‘H ”Mg“-s‘.‘on"- THE STIMULATION OF VERBAL BEHAVIOR IN CULTURALLY DISADVANTAGED THREE-YEAR-OLDS THROUGH A PROGRAM 0F MOTHER-CHILD INTERACTION AT HOME. USING CHILDREN’S BOOKS Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY MARY-CLARE BOROUGHS 1970 * ..i ‘11- “(g _' . J _ IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII I 1112' (“10528 0022 Michigan State THE .1 S univerSity This is to certify that the thesis entitled TIE STIMULATION OF VERBAL BEHAVIOR IN CULTURALLY DISADVANTAGED THREE-YEAR- OLDS THROUGH A PROGRAM OF MOTHER-CHILD INTERACTION AT HOME USING CHILDREN'S BOOKS presented by Mary-Clare Boroughs has been accepted towards fulfillment of the requirements for iii]; degree mm ogy Date 12, WMICHO 0-169 ABSTRACT THE STIMULATION 0F VERBAL BEHAVIOR IN CULTURALLY DISADVANTAGED THREE-YEARrOLDS THROUGH A PROGRAM OF MDTHERrCHILD INTERACTION AT HOME USING CHILDREN'S BOOKS By Mary-Clare Boroughs The purpose of this study was to determine whether the effects of cultural deprivation on language skills can be remediated through experimentally increasing verbal interaction between mothers and their three-year-old children. Research indicates that language skill deficiency exists as early as 18 months and that disadvantaged mothers interact verbally less with their children than advantaged mothers do. Current projects rarely center on three-year-olds. They also use professional trainers more often than they use the mothers themselves. This study was designed to assess the potential of a simplified book-reading program for three-year-olds utilizing the potentials within the home as well as available community resources. It was hypothesized that children in the experimental group with treatment designed to increase mother-child verbal interaction would show greater language skills and attention than a comparable control group. In order to examine the effects of experimental con- ditions a placebo control group received nutrition materials. It was hypothesized that verbal interaction treatment would have an effect Mary-Clare Boroughs that would also be greater than any experimental effect on the placebo control group. Forty-seven disadvantaged three-year-olds, Negro and white, boy and girl, completed the study. These children were randomly assigned to three groups balanced by race and sex. The experimental group of 19 was committed to three months of a daily 10-15 minute period of interaction using appropriate library books and inexpensive gift books delivered weekly for three months by paraprofessional Family Helpers. Family Helpers also delivered nutrition materials weekly to the 14 members of the placebo control group. Another 14 children had no contact. Children were studied both before and after the experimental period using known tests of vocabulary and intelligence, expressive responses to toys and pictures, mean sentence length, four functional categories of responses and an attention score. Only the first five of these were sufficiently reliable to analyze the data.. Results of statistical analysis showed that the Binet scores gave complete support to the hypothesis at the .05 level for the placebo and the .005 level for the control group differences from the experimental group. In addition the experimental group scored signifi— cantly higher than the control group (p < .05) on three of the other four measures and significantly higher than the placebo control group on one of the other five measures. For one additional measure experi- mental group girls scored significantly higher than placebo control group girls. In general, therefore, the results of this study showed that a three month program designed to stimulate verbal interaction between Mary-Clare Boroughs disadvantaged mothers and their three-year-olds through the medium of children's books did have an effect on verbal skill greater than that found in placebo or control groups. There was some indication that the placebo situation also had an effect on the children, but to a lesser degree. This research indicates the usefulness of a book distribution program administered through available paraprofessional workers in the community involving disadvantaged mothers reading to their own pre- school youngsters. THE STIMULATION 0F VERBAL BEHAVIOR IN CULTURALLY DISADVANTAGED THREE-YEARrOLDS THROUGH A PROGRAM OF MOTHERrCHILD INTERACTION AT HOME USING CHILDREN'S BOOKS By Mary-Clare‘Boroughs III III- I ’-. '5‘: A THESIS Submitted to Michigan State university in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1970 v. s”: I i DEDICATION To all the boys and girls who made this project possible. 11 ACKNOWLEDGEMENTS This research was carried out with the assistance of many people in many places. Without each and every one of them it could not have been completed. The initial arrangements were made through the cooperation of Dr. Edward L. Remick, research director for the Lansing Public Schools, Mr. James Kernen, consultant for continuing education, and Mrs. Catherine Jaffee, director of the Family Helpers. The major work of the project was accomplished by the Family Helpers under the super- vision of Mrs. Barbara Kirk and Mrs. Lillian McFadden. The secretaries of the two centers were most helpful in the management of testing appointments - Mrs. Mittie Keys and Mrs. Nancy Hause. Finally most of the workers in the two centers served children in the project Mrs. Louwillie Baugus, Mrs. Dorothy Clark, Mrs. Elizabeth Clark, Mrs. Florence Cross, Mrs. Pearl Davenport, Mrs. Fannie Eaton, Mrs. Lucy Lilly, Mrs. Lola Morgan, Mrs. Louise Rhodes, Mrs. Elvira Rivera, Mrs. Esther Silvas, Mrs. Mary Ashworth, Mrs. Jessie Kimball, Mrs. Christina Pegueno, Mrs. Betty Flowers, Mrs. Florence Hollingsworth, Mrs. Pauline Kelley, Mrs. Joan Lance, Mrs. Dorothy Lee, Mrs. Doris Luis, Mrs. Sara Magana, Mrs. Pinkie Moore, Mrs. Odelia Nieto, Mrs. Shirley Parrott, Mrs. Jackie Pennell, Mrs. Esther Polloch, Mrs. Carole Prince, and Mrs. Joyce Young. iii The library books for the project were borrowed from three libraries: The Lansing Public Library, The Ingham County Library and the Michigan State Library. Mrs. Margaret Hendrix, Mrs. Helen Miller, and Mrs. Norene Gentry assisted with the loan arrangements at the three libraries. Mrs. Winnifred Crossley, School Library Consultant with the State of Michigan, was also most helpful, as were Mr. Marvin Mounce of the county library and Mr. W. J. Kimbrough of the Lansing Library. Mrs. Hendrix graciously offered the use of the children's section reading room for small group meetings and was on hand for these meetings as well. Mrs. Kathy Dunlap and Mrs. Sue Miller of the Lansing Head Start provided consultation as well as leadership for the small group meetings for the experimental group. Mus. Annabelle Pangborn, public health nutritionist, volunteered her services to select literature and lead the meetings at the North Side Action Center for the placebo control group. Dr. Lawrence LeZotte, a consultant in educational research, was the primary source of help in setting up the data for the computer analysis. He was assisted by Mr. Bill Brown. To all of the above my thanks - and last but not least to all the members of my committee - to Dr. Dozier Thornton, Dr. Gary Stollak and Dr. Lawrence Messe for their ideas and support - to Dr. Ellen Strommen for time and concern for some of the difficult details in testing and scoring procedures, to Dr. Robert Boger for his initial interest and for practical assistance at several junctures and finally, to Arthur Seagull, Chairman, who encouraged me to pipe-dream and then iv lent his moral support to the project at the stages of planning as well as through the long winter months of the experiment. Acknowledgements would not be complete without some reference to my family. They lived with a distracted and sometimes distraught wife and mother, with short order, late order meals and with a dis— organized household-—my thanks to Art and with him to Becky, Debbie and Don. TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . . . . PROBLEM . . . . . . . . . . . . . . . . . . . . . REVIEW OF THE RESEARCH . . . . . . . . . . . . . . A. Differences in Language Development between the Social Classes . . . . . . . . . . . . . B. Remediation of Deficits in Deprived Infants and Young Children . . . . . . . . . . . . . C. Toward a Definition of Verbal Skill . . . . momESES O O O O I O O O O O O O O O O O O O O O PROCEDURES 0 O I O O O O O O O O O O O O O O O O 0 Setting . . . . . . . . . . . . . . . . . . . . Subject Population . . . . . . . . . . . . . . . Dependent Variables . . . . . . . . . . . . . . Independent Variables . . . . . . . . . . . . . T1--Experimental Group. . . . . . . . . . . . T2--Placebo Control Group . . . . . . . . . . T3--Control Group . . . . . . . . . . . . . . RESUI‘TS O O O O O O O O O O O O O O O O O O O O 0 Reliability . . . . . . . . . . . . . . . . . . Treatment Group Mean Differences. . . . . . . . Stanford Binet Intelligence Scale . . . . . . Peabody Picture Vocabulary Test . . . . . . . Expressive Measure #1 . . . . . . . . . . . . Expressive Measure #2 . . . . . . . . . . . . Mean Sentence Length . . . . . . . . . . . . . DISWSSION I I O O O O I I O O I O O I O O O O O 0 Implications for the Compensation of Cultural Deprivation . . . . . . . . . . . . . Implications for Further Research . . . . . . . Implications for Research Measurement . . . . . vi Page viii ix 15 17 l7 17 19 25 25 27 28 29 29 29 3O 31 32 32 32 33 33 35 36 Reliability 0 O I O O O O O O O O I I 0 Specific Measures . . . . . . . . . . . SUMMARY . . BIBLIWRAPHY O O O O O O O O O O O O O O O 0 APPENDIX A: APPENDIX B: APPENDIX C: APPENDIX D: APPENDIX E: DESCRIPTION, ADMINISTRATION AND SCORING PROCEDURES FOR EXPRESSIVE MEASURES #1 AND ’ 2 O I O O C O O O O O I 0 BOOK TITLES O D O O O O O O O O SAMPLE SUGGESTION BOOK CARDS . PARENT MATERIALS . . . . . . . TABLES 3-12: POST-TEST CELL MEANS FOR THE TEN ORIGINAL MEASURES . . . . . vii Page 36 38 40 43 46 50 55 59 66 Table 10. ll. l2. LIST OF TABLE Means (Adjusted) for Treatment Gro Five Reliable Dependent Measures Means (Adjusted) for Treatment Groups Separately by Sex for the PPVT . . . . . . Post-Test Cell Means Adjusted for the Stanford Binet Intelligence Post-Test Cell Means Adjusted for S ups for the Pre-Test on Scale . . . Pre-Test on the Peabody Picture Vocabulary Test . . . . Post-Test Cell Means Adjusted for the Expressive Measure #1 . . . Post-Test Cell Means Adjusted for the Expressive Measure #2 . . . Post-Test Cell Means Adjusted for the Mean Sentence Length . . . Post-Test Cell Means Adjusted for the Inattention Score . . . . . Post-Test Cell Means Adjusted for the Birch Response #1 . . . . . Post-Test Cell Means Adjusted for the Birch Response #2 . . . . . Post-Test Cell Means Adjusted for the Birch Response #3 . . . . . Post-Test Cell Means Adjusted for the Birch Response #4 . . . . . viii Pre-Test on Pre-Test on Pre-Test on Pre-Test on Pre-Test on Pre-Test on Pre-Test on Pre-Test on Page 30 31 66 67 67 68 68 69 69 7O 7O 71 LIST OF FIGURES Figure Page 1. The Scheme of Analysis of Response Styles to Cognitive Demands . . . . . . . . . . . . . . . . . . . 22 ix PROBLEM A common psychological position states that the behavior of a child reflects a combination of heredity and environment. The environ- ment of the lower class child lacks many of the intellectual stimuli of the environment of the middle class child. The question therefore is--Is it possible to compensate for environmental deficiencies by intervention in the early years? Because lack of financial resources, tested procedures and trained personnel make immediate wide-scale pro- grams impractical, there is a need to investigate the possibilities of small scale intervention using existing resources. The developmental area selected for this study was that of language with related aspects of attention and intelligence. REVIEW OF THE RESEARCH A number of research studies have been designed to examine the differences in language development and environment among social classes. In addition more recently there has been an increase in pro- jects which have set out to remediate verbal deficits in deprived in- fants and young children. These studies and projects are reviewed in Sections A and B of the pages that follow. In addition verbal skill can be conceptualized in more than one way. Three conceptual viewpoints are presented briefly in Section C. Particular emphasis is placed on the aspects of these viewpoints that might relate to measurement procedures. A. Differences in Language Development between the Social Classes There is considerable evidence for social class differences in language develOpment. In a review of the literature up to 1946 McCarthy (1946) reported a consistent positive correlation between socioeconomic status and language proficiency. More recent studies have attempted to describe these differences more fully. Starting with the home environment, some reports refer to stimulus deprivation or sensory deprivation. Deutsch (1964) describes this stimulus deprivation as the lack of educational tools as well as the lack of variety in household tools such as kitchen ware. 3 Bereiter and Englemann (1966) discredit this theory on three counts: (1) Studies of children under two years of age show few if any dif- ferences in development including sensory and motor development, (2) Sensory deprivation research drastically reduces stimulation, a condition which is gg£_paralleled in slum life, and (3) Organisms are known to seek optimal stimulation given the oppor- tunity, and lower class children are not locked in sound-proof cages. These authors refer instead to language deprivation. They point out that studies of blind children reveal average verbal IQs, whereas deaf children, like the culturally deprived, function below average in intellectual and academic development. Studies of deaf children also show that socio-economic status is not related to IQ scores as it is to the IQs of normal children. In the description of the initial stages of language develop- ment, John (1963) notes the lack of sufficient feedback in the cul- turally deprived home. It appears that in many such homes relatively little attention is paid to the early language of the child (a factor which also correlates with the higher child/adult ratio). Thus he may fail to receive the appropriate feedback to reward and develop his language patterns fully. Although, many studies of young children seem to show no dif- ferences in early development, Irwin (1948, 1948) reports differences in sound frequency (frequency of sound) and sound types (number of types of sound). In two rather carefully designed studies he shows that at about 1 1/2 years development in both number of sound types 4 used and frequency of sound diverges significantly for two groups-- laboring and non-laboring. At 29-30 months the mean number of sound types for an infant in a non-laboring family was 28.1, for laboring 25.2. The difference is even greater for sound frequency where the mean for non-laboring was 170.6 contrasted with 123.8 for laboring. Cynthia Deutsch (1964) describes the importance of the right signal-to-noise ratio to allow the infant to recognize the word signal. In the more crowded lower class home there is generally a higher level of noise with less clear signal directed to the individual child. It is possible that this signal—to-noise ratio in the crowded home may not allow the child to make accurate sound discriminations on repeated occasions. As language development proceeds, not only feedback in the sense of the initial responses to language, but also meaningful dialog with adults becomes an important factor. John and Goldstein (1964) found that only one-half of the lower class Negro families have even one common meal a day. There also seems to be clear evidence of little home planning and less shared activities in the lower class home (Deutsch, 1964). Several aspects of language development show the role that dialog may play. For example, John g£_gl, (1964) found that three classifications of words were less known by lower class children than by middle class children. Two of the classes were nouns that lower class children probably had not experienced. The third group was familiar action words.1 Use of these action words, she concludes, 1Loban's report (1964) from his longitudinal study of high and low ability groups shows that the "linking verb" is more often absent in the language of low ability school children. 5 depends on conversational dialog, and the lower class child may have heard more abrupt commands and incomplete phrase—type sentences in his less frequent encounters with adult language. She also points out that categories, a more developed form of thought and language, require the multiple experience of labels in which both labels and contexts vary. Where dialog is less prominent and where multiple experiences of stimulus variety in which labels are applied may be lacking, the groundwork for accurate categorization may not occur. Thought appar- ently develops with language, and thought too is thus dependent on dialog. In addition to difficulties with action words, categorization, and thought development, problems in attention and memory may also be related to lack of dialog. Whereas the typical middle class child has gradually learned to attend to extended descriptions and explanations (Bernstein, 1961), the lower class child may not have heard adults talk to him for longer than a few seconds. When the child enters school it appears that preparation neces— sary for the child to benefit from our educational system is weak. For example, a study of the ability to comprehend teacher and peer speech (Cherry, 1964) showed that young lower class school children missed a great deal of what is said in the classroom. Whereas middle class children had learned to attend and were adept enough with adult speech patterns found in the classroom to fill in words they did not hear or comprehend, lower class children were much less capable of this necessary task, and thus lost the total meaning of teacher language in a typical classroom situation. 6 Auditory discrimination also seems to show up as deficient when disadvantaged children progress through school. Cynthia Deutsch (1964) studied this with first and fifth grade Negro children of lower socio— economic status who were good and poor readers. Whereas among middle class children difficulties with visual perception are usually con- sidered to be more involved in reading problems, the major discrimi- nating variable among these lower class children was auditory discrimi- nation. Mrs. Deutsch refers to the signal-to-noise ratio and lack of feedback as well as learned inattention to help explain this difference. She also points to its relevance to remediation and early school training. Hess and Shipman (Hess and Shipman, 1965, undated, 1966; Shipman and Hess, 1966) have studied 160 Negro mothers and their four-year-old children representing four socioeconomic levels. They used interviews, tests, and observed mother-child interactions. Marked class differences appear in the results. Lower class mothers use more imperative control, whereas middle class mothers use subjective-personal or cognitive- rational methods. In assisting their child with a task, lower class mothers give vague orders and use negative reinforcement, whereas middle class mothers provide information, orient the child to the task, rein- force with praise, seek and respond to feedback. Lower class mothers encourage passive submissive behavior and view it as appropriate for school. For themselves lower class mothers are characterized by feelings of powerlessness and inability to effect change in their lives or in the school experience of the child. Olim, Hess and Shipman (1965) following the lead of Bernstein (1961) studied the language differences between middle and lower class 7 mothers. They found that lower class mothers use a more restricted language pattern with less "abstract language behavior." It was this last factor that appeared to be most closely related to the child's cognitive level. There was evidence in the research that this is not necessarily an ability factor, but one of habitual behavior patterns. Milner's study of reading readiness and parent-child interaction (1951) is tangentially related. Although this was not a study of class differences, she found that high scoring children came from homes which provided more verbal enrichment, more family activities, and more parent-child interaction. In addition mealtime conversation was encour- aged as well as self—responsibility. In contrast, low scorers experi- enced more restricted and controlling interaction. Significantly more high scoring children came from middle class homes than did low scoring children. In brief, it seems that the culturally disadvantaged child comes from an interpersonal environment which does not encourage adequate meaningful verbal development. Experience in rewarding dialog communi- cation with adults is limited. This results in speech patterns which begin to deviate from the middle class standard as early as 1-1/2 years and result in increasing deficiencies. When compared with middle class children speech is limited in sounds, number of words and complexity of structure. These are problems which seem to be related later to lower IQs and reading retardation. 8 B. Remediation of Deficits in Deprived Infants and Young Children Attempts to remediate the verbal deficit of the culturally dis- advantaged child had their start within the established school system through attempts to upgrade the poorer quality of education found in urban slum areas (Sexton, 1961). This has meant increasing special services to target schools throughout the school years. Increasingly, however, educational efforts have turned to the preschool years. Hunt (1964) describes the psychological basis for this in some detail. He points out that there have been changes in common views of intelligence and deve10pment. Some of the more current beliefs include the belief that intelligence is not a fixed entity, the belief that deve10pment is not totally predetermined, the belief that the human brain is not a static switchboard, the belief that even pre-verbal experience is important, the belief that the experience of variety is important to the human organism, and the belief in the role of in- trinsic motivation in learning. As a follow—up to his earlier research, Irwin (1960) reports a study in which working class mothers were instructed to read to their infants 15 minutes a day. The total experimental period was 18 months, but significant differences between the experimental and control groups (using tape-recordings that were analyzed for phoneme frequency) appeared as early as the 4th month. It is not clear whether this re- search was a part of any on-going remedial program. Project Head Start, first as a summer program, later as a pre- school age year-round experience was added to the usual succession of 9 grades as educators became aware that school remediation was often too little and too late (Bereiter and Engleman, 1966; Pines, 1966). In general, Head Start programs with four year olds have resulted in IQ gains. However, early enthusiasm has waned as these gains were erased in kindergarten where Head Starters made no new gains and others caught up with them. As a result a variety of projects have been developed moving in several directions. More structured training and integration of middle and lower class children are being incorporated into experimental pro- grams with four-year-olds. Some projects have attempted to begin with the infant rather than waiting until age four. Such projects exist both in day care centers and in home tutoring programs. Home training of mothers is being attempted with their infants or with older children already three or four years of age. A shift from total emphasis on the child to training of mothers-~either for home intervention or as para- professionals, or in combination--has come about. With the lack of highly trained personnel, the use of paraprofessionals other than mothers has also become an interest. The following pages will describe several such studies. Gordon (1968) at the University of Florida reports a study in which paraprofessionals were used to train mothers and their infants in their homes. He used two control groups, a group receiving visits without training procedures and a group receiving no visits of any kind. The 1966-67 groups were three months to one year in age. Results of tests at the first birthday show that infants in the experimental group surpassed infants in control groups on tasks from the training series as well as relevant dimensions of a developmental scale. Control 10 group infants did not differ significantly from each other. Therefore, to date, the use of paraprofessionals, the inclusion of the mothers in training and the focus on the infant appear to have been combined into an effective program. Further longitudinal studies are in progress. Project Know-how, directed by Richard Dunham at Florida State University, is using mothers as assistants in a class program at a center for pre-school training for ages 1-6. The mothers are employed at the center and the focus of their training is on "homemaking skills." It appears from the available brief report that this project is attempting to train mothers to provide more attention and stimulation for their children. Earl Schaefer (1965, 1967, 1968) attempted to tutor children during the period of fifteen months to three years through daily home visits by trained college students emphasizing verbal stimulation of the child by the tutor. Results indicate that the tutoring did make a significant difference in early verbal development in spite of repeated observations of extreme deprivation in the homes. Bettye Caldwell (Pines, 1966) at Syracuse University has a re- search program studying the effects of stimulating infants from six months to three years in a day care setting. The focus is on the child, with little or no contact with mothers. The adult-to-child ratio is arranged to provide frequent one-to-one interaction between an adult and the young child. The results in the children's emotional, social and intellectual development seem to indicate that day care can be a stimulating child-rearing method even for very young children. One of the more extensive current programs is DARCEE, directed by Susan Gray (Gray and Klaus, 1966; Gray, Miller, Hinze and Schoggen, 11 1967; Horton, undated; Rubow, 1968; Wilcoxon, 1967). Three treatment groups were included in this project--all comprised of four year olds. Maximal Impact included both mother and child in a pre-school program. The Curriculum Group is the more traditional child-only program. A third group is a Home Visitor program in which mother and child are trained with the mother committed to follow up during the week. DARCEE trained class participating mothers for more effective motherhood as well as classroom responsibilities. Results indicate that mothers can be trained, that the training enhances their self-concept and their ability to mobilize themselves to make changes in family life. In addition there is considerable evidence of diffusion--both to siblings and to friends and neighbors (including the control groups!). Their reports would indicate that given effective training and a good role model in the teacher, the mothers are eventually both ready and eager to assume an active role in their child's development. Both Weikart (1967) and Levenstein (1968, 1969a, 1969b) are involved with programs in which trained professionals have visited homes to train the mother to provide cognitive stimulation for their four-year-olds. Preliminary results indicate that mothers can be trained for this role. The mothers first trained in the project Levenstein reported are showing promise in a new study in which they train other mothers in the place of professional social workers. Concerns which are emphasized repeatedly in these reports include the behaviors disadvantaged mothers reward and do not reward. It is clear that disadvantaged children are most often "rewarded for passivity" (Gray and Klaus, 1966» and receive little positive reinforcement for language and assertive behavior. There is evidence of the impact on 12 the child of the household in which time and space as well as sound have no order or meaningful sequence in contrast with the use of order in the more effective classrooms (Gray, Miller, Hinze and Schoggen, 1967). Before training mother—child interactions tend to be limited to restricted orders by the mother. As researchers train the mother to interact more adequately with her child, they assist the mother to use positive reinforcement for the child's verbal behavior. They also teach her ways to interact so as to add information, encourage reflec- tion and produce a variety of responses. The variety of studies reported indicate that early language deprivation, including inadequate home language and methods for control of verbal and cognitive skill development, is a critical deficiency related to poverty. Training programs show that given a qualified pro— gram children can be trained in either group or home settings. In addition efforts to use mothers in both home and group settings to assist in groups and/or apply new methods at home have shown consider- able promise. The majority of the current research and action appears to be centered upon four-year-olds. In programs with both four-year-olds and younger pre-schoolers, whether at home or school, the interaction with adults has more often been with professionals than with family members themselves. .It appears that the potential resources for language education within the family have not been determined; particularly not for three-year-olds. 13 C. Toward a Definition of Verbal Skill Many of the studies dealing with culturally disadvantaged children speak of verbal skill but do not define it. In the majority of the studies the investigators use the Stanford-Binet Intelligence Test and the Peabody Picture Vocabulary Test. They do not generally define their choice of test by citing the connection between their definition of verbal skill and the aspect of verbal skill tested. Since different tests do evaluate different aspects of verbal skill, one must have a definition or description of the verbal skills sought through intervention in order to determine measures of change. Language skill has been measured in an increasingly varied number of ways in recent years. The ITPA (Illinois Test of Psycho- linguistic Abilities, 1968), for example, seeks to test twelve aspects of language ability in three areas: decoding (receptive), associative, and encoding (expressive). Samuel Kirk and Barbara Bateman are key researchers in this area. Speech therapists also center their attention on receptive as well as expressive language considering both verbal and non-verbal components. Receptive, associate and expressive language with both verbal and non-verbal components includes as much broader range of language characteristics than simpler views of lan- guage as verbal fluency or vocabulary level. The psycholinguists such as Roger Brown or Susan Ervin-Tripp focus on the development of grammar, on the complexity of sentence structure from one age to the next. They measure such elements as parts of speech, transposition of sentences, mother-child interaction patterns and other elements of spoken language. Brown and Bellugi 14 (1964) report that mean sentence length is the best single measure of level of speech development. The functional aspects of spoken language, the nature of its use in every day discourse, have received somewhat less emphasis in current research in the U.S.A. Bernstein (in England) and Hess and Shipman represent this tradition. From the point of view of efficient human behavior whether at home, at school, in the community, or on the job, a functional approach to language is useful. The question here centers around the ability to put language to use effectively in inter- actions with the environment. It appears that in the studies of the effects of cultural deprivation on language this is closest to the meaning of verbal skill implied by many of these studies. Thus there appear to be at least three somewhat different approaches to the area of verbal skill. The approach typified by the ITPA includes a variety of subcategories based on receptive, associative and expressive areas. The psycholinguists focus on sentence structure development. The functional approach to language observes verbal skills in action, serving a purpose. In this study an attempt was made to incorporate aspects of all three points of view in measures to test changes in "verbal skill." HYPOTHESES The research indicates that children from disadvantaged homes grow up with less opportunity for parent-child verbal interaction and that this lack is related to deficient language skills beginning as early as 18 months. Several attempts to remediate this lack of verbal skill among disadvantaged children have shown some promise. From the above the following major hypothesis of this study has been derived: 1. Children who have increased verbal interaction with their mothers will show greater verbal skill than children for whom mother-child verbal interaction is not increased. In addition children who have increased verbal interaction with their mothers will show greater verbal skill than children whose mothers have received equal atten- tion but on a topic not geared to enhancing interaction with their children. In addition to evidence regarding verbal skill there are indi- cations that inattention may be a factor in the deficitis of language and related behavior. Therefore a further minor hypothesis was also presented. 2. Children who have increased verbal interaction with their mothers will show better attention skills than children for whom mother— child verbal interaction is not increased. In addition, children who have increased verbal interaction with their mothers will show 15 16 better attention skills than children whose mothers have received equal attention but on a topic not geared to enhancing interaction with their children. PROCEDURES Setting This study was conducted with the assistance of the Family Helper Service of Lansing, a school-related, state—funded agency pro- viding paraprofessional services to problem poverty families in the community. Thirty helpers working out of two regional centers served over 500 families at the time of this study. Most of these families were referred by schools when children had school problems which appeared to have a basis in poverty-related home difficulties. The study was designed to use children served by the Family Helpers and to use the helpers as the liaison between the experimenter and the subjects. Pemily Helpers explained the project, transported mothers and children to testing and meetings and delivered experimental books and placebo control literature. In addition one of the two regional centers served as the testing site. Subject Population Of the families served by the Family Helpers there were 145 families with 151 children (3 sibling pairs, 3 sets of twins) born between 12/1/64 and 12/31/65. These families included 35 Spanish speaking families which were eliminated from the sample. Since most of these families speak only Spanish at home, the Spanish speaking children would not have been 17 18 familiar with the test language or with that of the children's books essential to the research. A number of other families were also eliminated for any one or more of the following reasons: 1) Child enrolled in Head Start (Lansing or Michigan State University). 2) Child served by one of a small group of volunteers working weekly one to one with selected disadvantaged three-year-olds. 3) Mother not sufficiently literate to read books to child. 4) Mbther known to be working full time.2 5) Families known to be uncooperative or in a state of current crisis. The remaining 85 families were invited to participate in the project and to report for testing during January and early February. In early January an orientation meeting was held to explain the project to the Family Helpers. Mothers were to be told that the project would be a study of three-year-old development and that all the children would be seen with their mothers in January and in May to observe their stage of growth and learning. It was explained that some children would have an at-home project in the interim. Initially the Family Helper responsible for each family was enlisted to contact the family, to explain the project, to arrange for test appointments and to bring mother and child to the assigned center for testing. Children were randomly assigned to days and hours for pretesting. Posttesting 2Since a few mothers went to work just prior or following en- rollment in the project, there were also a few working mothers in the sample. l9 appointments were again made randomly assigning days and hours, but keeping groups by weeks in a weekly order. (This latter arrangement was to assure that the time between pre-and post-testing was approxi- mately the same for all children.) Books were delivered for 12-13 weeks, but the total time period between pre-and post-testing was close to 15 weeks. Fifty-six children came in for pretesting. All children co- operated sufficiently and were included in the sample. Forty-nine of these 56 children returned for post-testing in May. Two children were eliminated from this final sample--one in the experimental group for lack of cooperation in both the project and testing and one in the control group who was discovered to be ineligible due to weekly visits from a volunteer.3 Therefore the final sample included 47 children. There were 22 boys and 25 girls, 23 white and 24 Negro distributed as follows: 11 white boys and 11 Negro boys, 12 white girls and 13 Negro girls. Dependent Variables The dependent variables included measures of the children's behavior before and after the experimental period. The measures chosen reflect a concern with using common, uncommon and original measures of verbal skill, the desire to measure more than one aspect of verbal ability and the need to design ways to study three-year—olds, given their stage of develOpment in language, attention and cooperation. 3There was a marked difference between pre and post-testing,be- havior of this child. Her Binet IQ score went from 83 to 111 and the Peabody from 77 to 97. Other scores were similarly improved. The mother described the volunteer as a person who took the child on trips for about two hours a week and returned leaving books for the mother to read in the interim before the next visit. 20 The Stanford Binet Intelligence Scale (Binet) has been used in most studies in this general area of research. While a number of items included at the pre-school level are non-verbal in character, the ability to attend to and to follow verbal instructions primarily a receptive ability is basic even for these items. Verbal items test expressive ability. Therefore the IQ score on this test is generally considered to be primarily a verbal IQ. In accordance with the major hypothesis it was predicted that the scores for the experimental group would exceed those of the two control groups. The Peabody Picture Vocabulary Test (PPVT) has also been used in many related studies. It uses pictures and non-verbal pointing re- sponses to the examiner's presentation of a single word to be discovered among four pictures. A measure of receptive vocabulary, it can be used with children as young as two years of age. The two forms, A and B, were randomly assigned for pre-testing and reversed during post-testing. In accordance with the major hypothesis it was predicted that the scores for the experimental group would exceed those for the two control groups. This test provides pictures stimuli indicating a vocabulary level for the young child. A pilot study was made to test out the use of objects selected from the items pictured at the lowest level of Peabody pictures. These objects were then used to elicit expressive responses. The pilot study also indicated that expressive responses could be elicited to Nursery School pictures. These protocols were scored using an elaboration of the ITPA scoring standards of the Vocal Encoding subtest. The directions for scoring are provided in detail in Appendix A. These two measures are designated Expressive 21 Measure #1 (toys) and Expressive Measure #2 (pictures) (also Expressive 1 and Expressive 2). Expressive Measure #1 consisted of a series of objects: a red toy car, a 10 inch baby doll, a grey plastic cup, a green plastic ball, a common plated teaspoon, and a small white baby shoe. The child was instructed to "Tell me what it is" and then "Tell me all about it" two times followed by a third question, "Tell me what you do with it." While these questions were elaborated somewhat to elicit responses from the very shy children, the general format was adhered to rather care- fully. In accordance with the major hypothesis it was predicted that the scores on this measure would be higher for the experimental group than for either of the two control groups. EXpressive Measure #2 consisted of three Nursery School pictures depicting common scenes: 1) Firemen running to their truck, 2) Two children (Negro) getting mail from a mailman and 3) A playground with many active children (racially mixed) and e.city scene in the background. In accordance with the major hypothesis it was predicted that the scores on this measure would be higher for the experimental group than for either of the two control groups. A reliability study was conducted for scoring on both expressive measures. After one short training session using sample protocols percent agreement for total scores on a new protocol was calculated. For Expressive 1 the percent agreement was 94 percent and for Expres- sive 2, 93 percent. 22 A complete record of utterance used by each child was made with the help of tape recordings. This record was compiled to compute mean sentence length (MSL). This was selected as the sole measure of lan- guage structure as suggested by Brown and Bellugi (1964). In accordance with the major hypothesis it was predicted that the scores on this measure would be higher for the experimental group than for either of the two control groups. Hertzig, g£.al. (1968) developed a method to examine behavioral style in the face of cognitive demands including the functional use of language. The basic categories are verbal and non-verbal responses. These measures were used successfully to compare styles of middle class and lower class Puerto Rican three year olds. These so-called Birch response categories are pictured below in Figure l. DEMAND WORK NOT-WORK Verbal Non—Verbal Verbal Non-Verbal Spontaneous :Delimitation Competence.~ Substitution Passive Extension Negation Aid Figure 1.--The scheme of analysis of response styles to cognitive demands (Birch, from Hertzig 1968, p. 13). A detailed interpretation of this schema is available in the Hertzig publication. The 1968-1969 national Head Start evaluation has employed this method, revised in part to meet their needs. For the purposes of this study both the original and the Head Start methods were examined and a further revision of a response system was drawn up. 23 The following aspects of behavioral style were selected for this study: 1) The proportion of verbal work responses (attempted work on verbal items) to total verbal items (Vw/Tv), 2) The proportion of brief verbal utterances accompanying non- verbal work responses (right here, over there, etc.) to total non-verbal items (WNVV/TNV), 3) The proportion of verbal labelling responses (naming of items) to non-verbal items (VE/TNV), and 4) The proportion of verbal comments (this includes several categories of the Birch schema combined) to the total number of items administered (VC/T). In accordance with the major hypothesis it was predicted that these proportions would be higher for the experimental group than for either of the two control groups. No formal reliability studies have been reported for either the original Birch-Hertzig measures or the Head Start adaptation. For this research a check on examiner and scorer reliability was made. One examiner-observer training session was held using sample protocol re- sponses. Then a child was tested and the observer scored the child's behavior while the examiner scored the protocol immediately after the testing session. Agreement between the examiner and the observer item by item was 80 percent. This figure was accepted readily, since it was clear that a combination of further experience and a re—run of a tape— recording as planned for the research would increase this percent agreement considerably. 24 Observations of behavior were also used to derive an attention (or more accurately an inattention) score. This score was the number of inattentive behavior units (looking away intently, moving out of seat, interrupting the examiner etc.) during the first 20 minutes of the administration of the Stanford Binet Intelligence Test. Except for gross inattentions that slowed up testing procedures, inattentions were only those during item administration, and did not include slight lapses from the testing situation during the time between items. In accordance with the second hypothesis for this study it was predicted that the score on this measure would be lower for the experimental group than for either of the two control groups. While this is the only measure of the ten that were selected that specifically measures attention, it is obviously not the only score which reflects attention. With the exception of the Birch system, the other measures all tap attention as well as verbal skill. However, this is the only measure that can be said to specifically test this second hypothesis. In order to increase the accuracy of all these measures and to allow the examiner to focus on the interaction rather than on note- taking, each test interview was tape recorded. Notes from tapes were transposed onto full protocols for a complete record of child verbal behavior. The following order of administration was used: 1) Peabody Picture Vocabulary Test 2) Expressive Measure #1 (Toys) 3) Expressive Measure #2 (Pictures) 4) Stanford Binet Intelligence Scale (Form LM) 25 The PPVT was used first since it requires the least response from the child (pointing only). Since the expressive measures had high interest value, and the Binet tends to press a child to the limits of his endurance, the two expressive measures were administered before the Binet. In addition, to maximize cooperation during the 3/4 hour session m chocolate candies were used. These were distributed primarily in connection with each subtest of the Binet, although they were used occasionally during the administration of early measures. The reward was given, not for success, but for response. For a few children this began with the response of chair-sitting, but for most it was any reasonable response to test items (including I don't know). It is the opinion of the experimenter that the M & Ms were most helpful in main- taining attention and therefore in increasing success as well as the reliability of the measures. This was in part responsible for the minimal problems in cooperation as compared with the 20 percent drop-out rate predicted for this age due to failure to cooperate. Independent Variables Tl—-Experimental Group The experimental condition, designed to increase mother-child verbal interaction, used a daily story-hour. The Family Helpers assisted the mothers by delivering books, encouraging their participa- tion in group meetings and generally in the project. weekly meetings were held with the Helpers to distribute books and discuss problems. Family cooperation was simply requested from a randomly selected experimental sample. With the exception of one mother, all families 26 cooperated in some degree. To assist the mother to plan reading time a heavy manilla envelope was provided to hold the books and with a weekly check list for daily reading records. It was hoped that this list would increase incentive and aid in records, but it proved to be beyond the scope of either Helpers or mothers or both. In cases where the mother was having difficulty scheduling reading, an effort was made to have other family members participate. Each week the families received two library books from a selec- tion of approximately 130 books on loan from three libraries in the area. These books had been carefully selected to be both appropriate for three-year-old children and easy-to-read for mothers with a low level of reading skill. Each book was provided with a 3x5 card with suggestions to expand interaction in the use of the book-picture study, story discussion, related events, suggested simple neighborhood trips etc. The complete list of books is found in Appendix B. Sample sug- gestion cards are included with Appendix C. In addition for nine weeks each child received a book to keep. The majority of these were Golden Shape Books. (All children received a book reward at the post—testing session--one last book for the experi- mental group and the set of ten for the control groups.) Books were chosen with a focus on vocabulary building, that is titles representing some variety of vocabulary were selected. These books are colorful, brief and appealing to three year olds. A card with interaction ideas was also provided for these books. A complete list of books and sample idea cards is included in Appendix C. Two further educational approaches were also used. Each mother was invited to attend two small group meetings led by Lansing Head 27 Start personnel: Mrs. Sue Millar, Acting Director and Mrs. Kathy Dunlop, Coordinating Teacher. These meetings included a brief welcome, state- ment of purpose and sample reading time to serve as a model for mothers and using a Head Start child. Both white and Negro models were selected for the children. Nine mothers attended at least one meeting. Three of these mothers attended two meetings. Following up from these meetings a series of simple information sheets were developed emphasizing various aspects of book use and high- lighting interaction. Copies of these sheets are included in Appendix D. Sheets were printed in primary type to maximize readability. Twenty children were assigned to this experimental group and 19 completed the study adequately. These 19 included five white boys, four Negro boys, five white girls and five Negro girls. T2--Placebo Control Group The placebo control group was designed to provide families with equal time from workers as well as literature to the home. Workers were instructed in such a way as to present this as a group of equal importance in the study. As with the experimental group Family Helpers assisted and attended a weekly meeting to distribute materials and discuss problems. The placebo control mothers received a weekly pamphlet on the subject of foods and nutrition. In addition mothers were invited to attend small group meetings similar to those held for the experimental groups. Literature and groups meetings were organized by a trained public health nutritionist, Mrs. Annabelle Pangborn. Nine mothers and one father attended one meeting each. 28 Nineteen children were assigned to this group and 14 completed the study. These included two white boys, four Negro boys, four white girls and four Negro girls. T3——Control Group The control group received no assigned contact during the experi- mental period, although workers continued to have occasional contacts related to their regular duties. Seventeen children were assigned to this group and fourteen completed the study. These included three white boys, three Negro boys, four white girls and four Negro girls. RESULTS Reliability Since only the Binet and the PPVT are standardized measures, pre- and post-test correlations were first examined for all measures as a rough estimate of the reliability of the instruments. The cor- relations between the pre— and post-test scores of the Binet, the PPVT, Expressive 1 and 2 and MSL were .92, .75, .67, .71, and .70 respectively. All five of these lie in the acceptable range for such measures. How- ever, the correlations between the pre- and post-test scores of the revised Birch l, 2, 3, and 4 and the Inattention score (.07, .07, .42, .17, and .26 respectively) were substantially lower. Therefore, only results for the following five measures are reported below: Binet, PPVT, Expressive l, Expressive 2, and MSL. This change also eliminated testing the minor hypothesis 2. Treatment Group Mean Differences The overall multivariate E for the ten tests was 2.48 (df_- 20,32; p_< .01). For this reason and because there were §_priori hypotheses with respect to the effects of differential treatment for each of the five selected tests, the Dunnett t-test (Winer 1962, pp. 89 ff.) was employed. Table 1 presents the means (adjusted for pretest scores) for the five tests. A 2(Sex) x 2(Race) x 2(Treatment) analysis of covariance was applied to the post-test data using the least squares solution for 29 30 unequal cell frequencies.4 Correlations with the following variables were examined to select the necessary covariates: pre-test scores for each measure, age, number of siblings, presence of a father in the home, Opportunities for enrichment outside the home, working mother, and place and level of mother's education. The only significant cor- relations were the pre-test measures, and these were selected as the covariates. A series of tables in Appendix E presents the total array of adjusted cell means for all ten measures. Since the focus of this study is on treatment, any significant main effects for race and sex will not be presented. Moreover, only one interaction was significant and that will be presented below. Table 1: Means (adjusted) for Treatment Groups for the Five Reliable Dependent Measures Measure Treatment Condition Binet PPVT Exp. #1 Exp. #2 MSL Experimental 95.64 82.84 45.20 47.44 4.32 Placebo 91.10 79.75 32.20 40.17 4.26 Control 90.17 73.31 36.68 32.78 3.72 Stanford Binet Intelligence Scale As Table 1 indicates, the mean Binet scores for the three treat- ment groups were in the predicted orders Moreover, the results of the Dunnett's tests indicate that the predicted differences were significant: 4When the data were assembled it was found that three children were missing a total of six pre-test scores (one lacked three, one two and one one only). It was decided to retain these children in the. sample, particularly since dropping them would have made computer pro- cedures difficult if not impossible. Therefore, a mean replacement score was made using the score for all children of the same race and sex. 31 comparing T1 and T2, c = 3.08 (g = 34; p < .005). t = 2.56 (d£_= 34; p_< .05); comparing T1 and T3, Peabody Picture Vocabulary Test As Table 1 indicates, the mean PPVT scores for the three treat- ment groups were in the predicted order. However, only the difference between the experimental and the control groups was significant, t = 2.28 (g§_= 34, p_< .05). Moreover, since (as indicated above) the interaction between Sex and Treatment was significant (F.- 3.83, g§_= 34; p_< .05) for this test, the simple main effects were also examined using Dunnett's t-test. Table 2 presents the means for treatment condition separately for sex. Table 2: Means (Adjusted) for Treatment Groups Separately by Sex for the PPVT Sex Treatment Group Male Female Experimental 75.49 89.46 Placebo 83.91 74.34 Control 70.43 75.72 Table 2 indicates that for males only the difference between the experimental and the control groups was in the predicted direction. Moreover, the results of the Dunnett's test indicate that no differences were significant. On the other hand, the mean PPVT scores for females were in the predicted order. Moreover, the results of the Dunnett's 32 tests indicate that the predicted differences were significant: com- paring T with T2’ t - 3.62 (d£_- 34; p_< .005); comparing T with T 1 l 39 c = 3.29 (g_f_= 34, p_ < .005). Expressive Measure #1 As Table 1 indicates, the mean Expressive 1 scores for the three treatment groups were in the predicted order. However, only the dif- ference between the experimental and the placebo groups was significant, c = 2.00 (g- 34, p < .05). Expressive Measure #2 As Table 1 indicates, the mean Expressive 2 scores for the three treatment groups were in the predicted order. However, only the dif- ference between the experimental and the control groups was significant, c - 2.23 (933:- 34, p < .05). Mean Sentence Length As Table 1 indicates, the mean MSL scores for the three treat- ment groups were in the predicted order. However, only the difference between the experimental and the control groups was significant, t - 2.31 (_d_f_ - 34, p_ < .05). DISCUSSION Implications for the Compensation of Cultural Deprivation It is very clear from these results that by this relatively simple intervention measurable language improvement has taken place. Irwin (1960) found his first significant differences after four months and the differences between the three groups widened appreciably over the period of the study (18 months). It can therefore be predicted that the changes appearing here after three months would also widen over a longer period of time should a long-term program be initiated. Furthermore, the high level of cooperation in the experimental group (a five percent loss compared with 25 percent for the two control groups) shows that such a program would appeal to mothers. The program does require a commitment of time and personnel, but this is minimal compared with the time and professional services necessary for a program such as year-round Head Start classes. In addition, since the mother 1e_the teacher here, changes in the patterns of verbal exchange between mother and child must take place, if only during the reading hour. Like the Gray study (1966) diffusion to younger siblings could also be expected. Furthermore, it is reasonable that some community agency or service group could serve as the book delivery teams, possibly on a two-week basis, rather than the regular once-aeweek program described here. In time mothers might also be trained to make use of a bookmobile service. 33 34 The loan books could probably be arranged for a project of this kind. All three libraries cooperated readily and the local librarians as well as a state children's book consultant were most helpful in selecting books for the age group in this study. While the project would involve an initial outlay of books, the families can be counted on to care for books. A total of only five books were lost and only two or three others damaged over the four month period with almost 150 books in circulation. In general these libraries seemed to be designed to serve a middle class population--by their location and their formal atmosphere. However, all the libraries indicated an interest in the project, and one of the three has just opened a new branch in a poverty area, a branch library planned to attract lower class families. Even the gift books could well be obtained through local resources if not through government funds. In a long-term program these books might be distributed periodically during the early months. This study was with three-year-olds. Irwin's research (1960) was with an even younger group. Such a program might well be combined with Head Start by beginning book distribution when children are two and continuing it up to and during the four—year-old Head Start and kindergarten until the school library program takes over from there. This would have the further advantage of involving mothers at home while children are quite young so that they would be better prepared to gain further skills for themselves through Head Start parent programs. If the mothers came to perceive themselves as successful teachers of their children, other kinds of enrichment activities might gradually be introduced to broaden the scope of the program of training and in Levenstein's program (1968, 1969a, 1969b) with less overall preparation 35 and professional leadership than either Schaeffer (1965) or Levenstein (1968) originally suggest. Therefore, the program suggested by this study could be carried out over an extended time and expanded both for the age of the child and the content of the program. All this can be implied from this brief research using three small groups of three- year-old disadvantaged children in a relatively simple, practical at- home book reading program. Implications for Further Research The above discussion provides a basis for a further study or studies expanding this program and modifying it in one way or another. Modifications might include different systems of delivery--midd1e class volunteers, bookmobile services, school library services-~to name three. Expansions would be by the extension of ages, both down and up, as well as broadening the scope of the program from books to other activities. Another area which this study only touches upon might well be the differential effects of trainers. Here the mothers read selected simple books to their children. We know that many of these mothers do not have adequate reading skills and the books were chosen to be at a simple grammatical level. The Bernstein research (1961) would also lead us to predict that the language patterns of these mothers are somewhat limited. The gains by the placebo group may reflect the simple effects of attention, but it could be important that the experimental and placebo group means were closest for mean sentence length. The question may be raised as to whether the reading had less effect on MSL than did the weekly visits of the Family Helpers. These Helpers have the advantage of being close to the families they serve by virtue of their 36 own origins. They are, however, women who have coped successfully with their environment, and whose language patterns can be predicted to correlate with their higher class level. It would be of considerable interest to study in some depth the differential effects of training using personnel with measurable differences in grammatical level and language style. Implications for Research Measurement A number of different problems were encountered with the measures of this study. Because a large number of studies are being conducted in this field, and because the available standardized measures are limited, some of these problems will be further pursued here. Reliability The necessity of eliminating one half of the measures in this study was an unfortunate development. There are three possible explana- tions: (1) that the behavior of three-year-olds in the areas of these measures is extremely variable, (2) that the sample of data for each measure was insufficient to provide a reliable record of that behavior, or (3) that the method of measurement was inadequate to measure the be- havior. Each one of these possibilities can be considered separately. (1) It should first be noted that the behavior of these five measures falls in a category closer to behavioral style than to verbal skill, that is, the way of responding is measured rather than level of suc- cessful verbal response. This style includes verbal as well as non- verbal components, but it does not look at verbal behavior strictly as a skill or ability. The other five measures, on the other hand, are all designed to measure ability, four of them language ability 37 alone, and the fifth (the Binet) some combination of language abilities with other abilities. What this test-retest reliability data may say is simply that the three-or four-year—old child does not yet have a consistent behavioral style in this test situation. Whether or not his style is more consistent in other areas is beyond the scope of the data of this study. (2) It is also likely that the sample of data was insufficient. The range of scores for most of these measures was narrow. Whereas Binet IQs ranged from 62 to 129 with a normal distribution, inattention scores, for example, ranged only from 0 to 9 with a definite skew toward lower scores. It is not clear why the ratio of verbal labelling to total non-verbal items (VB/TNV) had a test-retest correlation of .42, while a measure like the ratio of verbal comments to total items was much lower (.17). It may be recalled that VE/TNV was the labelling behavior exhibited during non—verbal administration. Possibly for some children this was perceived as the test demand and in some sense, since these labels were correct language use in most cases, might be seen as an ability, perhaps this measure falls more toward the area of the five verbal skill measures per se. (3) The failure of the measure to actually record the behavior it set out to record may also have been involved. This appeared to be the case with the inattention Score. The boundaries for units of inatten- tion were quite visible specific behaviors. It was the observation of the examiner that children had other methods of failing to pay atten- tion which did not compete with test administration as did these specific behaviors, and which would have been most difficult to measure under these circumstances. Passive inattention (with the glassy-eyed 38 look) or darting glances were not recordable on this measure and may well have been more frequent or better indicators of inattention than the infrequent gross behaviors that were measured. One further comment is in order regarding the reliability data for the Binet and the PPVT. The test manuals for each report relia- bilities for young children. For the Binet reliabilities for ages 2-1/2 to 5-1/2 range from .83 (IQs 140-149) to .91 (IQs 60-69); the reliability for this group was .92. For the PPVT reliabilities for ages 3—0, 3-6 and 4-0 are .75, .81, and .77 respectively; the relia- bility for this group was .75. Since these were children from dis- advantaged families with particularly visible problems in the community, one might well have predicted more variable behavior than a normal sample. The success of measurement may be attributed to the very good testing situation within the informal family helper offices and the use of M & Ms to reward responses, thus increasing attention and decreasing variable behavior. §pecific Measures Some comments can be made about the use of the PPVT of this study. All children made responses (one child had to be trained to respond using M & Ms), but some children very clearly found the pictures a poor attention-getter. While all children responded to the Binet with some degree of interest, a fair number of children simply were not "with it" on the PPVT. It was expected that pictures might prove a poor method for this group of children since deprived children generally have had relatively little contact with pictorial material or with the objects that may be pictured. The mean IQ for the PPVT 39 probably demonstrates this. Whereas the mean post-test IQ for the Binet was 91.99, the mean post-test IQ for the PPVT was 78.44. In addition a few children were below the norms. This was not true for the Binet-—in the sense that all children passed the items at the basal age of II or above. Two of the measures specifically created for this study were Expressive 1 and 2. These did appear to be useful for the measurement of expressive language. The test-retest correlation as a rough esti- mate of reliability was adequate (.67 and .71) and both measures showed treatment effect. From the evidence here no conclusions can be made of the relative effectiveness of objects versus pictures as stimuli to elicit language. This had already been observed in the pilot study. It was evident during the test administration that the pictures for this measure were much more attractive to the children than those of the PPVT. The difference in pictures--the black and white line drawing of single objects compared with the full color of action scenes--undoubtedly accounts for this. It is likely that the lower class child needs a picture with higher stimulus impact in order to make the responses required by the PPVT. SUMMARY The purpose of this study was to determine whether the effects of cultural deprivation on language skills can be remediated through - experimentally increasing verbal interaction between mothers and their three-year-old children. Research indicates that language skill deficiency exists as early as 18 months and that disadvantaged mothers interact verbally less with their children than advantaged mothers do. Current projects rarely center on three-year-olds. They also use professional trainers more often than they use the mothers themselves. This study was designed to assess the potential of a simplified book-reading program. for three-year-olds utilizing the potentials within the home as well as available community resources. It was hypothesized that children in the experimental group with treatment designed to increase mother-child verbal interaction would show greater language skills and attention than a comparable control group. In order to examine the effects of experimental condi- tions a placebo control group received nutrition materials. It was hypothesized that verbal interaction treatment would have an effect that would also be greater than any experimental effect on the placebo control group. Forty-seven disadvantaged three-year-olds, Negro and white, boy and girl, completed the study. These children were randomly assigned 40 41 to three groups balanced by race and sex. The experimental group of 19 was committed to three months of a daily 10-15 minute period of interaction using appropriate library books and inexpensive gift books delivered weekly for three months by paraprofessional Family Helpers. Family Helpers also delivered nutrition materials weekly to the 14 members of the placebo control group. Another 14 children had no contact. Children were studied both before and after the experimental period using known tests of vocabulary and intelligence, expressive responses to toys and pictures, mean sentence length, four functional categories of responses and an attention score. Only the first five of these were sufficiently reliable to analyze the data. Results of statistical analysis showed that the Binet scores gave complete support to the hypothesis at the .05 level for the placebo and the .005 level for the control group differences from the experimental group. In addition the experimental group scored signifi- cantly higher than the control group (p < .05) on three of the other four measures and significantly higher than the placebo control group on one of the other five measures. For one additional measure experi- mental group girls scored significantly higher than placebo control group girls. In general, therefore, the results of this study showed that a three month program designed to stimulate verbal interaction between disadvantaged mothers and their three-year-olds through the medium of children's books did have an effect on verbal skill greater than that found in placebo or control groups. There was some indication 42 that the placebo situation also had an effect on the children, but to a lesser degree. This research indicates the usefulness of a book distribution program administered through available paraprofessional workers in the community involving disadvantaged mothers reading to their own pre- school youngsters. BIBLIOGRAPHY BIBLIOGRAPHY Bereiter, C. and Englemann, S. Teaching Disadvantaged Children in the Preschool. Englewood Cliffs, N.J.: Prentice Hall, Inc. 1966. Bernstein, B. Social class and linguistic develOpment: a theory of social learning. Education, Economy and Society, A. H. Halsey, J. Floud and C. A. Anderson (eds.) New York: Free Press of Glencoe, Incorporated, 1961, 288-314. Brown, R. and Bellugi, Ursula. Three processes in the child's acqui- sition of syntax. Lenneberg (Ed.) New Directions in the Study_of Language, 1964. Cherry, Estelle. Children's comprehension of teacher and peer speech. Unpublished manuscript. Institute for Developmental Studies, Department of Psychiatry, New York Medical College, June 1964. Deutsch, Cynthia. Auditory discrimination and learning: social factors. Merrill Palmer Quarterly, 1964, 19, 277-296. Deutsch, M. The disadvantaged child and the learning process. Education in Depressed Areas. A. Harry Passow (ed.) New York: Teachers College Press, 1963, 163-180. . Facilitating development in the preschool child: social and psychological perspectives. Merrill Palmer Quarterly, 1964, 19, 249-263. Dunham, R. M. Project Know-How. Unpublished project description. Institute of Human Development. Florida State University. Gordon, I. Early child stimulation through parent education: Research report: Infant Performance. Unpublished manuscript. Institute for Development of Human Resources. University of Florida, 1968. Gordon, I. and Lally, J. R. Early child stimulation through parent education: The training program for parent educators. Unpub- lished manuscript. Institute for Development of Human Resources. University of Florida, 1968. Gray, Susan W. and Klaus, R. Deprivation, development and diffusion. Presidential Address, Division of School Psychology, APA, Sept. 4, 1966. 43 44 Gray, Susan W., Miller, J. 0., Hinze, R. H. and Schoggen, Maxine. Research, change and social responsibility: An illustrative model from early education. DARCEE Papers and Reports, Vol. 2, #3, 1967. Hertzig, Margaret E., Birch, H. G., Thomas, A. and Mendez, Olga. Class and ethnic differences in the responsiveness of preschool children to cognitive demands. Mbnographs of the Society for Research in Child Development, Vol. 33, No. 1, 1968. Hess, R. D. and Shipman, Virginia. Early blocks to children's learning. . Cognitive elements in maternal behavior. Unpublished manu- script. University of Chicago, undated. . Maternal attitude toward the school and the role of pupil: some social class comparisons. Fifth Work Conference on Curriculum and Teaching in Depressed Areas at Columbia Univer- sity, Teachers College, June-July 1966. Horton, Della M. A training program for mothers. Unpublished manu- script. DARCEE, undated. Hunt, J. McV. The psychological basis for using pre-school enrichment as an antidote for cultural deprivation. Merrill Palmer Quarterly, 1964, lg, 209-248. Irwin, O. C. Infant speech: the effect of family occupational status and of age on sound types. Journal of Speech and Hearing:Dis- orders. 1948, l}, 224-226. . Infant speech: the effect of family occupational status and of age on sound frequency. Journal of Speech and Hearing_ Disorders. 1948, l}, 320-323. . Infant speech: Effect of systematic reading of stories. Journal of Speech and Hearing Research, 1960, 3, 187-190. John, Vera P. The intellectual development of slow children: some preliminary findings. American J. Orthopsychiat., 1963, 33, 813-822. John, Vera P. and Goldstein, L. S. The social context of language acquisition. Merrill Palmer Quarterly, 1964, l9, 265-275. Levenstein, Phyllis. Aiding cognitive growth in disadvantaged pre- schoolers. Unpublished manuscript. Nether-Child Home Program. Progress Report, Feb. 1968. Loban, W. Language Ability: Cooperative Research Project 1131. Berkely: University of California, 1964. 45 McCarthy, Dorothea. Language development in children. Manual of Child Psychology, Carmichael, L. (ed.) New York: Wiley, 1946, 476-569. Milner, E. A study of the relationship between reading readiness in Grade 1 and patterns of parent-child interaction. Child Develop- Olim, E. G., Hess, R. D. and Shipman, Virginia C. Maternal language styles and their implications for children's cognitive develop- ment. Expanded version of "The Effect of maternal behavior on cognitive development and impulsivity." APA Chicago, Sept. 1965. Pines, Maya. Revolution in Learning, Harper and Row, New York, 1966. Rubow, Carol L. The effectiveness of three training methods for teacher aides working in pre-school classrooms. DARCEE Papers and Reports, Vol. 2, #4, 1968. Schaefer, E. 8. Intellectual stimulation of culturally deprived in- fants. Excerpted from Mental Health Grant Proposal No. MH - 09224-01. June 1965. Schaefer, E. S. and Fluffey, P. H. Intellectual stimulation of cul- turally deprived infants during the period of early verbal development, Unpublished manuscript. June 1967. Sexton, Patricia. Education and Income: Inequalities of Opportunity in our Public Schools. New York: Viking, 1961. Shipman, Virginia and Hess, R. D. Early experience in the socializa- tion of cognitive modes in children: A study of urban negro families. A paper presented at the fifth annual conference on family and society at the Merrill Palmer Institute, April, 1966. Weikart, D. P. and Lambie, Delores. Preschool intervention through a home teaching program. Unpublished manuscript. June 1967. Wilcoxon, Reba. The Picture Story of DARCEE, DARCEE, 1967. Winer, B. J. Statistical Principles in Experimental Design, McGraw- Hill Book Co., New York, 1962. APPENDICES APPENDIX A DESCRIPTION, ADMINISTRATION AND SCORING PROCEDURES FOR EXPRESSIVE MEASURES #1 AND #2 APPENDIX A DESCRIPTION, ADMINISTRATION AND SCORING PROCEDURES FOR EXPRESSIVE MEASURES #1 AND #2 #1 Expressive Measure, Toys Description of Toys Red Plastic Car Small Doll in Dress Grey Plastic Coffee Cup Green Plastic Ball Silver Plated Teaspoon White Infant Shoe Administration The toys were kept out of sight and presented one by one in order. Boys received the car first; girls received the doll first. The last four items were presented in the order above. As the toy was presented the examiner asked "What is it?" The question was followed by two efforts to elicit descriptive responses usually, "Tell me about it." Very reticent children were encouraged somewhat more to get them started. A final question was "What do you do with it?" This was asked of every child - whether or not he had already made some doing response. Scoring Instructions General: A. Score leniently. B. Name of object 8 1 point. Car - car Doll - doll, baby etc. 46 47 #1 Expressive Measure, Toys (Cont.) C. Modifiers - 1 point. Igy_car, baby_doll, baby_shoe, coffee cup D. Descriptive details = 1 point. Car: wheels, roof, trunk etc. Doll: dress, legs, arms, etc. E. Uses and activities - both nouns and verbs a 1 point. Car: ‘M§n_can.d£iyg it; you pg k'it in the garage; you giyg_ Doll: the dolly a bottle Cup: you drink with it (milk, water etc.) F. Synonyms - 1 point. Doll - baby Car - buggy G. Negatives a 1 point Ball: This is not a football. Special Scoring Notes 1. Repeated description Score on the first time only per toy as in "baby shoe, shoe for a baby." 2. Repetitions from toy to toy in a vague form Score on the first use only as in "Take it home" or "Put it back." 3. Vague descriptive responses Score on the first use only per set as in "big doll, big car." 4. Partial inaccuracies - 1/2 credit only Ex. Car - truck; cup - glass, ball - apple. If the correct response is made later, score both. 48 #1 Expressive Measure, Toys (Cont.) 5. Personal associations Some children make repeated personal associations. Much of this receives no score, but when words are used which would be scored if descriptive, score as indicated. Ex. My Johnny has new shoes; they have laces like these. 6. Responses to the question "What do you $2 with it?" receive half credit for the first verb response that is a doing verb. Other parts of this response receive full credit. 7. No Score: a. Vague nothings such as: "see this" "that and that" b. Clear inaccuracy It's an apple (score 1/2), you eat it (0) All scores were doubled before recording to eliminate 1/2 scores. #2 Expressive Measure, Pictures Description of Pictures 1. Firemen in outfits running toward a red firetruck with driver. 2. Postman delivering mail to two children (Man is white, children are Negro). 3. Park scene - interracial group in many activities - sandbox, tricycle, children running, mother with carriage. Pictures used for the measure Expressive #2 were published by the David C. Cook Publishing Company in Elgin, Illinois. Pictures of the firemen and postman were from the series A 890: Community and Home Helpers. The playground scene was from the series A 892: Variety Subjects. Administration The pictures were kept out of sight and presented one by one in order, using the order above for all children. As the picture was presented the examiner said, "Look at this picture and tell me all about it." This question was followed by two efforts to elicit further reSponses - usually "Tell me more about it" or "What else can you gell me about it." Very reticent 49 #2 Expressive Measure, Pictures (Cont.) children were encouraged somewhat more to get them started. A final question was "What are they doing?" This was asked of all children whether or not they had already made a doing response. Scoring Instructions General A. Score leniently. B. Nouns for items in_the picture-:1 point each. fireman, truck, boy, spoon, etc. C. Verbal descriptions - 1 point each climb on it, getting something, playing etc. D. Adjectives B 1 each big truck, little boy, red dress, two shovels etc. E. Accurate negatives - 1 point each This is not a swimming pool, is it? etc. Special Scoring Notes 1. Repetitions: Score on the first time only per picture "Little boy and little girl" "N53 shoes and new shoes and new shoes" Partial inaccuracies - 1/2 Fireman - indian firetruck - car mailman - police painting for sandbox scene Responses to the question "What are they doing?" receive 1/2 credit for the first verb that is a doing response. No score for a. Vague nothings b. Clear inaccuracy (rare) c. The verb got when used for "has" All scores were doubled before recording to eliminate 1/2 scores. APPENDIX B BOOK TITLES APPENDIX B BOOK TITLES The following list gives the titles of the books used for this study. Bannon, Laura Red Mittens Boston, Houghton Mifflin Co., 1946. Toby's Friends Chicago, A. Whitman, 1963. Barbaris, Franco Would You Like a Parrot? New York, Scroll Press, 1964. Baum, Arline and Joseph One Bright Mondangorning_ New York, Random House, 1962. Begley, Evelyn M. My Color Game Racine, Wis., Whitman Pub. Co., 1966. Brinbaum, Abe Green Eyes Irvington, N.Y., Capitol Pub. Co., 1953. Bright, Robert The Friendly Bear Garden City, N.Y., Doubleday, 1957. I Like Red Garden City, N.Y., Doubleday, 1955. My Red Umbrella New York, W. Morrow, 1959. Brown, Margaret Wise The Country Noisy Book New York, W. R. Scott, 1940. The Golden Egg Book New York, Simon & Schuster, 1947. Goodnight Moon New York, Harper, 1947. The Indoor NoisyiBook New York, W. R. Scott, 1942. The Noisy Book New York, W. R. Scott, 1939. The Summer Noisy Book New York, Harper, 1951. The Winter Noisy Book New York, W. R. Scott, 1947. Browner, Richard Everyone Has a Name New York, Henry 2. Walck Inc., 1961. Brustlein, Daniel The Elephant and the Flea New York, McGraw-Hill Book Co., 1956. Buckley, Helen Josie and the Snow New York, Lothrop, Lee & Shepard, 1964. Where Did Josie Go? New York, Lothrop, Lee & Shepard, 1962. 50 51 Daugherty, James Henry Andy and the Lion New York, The Viking Press, 1938. DeRegniers, Beatrice Schenk May I Bring a Friend? New York, Atheneum, 1964. Ets, Marie Hall Gilberto and the Wind New York, Viking Press, 1963. Just Me New York, Viking Press, 1965. Play with Me New York, Viking Press, 1965. Fenton, Edward The BiggYellow Balloon Garden City, N.Y., Doubleday, 1967. Flack, Marjorie Angus and the Cat Garden City, N.Y., Doubleday Doran & Co., Inc., 1931. Angus and the Ducks Garden City, N.Y., Doubleday Doran & Co., Inc., 1930. Angus Lost Garden City, N.Y., Doubleday Doran & Co., Inc., 1932. Ask Mr. Bear New York, The MacMillan Co., 1932. Foster, Doris Van Liew Tell Me Little Boy New York, Lothrop, Lee and Shepard, 1953. Freeman, Don Corduroy New York, Viking, 1968. Friskey, Margaret Chicken Little, Count to Ten Chicago, Children's Press, 1946. Garten, Jan The Alphabet Tale New York, Random House, 1964. Gay, Zhenya Look New York, Viking Press, 1952. Green, Mary McBurney Is it Hard, Is it Easy? New York, W. R. Scott, 1948. Grossbart, Francine B. A Big City New York, Harper and Row, 1966. Hawkinson, Lucy Ozone All in One Day Chicago, Whitman, 1955. Days I Like Chicago, A. Whitman, 1965. Hoff, Sydney Lengthy New York, Putnam, 1964. Hoffman, Beth Greiner Red Is for Apples New York, Random House, 1966. 52 Jacobs, Leland Goodnight Mr. Beetle New York, Holt, Rinehart & Winston, 1963. Keats, Ezra Jack Peter's Chair New York, Harper & Row, 1967. The Snogy Day New York, Viking Press, 1962. Whistle for Willie New York, Viking Press, 1964. Krasilovsky, Phyllis The Very Little Boy Garden City, N.Y., Doubleday, 1962. Krauss, Ruth The Bundle Book New York, Harper, 1961. The Carrot Seed New York, Harper & Bros., 1945. I Can Fly New York, Simon & Schuster, 1950. Krum, Charlotte The Four Riders Chicago, Wilcox & Follett Co., 1953. Langstaff, John M. Over in the Meadow New York, Harcourt, Brace, 1957. Lionni, Leo Inch by Inch New York, I. Obolensky, 1960. McCloskey, Robert Blueberries for Sal New York, Viking Press, 1948. McKie, Roy and Eastman, P. D. Snow New York, Beginner Books, 1962. Miles, Betty A House for Everyone New York, Knopf, 1958. Munari, Bruno ABC Cleveland, World Publishing Co., 1960. Nodset, Joan Who Took the Farmer's Hat New York, Harper & Row, 1963. Palazzo, Tony Animal Babies Garden City, N.Y., Garden City Books, 1960. A Little Red Hen Garden City, N.Y., Garden City Books, 1958. Petersham, Maud and Miska The Box with Red Wheels New York, MacMillan Co., 1949. The Circus Baby_ New York, MacMillan Co., 1950. Piper, Watty The Little Engine that Could New York, The Platt & Munk Co., Inc., 1930. 53 Preston, Hall and Catherine Barr Snopp Waits for Dinner New York, Oxford University Press, 1952. Robinson, Irene Bowen Picture Book of Animal Babies New York, The MacMillan Co., 1947. Rojankovsky, Feodor Animals in the Zoo New York, Knopf, 1962. Animals on the Farm New York, Knopf, 1967. Schlein, Miriam Billy, the Littlest One Chicago, A. Whitman, 1966. City Boy,_Country Boy, Chicago, Children's Press, 1955. Snowtime Chicago, A. Whitman, 1962. Seignobosc, Francoise Jeanne Marie Counts Her Shegp New York, Scribner, 1951. Shapp, Charles Let's Find Out What's Bigfand Small New York, F. Watts, 1959. Slobodkin, Louis One Is Good, But Two Are Better New York, Vanguard Press, 1956. Slobodkina, Esphyr Caps for Sale New York, W. R. Scott, Inc., 1940. Smith, Eunice Young The Little Red Hen Chicago, A. Whitman, 1961. Tensen, Ruth M. Come to the Farm Chicago, Reilly & Less Co., 1949. Come to the Pet Shop, Chicago, Reilly & Less Co., 1954. Come to the Zoo Chicago, Reilly & Less Co., 1948. Tresselt, Alvin R. Hide and Seek Egg New York, Lothrop, Lee and Shepard, 1965. Rain Drppsg§plash New York, Lothrop, Lee and Shepard, 1946. Sun Up New York, Lothrop, Lee and Shepard, 1949. Wake Up, City_ New York, Lothrop, Lee and Shepard, 1957. Wake Up, Farm New York, Lothrop, Lee and Shepard, 1955. Udry, Janice May A Tree Is Nice New York, Harper, 1956. Ward, Lynd Kendall The Biggest Bear Boston, Houghton Mifflin, 1952. Wildsmith, Brian Mother Goose New York, F. Watts, 1964. Zion, Gene Harry the Dirty Dog New York, Harper, 1956. 54 Books distributed as gifts are listed below. All of them are published by the Golden Press of New York. Series Title: "A Little Golden Book" Brown, Margaret Wise The Color Kittens DeWitt, Cornelius The Little Golden ABC Rojankovsky, Feodor The Three Bears SEries Title: "A Golden Shape Book" Davis, Daphne The Baby Animal Book Dugan, William The Truck and Bus Book Kaufman, Joseph The Snowman Book Miller, John The House Book Pfloog, Jan The Farm Book The Zoo Book Shortall, Leonard The Hat Book APPENDIX C SAMPLE SUGGESTION BOOK CARDS APPENDIX C SAMPLE SUGGESTION BOOK CARDS My House Book Read this book together. See if your child can name the things in the pictures. If not you say the name and see if he can find it. Don't forget those tools in the basement or the garage. Play a game in your house. Say: What room is this? Then see if he knows the names of all the things in the room - table, chair, light, clothes, and so forth. With two or three children you could make a game to see who knows the most names. Maybe you could make a list and write down each new name he learns. Be sure to help him when he needs help - he will learn. Tell him how well he does when he learns a new name. 55 56 Three Bears Read the story for fun. Then look at the pictures the next time. Help your child find the big bear, the middle-sized bear and the little bear -- the big chair, the middle-sized chair and the little chair, and so on. Try pig and little as a game this week. Can he tell you which is the big spoon and which is the little spoon -- or which is the big chair and which is the little chair at your house? After he knows this story well can he help you tell it? Stop on a page and see if he knows what happens next. Always help him when he needs help - do not scold - he will learn. Tell him how well he does when he can answer or tell the story. 57 The Hat Book This little book shows lots of hats and tells why they are used. Look at the pictures as you read and see the hats used in each picture. Can your child tell you what the hat animals are doing? Help him. Can he learn what color the hats are? Help him. Colors are hard for some children, but your child may be ready to learn a few colors. Next time you go for a walk or a trip plan to look for hats. Can you see a fireman or a mailman or a worker with a special hat? Be sure to show your child and talk to him about the hats. One of the important ways your child learns is by what you show him and tell him about. Show him hats and show him many other things at home and when you go out. Talk to him about what you see - cars on the street, food in the grocery store, the weather - whatever you see. 58 The Zoo Book Here are the zoo animals. Can your child look carefully at them and learn their names? Then play this guessing game. Say "I know an animal that is like a horse and has black and white stripes - what is it?" Can your child guess. Help him. You can also teach him the noise the animal makes and play a fun guessing game with the noises. "I go meow - who am I?" Can you visit the Potter Park Zoo?? When you do talk about the animals - what they are doing, what they are eating, where they live. Take a picnic and play on the swings too! APPENDIX D PARENT MATERIALS APPENDIX D PARENT MATERIALS The following pages are the materials delivered with books to mothers in the experimental group at intervals during the first two months of the study. The original materials were typed with large primary type on 8-1/2 x 11 sheets of paper. 59 60 #1 Thank you for bringing your child to visit with me this past month. I am looking forward to seeing you again in May. Mary Clare Boroughs NOW- You and your child are now members of a book group in the study of 33 and 43. Reading from storybooks is one very good way to help your child learn and to have him ready for school. You will get two children's library books each week for 12 weeks. Your Helper will bring the books to you each week and the next week she will change them for 2 more. Starting in two weeks your child will also get one little book a week to keep. This will be your child's prize for helping with this study. There will also be two short meetings for mothers in this group. At these meetings a Head Start teacher will talk and show about reading to your child. We are asking you to EEZ.t° read to your child for 10-15 minutes a day. Your other young children may join in booktime too. Taking Care of Books Books are for reading and for looking. They can be torn or broken. Keep your books in the big folder and put them away each day. We must return them for someone else to read next week. Tell your child that a book is to keep safely; so he can read it again. Each time you read make a check on the list on your folder. This will give us an idea of how often you were able to read aloud together. 61 #1 (cont.) Starting Out First look at the books yourself - look at the pictures and read the story. When you read to your child you can read the s tory or tell the story orlook at the pictures. In a pocket at the front of the book or at the back of the book are some helps for you with your child. Leave the card in the pocket for the next family to try. When to Read Try to read the same time each day. Pick a time when the TV is off and your child feels quiet. Some mothers like bedtime but for others just before or after a nap is better. Maybe right after lunch when older children go back to school is a good time. Or right after breakfast might be best for your child. Where to Read A big chair near a light is good. If you read to 2 or 3 children try a couch or a bed. One child may sit on your lap. Be sure they all ggg_the pictures so you can talk about them. 62 #2 CHILDREN'S BOOKS ARE A STORY The books you read have a story. After you read see if your child understands the story? Can he tell you what they did? Does he know their names? Talk about it a minute. Tomorrow after you read the first part see if he can tell you how the next parts goes. CHILDREN'S BOOKS ARE PICTURES Pictures tell a lot. Talk together about the pictures. Can he tell you what he sees? Ask what they are doing in the picture. Try these picture time games: 1. Show Me: Look at a picture. Say "Show me the ___," When he finds it say, "Good." If he can't find it help him. When he can find things in the pictures well, try this- 2. Tell Me: This time point to something in a picture and say, "Can you tell me what this is?" or "Tell me about this picture." Help him to play the Tell Me game. Is reading time a happy time? Have fun as you talk with your child. When he is tired put it away for another time. 63 READING AND TALKING TOGETHER Reading aloud is fun because children like stories. Children also like talking with you and being near you. A child learns when he talks with his mother A child who talks with his mother learns many things that will help him talk and think better - and help him talk with his teacher about stories and books when he gets to school. This means YOU help him talk and think. Maybe you have been reading about a dog. Was the dog like your dog? Was it like the dog next door? Ask your child? Talk about what the dogs are like. Next time you see a dog be sure he sees it too and talk about it again. Is it big or little? What color is it? How old? What is doing? Was the story about a boy - or some hats or a truck? Talk about that. Helping Cards At the back of each book is a card with ideas for you to talk about. You may want to talk about these at reading time. Maybe you can talk about them later in the week. Can all the family talk about some of them? #3 it 64 #4 BOOKS GO OUT WALKING A book can lead you and your child on a trip. They can add fun & interest to life. Did you read about animal babies? Find some kittens or puppies near you. Visit them and talk about how little they are, what color they are, and what they are doing. Did you read about snow or spring. Look outside and see the weather. Talk about it. Go for a walk. Is it sunny or dark? Did the rain leave a puddle? Is there snow on the ground? Are their leaves on the tree? Show your child. Did you read about a truck or a fire engine? Can you walk to see one? - an oil truck, or a power digger, or a fire truck? Put up a curtain and see what you can see on your street today. Talk about it. Did you read about the zoo? Can you plan a family trip to Potter Park? Talk about the animals. Do your children know their names? Be sure they see what the animals do. Do they have a tree or rocks or a cave or water. What do they eat? DON'T FORGET THE IDEA CARD - at the back of each book. Try to do something from the card for each book. 65 #5 BmKS FOREVER Keeping_Books Find a special place to keep books. A shelf or even a cardboard box will do. Teach your child to care for his books. Older chil- dren need a safe place for school books too. You can add paper and pencil or crayons. (Buy washable crayons at first.) Or let him cut up old magazines. Be sure you tell him how good his work is! Have him tell you all about what he has made. Buying:Books Little books are not costly. There are good little books for 19c and 29¢. Look at them and pick what your child would like. Or buy a magazine for yourself that a child can share. After you read it look at the pictures with him. Talk about food and animals and what the people are doing. Later he can cut out the pictures he likes. Using the Library Libraries are for you. Do you know your nearest library? or bookmobile stop? Do your children get books at Head Start? at school? Older children can share their story and pictures with little ones. Can we help you find a library? Would you like a trip to learn about the library? APPENDIX E TABLES 3-12: POST-TEST CELL MEANS FOR THE TEN ORIGINAL MEASURES TABLES 3-12: APPENDIX E POST-TEST CELL MEANS FOR THE TEN ORIGINAL MEASURES Table 3: Post-Test Cell Means Adjusted for Pre-Test on the Stanford Binet Intelligence Scale Treatment Experimental Placebo Control Sex Race (N - 19) (N - 14) (N - 14) Negro 93.64 89.61 86.39 Male (N - 11) (h ' 21) White 92.16 85.77 91.27 (N = 10) Negro 95.95 92.13 85.77 Female (N . 13) (N ' 26) White 100.41 94.23 96.59 (N - l3) 66 67 Table 4: Post-Test Cell Means Adjusted for Pre-Test on the Peabody Picture Vocabulary Test Treatment Experimental Placebo Control Sex Race (N - 19) (N - 14) (N - 14) Negro 75.95 81.43 71.57 Male (N = 11) (N g 21) White 75.13 88.86 69.30 (N a 10) Negro 89.82 70.58 76.80 Female (N - 13) (N 3 26) White 89.10 78.11 74.63 (N - 13) Table 5: Post-Test Cell Means Adjusted for Pre-Test on the Expressive Measure #1 Treatment Experimental Placebo Control Sex Race (N - 19) (N = 14) (N - l4) Negro 48.20 35.52 42.21 Male (N - 11) (N ' 21) White 40.63 28.61 29.83 (N = 10) Negro 50.43 35.52 42.21 Female (N - 13) (N ' 26) White 31.11 27.62 42.14 (N - l3) 68 Table 6: Post-Test Cell Means Adjusted for Pre—Test on the Expressive Measure #2 Treatment Experimental Placebo Control Sex Race (N - 19) (N - 14) (N a 14) Negro 52.29 33.07 40.55 Male (N - 11) (N ' 21) White 42.79 45.51 33.98 (N - lO) Negro 49.29 32.86 29.41 Female (N a 13) (N a 26) White 46.38 51.90 29.44 (N - 13) Table 7: Post-Test Cell Means Adjusted for Pre-Test on the Mean Sentence Length Treatment Experimental Placebo Control Sex Race (N - 19) (N - 14) (N . l4) Negro 4.18 4.07 4.27 Male (N a 11) (N ' 21) White 4.41 4.43 3.75 (N - 10) Negro 4.35 4.26 3.88 Female (N - 13) (N ' 26) White 4.31 4.36 3.10 (N - l3) 69 Table 8: Post-Test Cell Means Adjusted for Pre-Test on the Inattention Score Treatment Experimental Placebo Control Sex Race (N B 19) (N = 14) (N 8 l4) Negro 1.41 1.75 2.18 Male (N a 11) (N ‘ 21) White 1.69 0.84 2.41 (N a 10) Negro 0.70 1.33 1.21 Female (N a 13) (N ' 26) White 0.80 0.24 0.26 (N - 13) Table 9: Post-Test Cell Means Adjusted for Pre-Test on the Birch A Response #1 Treatment Experimental Placebo Control Sex Race (N - 19) (N - 14) (N ' 14) Negro 0.95 0.92 0.91 Male (N . 11) (N ' 21) White 0.85 0.86 0.84 (N . 10) Negro 0.95 0.92 0.84 Female (N . 13) (N a 26) White 0.92 0.93 0.85 (N - 13) 70 Table 10: Post-Test Cell Means Adjusted for Pre-Test on the Birch Response #2 Treatment Experimental Placebo Control Sex Race (N I 19) (N I 14) (N ' 14) Negro 0.11 0.11 0.15 Male (N a 11) (N = 21) White 0.14 -0.09 0.10 (N = 10) Negro 0.02 0.21 0.07 Female (N - 13) (N a 26) White 0.12 0.09 0.10 (N - 13) Table 11: Post-Test Cell Means Adjusted for Pre-Test on the Birch Response #3 Treatment Experimental Placebo Control Sex Race (N I 19) (N I 14) (N I 14) Negro 0.05 0.15 0.09 Male (N . 11) (N ' 21) White 0.09 0.13 0.06 (N I 10) Negro 0.05 0.06 0.05 Female (N a 13) (N = 26) White 0.10 0.05 0.03 (N I 13) 71 Table 12: Post-Test Cell Means Adjusted for Pre—Test on the Birch Response #4 Treatment Experimental Placebo Control Sex Race (N I 19) (N I 14) (N I 14) Negro 0.03 0.01 0.04 Male (N . 11) (N a 21) White 0.06 0.04 0.03 (N I 10) Negro 0.06 0.03 0.05 Female (N - 13) (N " 26) White 0.07 0.03 0.02 (N I 13) "‘Willi#44989Es