214» .u‘ ‘4 ‘I-i’glfitxv (‘IIfi‘Vk’g ”a" ”‘31,. uF~T 31.7.} ‘3“ 533?, A P» 1 ‘5}‘1 :51" n J'"! I: ‘ umv‘ ‘Q- _ ~ A I") WU». ~ ' ',‘ t ‘ :E‘ . L.‘ - A - u ‘5‘" I 4121 ”-5-“: ‘ 3 (iv: 4: .“s' ‘14 ‘24, E o 4 1...}? . j“? ‘5 $93.: . Cc. 'h: :r ’1 ‘. 111% .. ‘2"; 1.54 1 "£3 C“ A? g; 1., I“ r' \u w :fi‘iécg. I}??? qrikIt's" . . li'f.‘ I] ‘V? " I > K‘Ai’u‘ “314.5“! .‘T‘Mfi «W In U4 J" 4‘21 ,. "‘<-~t"m¢ l'.‘":' . i I'x‘ 'f‘mfir E w a; ’“( £1,532qu “1} J‘! sap-":13 u. .3‘1‘! ,1 “7hr (44. .16“ #_ :Jf" £1." ’r - v n... r m ‘f:'.:,1'-" q 'yh‘m'!‘ {Iii/1:5 A43; $HO~ 2*: v I " ’ {6%} I. :5" v?! p21 f‘m- g, '2'; n I’ -‘:;IJ ‘, {55? 1" Alzll'”:1of: . D I LI '3', .- l N.— . .. if - - . ’2’?“ i- —-‘"4 ,, 5:41.51 m:(\bifl‘ . $"‘---¢' :‘IH‘ Jf:‘§w 'W ya {I 2% fix-.4 .. mi ‘ N‘. "‘52:qu £35.53. “I .'.,.CIIl.-. ‘kfrvu‘f‘ . ‘ sz'si. ._ 2 ' “1:1 'fiwhffl \‘ ‘3 Q, ' . {l'v-‘,. “a I'i’t :5? V.“- "; £ $15333 I: #4} ' 74%?! I)". “it/01' I‘V‘V'v 5- $321 “Yum-‘6‘ km: ".55" 9' ‘24,} “iv ‘ ”i‘ 3;), “35’. VA 1*" ' w, 1}» '. Mn): "l- ”3:11, ' fizz; wflq 'I;':I " . r 9:7’3 ;I."' J5 , “9' 5/” , "I a a \:,.{‘2.‘ 1;.75£::." 55%;": , v If 4 \‘f ‘34”; L " .>;' .. )1 A , M'- ‘4» 73': .4. - 453‘ :vlifl; o‘r vIVQ‘. I'IML‘ a 31% l ' U . 'I-b’nn-unn'nthhr aw} MI. Gui-Lit o mu ll Ill Il’fil‘fgljlyl lg @1761? l L mFS's This is to certify that the dissertation entitled PARENT INVOLVEMENT IN HEAD START presented by Margaret Ann Lieblein has been accepted towards fulfillment of the requirements for Ph.D. degreein Family and Child Ecology Mvjor professor Date November 10, 1988 M3 U is an Affirmative Action/Equal Opportunity Institution 0-12771 HERARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or botoro date duo. J r DAME DATE DUE DATE DUE ' l2 5. . ' . . 2001 _.__ mutt—[,2 51:13 9 LL" 7'3 !‘ 1‘":- c: 7 up ‘ A‘ M“;— W .4 "‘M L n 'gtm MSU In An Affirmative Action/Equal Opportunity Institution PARENT INVOLVEMENT IN HEAD START By Margaret Ann Lieblein A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHI LOSOPHY Department of Family and Child Ecology 1988 ABSTRACT PARENT INVOLVEMENT IN HEAD START By Margaret Lieblein This study investigated to what extent parent involvement in Head Start programs was influenced by differences in parent background, attitudes and behaviors, as well as differences in the background, attitudes and behaviors of the teachers with whom they worked. The study reflects the behavior and attitudes of parents and staff as they occured in natural settings. One-hundred-five parents of children who attended center-based classrooms in the Jackson County Family Development Head Start Program took part in the study. Twenty-eight teachers, assistant teachers and supportive service workers also were included. The study was initiated in September and October of the 1987-88 school year and continued through June, when the Head Start progrmn year ended. Correlations were used to assess the differences between various levels of parent involvement and parent, teacher and program characteristics. Multiple regressions were used to Further assess these differences and the ordering of those independent variables shown to be significantly correlated to parent involvement. The results of this study indicate that although a number of variables are related to parent involvement in Head Start classrooms and parent education groups, face-to-face contact by the supportive service worker (SSW) is most significant. Analytical evidence indicates that phone contact by the SSW, teacher, assistant teacher and SSW attitudes toward parent education and classroom involvement, and the parent's perception that the teacher, assistant teacher and SSW are warm and accepting are also related to parent involvement in the classroom. The same pattern holds true for involvement in parent education. This suggests that parent involvement behaviors are most strongly related to the arttitudes and behaviors of the three staff people who deal with their families on a regular basis. The strongest relationships between measures of parent involvement and teacher outreach were found for the involvement of the supportive service worker. ACKNOWLEDGMENTS Many individuals have contributed to the completion of this dissertation and I wish to express my deepest appreciation to them. Specifically, I wish to thank Dr. Robert Boger, my guidance committee chairman. From the very beginning of my work at Michigan State University, his encouragement, tolerance and faith in my ability to learn and grow have provided me with a base from which to move forward and complete this portion of my education. I also wish to acknowledge the influence of Dr. Nhrjorie KostelnH<. Her counsel has been helpful in giving direction and support to me in both nw study and research. Drs. Boger and Kostelnik have provided a model of the caring professional for which I am very thankful and hope to be able to emulate. Thanks also go to my corrrnittee members, Dr. Dennis Keefe and Dr. Hiram Fitzgerald, for their time and willingness to provide guidance. The staff and parents of the Jackson County Family Development Head Start Program also deserve.a special thanks for their cooperation during the yearlong data collection for this research. The cooperation and support from these men and women were essential in the completion of this project. iv Last, but most importantly, I express my sincere appreciaton to my family who patiently tolerated my absence and preoccupation during the past four years. My husband, Bob, has been strongly supportive throughout my educational career as well as a valued and tireless proofreader. TABLE OF CONTENTS LISTOF TABLESOOOOOO0.0.0.0000...OOOOOOOOOOOOOOOOOOOOOO LISTOF FIGURESOOOOOOOOOOOOOO0.0.000...OOOOOOOOOOOOOOOO Chapter I. II. IV. INTRODUCTIONOOOOOOOOOOOOOOOOOOOOOOOO Statement of the Problem..................... Research Questions........................... Need for the Study........................... Conceptual and Operational Definitions....... Theoretical Framework........................ Assumptions.................................. REVIEWOF LITERATURE...OOOOOOOOOOOOOOOOOOOOOO Head Start Research.......................... Program Effects on Parents................... Parent lnvolvement........................... Parent Education and Social Support.......... Adolescent Research As It Pertains To Head StartOOOOIOOOOOOOOOCOOOOOOOOOOOOOOOOO. sumarYOOOOOCOOOOC0.0.00000000000000000000000 METflomLmYOOOOOOOOOOO Design of the Study.......................... Variables and Measures....................... Research Hypotheses and Questions............ Data Analysis................................ sampleOOOOOOOOOOO0..OOOOOOIOCIOOOIOOOOOOOOIO. ProcedureSOOOIOOO0.00.00.00.00000000000000000 RESULTSOOOOOOIOOOOOOOOOOOOOOOOOOOOOO. Hypotheses Hypotheses Hypotheses Hypotheses Hypotheses Hypotheses Hypotheses Hypotheses One, Two and Three................ Four and Five..................... Six, Seven and Eight.............. Nine, Ten and Eleven.............. Twelve, Fourteen and Fifteen...... Thirteen.......................... Sixteen and Seventeen............. Eighteen and Nineteen..... vi Page viii X .— mas-kw 15 21 22 31 46 59 73 76 78 78 79 84 87 92 97 97 102 103 104 105 107 108 110 Questions Pertaining to Parent and Family Characteristics............................ Multiple Regressions to Order Independent Variables.................................. v. DISCUSSION0.0000.0.0.00.0...0..............0. Research Question Resea Research Resea rch Question rch Question Research Question Research Question Research Question Research Question Research Question Research Question Research Question VI . CONCLUSIONS , l0....0.00...0.0.0.....0... 20.000.00.00..0.00..00.00.. QuestionBand 4.....00000000000..0. 50.......0..0..........0... 63nd 7.00000..000000.0.... 8.0......000.0..000..0..... 90....00.00.....000.00..0 lo and 11000...0000..0....0 12 and 13.00.00.0.0.0...... 14 through 1900.00.00.00... 200.00.000.000.000.0000.00. IMPLICATIONS AND SUGGESTIONS .Fm FUTURE RESEARCH.000000.....0....0...... Conclusions.................................. Limitations of the Study..................... Implications and Suggestions for Future Research................................... APPENDICES A. Research Conditions and Agreements... Parents' Opinion Survey................. Description of Parents' Opinion Subscales HOME Inventory (Preschool).............. Parent Education-Involvement Attitude scale0000.0000.00.0.....000.0000.......0 Head Start Parent Involvement QueStionnaireooooooooooooooooooooooooooe REFERENCES......0.00.0.....0...0..0......0.00.0..0 112 11# 121 121 124 125 126 127 128 129 129 130 132 133 137 137 139 140 143 102 149 15# 156 160 162 10. 11. 12. LIST OF TABLES Literature Related to the Study Questions With Reference to Chapter II Literature Review..... Characteristics of the Sample and Population.... Data Collection Time-line for the Program Year.. Pearson Correlations and T-statistics for Teacher Outreach and Parent Involvement....... Pearson Correlations and T-statistics for Teacher and Assistant Teaching Team Outreach and Parent Inv01vement....0................... Pearson Correlations and T—statistics for Teacher and Assistant Teaching Team Face-to-Face Outreach and Parent Involvement.. Pearson Correlations and T-statistics for Supportive Service Worker Outreach and Parent Invalvement....0........0.00.00.0000... Pearson Correlations and T-statistics for Supportive Service Worker Face-to-Face Outreach and Parent lnvolvement............... Pearson Correlations and T-statistics for Teacher Outreach and Attitudes Toward Parent InVOlVementoooooooooooooooooooooooooooo Pearson Correlations and T-statistics for Parent Involvement with Parent's Past Experiences and the Degree of Satisfaction With Head Start............................... Pearson Correlations and T-statistics for Parent Involvement with the Number of Younger Children and the Availability of Child Care and Transportation............................ Pearson Correlations and T-statistics for Parent Involvement with Child-Rearing Attitudes and BehaVIOrSooooooooooooooooooooooooooooooooo viii Page 23 91 95 98 99 100 100 101 102 104 105 107 13. 14. 150 16. 17. 180 19. 20. 21. Pearson Correlations and T-statistics for Parent Involvement with Parent-teacher Difference Scores on the Construct Parent As TeaCher...0.0.0..000.0....0..000..00000....... Pearson Correlations and T-statistics for Parent Involvement with the Amount of Time Spent in School or at Wbrk.................... Pearson Correlations and T-statistics for Parent Involvement with Perceptions of Warmth and Acceptance of Teachers.................... Pearson Correlations and T-statistics for Parent Involvement with Perceptions of Parent Group Warmth and Acceptance................... Pearson Correlations and T-statistics for Parent Involvement with Parent Education and Family Income.......0..0...................... Parent Involvement in Head Start Education Components Related to Gender, Race, and Family compos'ltion.......0...0.........0...... Results of Multiple Regression Analysis Identifying Variables with the Strongest Relationships to Classroom Parent Involvement.. Results of Multiple Regression Analysis Identifying Variables with the Strongest Relationships to Parent Education Involvement.. Summary of Hypotheses Tested and Decision Rule for Research Questions......................... ix 108 109 110 112 113 114 115 120 1. LIST OF FIGURES Urie Bronfenbrenner's Ecological Model of Human Development..000.000.0000....0....0.00.0.0.0... 17 CHAPTER I I NTRODUCT ION Background of the Study From the onset, most Head Start programs viewed parents as the primary influence in their children's development. Consequently, they sought various ways to involve parents in several aspects of the program. Head Start Performance Standards specifically state that in order for children to maintain the developmental momentum generated while in Head Start, it is important for families to understand the developmental changes occurring in their children and provide continuity for them within the family and the community. The philosophy underlying the Head Start approach is based upon parent hmvolvement. It is theorized that if parents come into the classroom to volunteer, attend parent educathui group sessions and are active on policy committees, they will learn to be more effective parents and advocates for their children in the community. The underlying assumptions of this philosophy are based on the results of a number of child development studies that have been carried out in the past decade (Bronfenbrenner, 1974; McKey, Condelli, Ganson, Barrett, NkConkey, & Plantz, 1985). Increased parental involvement has been shown to relate positively to increased achievement in children (Weld, I973; Kinard, 1975; Monroe & McDonald, 1981). However, little is known about the factors that combine to motivate a parent to become involved in Head Start. This presents a problem to program personnel who have a mandate to involve all parents but are forbidden to require parent involvement as a condition for continued enrollment of the child in the program. Staff have to rely upon their instincts to attract parents to become and stay involved, but know little about what: specific situations or characteristics promote parent involvement. This weakness is evident in the fact that some programs and some individual teachers have had little success in involving parents even after repeated efforts by program administrators to emphasize this aspect of the program through training and supervision. Since 1965, parent involvement in Head Start has received a great deal of research attention, partly as a result of a continuing legislative emphasis on this aspect of the program. However, the vast majority of this research examined characteristics of Head Start families, such as how they were involved and the effects of involvement on the children, the effects of Head Start on parental attitudes, and the impact of the program on families. Little research has been done that speaks directly to the question of what influences parent involvement in the Head Start program. The challenge is not only to learn how parent involvement affects change but also to identify the factors that combine to motivate parents to become involved in Head Start. Statement of the Problem This study was designed to assess the relationship between parent, teacher and Head Start program variables, and the amount of involvement the Head Start parent exhibited during the child's one year Head Start experience. It investigated the relationships between the degree of parent involvement and parent age, gender, income, family composition, educational level, work or school involvement, past experience with Head Start, past experience with parent education groups, and attitudes about child rearing practices. Furthermore, the relationship of parent involvement, and parenting behaviors exhibited in the home and parenting self-concept were investigated, as well as the relationship of these involvements with attitudes concerning the parent involvement, parent education and teaching components of the Head Start program. Further investigation was undertaken to identify the relationships of these factors ‘H: the teacher's demographic characteristics, attitudes toward parent involvement/parent education, child-rearing attitudes, and subsequent outreach to parents aimed at eliciting their involvement. Teacher variables of interest included age, SES, gender, educational level, family composition, attitudes toward parent involvement, attitudes toward parent education, attitudes about child-rearing practices, and the amount and type of outreach targeted at involving parents in the educational components of the Head Start program. Program variables of interest included accessibility of the Head Start program to parents as a result of the availability of transportation and child care services. Research Questions The following research questions were addressed in this study. The educational components mentioned in some questions refer to the classroom component and the parent education component. Does the level of teacher involvement with the parent relate to the parent's degree of involvement in the education components of Head Start? Does teacher attitude toward parent education and parent involvement relate to the level of outreach/involvement that the teacher exhibits to elicit parent involvement in the education components of Head Start? Does the parent's past experience with Head Start relate to his/her degree of involvement in the education components of Head Start? Does the parent's past experience in parenting classes relate to his/her degree of involvement in the education components of Head Start? Does availability of transportation and child care relate t1) the parent's degree of involvement in the education components of Head Start? Do the child-rearing attitudes of the parent relate to his/her degree of involvement in the education components of Head Start? Does the degree of congruence of teacher childrearing attitudes and parent child-rearing attitudes relate to the parent's degree of involvement in the education components of Head Start? Do the parent's child-rearing behaviors relate to his/her degree of involvement in the education components of Head Start? Does the degree of parenting self-concept relate to the parent's degree of involvement in the education components of Head Start? Does the number of hours that a parent works outside the home relate to his/her degree of involvement in the education components of Head Start? Does the number of hours that a parent is in school relate to his/her degree of involvement in the education components of Head Start? Does the degree of warmth and acceptance of the teacher (as perceived by the parent) relate to the degree of the parent's involvement in the education components of Head Start? Does the degree of perceived warmth and acceptance of the parent education group relate to the parent's degree of involvement in the education components of Head Start? - Does family composition relate to the parent's degree of involvement in the education components of Head Start? Does the age of the parent relate to his/her degree of involvement in the education components of Head Start? Does the gender of the parent relate to his/her degree of involvement in the education components of Head Start? Does the parent's level of education relate to his/her degree of involvement in the education components of Head Start? Does the race of the parent relate to his/her degree of involvement in the education components of Head Start? Does family income relate to the parent's degree of involvement in the education components of Head Start? What is the relative strength of the relationship of parent involvement and the independent variables of interest? There are clear indications in the literature that the variables (If interest hi this study are related to parent involvement hi the child's classromn and in parent groups. These relationships are discussed extensively in Chapter II. The specific hypotheses of this research are pwovided in Chapter 111. Need for the Study By examining relationships between parent, teacher and program characteristics, and parent involvement in Head Start, program staff can be assisted in better understanding parent needs as they relate to parent involvement. In addition, analysis of these data can provide a base for planning and implementing a more effective linkage with parents, and subsequently enhance program effects on children and families. Conceptual and Operational Definitions ‘The following section provides the conceptual and operational definitions for the dependent variables. Parent Involvement in the Classroom Parent involvement in the classroom is conceptually defined as the extent to which a parent volunteers in his/her child's classroom. Parent involvement in the classroom is operationally defined as the sum of the parent's volunteer days in his/her own child's classroom for the 1987-1988 program year. Parent Involvement in Parent Education Parent involvement in parent education is conceptually defined as the extent to which a parent attends parenting group sessions. Parent involvement in parent education is operationally defined as the sum of the Head Start parenting sessions attended for the 1987-1988 program year. This will include all sessions offered by Head Start that address child development and parenting issues. Not included in this are policy committee or council meetings, or sesSions devoted to arts and crafts. Parent Involvement in Head Start Education Components Parent involvement in Head Start education components is conceptually defined as the extent to which a parent becomes involved iTI both the child's classroom and parenting group sessions. Parent involvement is operationally defined as the sum of the volunteer days and parenting sessions attended for the 1987-1988 program year. The following section provides the conceptual and operational definitions for the independent variables. Teacher Involvement lOut reach Teacher involvement/outreach is conceptually defined as the degree to which the teacher, teacher aide or supportive service worker reach out to entice the parent to volunteer in the classroom and attend parenting sessions. Teacher involvement/outreach is operationally defined in two different ways. The first definition is the total number of contacts instigated by the teacher, teacher aide and supportive service worker, used to entice the parent to volunteer in the classroom or attend parenting sessions. The second definition is the number and type of contacts instigated by the teacher, teacher aide and supportive service worker, used to entice the parent to volunteer in the classroom or attend parenting sessions. The type of contacts that will be used for this study are written communication, telephone, and face-to-face contact. Teacher involvement is also used as a dependent variable in one instance in this study. When teacher attitudes toward parent education and parent involvement are treated as independent variables, and teacher involvement with the parent is investigated in relationship to these attitudes, teacher involvement is a dependent variable. Parent Child-Rearing Attitudes Parent child-rearing attitudes of interest in this study are conceptually defined as the parent's acceptance of his/her child's right to express feelings, willingness to be a positive influence on his/her child's development, positive communication regarding expectations of appropriate behavior, and knowledge and positive attitude toward appropriate parenting techniques that promote optimal child development. Parent child-rearing attitudes are operationally defined as the parent's score on the following subsets of the Parents' Opinion Survey (POS): Parent as Teacher (P/T); Beliefs Regarding Communication (BRC); and Beliefs Regarding Discipline and Control (BRD/C). Child-Rearing Behaviors Child-rearing behaviors of interest in this study are conceptually defined as the parent's willingness to provide a home environnmht that offers a positive emotional climate and consistent contact with a supportive loving adult. Additional child-rearing behaviors of interest are: the provision of play materials which facilitate the coordination of the sensory-motor process and cognitive development; the provisicni for nweting basic physical needs, and the health and safety of the child; the fostering of development through rich and varied cultural experiences; the provision 10 of experiences that match the child's current level of cognitive, social and emotional development; and the promotion of appropriate and valued behaviors through the use of physical, verbal and emotional support of the child. Child-rearing behavior is operationally defined as the total score on the Home Observation for Nhasurement of the Environment (HOME). Parent Self-Concept Parent self-concept is conceptually defined as the parent's feelings of competence and control in his/her parenting role. Parent self-concept is operationally defined as the total score on POS subsets Perceived Competency (PC), Perceived Importance of Extra-Familial Influences (PIEFI), and Fatalistic versus Non-Fatalistic Outlook (F/NFO). Teacher Attitude Toward Parent Involvement Teacher attitude toward parent involvement is conceptually defined as the teacher's degree of positive feelings toward parents being involved in the child's classroom. Teacher attitude toward parent involvement is operationally defined as the teacher's score on the Parent Involvement Subscale of the Parent Education-Involvement Attitude Scale. 11 Teacher Attitude Toward Parent Education Teacher attitude toward parent education is conceptually defined as the teacher's degree of acceptance of the concept that parents should attend parent education sessions in Head Start. Teacher attitude toward parent education is operationally defined as the teacher's score on the Parent Education Subscale of the Parent Education-Involvement Attitude Scale. Teacher Child-Rearing Attitudes Teacher child-rearing attitudes of interest in this study are conceptually defined as the teacher's attitudes toward appropriate child-rearing techniques. Teacher child-rearing attitudes are operationally defined as the teacher's total score on the POS subsets Parent as Teacher, Beliefs Regarding Communication and Beliefs Regarding Discipline and Control. Congruence of Parent's and Teacher's Child-Rearing Attitudes Congruence of the parent's and teacher's child-rearing attitudes is conceptually defined as the amount of agreement or disagreement between the parent's and teacher's attitudes toward child rearing. Congruence of the parent's and teacher's child-rearing attitudes is operationally defined as the total of the 12 difference scores of the parent's and teacher's responses to the POS subsets Parent as Teacher, Beliefs Regarding Communication, and Beliefs Regarding Discipline and Control. Perceived warmth and Acceptance of the Teacher Perceived warmth and acceptance of the teacher is conceptually defined as the parent's feelings of acceptance and warmth from the education team personnel. Perceived warmth and acceptance of, the teacher is operationally defined as the total score on the subset Perceived Warmth and Acceptance of the Teachers (questions 2 through 5, and 7) of the Head Start Parent Involvement Questionnaire. Perceived Warmth and Acceptance of the Parent Group Perceived warmth and acceptance of the parent group is conceptually defined as the parent's feelings of acceptance and warmth from other parents attending the parent education group. Perceived warmth and acceptance of the parent group is operationally defined as the response to question 8 of the Head Start Parent Involvement Questionnaire. Availability of Transportation Availability of transportation is conceptually defined as the parent's perception that transportation is available 13 so that the parent and children can come to the Head Start facility to be involved in the education components of the program. Availability of transportation is operationally defined as the response to question 9 of the Head Start Parent Involvement Questionnaire. Availability of Child Care Availability (MI child care is conceptually defined as the parent's perception that there is appropriate child care available, either outside of (N' at Head Start, for siblings «of the Head Start child so that the parent can become involved in the education components of Head Start. Availability of child care is operationally defined as the response to question 10 of the Head Start Parent Involvement Questionnaire. The following section provides other definitions of interest to the study. Parent Characteristics Parent characteristics are conceptually defined as family composition, income, age, gender, educational level, race, work, school attendance, past experience with Head Start, and past experience in parenting classes. Operationally, these parent characteristics are defined as follows: 14 Family composition. Family composition will be defined in two ways. ‘The parent composition of the family will be defined as a single or two-parent family. The child composition of the family will be defined as the number of younger children in the family. Income. Income is defined as the family's total net income as reported on the Head Start enrollment form. Ag_. Age is defined as age at the last birthday. Educational level. Educational level is defined as the last grade completed in formal education. 5223. Race is defined as the race of the parent, as documented in program records. ‘Working. Working is defined as the average number of hours per week that the parent is employed outside the home. School attendance. School attendance is defined as the average number of hours per week that a parent spends in formal education. This includes high school completion, Adult Basic Education, and vocational training or college. Past experience with Head Start. Past experience with Head Start is defined as the number of years that the parent has had a child in any Head Start program prior to this year. 15 Past experience in parenting classes. Past experience in parenting classes is defined as the number of parenting sessions that the parent attended in the past. Teacher characteristics Teacher characteristics are conceptually defined as family composition, age, gender, educational level, and race. Operationally, these teacher characteristics are defined as follows. Family composition. Family composition is defined as a single or two-parent family. Ag_. Age is defined as age at the last birthday. Educational level. Educational level is defined in two ways. The first definition uses a continuous variable, the last grade completed in formal education. The second definition uses five discrete categories: CDA credentialed, bachelors degree or beyond in Early Childhood Education, bachelors degree or beyond and no Early Childhood Education, and associates degree. Race. Race is defined as race listed in program records. Theoretical Framework A human ecological framework was used to develop a model for describing relationships between Head Start parent char- acteristics, perceptions, attitudes and behaviors, and l6 ecological factors. This framework assumes interrelation- ships between humans and their interpersonal and physical environments. Emphasis is placed on examining phenomena related to the parent's involvement in Head Start from a total transactions perspective. Interrelationships within the parent's ecosystem are considered, rather than linear cause and effect relationships of isolated aspects of the ecosystem. hi this study, parent involvement is assumed to be related to ecological variables that include teacher's characteristics, attitudes and behaviors, and the Head Start program's services designed to support parent participation. Bronfenbrenner's (1979) ecological model of human development was used as a model for this study. Within this framework, development is related to factors in the ecological environment. The ecological environment consists of layers of settings and systems that reach far beyond the immediate setting of the individual and the family. Bronfenbrenner conceptualized four basic ecological systems in the developing child's environment. They include the microsystem, the mesosystem, the exosystem, and the macrosystem. IA graphic presentation of this model is available in Figure 1. Each level 13f structure in Bronfenbrenner's model consists of settings and characteristics that are unique to each individual and family systenu ‘The child's microsystem is defined as the immediate setting where the developing 17 Sodoeconomlc Status Ideology (M acrosysremS) Figure 1: Urie Bronfenbrenner's Ecological Model of Human Development child is functioning at a particular time, such as the school «or the home. It has three dimensions: physical space and materials; people in roles and grelationships to the individual; and activities that people engage in with or without each other. Contextual variables representing the microsystem for this study are demographics of the parent, demographics of the teachers, teacher outreach, and program support characteristics. 18 The mesosystem consists of the interrelationships between two or nmre of a.person's microsystems, such as the home and the school. The exosystem consists of systems or settings in which the developing individual is not an active participant, but which affect him/her indirectly through impact on other members of microsystemic dyads in which the child is involved. Because of the relationship between parent involvement and child development outcomes, the focus of this study is on the parent's relationship to the Head Start child's classroom and the parent education group. The parent's classroom involvement represents the mesosystem, and parent education involvement represents the exosystem. Work, education, and care of younger children are also considered factors representing the exosystem for the purpose of this study. These factors influence how the parent relates to the Head Start program, as well as how s/he relates to the child. The ideology of the society in which the individual lives and the cultural influences on the individual are aspects of the macrosystan. The Head Start policnr that requires parent involvement, but does not deal with methods of implementation, are an example of societal ideology. Ethnicity of the parent is one variable addressed in this study, however, other macrosystem influences were not addressed. In this study, parenting perceptions are conceptualized as influencing parenting behaviors through the decision 19 making process. Paolucci, Hall, and Axin (1977) contend that how individuals react to a given situation is a direct result «of how they perceive that situation, as well as what their behavioral dispositions are at that point in time. They postulate that values and perceptions play a major role in the decision making process. Assumptions ‘The following assumptions are supported by the literature review and form the basis for the research questions in this study. 1) Parent involvement in Head Start is a function of the interaction of a diverse set of variables representing characteristics, attitudes and behaviors of the parents, the teachers and the program. 2) Head Start parents are not homogeneous with regard to attitudes, behaviors or demographic characteristics and can be categorized into groups with distinct characteristics. 3) Head Start teachers are not homogeneous with regard to attitudes, behaviors or demographic character- istics and can be categorized into groups with distinct characteristics. 4) Teacher attitudes about the importance of parent involvement and parent education have been developed 5) 6) 20 as a result of personal experiences and, therefore, are different for individual teachers. Teacher attitudes about parent involvement and parent education will influence the teacher's efforts to involve parents in the educational components of the Head Start program. In order to understand the motivation behind parent involvement in Head Start, the variables associated with parent attitudes, behaviors and characteristics, and those associated with teacher attitudes, behaviors and characteristics must be studied and related to the involvement of the various diverse groups of parents. CHAPTER II REV 1 EW OF LITERATURE hi recent years, there has been increased interest in parent involvement in the educational lives of Ammrican children. As society has searched for ways to improve child development outcomes resulting from the school experience, local and national efforts have moved from focusing on the basics to changing the way reading and math are taught, and to working more closely with the child's closest microsystem, the family (Ford, 1979; Herman & Yeh, 1980; Johnston & Slotnik, I985). The assumption is that by working with the parents, their interest and influence on their children will assist the schools in the education process. Parental interest and influence are affected by attitudes and behaviors that have, in turn, been affected by the parent's life experiences. Researchers from a variety of theoretical approaches have studied these factors, however, such research is not easy to design or interpret because of the nmny factors that influence behavior (Bronfenbrenner, 1979). Irving Sigel (1982) suggested that ‘we'must find a way to translate our assumptions and intuitions into quantifiable data by proceeding cautiously and defining "areas that appear to have conceptual quasi- independence and that can be treated as a sub system of the 21 22 whole" (p.l). There is a need to look at assumptions and pull together the various threads of research that address the issues surrounding parent involvement in the child's development and education. Although parent involvement has become a focal concern of American schools, only a limited number of studies have investigated the effects of involving parents in education. Even fewer studies have touched on the factors that promote parent involvement. The following review provides some insights into these factors. It will also explore the effect of parent involvement on the developing child and the school experience, with a special focus on Head Start. An outline of this discussion as it relates to the research questions is provided in Table 1. Head Start Research Project Head Start began in 1964, when the Office of Economic Opportunity started planning a child development program designed to break the cycle of poverty in which large numbers of American families were trapped. The intent of the program was to create an environment for young children of low-income families which would help them reach their full potential. The Head Start Policy Manual was introduced in 1969 to establish guidelines for program operations. This first manual stressed the concept of the family as fundamental to the child's development. The 23 Table 1 Literature Related to the Study Questions With Reference to Chapter 11 Literature Review Study Questions ‘— Literature Review Page Dbes teacher involvement relate to parent involvement? Does teacher attitude toward parent involvement and parent education relate to teacher envolvement? Does parent past experience in Head Start relate to parent involvement? Does parent past experience in parenting classes relate to parent involvement? Does transportation and child care relate to parent involvement? Do child-rearing attitudes relate to parent involvement? Does parenting behavior relate to parent involvement? Related Literature Duncan, 1969 Morris, 1974 Windle, I988 Shelton & Dobson, 1979 Boger et. al., 1986 Morris, 1974 Epstein & Becker, 1982 Epstein, 1985 Kearney, I969 Colletta, 1983 Rebman, I983 Dickie & Gerber, 1980 Andrews et. al., 1982 Travers et. al., 1982 Clarke-Stewart, I983 Colletta & Lee, 1983 Morris, 1974 Juarez & Assoc., 1982 Boger et. al., 1969 Midco Ed. Assoc., 1972 Weld, I973 Hock & Lindamood, I981 Rebman, 1983 Roosa, 1984 Williams et. al., 1984 Silverman, 1974 Johnson & Peevers, 1979 Smith, 1980 Andrews, 1982 McKey et. al., 1985 Seitz et. al., 1985 Boger et. al. 1986 54 49 56 54 69 49 55 54 33 65 39 63 30 50 35 35 36 31 31 59 32 61 51 24 Table l (cont'dJ Does parent self concept relate to parent involvement? Does the amount of hours that a parent works relate to parent involvement? Does the amount of time a parent spends in school relate to parent involvement? Does family composition relate to parent involvement? Do parent characteristics such as age, gender, level of education, and income relate to parent involvement? race Does the warmth and acceptance of the teacher and/or the parent group relate to parent involvement? parent role was developing policies, establish and Performance Standards Register in 1973, minimum requirements designed as making decisions, implement program procedures. and are still for Chilman, 1973 Adams, 1976 Colletta & Lee, Belsky, I984 Boger et. al., 1983 I986 HEW Ser. Del. Assess., Long & Long, 1983 McKey et. al., 1985 1977 HEW'Ser. Del. Assess., 1977 HEW Ser. Del. Assess., 1977 Radun & Weikart, I967 Bronfenbrenner, I974 HEW Serv. Del. Assess., Obgu, 1982 Cochran & Henderson, Olmsted & Rubin, 1985 Phenice et. al., 1986 1977 1985 Windle, I977 Epstein & Becker, Epstein, 1985 1982 one of participation and helping 46 3O 65 65 69 49 49 49 49 49 46 45 49 46 65 48 46 56 55 54 in to The Head Start were published in used today as the program operations. the Federal basic These 25 standards define the minimum level of service that must be achieved in the program component areas of education, health services, social services and parent involvement. The standards ialso define parent involvement as including volunteering int the classroom, attending parent group sessions, being considered for entry level jobs, and membership in center policy committees and councils (Dawson, 1981). There are many descriptive studies concerning Head Start. However, very few deal directly with parent involvement and the benefits derived from this aspect of the Head Start experience. A large number of these studies have focused on the success of Head Start in improving children's health and physical abilities, encouraging self-confidence, enhancing mental processes and skills, establishing expectations of success for the child, enhancing a sense of dignity and self-worth within the child and the family, and increasing the ability of the child and family to relate to each other and to others (McKey, Condell, Ganson, Barrett, McConkey & Plantz, 1985). Early Research Early research focused on cognitive development of children based on IQ tests, and tests of general ability and learning readiness (Datta, I969; Westinghouse, I969; Hertz, 1977). One conclusion of these studies, the major 26 one of which was the Westinghouse Report, was that full-year Head Start attendance temporarily produced large and significant cognitive gains that reached the national average. However, they also found that the gains tended to "wash out" within a few years after the end of the program, and by third grade there was no discernable difference between Head Start children and non-Head Start children from similar backgrounds. The Westinghouse Report concluded that the summer Head Start program was totally ineffective and even had negative effects, while the full-year progrmn was only marginally effective. However, the design of the study contained questionable methodology. For example, the research was conducted only after the children completed Head Start. The comparison group, identified at the time of the post-test, was selected by matching characteristics with the children who had been in Head Start. Children were tested for cognitive outcomes only. ‘The primary method of analysis was covariate analysis that used a constructed index of socio-economic status as 'Hu: only covariate (Brown, 1985). Problems that might have arisen from these deficiencies included: inappropriate matching due to the lack of baseline data; inappropriate outcome measurements due to the short duration of the intervention; and lack <3f variation iii the covariate, socio-economic status. Shortly after the results of the study were released, a number of researchers contended that it contained 27 insufficient facts, and the methodology was sloppy. In the next ten years, many early childhood experts reanalyzed the data and pointed out the deficiencies cited above (Campbell & Erlebacher, 1970; .flnith & Bissell, 1970; Barnow, I973; Magidson, 1977). Magidson even demonstrated that the study actually showed that Head Start was effective. Unfortunately, the Westinghouse Report, with all of its deficiencies, had a devastating effect on Head Start for many year 5 o The Second Decade In an effort to take a more rational look at the data, an independent analysis group, the Consortium for Longitudinal Studies, was formed by Irving Lazar at Cornell University in 1975. The goal of this research was to pool data from the earlier major studies and combine it with a new round of data using the same set of measurements. The <30nsortium used only the data from well-run early intervention programs that had carefully controlled experiments. Although only a few of these were Head Start programs, the others were similar in clientele and purpose. They served economically disadvantaged children and were designed to improve the social and intellectual competence of the children they served. The Consortium found that high-quality infant and preschool services inmrove the ability of low-income children to meet the minimal requirements of further schooling over the long run. These 28 programs significantly reduced the number of children assigned to special education classes and the number retained in grade. They also significantly increased the IQ and school achievement of low-income children at least through the critical early primary years (Consortium, 1979, I983; Lazar, Hubbell, Murray, Rosche & Royce, 1977; Lazar, Darlington, Murry, Royce & Snipper, 1982). Lazar (1981) stated that there were five interrelated program characteristics that were related to these long term positive outcomes: intervention at an early age; low adult-child ratio; a larger number of home visits; direct participation of parents; and services for families in addition to educational services for children. ht a more recent study of a model program, included in both of the earlier research projects, the High/Scope Perry Preschool Project was found to have benefits that lasted into early adulthood. Not only was the program linked to improved school success, but it was also associated with the prevention of delinquency and teenage pregnancy, and improved the likelihood of employment (Berrueta-Clement, Schweinhart, Barnett, Epstein & Weikart, 1984). One of the major hypotheses offered by some researchers concerning the reasons for these positive effects on child- ren's school achievement is that, through their contact with the progrmn, parents become more effective in encouraging children to become motivated to learn. Not only has it been 29 shown that (fliildren do better when their parents are involved, lnrt there is also research to suggest that preschool intervention can succeed only when parents are participants in the educational process (Brofenbrenner, 1975; McKey, Cbndell, Ganson, Barrett, NkConkey & Plantz, 1985). It appears that Head Start increases participating parents' self-confidence and sense of control over their lives, which in turn seems to contribute to their children's sense of self-esteem (Adams, 1976; DHHS, 1980). A recent comprehensive analysis of 210 reports of Head Start research was conducted by CSR, a consulting firm working for the United States government. The resulting report, The Impact of Head Start on Childreq, Families and Communities, is distinguished from other reviews in that it includes all Head Start research, both published and unpublished, and uses nmta-analysis to produce numerical estimates of Head Start's effects (McKey, Condell, Ganson, Barrett, NkConkey & Plantz, 1985). The report states that "children enrolled in Head Start enjoy significant immediate gains in cognitive test scores, socioemotional test scores and general health. hi the long run, cognitive and socioemotional test scores of former Head Start students do rurt remain superior to those of disadvantaged children who did not attend Head Start. However, a small subset of studies was reported to find that former Head Start children 30 are more likely"n3 be promoted to the next grade and are less likely to be assigned to special education classes. Head Start also has aided families by providing health, social and educational services and by linking families with services available in the community. Finally, educational, economic, health care, social service and other institutions have been influenced by Head Start staff and parents to provide benefits to both Head Start and non-Head Start families in their respective communities" (p.1). The findings of this comprehensive analysis concerning the impact that Head Start has on families were based on 75 studies. ‘These studies were judged to contain enough pertinent information to enable the researchers to use meta- analysis. The researchers found that, although sizeable proportions of parents participated in various paid and volunteer capacities in the programs, the extent of involvement was uneven. A core of parents contributed a disproportionate share of time. They also found that children whose parents are highly involved perform better on cognitive tests than children whose parents are less involved. However, it was unclear whether the differences were actually a result of Head Start involvement, or if such factors as parental concern for the child's education lead to both parental involvement and child achievement. These findings have been questioned because of several important issues. For example, the studies included in the 31 synthesis were not based on a representative sample of Head Start sites. Adso, in order to obtain the numbers needed for meta-analysis, results were obtained from both low- quality-of-design and high-quality-of-design studies (Schweinhart & Weikart, 1986). Despite these shortcomings, the findings of this analysis are provocative, providing a comprehensive description of the effects of Head Start programs on the family and community, as well as the effects on children's intellectual development, socioemotional development and health status. Program Effects on Parents One (xi the basic assumptions of the Head Start philosophy is that by being involved, parents will improve their child-rearing attitudes and practices. However, changes ht parental attitudes, behavior or status has not been a central concern of Head Start researchers. This is unfortunate, especially now that the importance of a child's human ecological environment is being recognized as critical to development (Bronfenbrenner, 1979). There have been several reviews over the past ten years that have examined the effects of Head Start on families (O'Keefe, 1979; Grotberg, 1980). Most of the studies included in these reviews have focused on the impact of Head Start involvement on parental attitudes toward child-rearing behaviors, parent education, and changes in 32 their lives within the community. The authors generally reported limited benefits to parents in the areas of increased life satisfaction, and improvement of life skills, job training and employment. Head Start Parent Involvement Research Midco Educational Associates (1972) investigated the effects of parent involvement in Head Start and found that parents who were highly involved felt more successful, happier, and more satisified than parents who were less involved. In general, life satisfaction levels of the involved parents increased more than that of the less involved parents during the time that their children were enrolled in the program. Highly involved parents felt more control over their own lives and better able to influence the schools and their child's education. An interesting secondary finding was that centers classified as low on parent involvement were comprised of parents with greater incomes and educational levels than centers with high parent involvement. Similar findings were reported by Adams (1976) hi a study that probed the differential effects of parent involvement in the Dane County, Wisconson Head Start program. Twenty-five parents who served on the advisory council of the Dane County program were questioned in loosely-structured tape-recorded interviews. Predictions had been made that parents who had been involved at many 33 different levels of adult participation in Head Start would exhibit the most change, both in terms of personal evaluation and in actual economic status. Personal changes were reported by nearly every respondent. Adams concluded that Head Start parents who are involved in the progrmn showed growth in the areas of increased self-confidence, control over their lives, and community participation. Child-Rearing Practices In relaticni to child-rearing practices, Johnson & Peevers (I979) found that after involvement, Head Start mothers were less strict in terms of general control of the chiltl and more accepting of "contemporary rather than autocratic attitudes". However, at the end of the program year, mothers were stricter with their children regarding accidental breakage of items as well as dealing with parent- chilcl conflict. Several other studies have reported that the researchers were unable to find significant changes in parental attitudes or behaviors over the short one-year Head Start intervention (Silverman, 1974; Smith, 1980). This is not surprising, since parenting behaviors are learned slowly over the years and show a lack of immediate response to many intervention'programs. In the area of health and nutrition, some studies have found that children in the Head Start program have better diets at home than children not in the program (Abt, 1984; McKey et. al., 1985). Other studies have reported 34 no difference between the groups (Owen, 1970). There is an indication that the key to better home nutrition may be the level of parent involvement (Abt, 1984; McKey, Condell, Ganson, Barrett, McConkey & Plantz, I985). Furflunhmre, studies generally show that although Head Start is successful in improving existing health care within communities, providing needed health care for enrolled children, and improving the general health of the children, it is less successful in involving parents in health education. It is important to note that it was also indicated that there is no significant difference in health behavior practices at home between parents who participate in health education programs and those who do not participate (McKey, Condell, Ganson, Barrett, NkConkey & Plantz, 1985). Parent Attitudes While some studies report significant gains by children 'whose parents are involved in the program, others show no changes. hi a project that examined the relationship of family attitudes and characteristics to the amount the child gained from Head Start, Weld (1973) found that the children's residual gain scores on several cognitive tests related significantly to Head Start attendance, and attendance was significantly related to familly perception of the value of education. However, evidence is lacking to show that Head Start has much effect on parents' attitudes 35 toward the value of education. Part of this lack of evidence is due to the quality of the studies that have been undertaken. Kearney (1969) studied 160 parents in York, Pennsylvania, to determine whether prior enrollment in Head Start was related to their attitudes toward education. She compared parents having no past experience in Head Start with those who had one, two and three years of experience. All groups were from the same community and the same socioeconomic class. It was found that the no-experience parents had significantly higher attitudes toward education. However, only five percent of the parents from any of the four groups had ever visited the Head Start classroom. This lack of involvement for most parents makes any analysis of the results of this research very questionable if the assumption has been made that parent involvement affected parent attitude. Unfortunately, this problem of low involvement is exhibited in many of the studies. Because of this no correlations can be made with any degree of reliability between parent attitudes and behaviors, and program involvement. In studies outside of Head Start, differences in parental attitudes and beliefs have been related to demographic variables, including socioeconomic status, occupation, educational level, nationality, and religion. Perhaps the most interesting result of this body of research has been Kohn's (1969) finding that lower-class parents give 36 higher priority to values reflecting behavioral conformity, including obedience, honesty, neatness and cleanliness. Segal (1985) reported the results of a study that provided a better understanding of the relationship between childrearing attitudes and family demographics. In this preschool home visit program, the Ready For School Project, home visitors helped parents learn skills that would enable them to become effective teachers of their own children. It was found that mothers in single-parent families :rtressed obedience and competition more than mothers in two-parent families. Income and educational level were associated with childrearing values; high-income and better educated parents gave greater emphasis to process goals than parents with less income and lower education. These findings along with Kohn's have implications for programs. By having prior information about the demographics of the parent, service delivery can be more sensitive, economical and effective. The above research findings, as in Head Start, have been inconsistent with respect to the impact of the intervention programs on parenting attitudes. However, the degree of continuity over time in parental childrearing attitudes without any intervention has been found to be somewhat stableu Hock and Lindamood (1981) used the Maternal Attitude Scale (MAS) to evaluate parenting attitudes with 57 mothers, representing all SES groups. The 37 scale was administered when each child was eight months old and again at 3 1/2 years of age. Over this three-year period, researchers found little change in attitudes about appropriate: control over the child's aggression, encouragement of reciprocity, feelings of competence in perceiving aunt meeting the child's needs, appropriate closeness with the child, and acceptance of emotional complexity in child rearing. Hock and Lindamood attributed this stability to the fact that parenting attitudes reflect critical functions of the ego. They identified the specific functions as appraising and labeling interpersonal transactions, and facilitating organization of the interpersonal environment. In a study of short-term impact of parenting programs on pregnant teens, Roosa (1984) found that attitudes concerning child development and reproduction were stable over time when not influenced by intervention programs. Furthermore, short term increases in knowledge attributable to a parenting program for these teens did not show significant changes in the three subscales of the Maternal Attitude Scale (MAS), used to study attitude changes. Scores indicated that these teenagers had a relatively low level of anxiety about child rearing, were relatively neutral on the maternal satisfaction vs. hostility scale and felt that young children tended to make demands which led to unhappiness and frustration. Roosa suggested that although 38 teenagers may acquire information regarding a child's developmental needs, which include those of a warm, loving, caring environment, their own developmental level and egocentric nature may preclude putting this knowledge into practice with any consistency. Many Head Start parents were adolescents art the time of their child's birth. Although most are not adolescents at the time the child is in Head Start, the effect of adolescent parenting may still be apparent. Sameroff & Feil (1985) suggest that the way parents interpret the behavior of their children is related to the parent's level of understanding development, combined with the characteristics of the child and the care-giving environment. Their findings appear to support some of the outcomes of Roosa's adolescent research. Williams, Omizo & Abrams (1984) studied the effects of participation in Systematic Training For Effective Parenting (STEP) on parents' attitudes and on their learning disabled children's locus of control. The sample used in this study was white and middle to upper-middle class. After completicni of the nine parenting sessions, the Parent Attitude Survey (PAS) was used to record changes in attitudes about acceptance, trust and causation of behavior. Half of the 38 parent volunteers were randomly assigned to a treatment group and half to a control group. Significant changes were found in treatment parent's attitudes in 39 accepting and trusting, and in the perception that their behavior had a greater influence on the children's behavior. However, attitudes about confidence in parental ability, understanding and communication did not change significantly. The authors suggested that these attitudes were more cognitively affected and would thus require longer intervention to produce change. The child's behavior change was indicated in the attitude toward locus of control being more internal than external. Inherent in the STEP curriculum is the underlying assumption that without change iri the behavior of the parent there would be no change in the behavior of the child. Williams indicated that further research should be undertaken to assess the relationsipi between parenting education, parent attitude change, and parent and child behavior change before any conclusions could be drawn about the relationships. It is important to realize that the STEP curriculum is very specific: ht that it teaches parents a method of assessing the child's behavior and then planning to act in a way that helps the child learn more acceptable and effective ways of dealing with problems and emotions. Some Head Start programs offer STEP as a separate curriculum, some imbed many of the concepts and methods into the regular parent education curriculum and many do not use any of this material. Because of this diversity, findings from research of specific curriculums such as STEP must be viewed as an 40 indicaticni of possible effects rather than indicating effectiveness in changing parenting attitudes and behaviors. A variety of studies that have looked at the effects of parent behavior (M1 child functioning have highlighted the role that warm, supportive, sensitive, and consistent care- giving plays in healthy development (Clark-Stewart, I973; Beckwith, Cohen, Kopp, Parmelee & Marci, 1976; Belsky, 1980). In a study of 120 two-parent families, Skinner (1985) looked at the relationship between parent beliefs and parent sensitive and responsive behavior for families at two SES levels. She found that mothers who believed that children develop and learn through their own self-initiated action allowed children more options in their task oriented behaviors. However, they also showed high directiveness when it came to explaining the task. It was also found that high-SE5 mothers reported more attitudes that supported taking the child's viewpoint, and saw the child as more active iri learning and development than their low-SES counterparts. However, Skinner found, using analyses of variance, that SES groups did not differ significantly on any of the nmther behaviors. She stresses the points that these mothers differed in attitudes but not in behaviors. One problem with this research is that Skinner used a low-SES group mean income of $19,000. This is quite high compared to most low-income studies. 41 Relationship Between Parent Attitudes and Parent Behaviors Two studies were located that indirectly addressed the effect of parent attitude change on parent behavior. Rebmen (1983) studied the effect of parental attitudes and program duration upon parental participation patterns in a home- based pre-school education program. Thirty-six parents were studied, 18 reflecting non-responsible and 18 reflecting responsible attitudes toward early educational involvement. Post assessment Showed that the non-responsible group had significantly changed behavior after twenty weeks of involvement with the program, reflecting a more positive attitude toward responsibility in their children's education. The program did not succeed in changing the non- responsible group to equal the attitudes and behaviors of the responsible group. However, the magnitude of the change that did occur indicated more than a simple regression toward the mean. Bergan,INeumann & Karp (1983) found that there was some change in parental behavior as a result of a brief parent training session. They investigated the effects of using a training packet to teach parents to use Inore modeling, reinfowcement and physical prompting while instructing their children. A time series design was used with experimental and control groups. The sample, consisting of two groups of 28 parents of 2 1/2- to 5-year- old children, was primarily white and middle- to upper- middle-class. Parent teaching behaviors were observed using 42 a momentary time-sampling procedure from behind one-way mirrors. The experimental group did show positive changes not seen in the control group but the researchers were cautious about making any statements or major conclusions about the relationship between attitude and behavior changes. They suggested that much more research was needed in order to make valid inferences about the relationship. Head Start Research There have been two recent evaluations of Head Start that in part deal directly with change in parenting attitudes and behaviors as as result of volunteering in the child's classroom or being involved with parent policy «councils or committees. Both studies deal with the Child and Family Resource Program (CFRP), a ten-year pilot project that was linked to regular Head Start programs. The major difference from regular Head Start was that staff worked with the total family from the birth of the child until s/he was eight years of age. Head Start usually only works with the child and family for one or two years directly preceding enrollment in kindergarten. The Government Accounting (foice (GAO) reported significant changes in parent attitudes and behaviors after participation in the CFRP. The GAO concluded that CTRP was highly effective and recommended the program to Congress as an intervention strategy with great potential for improving the functioning and well-being of low-income families. Abt Associates 43 conducted a five-year longitudinal study of CFRP. The study was designed to describe the eleven CFRPs and their operations, and to link family outcomes to participation or nonparticipation and particular aspects of treatment and family characteristics. A pretest, posttest, follow-up design was used which involved random assignment to a treatment or control group. ‘This was supplemented with an ethnographic study of the programs. The Abt study partially confirmed the GAO findings, but also identified several hnportant areas in which the program's operations and effects fell short of expectations. Parents were assisted in moving into work, school or job training which improved their chances for economic self-sufficiency. They were also offered group sessions which increased their feelings of personal efficacy as well as their knowledge and skills in child rearing. However, there were no measurable changes in infant or toddler cognitive or social development. This lack of cognitive growth was an important outcome of the study because that appeared to be the nmjor factor that influenced decision makers at the national Head Start offices to discontinue efforts to redesign Head Start along the lines of a CFRP. They were not interested in the changes in family functioning; they were interested in significant cognitive growth in the child. Two of the problems uncovered by the study were: low rates of parent participation, even though there was a strong emphasis 44 placed on this; aunt poor staff training and supervision, especially in the area of child development. Evaluators pointed out that program operations and effectiveness varied markedly from site to site; several of the programs showed significantly better results than others. The Jackson, Michigan progam was one that showed more significant improvement hi the area.Home Parent meetings to inform parents Individual Conferences Parents provide snacks or supplies for events Level 2: assisting School< ----- Home Act as aides to relieve teachers to perform more educa- tional tasks Level 3: interaction School<---->Home Parents serve on committees and boards Community Parents teach their own children However, it activities and interest at home, in school and classrooms is clear that parental and positively influence achievement. This is true even after the student's ability and family socioeconomic status are taken into account. Students gain in personal and academic development if their families continually emphasize schooling over the school years. 56 Communication between school personnel and parents has been shown to increase rates of children's attendance (Duncan, 1969; Sheats & Dunkleberger, 1979), to improve school performance (Shelton & Dobson, 1973), and to increase parent-initiated contacts with the school (Duncan, 1969). In Duncan's study (1969), school counselors met individually with parents of students entering junior high school. These students were compared to students from the previous year whose parents had only been invited to attend an orientation session before the school year began. At the end of three years in junior high school, it was found that: the average daily attendance was significantly higher for the group *whose parents had individual face-to-face contact; there were fewer school drop-outs and the grade point average was higher for the parent-contact group; and parents of the contact group came to school more often. It was reported by Epstein (1985) that three factors strongly influence the teachers' use of parent involvement. The grade level was the most important influence. Teachers of younger students made more frequent use of parent involvement than did teachers of older students. Secondly, parents active at the school influenced teachers to use more parent involvement in learning activities at honmu Finally, teacher commitment of time and energy to interact in a meaningful way with parents increased the teacher's use of parent involvement in learning activities 57 at home. The data does not support the common belief that teachers only use parent involvement activities with better- educated parents. The studies also found that married parents spent more time helping teachers at school. However, single and married parents were equally interested and willing to help their children with learning activities in the home. One of the issues that has been addressed by several studies involving families of older students is teacher willingness to reach out to involve parents. Epstein & Becker (1982) contend that many teachers leave their respective training institutions with attitudes and practices that keep parents out of the learning process and even curt of the classroom. Unfortunately, there has been little research done to link specific teacher practices with the parents who experience them or to measure differences in attitudes and reactions of parents to teachers with different parent involvement philosophies and practices. Almost 2311 of the studies that address parent involvement and teacher outreach do so from the perspective of the gains made by the child as the result of the intervention (Keesling & Melaragno 1983; Leler, 1983). There are a limited number of studies that address the relationship between teacher efforts and parent involvement. It has been left to the individual teacher to decide whether to invest time in parent involvement practices and whether the time 58 involved is wemthwhile (Epstein & Becker, 1982; Epstein, 1985). A study by Windle (1977) highlights the role of the principal and the teachers in relationship to the level of parent participation in the school. Windle conducted 100 structured interviews with principals and early childhood education staff in ten schools from eight districts in Los Angeles. This information was used to prepare recommendations for improvements of parent partuflpation programs in early childhood education schools throughout California. Some of the findings and conclusions were: the principal's communication skills and enthusiasm are the most important competencies and characteristics for effective work with parents; the involvement of parents in the evaluation of the program contributed heavily to its effectiveness; parent assistance in the subject areas of reading, mathematics, and language contributed to program effectiveness; and parents who volunteered their time and talent without pay contributed more to program effectiveness than parents who were compensated. Windle stated that this study indicated that one of the crucial needs is for more training of school personnel in the area of parent education and involvement. In summary, Bronfenbrenner states that parent involvement is an essential ingredient in program effectiveness. A number of studies found that encouragement, 59 interest and parental participation had positive effects on children's school experiences. Kinard found that the length of enrollment related to this effectiveness. Several studies indicated that sending notes home, calling parents, providing transportation and baby sitting, and locating the center in the neighborhood all promoted parent involvement. ‘Work, school and attitudes also had an impact. However, there is only a limited amount of research that addresses the relationship of parent involvement to outreach and support services from schools. The studies that have been done point to the skills and enthusiasm of the school personnel in the involvement of parents in their child's education. Parent Education and Social Support There are a number of family characteristics that tend ‘ha be fairly stable over time. Sinclair and Ghory (1980) identified cultural background, socioeconomic status as measured by parental education, income or occupation, and family size as being less changeable. However, they also stated that other characteristics were more important when it came to influencing child development. Parent stimulation of the child was the major characteristic that they identified as being important to intelligence, motivation and achievement. Iverson and Walberg (1980) 60 conducted a quantitative synthesis of 18 studies of family environments of over 5000 children in eight countries. Their research supported Sinclair and Ghory's findings. The literature is filled with assertions that the (fliild's family and home environment are more important to academic achievement than the teacher or school (Dobson & Dobson, 1975; Mize, 1977; Walberg & Marjoribanks, 1976). In a study of Head Start children, Ware and Garber (1972) identified the following factors as being related to school achievement: materials for learning in the home; parents' awareness of the child's development; rewards for intellectual attainment; expectations parents have for children's schooling; and activitiies focusing on pre- reading skills. Hess (1969) enlarged on this to include mother's teaching behaviors, the experiences mothers provide and the models they offer. Griffore and Bubolz (1986) suggest that parents can be helped through parent education programs to emhance the family's role in education and to equalize educational opportunities for all children. Program Effectiveness There are currently many diverse parent education and support: progranm hi this country. Although some groups, of which Head Start is one, have been working in the area of parent education and social support since the early seventies, there has only been a major nmvement in this direction in the past eight to ten years (Weissbourd, 1983). 61 Even though research has not kept pace with the rapid development of these programs, there is a growing body of evidence which addresses program effectiveness. The recent interest in this area seems to stem from several forces: the heightened concern about pressures on families in light of greater participation of mothers in the work force; geographic mobility that disrupts extended families and friendship ties; divorce rates; economic uncertainties; reports that early intervention programs are more effective when parents are involved; and research that deals with influences on child development, family child rearing, and child abuse and neglect (Powell, 1986). IAlthough program effects on parents have not been studied as often as those on children, there is evidence of immediate positive effects on maternal behavior (Andrews, Blumenthal, Johnson, Kahn, Ferguson, Lasater, Malone & Wallace, 1982), parental competencies in reading infant cues (Dickie 6r Gerber, 1980), use of positive and facilitative language interactions with the child (Lambie, Bond & Weikart, 1974), Open and flexible child-rearing attitudes (Slaughter, 1983), and awareness of roles as educators (Travers, Nauta & Irwin, 1982). One of the questions that is still being asked is which of the many approaches to parent education is more effective. Parent education programs have varied foci, formats and formulations. Such programs occur in hospitals, 62 homes, schools and human service programs, and use printed materials, modeling, reinforcement, videotapes, group discussion, formal classes (H' tutorials. Unfortunately, researchers have nearly always focused on overall program effectiveness and not CHI variation in program dimensions such as length, intensity, or instructional method. They have neither addressed subject dimensions such as age, SES or background, nor delivery dimensions such as time, place or target group. Even the ambitious work of the Consortium on Developmental Continuity could not establish the effect of particular dimensions of programs on parenting outcomes (Lazar,Hubbell, Murray, Rosche & Royce, 1977). Clarke- Stewart (1983) contends, after reviewing a variety of parent education efforts in the United States, that parent education in any form may provide a valuable service to children and the country. However, we need further research to put "parent education on a firmer empirical foundation in the future" (p. 271). Another major question that is being asked is whether parent education and support programs have long-term effects. Some positive results have been found mid-term in the CFRP research. The increase in parent's feelings of control over their lives as a result of program participation continued throughout the ten year project (Travers, Nauta & Irwin, 1982). It was also found that the Yale Child Welfare Project, after working intensively with 63 very low-income mothers, showed an impact on family circumstances ten years after the intervention ended. Mothers who had been in the Yale program were more likely to be self-supporting, more educated, and have smaller families than mothers an”) had not been in the program (Seitz, Rosenbaum & Apfel, 1985). The Parent Child Development Centers (PCDC) were created hi the early 1960's to provide model parent education programs for low-income parents with children from birth to three years of age. The careful planning that went into this project drew from the understanding of development and family systems. Planners were concerned with: the emerging understanding of infant development; the recognition of parental influences on child development; the determination that parents need and can profit from the latest information on child rearing; the belief that developmental programs are particularly needed for poverty families during the child's earliest years; and the understanding that adult education with a parent focus can have favorable consequences for child development and the economic condition of the family. It was felt that due to a variety of income-related factors, stress in low-income families was such that developmentally stimulating experiences were often lacking. Consequently, the programs were developed to encourage positive parent-child interactions at the earliest phases of development with the hOpe that this might lead to enhanced 64 development of both parents and children. Unlike many other early intervention programs, PCDCs put primary emphasis on the education of mothers and on other important family members when available. Adult growth and development were central, with an emphasis on the understanding that parents change their styles and beliefs over time to suit their own needs. The five-year evaluation of the three original PCDC sites, serving 60 to 70 families each, centered around the social competence concept. Zigler and Trickett (1978) argued stongly that human development programs should be evaluated by this concept. They also emphasized looking at longer-range and policy-related indicators such as rate of school dropout, grade retention in school, placement in special education, juvenile delinquency rates, teenage pregnancy rates, and participation in the welfare system. The evaluation addressed the physical well-being, cognitive ability, self-image, and achievement of the child. Researchers found that although the PCDCs differed in program intensity, they were generally effective in providing mothers with increased skills for child rearing, but turf as impressive in affecting child development outcomes (Dokecki, Hargrove & Sandler 1985). The Systematic Training for Effective Parenting (STEP) program has been used since 1976 as a model for parent education in some Head Start programs as well as other 65 community organizations. It is based on Adlerian concepts that stress democratic methods of child rearing. This approach uses such techniques as encouragement and helps parents understand why children behave or misbehave. McKay (1976) conducted a study of the effectiveness of STEP. She used 20 middle- to upper-middle class mother volunteers who had a child between four and 13 years of age. They were randomly assigned to a treatment and non-treatment group. McKay found that STEP was effective in changing the mothers‘ perceptions of their child's behavior. However, there were no significant differences in the two groups of mothers in number of facilitating and non-facilitating statements made to children after the nine week program period. In the Parents Are Teachers Too Program, parents from a rural Michigan Head Start program were involved in a home language intervention project aimed at deternflning the effectiveness of a parent-as-change-agent intervention model (Boger, Kuipers & Berry, 1969). Parents took part in language training programs Specifically designed to help them increase interactions with their children in the home with a focus on increasing children's language skills. The study incorporated a pretest and posttest design to ascertain the effects of a short-term parent training program. Children's linguistic, intelhxnual, and self- concept performances were measured. The researchers found that children whose parents participated in the training 66 increased their language skills, developed more positive self-perceptions and perceptions of their mothers' views of them, and performed better on general intellectual tasks than chidlren whose mothers did not participate in the training. Mothers who participated increased their own verbal and linguistic skills and generally improved the teaching quality of their interactions with their children. It is important to recognize that there are considerable differences among parents in terms of their program participation patterns and what they gain from the program. Some parents relate mostly to the staff, while others focus more on other parents (Powell, 1985). Some mothers accept ideas and activities more readily than others (Lambie, Bond & Weikart, 1974). Personality factors may influence parents' participation. The CFRP was nmst successful with parents who felt confident about their ability to control events in their lives. These parents were more likely to enroll in school, experienced less difficulty in obtaining health services, and were more flexible in their attitudes toward child rearing (Travers, Nauta & Irwin, 1981). A variety of conditions in a parent's life may influence participation. Mothers with extensive family and friendship ties have been found to be more responsive to home-based parent educators and more likely to form friendships with other parents in the program (Powell, 1983). Program effects have also been found to vary greatly 67 in relation to ethnic group and family structure (Cochran & Henderson, 1985). ‘The parent education component of Head Start not only offers classes on child development and parenting, but also provides social support for parents. Social support appears to have an important impact on parental functioning (Colletta & Lee, 1983). It provides emotional support for the parent, instrumental assistance to the family, and social expectations for individuals within the family. Such support as offering praise for handling a child-rearing situation or offering child-care assistance during important phases of the parent's life has a direct affect on parental behavior. Indirect effects on parental behavior may result from support aimed at showing a parent that s/he is loved and respected (Belsky, 1984). A long-term study of program participation showed that mothers who were experiencing a great deal of stress attended less often, were in the group for a longer time before they contributed to grOup discussions, and needed more time to form friendships with others in the group (Powell, 1986). Social Support, Isolation and Stress Social support has also been found to combat isolation and moderate the adverse effects of stress (Kohn & Antonucci, 1980). According to Bronfenbrenner (1979), forces that contribute to this isolation of the family are the breakdown of the neighborhood, occupational mobility, 68 consolidaton of school districts, delegations of child care to outside institiutions, and age groups segregating into different social patterns. Social networks are important for all families, but there needs to be a goodness-of-fit. The support desired and the support received need to match. There are a number of elements that make up social support: geographic proximity; interconnectedness, size, continuity, diversity and intensity of support; lines of transmission that go directltr to the child or are nmdiated through the parent; emotional and material support; access to opportunities; and the provision of models and sanctions for behavior (Cochran and Brassard, 1979). In a study by Crnic, Greenberg, Ragozin, Robinson & Basham (I983), mothers with greater levels of stress and less support were found to have had less positive attitudes and behavior, and were less able to recognize and respond to their infants' more subtle behavioral cues. The infants of these mothers were also found to be less responsive and less clear in the cues they provided, suggesting that a circular feedback loop may have existed. This finding supports Belsky's contention that it is the combination of the parent's and child's characteristics and the environment that produce the child development outcomes. In reviewing research on infant development Fitzgerald, Strommen & McKinney (1982) stated that it is important to 69 remember that at each organizational stage, various developmental pathways are available to the child. Which pathways will be followed will depend on each child's life experiences, including the level of maternal stimulation and the degree of sensitivity to the child's needs. ‘Without an attentive parent, stimulation and sensitivity cannot be achieved. Studies dealing with the patterns of attachment of infants to their mothers and subsequent child development outcomes have revealed consistent findings. Securely attached infants receive a solid foundation upon which to build social and personal skills with confidence and self- assurance (Sroufe, 1979). ‘These findings are consistent with the results of research on adult populations that have demonstrated that support is related to high self-regard, an internal «orientation and decreased stress reactions. The most common theoretical explanation for these relationships is that support serves as a mediating factor, buffering the effects of difficult life circumstances. Mbdes of Service Delivery When considering stress and support, it is important to remember that services to families can be delivered through a variety of modes: through parent education, support groups, (:linical «or individual casework services; through involvement hi child-oriented programs; or through family and friends. IAlthough there are nmny nmre opportunities than in the past for parents to learn more about parenting, 70 some researchers (Bobbitt & Paolucci, I975; Clarkson, 1975; Sparling & Lowman, 1983) indicate that families tend to turn to relatives or friends more readily than to other people. Sparling and Lowman's (1983) study was aimed at learning about parent information needs, problems, attitudes and preferences. They obtained 1,400 responses from 2,088 questionnaires that were mailed to a national stratefied random sample of mothers with newborn infants from birth to children three years of age. Che-hundred additional personal interviews were conducted with low-income parents because they were considered less apt to take part in a mailed data collection study. Sparling and Lowman found that parents want to receive information through relatives or friends. They stated that "program developers might do 'well to encourage parent-to-parent information flow and to help professional home visitors to become a friend of the fannlyfl' (p. 320). They also stated that certain authorities, such as doctors or teachers, might contribute in their recognized areas, but should not be the exclusive source of information in a parent education program. An example of a program that used peers to bring information to parents is the Perinatal Positive Parenting Program (PPP). The goal of PPP was to meet parenting support and informational needs of first-time parents. The program trains experienced parents to provide parent-to-parent support to new parents in the hospital and 71 at home. Parent support groups, facilitated by an experienced parent volunteer, are also offered to parents within three weeks after leaving the hospital (Boger, Richter 6r Weatherston, I983). Forty-eight first time mothers who delivered babies in a five month period in 1982 were randomly assigned to treatment and control groups. The treatment group participated in PPP, while the control group only received information normally given out at the hospital. Infant development, parental attitude, home envirwnunent, and parent-child transactions were evaluated. Evaluations of the program, utilizing direct observation of parent/child transaction in the home at fifteen months postpartum, indicated that treatment mothers exhibited more positive, supportive behavior and provide a higher quality of ruirturing environments for their children than did the control mothers (Boger, Richter, Kurnetz & Haas, 1986). The researchers state that "The program is holistic in nature and focuses upon the provision of specific information about infant capabilities and parent-infant transactions delivered within a sensitive and supportive social context. This context is created by the positive and powerful social process that results from a supportive dyadic relationship established between the first-time parent and a carefully trained, experienced parent volunteer." (p.142). The success of this program stimulated interest in applying the PPP model to high-risk parents and their infants. As a 72 result, the neonatal intensive care unit (NICU)PPP program was developed in 1984 to assist new parents who were coping witti the hospitalization of their newborn in an intensive care unit. The results of a study of this program indicated that mothers in the PPP treatment group showed significantly more maternal-child transactions at twelve months post- discharge than did the control group. In addition, treatment mothers exhibited significantly lower anxiety, anger, fatigue, confusion, and depression. Furthermore, treatment mothers maintained the same level of self esteem during the discharge to four months post-discharge period, while self esteem of the control mothers dropped significantly during that time (Boger, Roman & Haas, 1987). Levenstein (1970) and Gordon (1972) developed home visit models, like the PPP model, that carefully constructed colleague relationships with parents. In Levenstein's program, home visitors demonstrated verbal transaction techniques, then drew the mother into the activity and finally, took a secondary role as the mother repeated and elaborated the activities. Levenstein's research, used in both the McKey, Condell, Ganson, Barrett, NkConkey & Plantz (1985) and Lazar, Hubbell, Murray, Rosche 6r Royce (1981) studies, showed significant pcsitive outcomes for children and parents. These findings are consistent with Bronfenbrenner's (I974) conclusion that, because the participants (parent and child) remain together after 73 intervention ceases, the momentum of the system insures some degree of continuity for the future. In summary, it has been found that when individuals receive assistance from others, they feel positively valued and in control of their lives. These feelings of self- esteem and sense of control in turn result in lower levels of stress. However, there needs to be a goodness-of-flt between the support needed and that provided by the environment if there is to be assistance in dealing with difficult family circumstances. Parents tend to prefer receiving information through relatives or friends, not just from experts. ‘There are indications that short-term and mid-term effects have been accomplished by parent education and parent support programs. However, there are considerable variations among parent participation in these programs and, therefore, considerable variation in results. Adolescent Research As It Pertains To Head Start A substantial number of Head Start parents were adolescents when their children were born and have consequently experienced many of the problems attributed to adolescent parenting. In the past ten years, there have been a number of studies that have addressed the issues surrounding adolescent parenting. Although this body of research generally focuses on mother-infant issues, it has 74 implications for parent intervention at the preschool level because there tends to be a stability in the child's environment due to stable parental behavior. This stable behavior hi turn elicits similar or related behavior from the child at different ages. (Bradley & Caldwell, 1981) Although adolescent research presents conflicting findings, the majority of scientific opinion, based on the most impressive studies, agrees that children of teenagers fare worse than children of older parents. CMtcomes that have been measured are child health and development, mother- child interaction, sources of stress and support, and incidences of abuse and neglect. Early studies that have examined the quality of teenage parenting behavior have generally reported teen mothers as lacking the knowledge and positive attitudes necessary for adequate child rearing (de Lissovoy, 1973; Fhrstenberg, 1976; Sandler,. 1979). They found that parents who lack knowledge about what constitutes normal child development tend to both overestimate and underestimate development. Those who expect a child to perform a behavior earlier than is typical may become inappropriately concerned and abusive, believing that the child could perform, but chooses to do otherwise (Davoren, 1975). However, it has been found that most parents, not just teenagers, underestimate a child's abilities when they are wrong, indicating that they 75 are less aurt to abuse but nmre apt to understimulate (Klinman, 1985). Several of the recent studies, with representative samples and controls, have found that SES and educational level of the mother are the best predicters of child outcome. Cme study, assessing the childbearing and child- rearing experiences of teenage and older mothers to determine the sources of the developmental differences, found that SES was the most pervasive and strongest influence upon the predictors of infant development. However, when SES was controlled, young mothers and their infants rated higher on the predictors of child development than the older mother-infant dyads. They concluded that the negative effects of adolescent child rearing were due in part to the fact that young mothers had fewer material and nonmaterial resources to invest in their children's development (Roosa, Fitzgerald & Carlson, 1982). Kinard and Reinherz (1984) found while controlling for family structure and maternal education, that maternal education had a far greater effect (Hi the children's outcome measures of behavioral and emotional functioning. They also found no appreciable differences between children of 15—17, 18-19, and 20-24 year old mothers. It is important to remember, when considering these studies, that teenage parents are more likely to be in a lower social class throughout their lifetime because they tend to have truncated educations and 76 work in lower status and lower paying jobs (Alan Gutmacher Institute, 1981; Card 6: Wise, 1979; Roosa et al 1982). This fact alone makes many of these families eligible for Head Start because pcverty is the pmimary enrollment criteria. Summary It has been established that parent involvement makes a difference~ Hi child development outcomes. The literature supports the contention that parent involvement in the young child's school does have a positive lasting effect on child development. The length of the parent's relationship with the program, the type of program, and the type of outreach methods appear to make a difference in some cases. Recent research concerning adolescent parenting indicates that SES and educational level of the mother are the best predictors of child development outcomes for that population. This body of literature, along with studies that have used samples representing older populations, support the idea that social support moderates stress and is related to a higher level of parental functhnnng. The literature appears to indicate that parental attitudes can be changed through parenting education, and that some attitude changes may result in changes in behavior, although these attitudes and behaviors are somewhat stable over time. It was also indicated that parents tend to prefer receiving 77 informaticni through friends and relatives. These studies strongly suggest that there needs to be more work done in studying the effects of parent education on both attitudes and behavior. There are a number of areas related to parent involvement that have not been adequately researched: the relationship between teacher involvement and parent involvement; the ecological makeup of families in the involved and non-involved groups; the relationship between parent and teacher goodness-of-fit and parent involvement. Further study is needed on the effect of other factors, such as parent educational level, parent age, gender, SES of the household, employment status of the mother, the program's ability to reach out to the parent to lure him/her in to participate, and the types of participation that correlate to involvement. All of these factors must be included in one study in order to truly evaluate what drives parents to become involved. CHAPTER III METHODOLOGY This chapter describes the design, method, and procedures used to address the research questions. The chapter is divided into six sections. These sections are labeled Design of the Study, Variables and Measures, Research Hypotheses and Questions, Data Analysis, Sample, and Procedures. Design of the Study The investigation used a case study design. Data from one program was evaluated in order to establish the probable existence and description of variables that related to parents becoming involved in that program. Interview strategies were used to collect multiple data consisting of attitude, behavior and demographic informatHNIchuing the first month of the school year, outreach and involvement information throughout the school year, and program and outreach evaluation by the parents at the end of the school year. The data reflect the behavior and attitudes of parents and staff as they-occur in natural settings over time. The study investigated the relationship of parent involvement in Head Start programs to differences in parent background, attitudes and behaviors, as well as to differences in the background, attitudes and behaviors of the teachers with whom 78 79 they worked. The relationship of program variables and parent involvement was also studied. Variables and Measures Variables The dependent variables of interest hi this study were parent involvement in the Head Start child's classroom, parent involvement in parent education offered by the Head Start program, and a combination of the two. The independent variables of interest were: teacher involvement/outreach designed to elicit parent involvement in the classroom and parent education; parent child-rearing attitudes; parent child-rearing behaviors; parent self- concept; teacher child-rearing attitudes; teacher attitudes toward parent involvement and parent education; the warmth and acceptance of the teacher and the parent group as perceived by the parent; the availability of transportation and child care; and demographic characteristics of both the teacher and the parent. Measuring the Dependent Variables The Parent Involvement Log was used to record information concerning the dependent variable, parent involvement. For each parent in the study, all instances of 80 involvement in the classroom and the parenting sessions were gathered from program records and entered on this log. Nhesuring the Independent Variables The following instruments were used to collect information for the independent variables. The Teacher Involvement Log was used to tabulate contacts initiated by progrmn staff. The date and type of contact (written, phone, face-to-face) initiated by the teacher, assistant teacher or supportive service worker were detailed for each parent involved in the study. The Home Observation for Measurement of the Environment (HOME) is an observation/interview instrument developed to assess quantitative and qualitative dimensions of the child- rearing environment. This study used the version consisting of 55 items that was developed for use with families of preschool age children. All items were scored in a binary fashion and, therefore, inter-observer reliability was quite high. After a brief training period, an inter-rater rel.iabi.lity' of .9Cl or bet ter (was repcnrted in the instructional manual for the HOME (Caldwell, 1970). Caldwell used a sample of 232 parents, representing demographics similar to those found in Head Start, to compute reliability and validity. Internal consistency is reported as being .93 for the total scale using the Kuder-Richardson 20 formula. Stability is reported at .70. ‘This indicates that the scales are internally consistent and relatively 81 stable across time. The HOME appears to be significantly related to cognitive development. HOME scores obtained at 4 1/2 years of age displayed a strong associathMT with IQ, with a multiple correlation of .62. Multiple correlations ranged from .58 to .64 for the four achievement test scores of the SRA and the HOME (Caldwell & Bradley, 1984). Construct validity reports are limited. One published study (Bradley & Caldwell, 1981) reported moderate correlation between the HOME and the Infant Behavior Record from the Bayley Scales. Interrater reliability for this study was assessed prior 1x) data collection. ‘The three social workers who administered the HOME were first trained in the use of the instrument and then sent out in pairs to conduct six sessions in homes similar to those encountered in the study. Each interviewer had a mean percent of agreement over all 12 sessions of 98%, with a range of 83% to 100%. The Parents' Opinion Survey (POS) assesses parental beliefs regarding effective and appropriate chihbmearing practices and perceived competency (Luster & Rhodes, 1988). The original POS was developed for work with parents of infants and was adapted by Luster and Boger for use with parents of preschool-age children. This researcher developed an additional subscale, Parent as Teacher, for use in this research. The Parent as Teacher subscale and five of Luster's eight subscales are included in this study. 82 The scales and questionnaires that were developed by the author CH' adapted from other instruments, were reviewed by three child develomnent experts to check for appropriate conceptualization. Chly those items on which three of the four reviewers agreed (the author and three experts) were used in a pretest of the instrument. The pretest instrument was given to ten Head Start parents who were not part of the overall Study. After completion of the pretest, parents were asked to comment on any items that were confusing or inappropriate. Based on this conceptual review of the items, the final instrument was completed with an eye to utilization and relationship to the construct. Following the data collection, the internal consistency of each of the a priori clusters was estimated using Cronbach's alpha. Only one item (#12. One of the worst things a parent can do is insist that the child obey his/her every command.) lowered the reliability estimate of the cluster concerning beliefs regarding discipline and control, and was therefore deleted. The remaining items were summed to produce a score for each subscale. IAs described earlier, the revised POS was reviewed by three experts and pretested on Head Start parents before final item selection was made. The subscales included in the revised POS are Beliefs Regarding Discipline and Control, Beliefs Regarding Communication, Parent as Teacher, Perceived Competency, Perceived Importance of Extra-Familial Influences 83 and Fatalistic Versus Non-Fatalistic Outlook. Specific descriptions, which include a listing of the items for each subscale, are provided in Appendix C, along with the revised Parents' Opinion Survey in Appendix B. The Parent Education-Involvement Attitude Scale consists of three subscales: Parent Education Attitude Scale, Parent Involvement Attitude Scale, and Parent Education Techniques (Tudor, 1977). The subscale dealing with attitudes toward the various techniques used in parent education programs was not used in this study. The Parent Education Attitude Subscale was used to measure teacher's attitudes toward parent education. ‘The highest score represents favorable attitudes toward parent education and the lowest score represents unfavorable attitudes. The Parent Involvement Attitude Subscale was used to nmasure teachers' attitudes toward parents becoming involved in the child's classroom. The highest score represents favorable attitudes toward parent involvement in the Head Start classroom and the lowest represents the opposite. The items used in the study were reviewed by three early childhood education professionals for face validity. The Head Start Parent Involvement Questionnaire was developed by the author to measure parent perceptions of program and staff outreach efforts, and to gather information concerning transportation and child care availability during 84 the program year. Three early childhood education professionals reviewed the instrument for face validity. Research Hypotheses and Questions The following research hypotheses were tested. Where the evidence from previous studies is contradictory or unclear, non-directional or null hypotheses were used. The research questions are restated and the hypotheses are grouped by question. Does the level of teacher outreach/involvement with the parent relate to the parent's degree of involvement in the education components of Head Start? 111: There is a positive relationship between the overall number of teacher contacts and parent involvement in the education components of Head Start. 112: There is a positive relationship between the number of face-to-face teacher contacts and parent involvement in the education components of Head Start. 113: There is a positive relationship between the number of telephone teacher contacts and parent involvement in the education components of Head Start. Does teacher attitude toward parent education and parent involvement relate to the level of outreach/involvement the teacher exhibits to elicit parent involvement in the education components of Head Start? 114: There is a positive relationship between the teacher's score on the Parent Education Attitude Scale and the number of teacher outreach/involvement attempts. 115: There is a positive relationship between the teacher's score on the Parent Involvement Attitude Scale and the number of teacher outreach/involvement attempts. 85 Does the parent's past experience with Head Start relate to his/her degree of involvement in the education components of Head Start? H 6: There is a positive relationship between the number of years a parent has been involved with Head Start prior to this year and parent involvement in this year's education components. Il7: There is a positive relationship between the parent's degree of satisfaction with past experiences with Head Start and parent involvement in this year's education components. Does the parent's past experience in parenting classes relate to his/her degree of involvement in the education components of Head Start? H 8: There is a positive relationship between the number of parent education sessions that a parent has attended in the past and parent involvement in this year's education components. Does availability of transportation and child care relate to the parent's degree of involvement in the education components of Head Start? H 9: There is a negative relationship between the number of younger children in the family and parent involvement in the education components. IHO: There is a positive relationship between the availability of transportation to and from the program and parent involvement in the education components. lHl: There is a positive relationship between the availability of child care for younger children and parent involvement in the education components. [M3 the child rearing attitudes of the parent relate to his/her degree of involvement in the education components of Head Start? H12: There is a: positive relationship between the total score on the Parent's Opinion Survey subscales P/T, BRC & BRD/C and parent involvement in the education components. Does the degree of congruence of teacher child rearing attitudes aunt parent child rearing attitudes relate to the parent's degree of involvement in the education components of Head Start? 86 H13: There is no significant relationship between the degree txf congruence of parents' and teachers' child-rearing attitudes (as measured by the total score of the Parent Opinion Scale subscales of P/T, BRC & BRD/C) and parent involvement in the education components. Do the parent's child rearing behaviors relate to his/her degree of involvement in the education components of Head Start? H14: There is no significant relationship between the parent's child rearing behaviors as measured by the total score (Hi the HOME and parent involvement in the education components. Does the degree of parenting self concept relate to the parent's degree of involvement in the education components of Head Start? H15: There Ms a positive relationship between parenting self concept (as measured by the total score on the Parent's Opinion Scale subscales PC, PIEFI & F/NFO) and parent involvement in the education components. Does the number of hours that a parent works outside the home relate to the degree of involvement in the education components of Head Start? H16: There is a positive relationship between the number of hours a parent works outside the home and parent involvement in the education components. Does the number of hours that a parent is in school relate to his/her degree of involvement in the education components of Head Start? H17: There is a positive relationship between the number of hours a parent spends in school and parent involvement in the education components. Does the degree of warmth and acceptance of the teacher (as perceived by the parent) relate to the parent's degree of involvement in the education components of Head Start? IH8: There is a positive relationship between the parent's perception of warmth and acceptance from the teacher and parent involvement in the education components. 87 Does the degree warmth and acceptance of the parent educaticni group (as perceived by the parent) relate to the parent's degree of involvement in the education components of Head Start? IH9: There is a positive relationship between the parent's perception of warmth and acceptance from the parents in the parent education group and parent involvement in the education components. The following parent and family characteristics were of interest to the researcher and therefore were included as questions to be addressed by the study. Does family composition relate to the parent's degree of involvement in the education components of Head Start? Does the age of the parent relate to his/her degree of involvement in the education components of Head Start? Does the gender of the parent relate to his/her degree of involvement in the education components of Head Start? Does the parent's level of education relate to his/her degree of involvement in the education components of Head Start? Does the race of the parent relate to his/her degree of involvement in the education components of Head Start? Does family income relate to the parent's degree of involvement in the education components of Head Start? What is the relative strength of the relationship of parent involvement and the independent variables of interest? Data Analysis The overall data analysis consisted of several phases. First, interrater reliability was computed for all interviewers working with the HOME. Six duplicate results for the HOME for each pair of interviewers was used in 88 this analysis. Frequencies were then run on all sample parent demographics and total program demographics. These data were used to describe the degree to which the sample matched the population of parents who had children enrolled in the classroom option of the Jackson, Michigan, Head Start program. The second phase incorporated correlational and multiple stepwise regression analyses. Pearson product- moment correlations were used to show the strength and direction of the relationship for all variables except parent age, race, and family composition. The T-statistic was used to determine the significance of alpha chance probability levels for the correlational data. T-tests were used for age, race and family composition. Stepwise nmltiple regressions were used to test the relative strengths of each independent variable that was found to be significantly related to parent involvement. Hypotheses I through 19 were tested using Pearson product-moment correlations. Stepwise nuiltiple regressions were used to test the relative strengths of each independent variable that was found to be significantly related to parent involvement. This analysis helped to answer the question: What is the relative strength of the relationship of parent involvement and the independent variables of interest? 89 Sample One-hundred-thirty parents of children who attended center-based classrooms in the Jackson County Family Development Head Start Program were recruited for this study. One classroom from each of the thirteen teachers was randomly se:lec:ted for in<:lus.ion in ‘the stIIdy. IWTert: “Here approximately 258 children involved in these classrooms, some with two-parent families and some with single parents. Each teacher was assigned an assistant who worked with him/her during the school year. There were six supportive service 'workers assigned to the classrooms who facilitated parent education groups and provided referral and supportive services to families. These 32 staff persons (teachers, assistant teachers and supportive service workers) were asked to participate in the study. One teacher and her assistant chose not to be involved. In addition, another classroom teaching team was not included in the study because there was a series of substitute teachers in that classroom until late in the school year, confounding consistency in home-school communication data. IAll Head Start parents with children enrolled in the eleven classrooms were contacted by their classroom teachers during the first home visit in late September or early October. IAt that time the teacher explained the study and requested parent participation. Parents who agreed to 90 participate were asked to Sign a Research Conditions and Agreement form. Of the 130 parents who originally agreed to take part in the study, 105 completed all assessments. This represents slightly less than 20% attrition. Three parents were dropped from the sample because their children had been removed from the home by the Department of Social Services, making them ineligible for Head Start. Seven were dropped because they could not be reached for evaluation even after repeated tries. The other 15 moved out of the Head Start catchment area. The distributions of parents in the sample and in the Jackson Head Start program are described according to head-of-household data in Table 2. Characteristics compared are family composition, age, sex, education level, race, and income. The sample is by necessity an availability sample. There was no way to require parents to become involved in the research, therefore volunteers were used. A comparison of sample demographics and demographics of the total population of parents, whose children were enrolled in the Jackson Head Start progam, was performed. The two groups were compared by using head-of-household information because that was the only informaticni available from program records for the population. The assumption of equal variances was tested and the two groups were found to have about the same degree of variability. Using t-tests to compare sample and population 91 Table 2 Characteristics of the Sample and Population Household Sample Sample H.S. H.S. Characteristics N=92 % N=630 % Family Composition One parent 43 46.7 275 43.6 Two parents 49 53.3 355 56.3 Age 19 and under 1 l 65 10.3 20-29 58 63 418 66.3 30-39 32 35 139 22 40 and over I l 8 1.3 Sex Female 76 82.6 505 80.2 Male 16 17.4 125 19.8 Education level 9th grade and less 7 7.6 67 10.6 IOth-llth grade 23 25 127 20.2 H.S.graduate or GED 49 53.3 327 51.9 Some college 11 11.9 77 12.2 College graduate 2 2.2 30 4.8 Race White 70 76.1 503 79.8 Black 20 21.3 119 18.8 Other 2 2.2 8 1.3 Income 0-2,999 4 4.3 69 10.9 3,000-5,999 30 32.7 226 35.8 6,000-8,999 _ 34 36.9 180 28.6 9,000-ll,999 8 8.7 70 11.1 12,000-14,999 6 6.5 39 6.2 15,000 and over 10 10 9 46 7.3 Employment Unemployed 56 60.9 349 55.4 Employed 36 39.1 281 44.6 92 means, it was found that sample heads-of-household were significantly older than those found in the overall population of the Head Start program (p= .05). However, the sample was similar to the population in racial mix, single- parent versus two-parent families, sex, education level, income, and employment conditions. The figures for the overall population of the Head Start program included 140 home visit households which had younger children in the program. It is possible that these parents would be younger, on average, than the classroom parents whose children were older. ‘This would tend to skew the age of the population toward younger parents. Procedures During pre-service training, the classroom education and supportive service staff were given a brief overview of the study and asked to participate. Those who agreed were given a Staff Research Conditions and Agreements form to sign. This allowed the researcher to gather information about program outreach, teacher characteristics and teacher attitudes and opinions. Each teacher was given a packet of materials to be used while recruiting parents for the study. The packet included an explanation of the study, a parent agreement form, Parents' Opinion Survery, and stamped envelopes with the researchers address. A random selection 93 of one of the two classrooms for each teacher was made and designated the target group for the study. During the regularly scheduled home visits to enroll families in the program, each teacher gave target group parents a letter describing the study and requesting volunteers. Teachers read the letter to any parents who could not read, answered questions, and obtained signatures on the Research Conditions and Agreements Form for those who volunteered to participate in the study. This signed permis- sion allowed the researcher to gather information about the parent from program records, to come into the home to administer the Home Observation for Measurement of the Enviornment (HOME) and the Parents' Opinion Survey (POS) during October and November, and to administer the Head Start Parent Involvement Questionnaire in late May. The POS and a stamped addressed envelope was left with each parent volunteer who could read. Three assistants were hired to administer the HOME for all study parents and the POS for those who did not mail their POS back within four weeks. The interviewers were all social workers, experienced in dealing with low-income people. They were given a short training session that consisted of reviewing the instrument with the researcher and accompanying her (n1 a sample home visit to administer the HOME. 94 In order to ascertain interrater reliability for the HOME, each interviewer paired up with each of the other two and conducted six interviews per pair. The percent of agreement was figured for each item on the HOME for all 18 pairs of results. As soon as parents were enrolled in the study, an interviewer was assigned and appointments were made to administer the HOME. At this time parents who had not completed the POS were asked to do so. Gifts were given at this time to parents for their involvement in the study and to promote continued cooperation. The gifts consisted of crayons, scissors and paper for the children in the family. Demographic information pertaining to the teachers and sample parents was collected from program records during October and November. Teachers, teacher aides, and supportive service workers completed the Parent Education- Involvement Attitude Scale during November. Parent education and parent volunteer attendance information was collected from program records on a monthly basis. Staff involved in the study recorded their contacts designed to elicit parent involvement on a staff outreach form. The form contained the names of their parents involved in the study and outreach categories. At the close of the program year in late May, each parent volunteer was given the Head Start Parent Involvement Questionnaire by his or her classroom teacher during the 95 closing home visit. For those who did not return the completed questionnaire by mail within two weeks, follow-up phone calls and/or home visits were made. The time line of the study data collection is provided in Figure 3. Table 3 Data Collection Time-line For The Program Year Sept Oct Nov Dec Jan Feb Mar Apr May June Parent: Parent: demographics, attitude childrearing toward PI attitudes and component. behaviors. Teacher: demographics, PI & PE attitudes, child rearing attitudes. Monthly teacher outreach/involvement recorded = Total outreach Monthly parent involvement recorded = Total PI sept Oct Nov Dec Jan Feb Mar Apr May LJDne All information was kept confidential and results were treated so that the subjects remained anonymous. Program staff did not know which teachers provided the highest or lowest levels of support to parents. Therefore, there was no way that program administrators could use study results to reprimand or promote teachers because of their recorded 96 performance. Because of these precautions, there were no apparent risks to parents, their children or the Staff who took part in the study. In order to complete the statistical analysis necessary to address the research questions, Statistical Language for Microcomputers was used. The results of the analysis are offered in Chapter IV. CHAPTER IV RESULTS The main objective of this study was to develop a better understanding of the factors that nfight influence parent involvement in both the Head Start child's classroom and in the parent education groups. Chapter IV presents the results of the data analyses. 4A p<.05 level of significance was used and all .01 or .001 chance probabilities were noted. Pearson correlations were used to assess the significant relationships for all hypotheses. In accordance with the SL-MICRO design for analyzing data, the T-statistic was used ‘ho assess alpha levels for decision rule application. Statistical results are presented, and interpreted for each hypothesis in order of their presentation in Chapter III. Teacher Out reach H 1: There is a positive relationship between the overall number of teacher contacts and parent involvement in the education components of Head Start. H 2: There is a positive relationship between the number of face-to-face teacher contacts and parent involvement in the education components of Head Start. H 3: There is a positive relationship between the number of telephone teacher contacts and parent involvement in the education components of Head Start. 97 98 Teacher outreach was looked at from several perspectives. Total teacher outreach, the combination of teacher, teacher assistant and supportive service worker, is reported in Table 4. The Head Start classroom teacher and his/her assistant generally performed their outreach together rather than individually. Most face-to-face contacts were made as a team, and in most cases, notes were sent home from the classroom rather than an individual. Because of this, outreach analyses for the classroom teacher and assistant teacher are reported for the team only in Tables 5 and 6. Supportive service worker outreach analyses are reported in Table 7 and 8. Table 4 Pearson Correlations and T-statistics For Teacher Outreach and Parent Involvement Total Outreach Face-to-face Phone Classroom 0.2649** 0.3005** ~0.0703 Involvement 2.788 3.197 -0.716 Parent Ed. 0.1983* 0.2800** -0.0007 Involvement 2.053 2.960 -0.073 Total Parent 0.2835** 0.3448** -0.0561 Involvement 3.000 3.728 -0.570 *P<.05 **P<.01 n=105 Total outreach and face-to-face outreach were significantly correlated (p<.01) with all categories of parent involvement except for parent education involvement 99 and total outreach, which were significantly related at the p<.05 level. ‘The correlations between phone outreach and parent involvement were negative but not significant. Table 5 Pearson Correlations and T-statistics for Teacher and Assistant Teaching Team Outreach and Parent Involvement Total Classroom Parent Education Outreach Outreach Outreach Classroom 0.1288 0.0920 0.0798 Involvement 1.318 0.937 0.813 Parent Ed. -0.0005 -0.1590 0.2328** Involvement -0.051 -l.634 2.429 Total Parent 0.0951 0.0002 0.1583 Involvement 0.969 0.025 1.627 **P<.01 “=105 The