Dr. Marjorie Kostelnik and Dr. Tom Luster for their assistance and suggestions. The committee's support was invaluable. Special thanks are also extended to Dr. Praveen Kaur, Lynn Navin and especially Cheryl Baas for their participation in the data collection process. Thanks to Dennis Baese for the use of his computer. Also, thanks to Julie Curtis for her statistical expertise. Finally I would like to thank the administrators, teachers and children who served as subjects in the study. iv TABLE OF CONTENTS LIST OF TABLES CHAPTER I STATEMENT OF THE PROBLEM . Limitations II REVIEW OF THE LITERATURE III METHODS Subjects Research Design Instrumentation The Purdue Self-Concept Scale Early Childhood Environment Rating Scale. The Verbal Checklist. The Administrative Rating Form Research Hypothesis Data Collection Data Analysis IV ANALYSIS OF THE RESULTS Testing the Hypothesis. V DISCUSSION AND IMPLICATIONS Summary Suggestions for Further Research BIBLIOGRAPHY Page vii 16 16 18 19 19 20 22 23 24 24 25 27 27 30 32 33 35 APPENDIX APPENDIX APPENDIX APPENDIX DOW) APPENDIX E APPENDIX APPENDIX APPENDIX Im'l'l ADMINISTRATIVE RATING FORM VERBAL BEHAVIOR CHECKLIST . PURDUE SELF-CONCEPT SCALE HARMS EARLY CHILDHOOD ENVIRONMENT 'RATING SCALE ADMINISTRATOR CONSENT FORM TEACHER CONSENT FORM PARENT CONSENT FORM RAN DATA vi 38 4O 41 43 44 46 49 52 LIST OF TABLES Table Page 1 Sociodemographic Characteristics of the Sample . . . 17 2 Independent Samples T-Test on Purdue by DR . . . . . 28 3 Regression Analysis . . . . . . . . . . . . . . . 29 CHAPTER I STATEMENT OF THE PROBLEM Although much reseach exists regarding classroom environment and climate, the focus of these studies seems to be on such variables as classroom temperature, color scheme and physical layout (Moos, 1979; Smith, Neisworth & Greer, 1978). Relatively few studies examine the verbal exchanges which take place within the classroom. Adverse verbal encounters tend to make children feel inadequate, angry or confused (Kostelnik, Stein, Nhiren & Soderman, 1988; Hoffman, 1963) while positive verbal exchanges convey warmth, respect and acceptance to children (Kostelnik, Stein, Nhiren and Soderman, 1988). Therefore research focusing on classroom verbal interactions could serve as a valuable resource to educators, providing them with the information necessary to create the environment best suited for promoting self-esteem in children. Past childhood self-esteem research has primarily focused on school aged children or adolescents. Because researchers have found self rating measures difficult to use with preschoolers, few studies can be found which measure 1 2 the effects of environmental variables on self ratings in preschool children. Young children see themselves as being all good or all bad, and this assessment changes as circumstances change. Preschoolers also have difficulties understanding and verbalizing abstract ideas and internal processes like self-esteem (Marshall, 1989). Thus, administering a tool designed to measure such abstract concepts by employing items such as ”I'm not doing as well in school as I would like to" is developmentally inappropriate for preschoolers. In the past 10-15 years, new measures geared down to young children, using picture items or open-ended questioning strategies, have appeared in the literature, some with research support and others needing support for reliability of the tools. This study will use such a tool, The Purdue Self-Concept Scale for Preschool Children designed in 1980 by Victor B. Cicirelli. The purpose of this study is to examine the relationship between the verbal environment within the child care classroom and the self-esteem of the children in the class. From an ecological perspective, this study involves a bidirectional influence between the classroom microsystem and self—esteem of the classroom members. While self-esteem is an internal process, it is influenced by external environmental factors. The study will also examine individual characteristics of classroom members that may 3 influence self-esteem, such as: age of the child, parents' education and household income. VERBAL ENVIRONMENT: Conceptually the verbal environment is defined as the verbal exchanges and silences which take place in a given setting. Verbal environments can be positive or negative. Characteristics of a positive verbal environment include: 1. 10. Adults use words to show affection and interest in children. Adults focus their attention on children as they perform daily tasks and routines. . Adults speak courteously to children. Adults send congruent verbal and nonverbal messages. . Adults use children's interests as a basis for conversation. Adults take advantage of spontaneous opportunities to talk with each child informally. Adults avoid making judgmental comments about children within the child's hearing. Adults extend invitations to children to interact with them. Adults listen attentively to what children have to say. Adults refrain from speaking when talk would destroy the mood of the interactions. (Kostelnik, Stein & Whiren, 1988) 4 Negative verbal environments are characterized by the following: 1. 11. Adults show little or no interest in children's activities because they are busy, hurried or tired. . Adults talk more with peers than with children. . Adults pay superficial attention to children's verbalizations. . Adults are discourteous to children. Adults discourage children from expressing themselves. . Adults use baby talk when talking to children. Adults use judgmental vocabulary when describing children. Adults ask questions for which no real answer is expected or desired. . Adults rely on giving orders and making demands. Adults use children's names as synonyms for the words "no", "stop" and “don't". Adults dominate the verbal exchanges that take place each day. (Kostelnik, Stein & Whiren,1988) Both positive and negative verbal environments are operationally defined by scores given on the Administrative Rating Form (AR)(see Appendix A) and on The Verbal Checklist (see Appendix B). High scores on either 5 instrument indicate a positive verbal environment while low scores indicate a negative verbal environment. SELF-ESTEEM: Conceptually self-esteem is defined as the evaluative component of the self-concept. Self-esteem can also be positive or negative. Some characteristics of a positive self-esteem include confidence in physical abilities (running, catching, bike riding), confidence in intellectual ability (reading, counting, problem solving) and a happy emotional state (Cicirelli, 1980). Some characteristics of a negative self-esteem include lack of confidence in physical and intellectual abilities and a sad emotional state (Cicirelli, 1980). Operationally, self-esteem is defined by the child's own report of his/her self on the 40 item Purdue Self-Concept Scale (Cicirelli, 1980; see Appendix C). PHYSICAL ENVIRONMENT: Conceptually the physical environment is defined as the physical setting wherein the children interact. Characteristics of the physical environment include personal care routines, furnishings and displays for children, language and reasoning experiences, fine and gross motor activities, creative activities, social development and adult needs (Harms and Clifford, 1980). Operationally, the physical environment is determined by the score of the seven subscales of the Harms Early Childhood Environment Rating Scale (Harms and Clifford, 1980; see Appendix D). 6 DEMOGRAPHIC VARIABLES: Socioeconomic status is defined as the annual household income of the child's family, parental education is defined as the numerical grade level attained by both_parents individually, and age is defined as the chronological number of years the child has attained. LIMITATIONS Some limitations are inherent in any research design. One limitation of this study relates to the selection of the test variables. It is impossible to identify and control for all the possible variables, but several variables were controlled in this study including : physical environment, socioeconomic status, parental education, the age of the child and using teachers from the same center to control for location of the center and parental selection of the program. Another limitation relates to the measurement of self-esteem using self-reports by preschool children. The results of such a measure may actually measure the child's verbalization of his/her perceived self-esteem rather than his/her actual perception. In measuring the verbal environment, it is assumed that the session manager and the support adult observed verbally interact in similar ways with different children in the classroom and that their verbal behavior rating is consistent from day to day. CHAPTER II REVIEW OF LITERATURE This chapter will examine the past verbal interaction research as well as examine the previous studies on self-esteem in children. It will show the need for further research in both areas as well as the need for a study examining the relationship between the two areas. Past research of verbal environment has primarily been conducted using samples of junior high, high school and college classrooms. The units of analysis in such studies often were classroom lectures. Using techniques such as interaction analysis (Flanders, I970; Amidon & Hough, 1967) which measures teacher and pupil verbalizations using matrices, researchers have found that student growth is maximized when secondary teacher verbalizations are clear and simple (Harris & Swick,l985; Hiller, Fisher & Kaess, 1969). Harris and Swick (1985) recommended in order to maximize high school student growth, teacher training programs should emphasize verbal teaching behavior. They suggested programs focus on ways to reduce vague terms, 7 8 multiple level questioning behaviors and frequency of questioning behaviors and on ways to increase teacher wait times and lesson clarity. Gorham (1988) found teacher verbal clarity important in college students' perceptions of how much they can learn. She also emphasized the importance of immediacy behaviors, both verbal and non-verbal, in determining students' perceptions of how much they can learn. Verbal immediacy behaviors included such items as "uses personal examples", "addresses students by name", "asks questions that solicit viewpoints or opinions" and "discusses items of interest to the students”. Non-verbal immediacy behaviors included "touches students", "moves around the classroom while teaching" and "smiles at individual students in the class". Honig (1984) reported a positive correlation between Stanford Binet IQ scores and maternal involvement with child and emotional and verbal responsivity of the mother. Further, in the final summative evaluations of several parent-child development centers, mother-child verbal interaction training had a positive effect on Stanford-Binet IQ scores of the children (Andrews, Blumenthal, Johnson, Ferguson, Lasater, Malone & Wallace, 1982; Blumenthal, Andrews & Wiener, 1976; Johnson, Kahn & Leler, 1976) and on several developmental measures including the Purdue Self- 9 Concept Scale (Blumenthal et al.,1976; Johnson et al.,1976). Also, studies have revealed that children's overall competence is related to positive verbal exchanges with significant others. Clarke-Stewart (1973) found verbal exchanges between mother and child to be related to quality maternal care. Further, Levenstein (1970) found children eXposed to stimulating verbal interaction in mother-child dyads made significant cognitive gains in contrast to two comparison groups who were not exposed to such treatment, though these findings were not supported in replications. Levenstein's (1970) study indicates that one way young children gain knowledge is through verbal exchanges with adults. The literature also indicates children gain I self-knowledge through such interactions. Marshall (1989), in a review of the literature, identifies several verbal strategies which are likely to enhance self-concept. First, when adults listen attentively to what children say and ask children for their suggestions, they show children that their ideas are respected which helps children feel they are of value (Marshall, 1989; Kostelnik, Stein, Whiren & Soderman, 1988; Kostelnik & Stein, 1983). Next, when adults help children identify their positive and prosocial behaviors by giving children words to describe their actions, they help children see themselves positively and act accordingly. Marshall calls this a "self-fulfilling prophecy" (Marshall, 1989; Kostelnik, Stein, Whiren & 10 Soderman, 1988; Kostelnik & Stein,1983). Further, primary children benefit from adult use of reflective dialogue (Kostelnik, Stein, Whiren & Soderman, 1988; Schirramacher, 1986; Taunton, 1984). Also, children need to see for themselves that they can succeed. Some children need verbal reinforcement and encouragement, but several studies point to the limitations of verbal praise and persuasion (Dinkmeyer,1963; Potter, 1985; Madden, 1988; Hitz & Driscoll, 1988). Hitz and Driscoll (1988) reviewed several studies examining the effects of praise. They found that some forms of praise have negative effects such as diminishing a child's sense of worth and struggle for independence. The researchers also found that encouragement, which is specific, teacher initiated, non-comparative and sincere, helps students develop an appreciation for their own behaviors and achievements. Madden (1988) suggested that although use of praise is well intentioned, it is also judgemental and manipulative. When teachers deliver praise, student recipients view themselves as having worth. However, their feelings of worth then become contingent on teacher judgements. Madden (1988) also indicated praise is based on achievement and may cause students to be anxious, dependent, and competitive while encouragement is based on specific 11 growth and contributions and promotes within students self-reliance, self-direction and cooperation. Similarly, Potter (1985) suggests teachers stress areas of success while acknowledging areas which need improvement rather than praising incompetent efforts which probably is a disservice to children. Another verbal behavior which seems to foster self-esteem is providing children with choices (Marshall, 1989; Kostelnik, Stein, Whiren & Soderman, 1988; Miller, 1983). When adults offer children choices, children have an opportunity to practice decision making skills. Finally, adults who give rationales for their evalua- tions help children develop a vocabulary for standard setting (Potter, 1985; Kostelnik, Stein, Whiren & Soderman, 1988). Kostelnik and Stein (1983) identify characteristics of both good and poor language environments. Characteristics of a rich language environment include: adults speech furthers children's understanding of language, adults use language to increase children's vocabulary and adults provide opportunities for children to use language. Characteristics of a responsive language environment include: adults use language to respond to children and adults use silence to respond to children. Characteristics of a supportive language enviornment include: adults use language to demonstrate interest in children, adults use language to demonstrate appreciation of children, adults use 12 language to demonstrate acceptance to children and adults use positive language. Characteristics of an instructional language environment include: adults use language to help children become more aware of others, adults use language to help children resolve conflicts with others and adults use language to help children learn more about their world. At the opposite pole, characteristics of a poor language environment include: adults speech inhibits children's understanding of language, adults fail to communicate clearly and adults fail to provide opportunities for children to use language. Characteristics of a non-responsive language environment include: adults fail to respond to children's verbalizations and adults interrupt children's activities. Lastly, characteristics of a non- supportive language environment include: adults use language to shame children, adults use language to demonstrate non-acceptance to children and adults use negative language. While it seems the verbal environment impacts self-esteem, early affective experiences play a key role in determining a component of self-esteem, self-worth (Pelham & Swann, 1989). Before children develop the cognitive capacity to assess specific beliefs about themselves, they learn that their environments are either friendly and satisfying or hostile and frustrating. Children then translate such experiences into a basic sense of pride or shame. 13 Pelham and Swann (1989) claim self-worth is determined by children's positive or negative feelings about themselves, their specific beliefs about themselves and the way they frame these beliefs. Millar (1983) found that self-esteem is built by mastery experiences that provide children with positive feelings about themselves and make them feel they can accomplish tasks. Chance (1982) states that children's perceived competence has little to do with their actual natural ability. He indicates that a strong parent-child bond and a stimulating environment effect self ratings to a greater degree. Marshall and Weinstein (1984) indicate that feedback and evaluation procedures and information about ability were an important part of an optimal classroom environment which has a positive effect on self-evaluations. Bandura's social cognitive theory indicates that psychosocial functioning is regulated by an interplay between internal and external influences (Bandura, 1989). He attests that peoples' beliefs about their ability to attain self-set goals determines whether they view discrepancies between internal standards and external attainments as motivating or discouraging. He indicates that failure to a self-assured person is a motivation to intensify efforts and persist until he/she succeeds, whereas failure to a person with self-doubt is a discouragement and 14 dissuades him/her from persisting with the task (Bandura, 1989). Therefore, self-efficacy is a key internal motivational factor in development. Social cognitive theory involves a triadic reciprocal determinism model of causation (Bandura, 1989). Unlike unidirectional theories which state that behavior is either controlled by external environmental factors or internal processes, Bandura's model shows behavior, internal processes and environment bidirectionally influencing one another. Therefore, while self-esteem may influence behavior, models within the environment may influence self-esteem. This study will examine one dimension of the environment, the verbal dimension. Finding a relationship between this environmental dimension and self-esteem would indicate an external influence on an internal disposition. It was previously stated that very young children perceive themselves as being all good or all bad and this assessment changes as circumstances change. Regardless, even very young children have the cognitive ability to maintain a stable, continuous record of who they are (Eder, Gerlach & Perlmutter, 1987). In summary, it was found that several verbal behaviors of children's significant others seem to be related to children's acquisition of self-knowledge. This study will examine the effects of the verbal 15 environment in the child care classroom upon the children's self-esteem. It is important because relationships found between certain verbal behaviors and self-esteem could help teachers create the environment best suited for promoting self-esteem in children. CHAPTER III METHODS This chapter will describe the methods used in the study. The subjects and the research design will be described and the instrumentation will be discussed. The research hypotheses will be presented and the data collection and data analysis procedures will be described. SUBJECTS This study was conducted in the Lansing area of Michigan. Lansing, located in the south central part of lower Michigan, is the capitol city. It is an industrial center and houses Michigan State University. Six child care classrooms were selected in which to conduct the study. Stratified random sampling was used to select centers from a list of Ingham County centers provided by the Office for Young Children. Centers of over sixty students and centers for the economically deprived were deliberately included. Child care directors were asked to rate all of their classrooms to locate two classrooms that 16 17 exhibit extremes in verbal environment, one being positve and the other negative (see Appendix A). Ten children, aged 4-7, in each classroom were randomly selected to participate in the study. Nine parents and one child declined participation, however, bringing the sample size down from 60 to 50. Table 1 Sociodemographic Characteristics of Sample Total Site Site Site Variables Sample 1 2 3 (N=50) (N=16) (N=21) (N=13) Child's Age Mean 4.646 4.571 4.905 4.615 SD 0.812 0.452 1.019 0.487 Father's education Mean 14.830 11.786 15.550 17.000 SD 3.286 2.006 2673 3.113 Mother's education Mean 14.660 11.071 15.300 16.308 SD 3.192 5.574 1.487 2.839 Median Income $30,000- under $30,000- $30,000- 50,000 $15,000 50,000 50,000 18 Actual participants in the study had a minimum age of 4 and a maximum age of 7 with a mean age of 4.646 (SD= 0.81). Father's minimum grade attained was 9 and maximum was 22 (post-doctoral) with a mean grade attainment of 14.830 (50 = 3.29). Mother's minimum grade attained was 8 and maximum was 22 with a mean grade attainment of 14.660 (SD = 3.19). Household income had a minimum of under $15,000 and a maximum of over $50,000 with a median income of $30,000 - $50,000. The most common income level was over $50,000 which occured in one third of the responses. The majority of the subjects were caucasian with a few of African or Asian descent. These results are summarized in Table 1. RESEARCH DESIGN The main goal of this research was to gain more information about teacher verbal behavior and to examine the relationships between such behaviors and self-esteem in children. This study was a field research study. Data collection took place in the child care classroom. It was a two group comparison study. The primary goal was to determine whether or not children whose teachers have a high observed verbal environment rating have self-esteem scores greater than children whose teachers have a low observed verbal environ- ment when potentially confounding factors are controlled. 19 INSTRUMENTATION Four instruments were used in the study: The Purdue Self-Concept Scale, The Early Childhood Environment Rating Scale, The Verbal Checklist and The Administrative Rating Form. THE PURDUE SELF-CONCEPT SCALE The Purdue Self-Concept Scale assess the child's self-esteem. A test booklet with a drawing of a child that is performing a task well and one of the child that is not performing the task well was shown to the child. He/she was asked to listen to a story about the two children and asked to tell which child he/she is most like. Some of the 40 items the child was asked to rate included: catch a ball, count to twenty, climb high and clothes always look nice (for a complete list see Appendix C). The internal consistency reliability of the instrument was computed for various samples when it was initially administered. It was .80 for three year old whites, .89 for four year old whites, .82 for five year old whites, .86 of combined sample and .88 for four and five year old blacks (Cicirelli, 1980). A two week test-retest reliability was computed and the coefficient was .70. Both the mean and the standard 20 deviation remained relatively constant (Cicirelli, 1980). The reliability coefficient over a six month period was .52 indicating some stability of the measure over time (Cicirelli, 1980). THE EARLY CHILDHOOD ENVIRONMENT RATING SCALE The classrooms were rated using the Early Childhood Environment Rating Scale (Harms & Clifford, 1980). This instrument is used to asses each classroom environment on 37 items that are divided into seven subscales. Subscales included: personal care routines of children, furnishing and display for children, language-reasoning experiences, fine and gross motor activities, creative activities, social development and adult needs (for a complete list see Appendix D). Evaluations were completed on a 7 point scale for each item. In testing the validity of the instrument, the authors invited a panel of national experts in early childhood to rate each item in terms of importance to early childhood programs. Seventy eight percent of the ratings indicated high importance and only 1 percent indicated low importance (Harms and Clifford, 1980). As an additional check on the validity of the instrument trainers rated the quality of 18 classrooms. Scores using the instrument were compared with these rating 21 and a Spearman correlation of .73 was obtained (Harms and Clifford, 1980). In establishing interrater reliability the authors used two independent raters who observed 22 classrooms. A Spearman correlation of .93 was calculated. In a second test of 25 classrooms the correlation was .93. Interrater reliability by classroom was tested on three occasions by two independent observers, one on 22 classrooms, one on 18 classrooms and one on 25 classrooms. The Spearman correlations were .89, .79 and .88 respectively (Harms and Clifford, 1980). Also tests of internal consistency were made by the authors. Alphas for the total scale scores were .81, .93 and .91 for three sets of observations on 22 classrooms (Harms and Clifford, 1980). A test of interrater reliability and a test of test-retest reliability was conducted prior to data collection. Interrater reliability yielded 81 percent reliable with 30 items the same and 7 items only 1 number off. Test-retest reliability measures were taken one month apart. Eighty one percent reliability was found with 30 items the same, 4 items 1 number off, 2 items 2 numbers off and 1 item 3 numbers off. This discrepancy may be due to room restructuring by student teachers during the one month gap. 22 THE VERBAL CHECKLIST Observations were made to rate the verbal environment. The Verbal behavior of the head-teacher/session manager was examined as well as the verbal behavior of one support adult in the classroom. Observations took place on two separate days during free—play period. Verbalizations were recorded in five minute intervals. Interactions were categorized as occuring or not occuring during the five minute intervals. Also, a complete verbal script was recorded. The checklist included 25 items based on the operational definitions of positive and negative verbal environments. Positive items included:invites, listens says please and thank you, bases conversations on children's interests, speaks informally to children, takes advantage of spontaneous opportunities to speak with chidlren, uses silence, sends congruent verbal and non-verbal messages and uses words to show affection to children. Negative items included: talks to colleagues more than children, is busy or tired, pays superficial attention to children, yells, calls names, labels, hushes, demands or orders, asks rhetorical questions, uses children's names as synonyms for "no", "don't" and "stop", uses baby talk, dominates the exchange, uses judgements, parrots, ignores and uses sarcasm. 23 A test of interrater reliability was conducted and a rating of 80 percent was found. 0f the 6 time interval on the 25 items (150 total marks), 120 were rated the same. Eleven items were marked the same in all 6 intervals, 7 items had 5 intervals the same, 1 item had four intervals the same, 3 items had 3 intervals the same and 3 items had 2 intervals the same. THE ADMINISTRATIVE RATING FORM Administrators were asked to rate each of their classrooms on the basis of verbal environment. Directors received rating forms for each classroom in their center serving children ages 4-7. The rating form included 23 items based on the operational definitions of positive and negative verbal environments (see Appendix A). The researcher then coded each form and selected the two classrooms at each site which exhibited extremes in verbal environment in an attempt to control for location of the center and parental selection of the program. 24 RESEARCH HYPOTHESIS To meet the objective of this research the following hypothesis was proposed: H1 Children whose teachers provide a positive verbal environment have self-esteem scores greater than children whose teachers provide a less positive verbal environment when classroom physical environment, child's age, child's sex, father's education, mother's education and family income are controlled. DATA COLLECTION The data were collected over a six week period of time through observations, the Administrative Rating Form, the Early Childhood Environment Rating Scale and through administration of the Purdue Self-Concept Scale. Consent was obtained from administrators, teachers and parents before data were collected (see Appendices E, F and 6) Two raters were periodically used on the observational instruments to determine reliability of the observations, although scores 25 used in the data analysis were all measured by the researcher. The self-esteem measure was administered by the researcher and one other trained rater. One assumption that was made was the score on the Administrative Rating Form (AR) and the observed score on the Verbal Checklist (0R) would be congruent in pinpointing the most positive and the least positive classrooms in each center. This was not the case. The three classrooms rated highest by the administrator's had a mean observed verbal environment of 199 of a possible 300 while the three classrooms rated lowest had a mean score of 190. When dividing the classrooms by scores on the Verbal Checklist, the three classrooms scoring above 195 had a mean observed verbal environment of 209 while the three classrooms scoring below 195 had a mean score of 175. While AR scores were determined without systematic observations, 0R scores were determined from direct observation. Each of the high OR group were from three different sites, thus retaining the control for parental selection. Therefore, the data were analyzed using the 0R designations. DATA ANALYSIS In this study, the analysis of the data included both descriptive and inferential statistical procedures. 26 Descriptive statistics were used to describe the sample and the distribution of responses. Inferential methods used included an analysis of variance, Pearson's Product correlations, t-tests and regression analysis. Regression analysis predicts the value of the dependent variable from the value of the independent variable. Correlation analysis determines if there is an association between two variables. Analysis of variance determines if two groups differ. CHAPTER IV ANALYSIS OF RESULTS This chapter summarizes the results of the statistical analysis conducted on the data. The results will be reported as they relate to the research hypothesis identified in Chapter III. Conclusions and implications of the results and identification of other questions of interest will be discussed in Chapter V. TESTING THE HYPOTHESIS H1 Children whose teachers provide a positive verbal environment have self-esteem scores greater than children whose teachers provide a less positive verbal environment when classroom physical environ- ment, child's age, child's sex, father's education, mother's education and family income are controlled. It was found that children in classrooms with a positive verbal environment scored higher on the average (35.593; SD = 6.577) than those in classrooms with a negative verbal environment (32.273; 50 = 6.606) on the 27 28 Table 2 Independent Samples T-Test on Purdue by GR GROUP N MEAN SD Positive 27 35.593 6.577 Negative 22 32.273 6.606 Separate Variances T = 1.753 DF = 44.9 PROB = .087 Pooled Variances T = 1.754 DF = 47 PROB = .086 Table 3 Regression Analysis Dependent variable: PURDUE N: 47 Multiple R: .384 Squared Multiple R: .147 Adjusted Squared Multiple R:.108 Standard Error of Estimate: 5.301 Variable Coeff STD Error STD Coeff Tolerance T P Constant 30.955 4.825 0.000 6.416 0.000 Age 0.312 1.110 0.045 0.7470132 0.281 0.780 0R 3.994 1.793 0.359 0.7470132 2.228 0.031 Analysis of Variance Source Sum-of-Squares DF Mean-Square F-Ratio P Regression 213.317 2 106.659 3.796 0.030 Residual 1236.385 44 28.100 29 self-esteem measure, although this difference was not statistically significant (T=1.754; p=.086). These results are summarized in Table 2. Regression analysis indicated a significant difference in self-esteem scores of children in positive and negative verbal environments when the child's age was controlled ( F=3.796;p=.030). These results are presented in Table 3. Separate t-test were conducted pairing individual demographic variables with self-esteem scores and no significant differences were found. CHAPTER V DISCUSSION AND IMPLICATIONS The purpose of this research was to examine the relationship between the verbal environment within the child care classroom and the self-esteem of the children in the class. In order to accomplish this objective, the verbal environments of six child care classrooms were examined and self-esteem ratings of the children within the classrooms were taken. Being a pilot study, this research posed the question of the existence of positive and negative verbal environments. The findings indicate that there is a verbal environment in the child care classroom that can be either positive or negative. Further research to determine which factors are key in this differentiation is suggested. Also, refinement of the instruments used to detect positive and negative verbal environments is needed. Past research of teacher verbal behavior has focused on instructional verbalizations. This research indicates teacher/child interpersonal verbalizations are also very important to self-esteem development and therefore to 30 31 learning. It is suggested that teacher training institutions provide prospective teachers with interpersonal communication training and include practicum experiences where they can practice positive verbal exchanges with children. Significant findings were generated by the regression analysis with age and verbal environment as predictors of 3.796; p =.030). Age Purdue Self-concept scores (F—Ratio as a determining factor is an expected finding which concurs with the self-esteem literature (Marshall, 1989). As a matter of interest, an independent samples t-test on the self-esteem measure by age was also conducted. Results showed statistical significance (Pooled Variances T=4.038; p=0.000). This could indicate self-esteem is more constant or more measurable as a child ages. It could also indicate that younger children may elicit or be recipients of adult verbalizations more characteristic of a negative verbal environment. Another possible conclusion is the Purdue Self-Concept Scale more accurately measures (self-esteem in older children. In a study with a larger sample the relationship between age and self-esteem might become clearer. As mentioned in Chapter III, an examination of the results indicated some non-agreement in designation of the six classrooms as positive or negative verbal environments. The two instruments used to determine this designation were 32 both developed by the researcher. Althouh this discrepancy is not a primary consideration of this research, it is an interesting area to eXplore. Pearson's Correlation Matrix revealed no significant correlations with Harms Early Childhood Environment Rating Scores and OR scores. However, when Harms total scores and subscale scores were correlated with AR scores some moderately high correlations were found, specifically between AR scores and the Furnishings subscale (0.751) and between AR scores and the Language subscale (0.643). While it might be expected that Language subscale scores and verbal ratings are correlated, it is interesting that verbal ratings are correlated with the Furnishing sub- scale scores. Specially the Furnishing subscale rates the routine furnishings, the learning and relaxation furnishings, the room arrangement and child related displays. Directors may have been influenced by outward characteristics, such as a physically pleasing environment, when making ratings of the verbal environment. Other plausible explanations for the discrepancy between OR and AR ratings are: non-agreement between observational raters and directors may have occurred in deciding which behaviors were exhibited, inconsistent verbal behaviors may have occurred within a single classroom or directors may have based their verbal ratings on other teacher characteristics besides teacher/child interaction style. 33 SUGGESTIONS FOR FURTHER RESEARCH This research introduced the notion of positive and negative verbal environments. It is suggested that further research be conducted targeting specific elements of the verbal environment to determine which elements have the greatest effect on the differentiation between positive and negative verbal environments. Also, it is suggested that specific elements of the verbal environment be targeted to determine which elements have the greatest effect on self-esteem. ' Additional research should also be conducted examining the verbal environment in the home and its effect on self-esteem and the combined effects of home and school verbal environments on self-esteem. Further, the instruments used to detect differences in the verbal environment should be refined. It is also suggested that similar research be conducted using a larger representative sample. Finally, it is suggested that teacher characteristics such as education, experience with children and interaction training be examined as variables that may accentuate differences in verbal environments teachers provide. 34 SUMMARY These data do tend to support the hypothesis that there is a difference in self-esteem scores of children in positive verbal environments and children in negative verbal environments when controlling for child's age. Significant differences were found in self-esteem scores of children in positive verbal environments and children in negative verbal environments when child's age was considered. It seems plausible that the "verbal environment" does exist and that verbal environments can be positive or negative. Age was found to be significantly related to self-esteem scores. While no significant relationship was found between the physical and the observed verbal environment, moderately high correlations were found between the Furnishings and Language subscales of the physical environment and director's ratings of the verbal environment. BIBLIOGRAPHY BIBLIOGRAPHY Amidon, E.J. & Hough, J.B. (1967) Interactionaanalysis: Theory resaarch and application. Reading, Massachusetts: Addison-Wesley. Andrews, S.R., Blumenthal, J.B., Johnson, D.L., Kahn, A.J., Ferguson, C.J., Lasater, T.M., Malone, P.E. & Wallace, D.B. (1982) The skills of mothering: A study of parent child development centers. Monographs of the Society for Research in Child Development, 47, 1-82. Bandura, A. (1989) Social cognitive theory. In R. Vasta, Annals of Child Development. Greenwich, C.: JAI Press. Blumenthal, J.B., Andrews, S.R. & Wiener, G. (1976) Five year summary of the New Orleans Parent-Child Development Center. ERIC Document Reproduction Service No. ED 211-237. Chance, P. (1982) Your child's self-esteem. Parents, 57, 54- 59. Cicirelli, V.G. (1980) The Purdue Self-Concept Scale for Preschool Children. The Office of Child Development, Contract 50037. Clarke-Stewart, K.A. (1973) Interactions between mothers and their young children: Characteristics and consequences. Monographs of the Society for Research in Child Qayelooment, 38, 1-108. Dinkmeyer, D. & Dreikurus, R. (1963) Encouraging children to laarn: The encouragement process. Englewood Cliffs, New Jersey: Prentice-Hall. Eder, R.A., Gerlach, S.G. & Perlmutter, M. (1987) In search of children's selves: Development of the specific and general components of the self-concept. Child Develop; ment, 58, 1044-1050. Flanders, N.A. (1970) Analyzing teacher behavior. Reading, Massachusetts: Addison-Wesley. 35 36 Gorham, J. (1988) The relationship between verbal teacher immediacy behaviors and student learning. Communication Education, 37, 40-53. Harms, T. and Clifford, R. M. (1980) Early Childhood - Environment Rating Scalp. New York, New York: Teachers College Press. Harris, P.P & Swick, K.J. (1985) Improving teacher communications. The Clearing Hoaga, 59, 13-15. Hiller, J.H., Fisher, G.A. & Kaess, W. (1969) A computer investigation of verbal characteristics of effective classroom lecturing. American Educational Research Journal, 6, 660-675. Hitz, R. & Discoll, A. (1988) Praise or encouragement? New insights into praise: Implications for Early Childhood teachers. Young Children, 43, 6-13. Hoffman, M.L. (1963) Parent discipline and the child's con- sideration of others. Child Oevelonment, 34, 573-585. Honig, A. (1984) Risk factors in infants and young children. Young Children, 40, 60-72. Johnson, D.L., Kahn, A.J. & Leler, H. (1976) Final report of the Houston Parent-Child Develogment Center. ERIC Docu- ments Reproduction Service No. ED 135-459. Kostelnik, M.J. & Stein, L.C. (1983) Characteristics of tha language environment. East Lansing, Michigan: Michigan State University. Unpublished document. Kostelnik, M., Stein, L. & Whiren, A. (1988) Characteriatics of the Verbal Environment. East Lansing, Michigan: Michigan State University. Unpublished document. Kostelnik, M., Stein, L., Whiren, A. & Soderman, A. (1988) fipidinq children's social develonmant. Cincinnati, Ohio: South-Western. Levenstein, P. (1970) Cognitive growth in preschoolers through verbal interactions with mothers. American Journal of Orthopaychiatry, 40, 426-432. Madden, L. (1988) 00 teachers communicate with their students as if they were dogs? Langgage Arts, 65, 142- 146. Marshall, H. (1989) The development of self-concept. Young Children, 44, 44-51. 37 Marshall, H.H. & Weinstein, R.S. (1984) Classroom factors affecting students self-evaluations: An interactional model. Review of Educational Resaarch, 54, 301-325. Millar, T.P. (1983) How do you give a child self-esteem? Children Today, 12, 2-3. Moos, R.H. (1979) Evaluating educational environments. San Francisco: Jossey-Bass. APPENDICES APPENDIX A ADMINISTRATIVE RATING FORM 38 classroom # Directions: Read each statement and check the most appropriate response. Check one response for every statement. 00 not leave any statements blank. always usually sometimes occasionally never 1. Adults talk more with their colleagues than with the children. 2. Adults focus their attention on children as they perform daily tasks and routines. 3. Adults show little interest in children's activities because they are busy, hurried or tired. 4. Adults use words to show affection for children. 5. Adults pay superficial attention to what children have to say. 6. Adults send congruent verbal and nonverbal messages. 7. Adults yell at children. 8. Adults invite children to play with them. 9. Adults call children names. 10. Adults listen attentively to children. 39 page 2 11. 12. 13. 14. 15. 16. 17. 18. 19. ' 20. 21. 22. 23. Adults label children. Adults say "please" and "thank you“ to children. Adults discourage children from talking to them. Adults use children's interests as a basis for conversation. Adults give orders and make demands. Adults talk with each child informally. Adults ask questions for which no real answer is expected or desired. Adults use silence when talking would destroy the mood of the interaction. Adults use children's names as synonyms for the words "no", "stop" and "don't". Adults avoid making judgemental comments about children within their hearing. Adults use baby talk with children. Adults take advantage of spontaneous opportunities to talk with children. Adults dominate the verbal exchanges that take place each day. always usually sometimes occasionally never APPENDIX B VERBAL BEHAVIOR CHECKLIST CENTER INSTRUCTOR 1. OWVO‘UT-hwm NNNNNNHHHHHi—Ii—IHHH mthHommem-wah-io colleagues 4o scone ROOM PAGE TIME . pasyl tirea . superficial . yell; 3' | . _alls names bels n: shes . demands. orders rhetorical . judgements . sarcasm no. stop. don't . aaby talk dominates . parrots ignores invites listens please, thank yoa child'a interest informal . spontaneous silence conqraent . affection APPENDIX C PURDUE SELF-CONCEPT SCALE This child... 1. N i—Ii—Ii—Ii—I (JON 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 27. 28. HO \DmNO‘m-bw 41 40 Item Scale can catch a ball very well can count to twenty can't climb so high cannot write his name looks nice cannot make up good songs to sing is very smart clothes look old and dirty is trying to learn to read doesn't like to dance talks to other children knows how to make the toy work gets tired after doing a few things mother tells him he's good can put a very hard puzzle together is sad plays with other children teacher doesn't always like what what he does likes to talk to grownups mother does not pay attention likes to build things can't ride a two-wheeled bike gives a cookie to his friend mother knows he can do things for himself cannot find his toy does not like to swing high plays with other boys and girls mother is angry with him a lot This child... cannot catch a ball well cannot count to twenty can climb high can write his name does not look nice makes up good songs is not so smart clothes look clean is not trying to learn to read likes to dance doesn't talk to other children doesn't know how to make the toy work can do many things without getting tired mother doesn't tell him he's good cannot put a very hard puzzle together is happy does not play with other children teacher always likes what he does doesn't like to talk to grownups mother pays attention - likes to sit and watch can ride a two- wheeled bike does not give a cookie to a friend mother doesn't think he can do things for himself can find his toy likes to swing high doesn't play with other boys and girls mother is not angry with him very much 37. 38. 39. 40. 42 thinks he is weak has a lot of toys of his own has only a few friends needs help likes other children doesn't share breaks a lot of things tells stories children like cannot run faster than children people always know what he wants likes to look at picture books cannot call someone on the phone by himself thinks he is strong doesn't have many toys of his own has a lot of friends can do it alone doesn't like others shares doesn't break things doesn't tell stories children like runs faster people don't know what he wants doesn't like to look at books can call someone on the phone by himself APPENDIX D HARMS EARLY CHILDHOOD ENVIRONMENT RATING SCALE 43 HARMS SCALE CONTENT LIST Personal care: Furnishings/display greeting/departing ‘ routine meals/snacks learning nap/rest relaxation diapering/toileting room arrangement personal grooming child related display Language/reasoning Fine/gross motor understanding language fine motor using language supervision FM reasoning GM space informal language GM equipment GM time Creative activities Social development art space (alone) music/movement free play blocks group time sand/water cultural awareness dramatic play tone schedule exceptional pro- supervision (creative) visions Adults adult personal area adult opportunities adult meeting area parent provisions APPENDIX E ADMINISTRATIVE CONSENT FORM Kim Michaud MSU Child Development Lab 325 W Grand River East Lansing, MI 48823 (517) 355- 1900 School name School address City and state xxxxx Dear director: Preschool and day care teachers need information about the most effective ways to create the environment best suited for promoting self-esteem in the children they teach. Your school has been selected from a list of day care centers to participate in a research study which will examine the verbal behavior of preschool teachers and its relationship to self- esteem in preschool children. This study will enable me to complete the thesis requirement for my Master's degree in Child Development at MSU. Participation in the study would require minimal time on the part of your teachers and students selected. The research team would gather observational data on two separate occasions during a free-play period and children participating would individually see the researcher on another occasion to complete a 40 item verbal self-esteem measure. Written consent from each child's parent will be gained before the child is interviewed. Participation is completely voluntary. The results of the study will be treated with strict confidence and the students and the center will remain anonymous in any report of the research findings. Your participation in the study would enable the research team to gain more information about teacher verbal behavior and to examine the relationship between such behaviors and self-esteem in children. No two classrooms are alike. Each has various strengths and weaknesses as reflected by the staff in that room. In order to achieve some contrasting practices among staff, I will need your assistance in selecting two classrooms. 44 45 You indicate your voluntary agreement for your center to participate by completing and returning the attached forms. Please complete one form for each classroom in your center.Any questions you may have regarding participation in the study should be directed to myself at the above address and phone number or to my thesis advisor, Dr. Alice Whiren, at (517) 353-4450. Thank you, Kim Michaud Principal Investigator 0r. Alice Whiren Thesis Advisor APPENDIX F TEACHER CONSENT FORM Kim Michaud MSU Child Development Lab 325 W Grand River East Lansing, MI 48823 (517) 355-1900 Teacher's name School name address City and state xxxxx Dear teacher's name: Positive self-esteem enables children to explore and succeed. Preschool and day care teachers need information about the most effective ways to create the environment best suited for promoting self-esteem in the children they teach. Your center has been selected from a list of day care centers to participate in a research study which will examine the verbal behavior of preschool teachers and its relationship to self-esteem in preschool children. This study will enable me to complete the thesis requirement for my Master's degree in Child Development at MSU. Participation in the study will require minimal time on the part of you and your students. The research team will gather observational data on two separate occasions during a free-play period and selected children with parental consent would individually see the researcher on another occasion to complete a 40 item verbal self-esteem measure. There will be no personal benefits directly to you or your students. Neither are there any penalties if you do not wish to participate. The results of the study will be treated with strict confidence and the students and the center will remain anonymous in any report of the research findings. Your participation in the study would enable the research team to gain more information about teacher verbal behavior and to examine the relationship between such behaviors and self-esteem in children. You indicate your voluntary agreement to participate by completing and returning the third page of this mailing. Any questions you may have regarding participation in the 46 47 study should be directed to myself at the above address and phone number or to my thesis advisor, Dr. Alice Whiren, at (517) 353-4450. Thank you, Kim Michaud Principal Investigator 48 I agree to allow my class to participate in the research study. I understand that a research team will directly observe my classrooms and will interview individual children with parental consent in my classroom. I agree to be a participant in the research study and I understand my teaching behavior will be observed. I understand participation is completely voluntary. I have the right to refuse to participate and the right to withdraw my class from participation in the study at any time without penalty. (signature of teacher) APPENDIX G PARENT CONSENT FORM Kim Michaud MSU Child Development Lab 325 W Grand River East Lansing, MI 48823 (517) 355-1900 Parents name School name address City and state xxxxx Dear parents name: Positive self-esteem enables children to explore and succeed. Preschool and day care teachers need information about the most effective ways to create the environment best suited for promoting self-esteem in the children they teach. Past studies have shown positve verbal exchanges convey warmth, respect and acceptance to children while adverse verbal encounters tend to make children feel inadequate, angry or confused. Your child's school has been selected from a list of day care centers in Lansing to participate in a research study which will examine the verbal behavior of preschool teachers and its relationship to self-esteem in preschool chi dren. This study will enable me to complete the thesis requirement for my Master's degree in Child Development at MSU. Participation in the study will require minimal time on the part of your child's teachers and your child. The research team will gather observational data on two separate occasions during a free-play period and selected children with parental consent would individually see the researcher on another occasion to complete a 40 item verbal self-esteem measure. Your child will be invited to participate in the interview. No child will be pressured in any way and any child may leave the interview at any time. There will be no personal benefits directly to you or your child. Neither are there any penalties if you do not wish to participate. The results of the study will be treated with strict confidence and the students and the center will remain anonymous in any report of the research findings. 49 50 Your child's participation in the study would enable the research team to gain more information about teacher verbal behavior and to examine the relationship between such behaviors and self-esteem in children. You indicate your voluntary agreement to participate by completing and returning the third page of this mailing. Any questions you may have regarding participation in the study should be directed to myself at the above address and phone number or to my thesis advisor, Dr. Alice Whiren, at (517) 353-4450. Thank you, Kim Michaud Principal Investigator 51 I agree to allow my child to participate in the research study. I understand that a research team will directly observe selected classrooms at my child's school and will interview individual children with parental consent. I give my consent for my child to be interviewed. I understand I have the right to withdraw my child from participation in the study at any time. (signature of parent) PLEASE COMPLETE THE FOLLOWING: Your child's birthdate Father's highest grade completed Mother's highest grade completed CHECK THE BEST ANSWER: Combined family income under $15,000 $15,000-30,000 $30,000-50,000 over $50,000 APPENDIX H RAW DATA Columns Columns Columns Columns Columns Columns Columns Columns Columns Columns 52 RAW DATA CODING 1-2: Administrative Rating score (AR) 3-5: Verbal Checklist score (OR) 6-8: Harms total score 9-10: 11-12: 13-14: 15-16: 17-18: 19-20: 21-22: Column 23: Sex Column 24: Age Columns Columns Columns 25-26: 27-28: 29-10: Personal Care subscale score Furnishings/Displays subscale score Language/Reasoning subscale score Fine/Gross Motor subscale score Creative subscale score Social Development subscale score Adults subscale score Father's education Mother's education Purdue total score 53 8290888905884861.8857785357740056.1864705839069077 333402334333233333333333233333229.331243332433433 266932241222289826036469222237603626628688432289 111911111111.100111211119112111121211111111111211 nU22932252222209660657629222264685628630460832209 111.91111111111011211111911211111.1211112112111221 454944445544445554656575764444544444444455555555 000001011100011..1000100100001101001111011.001.1101...... 888888888888888844444444444444444444222222222222 22222222222227.2211111111.111111111111222222222222 233333333300000000000000000088888888333333888888 444444444444444422222222222211111.111333333111111 999999999977777755555555555544444444777777111.111. 333333333333333322222222222222222222444444333333 666666666622222277777777777722222222903333344.4442 333333333333333322222222222222222222333333333333 222222222222222211.l.11111111111111111222222111111 00000000004444.4422222222222255555555000000777777 333333333322222222222222222211111111333333111111 444444444466666666666666666644444444888888888888 222222222222222222222222222222222222222222222222 777777222288888833333303333377777777555555777777 222222777700000055555575555533333333111111666666 222222222222222211111111111111111111222222111111 777777777766666666666666666600000000444444444444 o00000000088888899999999999977777777999999222222 222222222211111111111111111111111111111111222222 222222222299999922222222222288888888222222000000 888888888855555588888888888877777777777777777777 "11111141414144I