Tthfié 1W{Hill/IUHIIHIINII)Will/{NIHIll/Wil/l/UM/M 3 1293 10417 6759 This is to certify that the dissertation entitled A PSYCHOLOGICAL STUDY OF PERCEIVED PARENTAL CAREGIVING BEHAVIORS OF ELEMENTARY SCHOOL CHILDREN FROM WORKING AND NON-WORKING MOTHER FAMILIES presented by Jaye Laurel Hamilton has been accepted towards fulfillment. of the requirements for Doctoral degree in Counseling Psychology i //;- W V Mgr prof/y sor 1/9/6232 / / MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 RETURNING MATERIALS: MSU Place in book drop tof remove this checkout rom LJBRCRJES your record. FINES will be charged if book is returned after the date stamped below. (3‘ \o ’xfiiflb o e \W! '~ jflfim’t .13: ,3;- W“ ‘ ‘ "" I. .\ . 5 ‘ ‘ A ‘ d. ”I" ‘ a 1 - 5,. ‘nqoj l-) . 2L7? (" .» ' ‘ NOV 1 5 2062 AMI--349 , .wvuo'Baég ‘-9""-, 1126 02 .... Q. . 1. . 2w- ? (7 1‘3 ;' ‘ "«_'_ APR 279995 p 07! 1 0 8’ A PSYCHOLOGICAL STUDY OF PERCEIVED PARENTAL CAREGIVING BEHAVIORS OF ELEMENTARY SCHOOL CHILDREN FROM WORKING AND NON-WORKING MOTHER FAMILIES BY Jaye Laurel Hamilton A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services and Educational Psychology 1982 a)!" ABSTRACT A PSYCHOLOGICAL STUDY OF PERCEIVED PARENTAL CAREGIVING BEHAVIORS OF ELEMENTARY SCHOOL CHILDREN FROM WORKING AND NON—WORKING MOTHER FAMILIES BY Jaye Laurel Hamilton The purpose of this study was to explore whether or not children's perceptions of their parents' caregiling bghgyigrs were affected by the family work status (working mother family or non-working mother family), the sex of the child (male or female), and the sex of the parent (mother scale or father scale).§jCaregiving behaviors were defined as loving, controlling, and punishing parent behav- iors as measured by the three subscales of the Cornell Parent Behavior Inventory.wj The sample for the study consisted of seventy "satis- fied" volunteer families with third, fourth, or fifth- grade children in them from the Lansing-East Lansing area. A causal comparative design was used to determine relation- ships between children's perceptions and the independent variables. The Cornell Parent Behavior Inventory was administered to children privately in their homes by trained interview- ers. Parents filled out questionnaires to determine family characteristics. An unweighted means repeated measures analysis of var- iance was used to test twenty-one research hypotheses. Hypotheses were tested at the .05 level of significance. Jaye Laurel Hamilton No evidence was found to indicate that children's per- ceptions of loving and punishing parental caregiving behav- iors were significantly affected by the family work status, the sex of the child, and the sex of the parent. A triple interaction effect was found between these independent var- iables and the controlling perception. Post hoc analysis could not locate the source of the interaction. This find- ing indicates a trend which warrants replication. The data suggest that children from satisfied middle- class families may not perceive mother's working absence as deprivation by either parent:: Also suggested is that par- ents from satisfied middle-class families relate to their children in the same basic pattern regardless of mother's work status. Mother's employment may not be as influential a factor in child rearing as has been assumed. Parental satisfaction with mother's role and with their marriage may be more influential factors on children's perceptions. Suggestions for future research were presented. Copyright by JAYE LAUREL HAMILTON 1982 ACKNOWLEDGMENTS I want to thank the members of my committee, Dr. Norm Kagan, Dr. Gary Stollak, Dr. Sam Plyler, and Dr. Al Aniskiewicz, for their continuous support and counsel during this dissertation project. I am grateful to each of them also in different ways for their contribution to my educa— tion and development as a psychologist. A special thanks to Dr. Kagan for his understanding and good judgment during this project. I am very grateful to Dr. Stollak for his practical advice, his availability, and his intellectual stimulation. My deep appreciation is extended to Dr. Plyler for his constant support and wisdom. I want to acknowledge the East Lansing and Bath Public School systems for their assistance in allowing me to con- tact students and parents through them. And, of course, I am extremely grateful to all the parents and children who were willing to participate in this project. To Dr. Jed Lewis much appreciation is extended for his patience and guidence in the statistical analysis. I want to express my gratitude to the members of my office for their "perspective" and concern. Finally, to John and Anna, thank you for your loving support and humor which helped me keep my life in balance. iii TABLE OF CONTENTS LIST OF TABLES .............. ....... .................... vii LIST OF FIGURES ........................................ ix Chapter I. INTRODUCTION ........ .............. . .............. 1 Statement of the Problem .............. ..... ...... 1 Need for the Study ..... . ................... ...... 2 Theoretical Framework ............ . ............... 5 Delimitations of the Study ................... .... 11 Definition of Terms ............................. . 12 Assumptions ................... . .......... . ..... .. l3 Hypotheses ....................................... 14 Overview ....... ' .................. . ............... 15 II. REVIEW OF THE LITERATURE AND RELEVANT THEORIES ... 16 Effects of Maternal Employment on Children ....... 16 Overview ............. ............... ... ....... 16 Effects of Maternal Employment on Infants ..... 19 Effects of Maternal Employment on Elementary School Children ..... ..... ..................... 23 Effects of Maternal Employment on Adolescents.. 36 Summary .... .............. ..... ....... ... ...... 41 Theory of Perceptual Psychology .................. 41 Children's Perceptions of Parental Behavior ...... 44 iv III. IV. V Children's Perceptions of Parental Behavior and Maternal Employment .............................. 54 Summary .......................................... 58 METHODOLOGY ...................................... 60 Design ........................................... 60 Population ....................................... 64 Description of the Sample ........................ 64 Selection of the Sample .......................... 65 Data Collection .................................. 66 Obtaining Children's Perceptions .............. 66 Training of Interviewers ...................... 67 Instrumentation ............................... 68 Mother's Questionnaire ........................ 73 Father's Questionnaire ........................ 75 Statistical Hypotheses ........................... 76 Hypothesis 1 .................................. 76 Hypothesis 2 .................................. 78 Hypothesis 3 .................................. 80 Methods of Data Analysis ......................... 82 Summary .......................................... 84 ANALYSIS OF RESULTS .............................. 86 Restatement of the Problem ....................... 86 Pilot Study ...................................... 87 Pilot Sample .................................. 87 Pilot Results ................................. 88 Pilot Discussion .............................. 92 Subjects ......................................... 92 Pearson Correlations .. ........................... 98 Hypothesis 1 ..................................... 100 Hypothesis 2 ..................................... 101 Hypothesis 3 ..................................... 106 Summary ...... . ....... .... ..... . ....... ..... ...... 109 V. SUMMARY, CONCLUSIONS, DISCUSSION, AND IMPLICATIONS FOR FUTURE RESEARCH .............................. 113 Summary . ............ . .................. . ...... ... 113 Conclusions ...................................... 117 Discussion ........ ...... . ........ ................ 121 Implications for Future Research .............. ... 131 Appendices A. LETTERS TO PARENTS ............................... 135 B. TRAINING MATERIALS ..... ............... .. ......... 138 C. INFORMED CONSENT ................................. 143 D. CORNELL PARENT BEHAVIOR INVENTORY ................ 145 E. PARENT QUESTIONNAIRES ............................ 151 F. PEARSON CORRELATIONS (PPMCC) ..................... 159 BIBLIOGRAPHY ........................................... 165 4.12. 4.13. 4.14. LIST OF TABLES Cornell Parent Behavior Inventory - Short Form Reliability Estimates ..... ....... ....... ......... 71 Pilot Analysis of Variance Summary: Scores for Loving Scale .... .............. ........... ........ 90 Pilot Analysis of Variance Summary: Scores for Controlling Scale ........ ............. ..... . ..... 91 Pilot Analysis of Variance Summary: Scores for Punishing Scale . ...... ..................... ...... 93 Frequency Distribution of Children's Ages ........ 96 Frequency of Distribution of Socioeconomic Status (SES) of Mothers . ............ .................... 97 Frequency of Distribution of Socioeconomic Status (SES) of Fathers ............................ ..... 98 Means and Standard Deviations for Loving Score ... 101 Analysis of Variance Summary: Scores for Loving Scale ....... . .................................... 102 Score ................. . .................. . ....... 103 Analysis of Variance Summary: Scores for Control- ling Scale ........ ............. ... ............... 104 Summary of Simple Two-Way Interaction Effects: Scores for Controlling Scale ............ . ........ 105 Means and Standard Deviations for Punishing Score 108 Analysis of Variance Summary: Scores for Punish- ing Scale ..... ................... .... ...... . ..... 110 Summary of Hypotheses and Results ................ 112 vii viii Pearson Correlations (PPMCC) Between Six Covariates Identified in the Literature and Perception Scores for All Groups .. ...... ... ....... ... ..... . ....... ..159 Pearson Correlations (PPMCC) Between Six Covariates Identified in the Literature and Perception Scores for Male Child Working Mother Families ............l60 Pearson Correlations (PPMCC) Between Six Covariates Identified in the Literature and Perception Scores for Female Child Working Mother Families ..........161 Pearson Correlations (PPMCC) Between Six Covariates Identified in the Literature and Perception Scores for Male Child Non—Working Mother Families ........162 Pearson Correlations (PPMCC) Between Six Covariates Identified in the Literature and Perception Scores for Female Child Non-Working Mother Families ......163 Pearson Correlations (PPMCC) Between Perception Scores for All Groups ............. ..... ...........l64 3.1. 4.1. LIST OF FIGURES Repeated Measures Design Model ........... Three-Way Interaction of Factors A, B, and C for Controlling Parent Behavior .............. ix CHAPTER I INTRODUCTION Statement of the Problem The purpose of this study was to determine whether or not children from working mother families and children from non-working mother families differ in their perceptions of their mother's and father's caregiving behavior. Differ- ences in perceptions were analyzed as a function of family work status, sex of the child, sex of the parent, and role satisfaction of parents. It has been shown that mothers may be interacting less with their children and/or in different ways when they work. It has also been demonstrated that the mother's working status changes the interaction between father and child (Gold 1979).] If these changes in interaction, by changing children's perceptual field, affect children's perceptions of their parents' behavior, then there could be differences in perceptions of parental caregiving behaviors between children from working mother families and non-working mother families. Since it has been demonstrated that perceptions of parental behavior are important determinants for chil- dren's behavior, differences detected between the groups would have important implications for child development. 1 2 Need for the Study ‘\ \_ . According to the 1980 census, over forty-three percent, of women with children under the age of six years work; the I percentage jumps to fifty-three percent when women with children under the age of eighteen are considered. In 1960, 18.9% of mothers with preschool children anui thirty-nine percent with school—aged children, worked; in 1971 these figures were 19.6% and 49.4% respectively, United States Bureau of the Census, 1972 (Etaugh 1974). The phenomenon of the working mother has grown rapidly in the past fifty years to the point where it is becoming the norm rather than the exception. Historically,women stayed at home and provided most or all of the child care and home care functions. The woman has changed her role to include out-of-the-home func- tions. It is assumed that this role change alters not only mother's role and functions in the family but also father's role and functions in the family. Knowing just what impact mother's employment outside of the home is having on various aspects of family life becomes an extremely significant con- cern. Since the role change is such a radical one and goes against the grain of societal patterns of hundreds of years, parents are often guilt ridden and conflicted about having mothers and/or both parents work (Hoffman 1974). Their most pressing concern is having adequate knowledge of the effects of maternal employment on children. Parents need to know what effects mother's employment will have on the rearing of children and on the emotional health and personalities 3 of their youngsters (Glueck and Glueck 1957; Hoffman 1974). Without this knowledge,parents cannot make informed choices concerning mother's work status and subsequent family life— style. Parents, especially mothers, are required to make career decisions and lifestyle decisions without adequate knowledge of maternal employment effects on children. Much of the research on the effects of maternal employ- ment on children has been conflicting, wrought with methodo- logical problems and devoid of theoretical framework or explanation. Recent reviews (Etaugh 1974; Hoffman 1974; and Wallston 1974) have noted these problems and have suggested that future research be based in theory with clearly defined parameters of the variables involved. This can promote com— parison and generalizability. More specifically,it has been hypothesized that the variable of maternal employment is too broad to be useful, and that lack of significant results can be attributed to this. Hoffman suggests future research be done on subgroups of working and non-working mothers (and families) according to such theoretically relevant variables as age and sex of child, socioeconomic status of the family, mother's satisfaction, and parents' attitudes towards work. Hoffman also suggests examining children's perceptions as an important mediating process between maternal employment and any child characteristic or behavior. Combs' theory of perceptual psychology maintains that behavior is a function of the perception of the individual. Combs indicates that a person's behavior is a direct result 4 of his perceptions at the moment of behaving. According to Combs people can behave only in terms of what seems to them to be so (Combs and Snygg 1974). So a person's perceptions become essential to his/her behavior in interpersonal rela- tionships. If this perceptual theory is valid, then a child's perception of his interpersonal climate would have a direct bearing on his behavior. It then becomes important to investigate how children of working parents perceive their parents and to see if there are differences between these perceptions and the perceptions of children from more traditional families where father works and mother remains home. The need to compare the perception of parental behavior characteristics between children from working mother families and non-working mother families subgrouped along theoreti- cally relevant variables with improved methodology is appar— ent. The present study corrects for methodological problems by clearly defining working mothers, by using a more widely used, more valid and reliable instrument to measure chil- dren's perceptions, and by examining mother's and father's satisfaction with mother's working status. The study also includes children's perceptions of fathers in working mother homes, a variable previously unexamined in studies on chil- dren's perceptions of caregiving behaviors of parents from working mother and non-working mother families. If chil- dren's perceptions affect their behavior and if the number of working mothers continues to increase, then the data 5 pertaining to differences in children's perceptions of par- ental behavior characteristics in traditional and working mother families would be of importance in understanding and predicting child behavior, and consequently in helping par- ents make informed decisions about careers and family life- style. The present study was an attempt to examine the effects of maternal employment on children in the little explored area of children's perceptions with improved methodology. The intent was to increase knowledge of the effects of maternal employment on children to aid parents in making informed choices regarding mother's work absence from the home. Theoretical Framework Initially child developmentalists and child behavior researchers assumed and saw the Working mother as having II negative effects on the child. The research prior to 1950 pointed mainly towards the negative effects of the working mother on her children, especially school—age children, where much research has focused because of subject availa- bility. Maternal employment was linked to juvenile delin- quency, poor academic achievement, increased anxiety and /“ problems with sexual identity (Hoffman 1974; Howell 1973). Investigators, assuming maternal employment to be harmful to children, often formulated hypotheses and phrased question- naires so that only ill effects could be demonstrated 6 (Howell 1974). This underlying assumption of the ill effects of working mothers on their children stemmed from the impact of Bowlby's writings and Spitz's research on the tragic consequences of maternal deprivation and the impor- tance of the mother-child bond (Bowlby 1969; Spitz 1945). The theory maintains that children need the permanent stable devotion of one person with whom he/she can form a // close attachment. Ego psychologists have expanded on this, especially Margaret Mahler, who has delineated the normal developmental stages a child goes through to reach object constancy and to have a secure "inner base" considered essential for healthy personality development. Without this bonding process and the subsequent separation and individu- ation phases, a child can withdraw, become depressed, and in’/ cases of deprivation described by Spitz, die (Mahler 1975). Mother's presence (or a primary caregiver) in specified ways is considered essential to this process of symbiosis, sepa— ration, and individuation. Mother's absence is thus seen as an interruption to normal development. It has been assumed that similar effects (depression, withdrawal, unhealthy personality development) were possible/‘ as a result of mother's absence while working. In reality, however, the kind of deprivation to which Spitz and Bowlby were referring could be very different from the situation of those working mothers who return daily to interact with their children and who also provide substitute care. Yet this theoretical stance and attitude, that disturbed 7 personalities result from maternal absence, prevails, and has had far-reaching consequences on research formulations and society's lack of acceptance of maternal employment. It has very directly impacted on the working mother's feelings / of guilt and anxiety about "leaving" her children routinely (Hoffman 1974). However, during the late fifties and early sixties there seemed to be a new theoretical outlook on maternal employment, as it became more widespread without seeming adverse effects on the children. The growth of the women's liberation movement with its different theoretical position challenged the deprivation stance. Sex role ideologists (Hartley 1960) hypothesized positive, not traumatic, effects of maternal employment on children because of the expanded role possibilities available through both mothers and fathers providing different models. There could be a break- down of stereotypic male and female behavior, so children could have more choice in role behavior. Subsequently, investigators expanded their assumptions and improved their methodology, most specifically in terms of controlling extraneous variables. However, the evidence from this body of research was confusing and without any clear trends. In a review by Hoffman and Nye (1960).it was indicated that maternal employment had no effects on the personality development of the child. In other words, chil- dren from working and non-working families seemed fairly similar. Yudkin and Holme (1960), Siegal and Haas (1960), ( and Stolz's (1960) reviews of the research in this area sug- gested that maternal employment should not be studied at all. However, Hoffman assumed that these vague findings may have more to do with improperly defined and explored research questions rather than there being no effect. She suggested that studies of maternal employment as a general concept would yield little, and that an examination of the effects under specified conditions would prove more fruitful. Since then, some investigations have borne these specula- tions out. Certain effects of maternal employment in cer- tain situations have been found for subgroups of the maternally employed. or of in if It has been determined that it is not a useful concept approach to try to determine whether or not the effects maternal employment are negative or positive for childrenl general. The questions instead would prove more fruitful formulated for specific effects in specific children in specific subgroups. Also taken into account must be theo- retical formulations as well as process considerations, rather than merely relating maternal employment to a single child characteristic. The typical study deals with only two levels-- the mother's employment and a child characteristic. The many steps in between--family roles and inter— action patterns, the child's perceptions, the mother's feelings about her employment, and child rearing practices are rarely measured. The dis- tance between an antecedent condition like mater- nal employment and a child characteristic is too great to be covered in a single leap. Several levels should be examined in any single study to 9 obtain adequate insight into the process involved. (Hoffman and Lippitt 1960) As Hoffman suggests children's perceptions may be an important area to investigate in the process between mother's employment status and the child's personality development. This area has gained considerable impact in the area of child personality development. Traditionally, theory on personality development revolved around the importance of child rearing practices and their influence on personality development. The accumulated literature, con- cerning the effects of child training methods on later per- sonality development, consists mainly of data obtained by questionnaires to and interviews with parents, or by obser- vations of parental behavior. The authors of these kinds of studies made assumptions about the relationship between what the parent said he did or was observed doing and personality factors in his children. Only a few studies focused on the child's perceptions of parental behavior or their feel- ings about it. In general there has been a lack of acknowl- edgment that it is not only the parental behavior to which the child responds but also his perceptions of parental behavior. Research shows that regardless of how the mother or father is, the child's perception determines the mother's impact on the child and his/her subsequent way of relating and internalizing the mother. A number of theorists in psy- chology have proposed that person perceptions are mediating variables between one person's actions and the other 10 person's responses, and thus influence interpersonal behav- ior (Taguiri 1974). Several writers have pointed out that a child may be more affected by how he or she perceives parent behavior than by the actual behavior that is emitted since child perceptions to some extent determine the impact of parent behavior. An analysis of perceived parental atti- tudes and children's ego structure led to the conclusion that although parental behavior is an objective event in the real world, it affects the child's ego development only to the extent that and in the form in which he perceives it. Hence the child's perception of parental behavior is in reality a more direct and relevant determinant of personal- ity than the actual behavior or stimulus to which it refers (Ausubel 1954). Research on person perception, beginning in the early part of the century, has demonstrated that the way people view others' characteristics and behavior can be quite dif- ferent from psychologists' best estimates of the reality of the object of perception. Hastorf, Schneider and Polefka (1970) conclude that the psychological processes in the per— ceiver play a large part in the manner in which he/she organizes his/her perceptual world. Thus the perceiver's role is an active one in which he/she (not always con- sciously) selects and synthesizes stimuli, often in idiosyn- cratic ways. Person perceptions, then, are not always accurate reflections of the actual characteristics and behaviors and vice versa. However, they are important and 11 powerful parts of the process of relating between parents and child (Dubin 1965). In summary, then, an exploration of children's percep- tions of parent behaviors as a function of subgroups of child and family characteristics could shed light on the effects of maternal employment on children. The literature suggests possible differential effects on perceptions by sex of child, sex of parent, family work status, and paren- tal role satisfaction. By examining children's perceptions as‘a function of these variables, effects of working mother families could be examined according to maternal deprivation and perceptual psychology literature. Theoretically, mother's work status would change children's perceptions of parents' behavior by altering the perceptual field. Accord- ing to ego psychology mother's work absence would be experi- enced as deprivation (unloving, punishing), and thus differences in children's perceptions of parent caregiving behavior would be expected between groups of working mother children and non—working mother children. Delimitations of the Study This study did not attempt to measure child rearing activities of mothers and fathers in the home or mother's and father's activity level with their children. The focus of this research was on children's perceptions of their par- ents' behavior only. Previous research has demonstrated a difference in activity level and time spent with children 12 between mothers and fathers in working mother homes and non— working mother homes. The study was limited by the use of a volunteer sample, which means that results were generalizable only to a volun- teer population. Refusals were investigated to determine how the sample may be baised. Only third, fourth, and fifth graders were sampled in this study, so results were also limited in generalizability to this age group. Different perceptions may occur at dif- ferent ages as a function of cognitive development and per- ceptual sophistication. However, this question was not investigated at this time. Measurement of children's perceptions was done in dne interview only with no follow up or longitudinal investiga- tion. Measurement of children's perceptions was limited to a paper and pencil inventory and children's report. No meas- urement of more covert processes was attempted. Though various trainer effects were controlled for (sex, standardized instructions, biweekly trainer discussion groups), there was no formal assessment of trainer behavior during their interviews with subjects. Definition of Terms Working mother - A mother who has worked outside the home for paid employment for thirty hours per week or more for at least two years prior to data collection. 13 Non—working mother - A mother who has not worked outside the home for paid employment for two years prior to data collection. Working father - A father who has worked outside the home for paid employment for thirty hours per week or more at least two years prior to data collection. Non-working father - A father who has not worked outside the home for paid employment for two years prior to data collection. Caregiving behavior - The behaviors of loving, controlling, or punishing as measured by the Cornell Parent Behavior Inventory (PBI). Parental role satisfaction - On the whole, parents report of being satisfied or not with mother's role of working mother or non-working mother (homeworker). Perception - Child's report of parent behavior on the Cornell Parent Behavior Inventory (PBI). Assumptions In carrying out this research, the following assump- tions were made. The first is that the parents and children in this study are typical parents and children and do not differ significantly from other third, fourth, and fifth- grade children and parents volunteer samples. The second assumption is that the relationship or some important relationship aspects of the parent-child relation- ship can be measured by a paper and pencil inventory. It is l4 assumed that children's report of parental behaviors is a reflection of some important interactional aspects of their relationship. It is also assumed that children's perceptions are con- stant, not trainer based, instrument based or situation based but are replicable within the age groups investigated. Hypotheses The following general hypotheses were tested in this study. These hypotheses are stated in testable form in Chapter III. H01: There are no significant differences in chil- dren's perceptions of loving parent behavior as a function of family work status (working mother family or non-working mother family), sex of the child (male or female), and sex of the parent (mother or father). H02: There are no significant differences in chil- dren's perceptions of controlling parent behavior as a func- tion of family work status (working mother family or non- working mother family), sex of the child (male or female), and sex of the parent (mother or father). H03: There are no significant differences in chil- dren's perceptions of punishing parent behavior as a func- tion of family work status (working mother family or non—working mother family), sex of the child (male or female), and sex of the parent (mother or father). 15 Overview Literature on the effects of maternal employment on children and children's perception of parental behavior is reviewed in Chapter II. Topical areas covered include: effects of maternal employment on infants, effects of mater- nal employment on elementary school children, effects of maternal employment on adolescents, theory of perceptual psychology, research on children's perceptions of parental behavior, and children's perceptions of parental behavior and maternal employment. The design of the study is described in Chapter III. The nature of the sample, instrumentation, and experimental design are discussed in detail. Research hypotheses are stated in statistical form and analyses are delineated. The data are presented and analyzed in Chapter IV. A discussion and conclusions, based on the results of the investigation, are stated in Chapter V. Implications for future research are also suggested. CHAPTER II REVIEW OF THE LITERATURE The review of the literature covers two basic areas: (1) the theory and research on the effects of maternal employment on children, and (2) a review of the literature on person perceptions specifically as regards children's perceptions of parental behavior. Effects of Maternal Employment on Children Overview Social concern for children of employed mothers has a long history. Professional and public concern began with the Industrial Revolution in England and Europe when mortal- ity statistics indicated that the babies of factory women died in alarming numbers. There was also evidence of increasing juvenile crime developing in the new overcrowded industrial areas (Stolz 1960). Since World War II, when the number of employed mothers increased in the US on an unprec- edented level, there has been an increasing interest shown in the influence a working mother has on child adjustment. The first studies on maternal employment done in the US were during the 1930's and 1940's and were mainly demo- graphic. They did not use sophisticated techniques of 16 l7 sampling and design, did not control for confounding vari- ables, and mainly examined data for argument's sake rather than knowledge (Hoffman 1963). Many of these early studies were done by economists and social workers and aimed at demonstrating the negative effects of working mothers to use in pressing for social reforms. After World War II, as concern for the children of working mothers gained momentum, many more sophisticated studies appeared. Theory, too, began to speak to and focus on the effects of maternal absence or "maternal deprivation" as it was called. In the Employed Mother in America (Nye and Hoffman 1963), the first book to try to integrate the research on maternal employment, studies dealing with children of employed and non-employed mothers were presented. About the same time,Stolz (1960) and Siegel and Haas (1963) also pro- vided excellent reviews of the literature on the effects of maternal employment on children. All of these reviewers came to a similar conclusion--that no generalization holds true for all working mothers or for all their children. The few significant differences reported could be accounted for by chance variations since they constituted a small propor- tion of the tests made and have not been replicated. It was determined that maternal employment is not so "potent a var- iable that it can be used without further specification and without examination of the data separately within different subgroups" (Hoffman and Nye 1963). It was suggested that 18 further research be examined in light of subgroups such as social class, age of child, sex of child, mother's attitude toward work, full-time versus part-time employment, etc., and that significant differences would then be detected. However, perhaps because of the lack of significant findings in earlier studies, few studies of maternal employ- ment were done during the late 1960's and 1970's. The ones that did arise, though, were more focused on relevant sub- group comparisons and found significant differences between children in these subgroups. These will be discussed in this chapter. The research trend has swung from demographic studies in the 1930's and 1940's demonstrating negative effects of working mothers on children to comparative studies of work- ing mother children and non-working mother children in the 1950's and early 1960's demonstrating few significant dif— ferences. During the late 1960's and 1970's studies of sub- groups of the maternally employed began to show significant differences between variables compared. The problem with some of these later studies and present-day research is their lack of grounding in theory and their lack of explana- tion of findings in terms of relevant theory. This review of the literature on maternal employment will discuss the most important studies illustrating these trends relevant to the present study. The review is organ- ized in the following manner: the effects of maternal employment on infants and preschoolers, the effects of 7% maternal employment on elementary-aged children, and the effects of maternal employment on adolescents. Effect of Maternal Employment on Infants Greatest concern is usually expressed about the effects of maternal employment on young children. Yet infants and preschoolers have been studied much less frequently than older children. Problems in data collection account for this paradoxical state of affairs. In fact, there are no studies dealing directly with the psychological effects of maternal employment on infants.[ However, extrapolation from} 1 (6ther studies sheds light on this area (Etaugh 1974). ‘ {Bowlby and Spitz stress the importance of attachment 2 and a one-to-one relationship in the early years (Spitz Z 1945; Bowlby 1958, 1969).] Although most of this research has been carried out in institutions under extreme condi- tions of maternal and social deprivation, it nevertheless seems clear that during critical periods, cognitive and affective inputs can have important ramifications that affect an individual throughout his life.) Research suggests that without a one-to-one relationship with an adult the losses an infant may suffer both affectively and cognitively may never be regained. [The importance of reciprocal inter- actions with an adult has also been particularly expressed, including evidence of a need for cuddling (Harlow 1966)]and 5 a need for environmental stiulation (Dennis and Najarian 1957). 20 Extending these findings to the maternal employment situation may be inappropriate, however. Depending upon mother's daily interactions with her child and substitute care, a child may not be deprived at all. Yarrow (1964) pointed out an important distinction--the difference between maternal deprivation and maternal separation. Margaret Mead (1954), too, has commented on the over empha— sis on the importance of the tie between the child and its biological mother and the unfounded insistence that any separation is damaging. "On the contrary cross cultural studies suggest that adjustment is most facilitated if the child is cared for by many warm friendly people." Ainsworth (1963), in a study of Ganda infants ranging from 0 to 24 months, reported that children cared for by multiple caregivers formed as secure an attachment to their mothers as children having a more exclusive relationship with their mothers. Additional evidence that the strength of attachment to the mother is unrelated both to maternal availability and to the number of caretakers comes from a study of mother attachment and stranger reactions of two- year old American children and kibbutz-reared Israeli chil- dren (Maccoby and Feldman 1972). Compared to the average American nuclear family, the kibbutz is characterized by lesser availability of the mother, who cares for her infant only the first six weeks after birth, and then slowly builds up to a full-time work within eight months. The mother spends 2-3 hours per day visiting her child at the child or 21 care center. A kibbutz child also usually experiences a great number of caretakers and more frequent changes of the chief caretaker. Despite these differences in child rear- ing, the American and kibbutz-reared children were found to be highly similar in their attachment reactions to their mothers and were equally likely to show distress over separation. Moore (1963, 1964, 1969) is one of the few maternal employment researchers to examine such potentially critical variables as the age of the child when first placed in sub- stitute care and the stability of this care. London chil- dren who had received substitute day care for at least one year before the age of five were compared at later ages with children who had been exclusively with their mothers before entering school. The groups were equated for sex, socio- economic status, birth order, maternal age, and amount of episodic (not daily) separation. Information was obtained from interviews with mothers and nursery school staffs and from tests and observations of the children. Few differ- ences were found at age six between children cared for by their mothers and those receiving stable substitute care though the latter was observed to be somewhat more self assertive (1963). At age eleven, boys who had received sub- stitute care were found to be more fearless and aggressive and less sensitive, fastidious, and conforming than boys receiving maternal care. Girls who had been in substitute care were less fastidious and more demanding of their 22 mothers' attention (1964). Moore concluded that children exposed to stable substitute care are no more disturbed than children who stay with their mothers (1969). Also demon- strated was the importance of examining results separately for boys and girls. Children in unstable substitute care showed more signs of insecurity than the other two groups. However, they also were from unstable families. The few children who were placed in substitute care under the age of two were more demanding of maternal attention at six than were those whose substitute care began between the ages of three and four (1969). This finding led Moore to conclude that substitute care under age three may be detrimental to the child. However, this conclusion is challenged by Caldwell and Smith (1970) who reported the results of a psy- chiatric examination of four-year olds attending day care. The psychiatrist, who did not know when the children had entered the day care program, found no differences in the adjustment of children whose attendance began before or after three years of age. Burchinal (1963) examined intelligence scores and school adjustment for a large sample of children in the 7th and 11th grades. Children whose mothers were employed when the child was three years or younger were compared with children whose mothers were never employed. Few statisti- cally significant results were obtained. Two conclusions can be drawn from the above results: (a) young children can form as strong an attachment to a 23 working parent as to a non—working one provided that the par- ent interacts frequently with the child during the times they are together in a stable family, and (b) stable stimu- lating substitute care arrangements are important for the normal personality and cognitive development of preschool children whose mothers work. The effects depend on the extent of mother's absence, the nature of the mother-child interaction, and the nature of the substitute care. Effects of Maternal Employment on Elementary School Children Direct studies are available on school-age children to a greater extent than for infants. More of the studies of elementary school children also avoid the methodological problems discussed in Chapter I. In general, many of the studies in the fifties and early sixties found few differ- ences related to mother's work status. Differences were again related to subgroups. Few studies were done during the seventies. Using observers unaware of mother's work status to rate kindergarten children from intact families, Siegel, et a1. (1963) found no significant difference in dependent and independent behaviors between children of working and non- working mothers. The samples were matched for sex, socio- economic status, number of siblings, and birth order. Working mothers had been employed for at least six months prior to the study, and non-working mothers had not been employed during the lifetime of the subject. The authors 24 concluded that maternal employment per se is not the over- whelmingly influential factor in children's lives that some have thought it to be. Post hoc inspection of the data sug— gests that the implications may be different for the two sexes, however. Had an analysis of variance testing this interaction been performed rather than twenty independent t tests, it is possible that more significant results might have been found. This study stresses the need for further research concerning maternal employment and sex typing in children, as in eight out of the ten behaviors studied, the differences were in the opposite direction for boys versus girls. This study underscores the necessity of analyzing data separately by sex of children which most studies after 1963 did. Hoffmann (1963), studying white intact families with children in the third to sixth grades, interviewed mother, got teachers' ratings, and had the children fill out a ques- tionnaire. She was interested in demonstrating differential effects of working mothers on children when subgrouping occurs and theory is used. Working mothers were classified as liking or disliking work, and she based her hypotheses on how guilt affects working mothers. The overall pattern of findings suggests that the working mother who likes working is relatively high on positive affect toward the child, uses mild discipline, and tends to avoid inconveniencing the child_with household.tasks; the child is relatively non- assertive and ineffective. The working mother who dislikes 25 working, on the other hand, seems less inyolved with the child altogether and obtains the child's help with tasks; the child is assertive and hostile. The results are consis- tent with the interpretation that the reason for the differ- ences is the presence or absence of guilt in the mother. The mother who likes work feels guilty so she asks less of her child and is more involved with her child to compensate. The study shows that maternal employment has a different effect on the mother-child relationship and on the child's behavior depending on whether or not the mother enjoys work— ing. The aim of this study was to demonstrate—that greater understanding of the effects of maternal employment will come about when working mothers are differentiated along a theoretically relevant dimension (liking work or not) and when theory is used. Further differentiating groups in a study on mothering adequacy and child adjustment, Yarrow (1961); Yarrow, et a1. (1962) divided both working and non-working mothers into those who were satisfied and those who were dissatisfied. Data on child rearing was obtained from mothers' reports in a sample of white intact families, matched for number, age and sex of children, mother's age; and education and to some degree socioeconomic status. Although employment was almost unrelated to childrearing patterns and to the summary meas- ure of adequacy of mothering (child rearing principles, limits of the child, sensitivity to the child's needs, warmth, emotion and satisfaction in the mother, child 26 relationships, etc.) the further breakdown according to sat— isfaction showed the dissatisfied non-working mothers to have the least adequate mothering score and least confidence in their child rearing. There were few differences between satisfied and dissatisfied working mothers, but satisfied working mothers showed slightly less adequate mothering than satisfied non-working mothers. Children of dissatisfied non-working mothers showed more problem behavior than chil- dren of dissatisfied working mothers, satisfied working mothers and satisfied non-working mothers. Regrouping by motives of family versus self fulfillment in working mothers and love versus duty in non-working mothers provided similar results. It is unclear from the write-up how much overlap there was between the satisfaction and motive categoriza- tion. This study again shows the importance of breaking down the maternally employed into subgroups, and it again demonstrates that one of the most relevant subgroups is mother's report of satisfaction. Woods (1972) researched children of maternal employment along another important subgroup variable-—supervision. She administered a series of psychological tests to fifth graders who reported that their mothers were employed (employment was undefined). The sample was largely lower- class blacks. When the sample was divided into those report- ing supervision (some sort of substitute care) and those without supervision, girls without supervision showed impov- erished cognitive and personal development. They also 27 perceived their mothers as less controlling and intrusive. Presence or absence of supervision was unrelated to boys' cognitive and personal development. Other differences between mothers who would leave children with or without supervision might have been influential in mediating the poorer condition of unsupervised girls, so it is difficult to determine the extent to which supervision per se is responsible for differences. No indication as to whether these children came from intact families was made, and this is also probably a relevant variable. This study focuses on an important but previously neglected aspect of maternal employment, the supervision of school-age children whose mothers work. Woods also found that positive maternal atti- tudes are associated with favorable adjustment of children, which is also in line with the Hoffman and Yarrow findings reported above--that maternal satisfaction is an important variable in child adjustment regardless of working status. Birnbaum (1971) did a dissertation comparing profes- sionally employed mothers with mothers who had graduated from college with distinction but had become full-time home- makers--women who could have pursued professional careers had they chosen to do so. Both groups were fifteen to twenty years past their bachelor's degrees at the time they were interviewed. The professional women were clearly higher in morale. The non-working mothers had lower self esteem, a lower sense of personal competence--even with respect to child care skills--felt less attractive, 28 expressed more concern over identity issues, and indicated greater feelings of loneliness. The non-working mothers were more insecure and unhappy in these respects than pro— fessional women who had never married. Asked what they felt was missing from their lives, the predominant answer from the two groups of professional women was time, but for the housewives, it was challenge and creative involvement. With respect to these women's orientation to children, the full- time homemakers felt that having children changed a woman's life by the sacrifices motherhood demanded. The profes— sional women more often stressed that their lives were changed by children in terms of enrichment and self- fulfillment. The non-working mothers also indicated more anxiety about their children. The professional women responded positively to the growing independence of their children while the homemakers indicated ambivalence, regret, and seemed to be concerned about loss of familiar patterns or their own importance. Although there is no direct data on the children themselves, the pattern of able educated, full-time homemakers suggests that they would have short- comings as mothers when the child needs a parent who can encourage independence and instill self—confidence. The anxieties and concerns of these women and their own frustra— tions might operate as a handicap in their mothering. Kilgler (1954) found that women who worked because of interest in the job were more likely than were those who 29 worked for financial reasons to feel that there was improve- ment in the child's behavior as a result of employment. The above findings and speculations again suggest the importance of looking at mother's satisfaction as well as motivations for working as important factors in mother's effect on her children's adjustment. Powell (1961, 1963) sent questionnaires to white urban middle-class mothers with nursery, kindergarten, second, sixth, ninth, or tenth-grade children. Employment was defined as working at least sixteen hours per week (which confounds part-time and full-time work). No differences were obtained between working and non-working mothers on attitude towards child rearing as measured by a portion of Wiley's Child Guidance Scale. However, this study is sub- ject to the problem of defensive self report and lack of breakdown according to relevant variables. The following four studies present somewhat mixed results regarding effects of maternal employment on school age children's adjustment. Schreiner (1963) conducted a four—year longitudinal study of thirty-five German children starting at age six. Sons of working and non-working mothers (not defined) did not differ on any behavioral measure, although boys whose mothers gave up employment became less active and less emo- tionally responsive. Girls with working mothers were more "differentiated" in the third year. Schreiner concluded 30 that these few differences were temporarily limited and reversible. In a study on the effects of maternal employment on children from rural areas, Nolan (1963) found no significant differences in academic achievement and peer acceptance as rated by their teachers. The population was composed of six to eleven year olds with either employed or homemaker mothers. Social class was not controlled, and data were not reported separately for each sex. Dits and Cambier (1966), on the other hand, obtained evidence that sons of working mothers are more poorly adjusted. They studied six to eight year old Belgian boys from working-class, intact families with two to four chil— dren. Twenty boys had mothers who had worked for at least two years while twenty controls had non-working mothers; all boys were supervised after school. The two groups differed in their responses to projective tests. While male figure drawings were equally differentiated, female figures were drawn more poorly by sons of employed women, suggesting problems of contact with the mother. In a sentence comple- tion task, working mother boys displayed more hostility and perceived mothers as more physically punitive. On the Children's Apperception Test, they were likely to choose adults or adult-child groups. Dits and Cambier interpret this finding as reflecting a greater isolation and independence from adults as well as a reduction of family life among sons of employed mothers from the working class. 31 Contradictory results were reached by Rabin (1965), who administered projective tests to ten-year olds reared either in the kibbutz (all with working mothers) or the moshav (where mothers usually remain at home). He concluded that the kibbutz children showed greater ego strength, more maturity, and better emotional adjustment. Since these studies were all done in different coun- tries, their differing results may be accounted for by cul- tural factors and may not be applicable to the US. However, each is important for different reasons—-the German study is large and longitudinal (one of the few longitudinal studies addressing this problem), the Belgian study is comprehensive and intensive in its testing techniques, the Israeli study examines the unique child-care arrangements of the kibbutz, and the rural study involves ratings of child adjustment outside of parental self report. For the present study the above projects again demon- strate sex effects and further indicate the importance of examining cultural and social factors in explaining results and interpreting data. A recent trend in maternal employment research has been to begin to examine effects of maternal employment on sex role ideology. Though most research has been focused on adolescent, college aged, and adult populations, a few studies have been done with elementary school—aged children. One main hypothesis about the effects of maternal employment on children pertains to changes in sex role 32 behavior. Because the mother is employed she, and possibly her husband, provide a different model of behavior for the children in the family. Children learn sex role behavior largely from their parents. To the extent that a different role is carried out by the working mother than by the non- working mother, the child has a different conception of what the female role is. The self concept of girls would be par- ticularly affected (Hoffman 1974). This theory also has implications for the expected differences in perceptions of boys and girls of their mothers and fathers in working mother or non-working mother families in the present study. Because of role changes, children would perceive differences in mother's and father's caregiving behavior in working mother families compared to non-working mother families most likely in terms of nurturance and punishing behaviors. The children of working mother families may tend to see mother as less nurturant and father as less punishing when compared to children from non-working mother families. There may be differences from the traditional stereotyped perceptions of mother as nurturer and father as disciplinarian. Hartley (1961) observed that one experience common to all children of working mothers is that they are "exposed to a female parent who implements a social role not implemented by the female parents of other children." Since the child learns sex roles from the observation of his parents, mater- nal employment influences his concept of the female role. More importantly, since one of the earliest statuses 33 assigned to the child is that of gender, maternal employ- ment presumably affects the female child's concept of her- self and the behavior expected of her. Some data on elementary school-aged children supports this theory (Hoffman 1964). Hartley (1961) found that elementary school daughters of working mothers compared with daughters of non-working mothers were more likely to say that both men and women typically engage in a wide variety of specified adult activ- ities, ranging from using a sewing machine to using a gun and from selecting home furnishings to climbing mountains. They saw women as less restricted to their homes and more active in the world. The daughters of working mothers reported more liking and less disliking of all activities—- household work and recreation. Daughters of working mothers also viewed work as something they will want to do when they are mothers (Hartley 1960). Dolores Gold and David Andres (1978) tested hypotheses specifying differential relations between maternal employ- ment and children's development, dependent on the sex of the child and the socioeconomic status of the family. An exam— ination was made of the sex role concepts, personality adjustment and academic achievement of 223 ten-year old girls and boys with either full-time employed or non- employed mothers from working-class or middle-class fami- lies. The data provided some support for the hypotheses. Children with employed mothers had the most egalitarian sex ”I . I / I :3 4 role concepts; however, this appears primarily related to their mothers' greater satisfaction with their roles. In general, the employed mothers are very much more content with their roles than the non-employed mothers, and it appears that the employed mothers who are most content with their roles have children with most egalitarian sex role concepts. Maternal employment status was partly related to the adjustment of the children. Middle-class boys with employed mothers had lower scores on language and mathemat- ics achievement tests than the other middle-class children. Employed mothers and their husbands reported more similar behavior patterns within the home and attitudes that dif- fered from those reported by non-employed mothers and their husbands. In general, when mothers were employed, the par- ents reported that they behaved more similarly in household functions and child supervision. In addition, the parents reported being more content with the wife's role and having more feminist attitudes. The authors concluded that mater- nal employment is heterogeneous in its influence on the development of children and that its effects are mediated by a number of factors and vary with the type of child behavior examined. Marantz and Mansfield (1977) found maternal work status and child's ages were equal predictors of the girls' sex typing of activities. Work status was the stronger predic- tor of sex appropriateness of career aspirations. Stereo- typing was maximally influenced by maternal work status at 35 ages 7—8; 5-6 year olds showed rigid stereotyping and 9-11 year olds little stereotyping regardless of lifestyle modeled by their mothers. Father's participation in house- hold chores, while associated with maternal employment, had no independent effect on daughter's sex role stereotyping. It was suggested that this may be different for sons. They also found that the more staisfied the daughter, the less her stereotyping. The age difference with regard to stereo— typing suggests that observational learning is mediated by the cognitive maturity of the observer. The child must be cognitively ready to attend to relevant behaviors and then to integrate what she observes if the model is to have a maximal impact on her perception. To summarize, these studies indicate that the adjust- ment of elementary school children generally is not adversely affected by their mother's working.] While one study reported a negative association between maternal employment and adjustment for boys, two others found a posi- tive association and one found no relationship. Discrepan— cies are difficult to interpret since the studies differ in several methodological respects, not the least of which is the diversity of cultural settings in which the data were collected: Germany, Belgium, Israel, a black US ghetto, and a rural US setting. The impact of maternal employment no doubt depends in part on cultural attitudes toward a mother's working. Consistent findings, however, from studies in both black and white settings are that data must 36 be analyzed separately by sex and that satisfied mothers whether working or not have the best adjusted children. Research on fathers in working mother families and its relationship to child adjustment was virtually nonexistent. Research on sex role conceptions and maternal employ- ment shows maternal employment has a definite impact by broadening both boys' and girls' conceptions of what is male and female. It seems that these results are again mediated by maternal satisfaction and also by cognitive development. Effects of Maternal Employment on Adolescents The adjustment of adolescents has been studied more than the adjustment of younger children, probably due to the availability of subjects as well as convenience in data col- lecting--using self report inventories administered to large groups of older children. Using the number of problems checked by white high school girls (aged 16-18) from intact families as an index of their adjustment, Whitmarsh (1965) found that daughters of employed mothers were better adjusted than daughters of homemakers; when the sample was further divided according to socioeconomic status, the finding was especially true for middle-class daughters but held in the lower class as well. However, Whitmarsh may have had a biased sample since only girls taking home economics were included. Nelson's study (1968) of 312 ninth graders from intact middle-class homes 37 found that boys whose mothers worked full time were better adjusted than those whose mothers worked part time or not at all as measured by the Minnesota Counseling Inventory. For girls, the pattern was less consistent, but the girls with non—working mothers tended to be better adjusted than those whose mother worked part or full time. Thus, Whitmarsh's findings were not supported, and in fact the reverse was found. This may be accounted for by different adjustment measures and biased sampling. Douvan (1963) observed that lower-class boys whose ,/’ mothers worked full time were somewhat more rebellious, less active and had poorer ego integration than lower-class sons of women employed part time and middle;class sons of women working full or part time. Douvan concluded that full-time maternal employment as such was not responsible for the poorer adjustment of lower-class boys,but rather that such employment indicated that the father was an inadequate pro- vider and hence an ineffective model for his son. Douvan also concluded that middle-class adolescent girls adjust better to full-time maternal employment than do lower-class girls. Adolescent daughters of working mothers were more likely than daughters of non-working mothers to name their mother as the person they most admired. Even when the lower-class father is an adequate pro- vider, however, the fact of the mother's working may dimin- ish the son's conceptions of his father's status. This was suggested in a comparison of lower-class boys of working and 38 non-working mothers (McCord, et a1. 1963) who did not differ with respect to father's occupation, employment history or general stability. The subjects were 149 boys from intact families who had been observed for five years in the 1930's when they were between the ages of ten and fifteen. Mothers of forty-eight boys had been employed part time or full time for at least six months during the first two years of obser- vation; the remaining boys had non-working mothers. Mater- nal employment was not related to mother-son or peer relationships but was associated with decreases both in sib- ling rivalry and in sons' approval of their fathers. Some effects of the mother's working status were found to depend on the stability of the family. (Instability was defined in terms of parental arguments, parental deviance, and lack of maternal warmth.) Maternal employment was related to sons' increased sexual anxiety in stable homes and to increased criminality and dependency in unstable homes.’ McCord hypothesized that in an unstable home maternal employment may be interpreted by the son as rejection whereas in a stable home it equalizes status between the sexes, thus increasing the difficulty of sex role adjustment and decreasing the son's perception of his father as a signifi- cant figure. Since maternal employment is now much more prevalent than when this study was done, these findings may no longer prevail. However, two more recent Canadian studies do report the same pattern; Kappel and Lambert (1972) in their study of nine to sixteen year olds found 39 that the sons of full-time working mothers in the lower class evaluated their fathers lower than did the sons of other full-time working mothers and lower than did the sons of part-time or non-working mothers in any class. Propper (1972) found in a predominantly working—class sample that the adolescent sons of full-time working mothers were less likely than were the sons of non—working mothers to name their father as the man they most admired. In a study by Vogel, et al. (1970), sex role percep- tions of male and female college students were studied and their relationship to their mother's employment. Sex role perceptions were measured by having subjects describe the typical adult male and female by checking a point along a continuum between two bi-polar descriptions. Previous work with this scale indicated which descriptions were more typi- cally assigned to each sex and also which were seen as posi- tive or negative. In general, the positively valued stereotypes about males included items that reflected effectiveness and competence; the highly valued female asso- ciated items described warmth and expressiveness. Both male and female students with employed mothers perceived signifi- cantly smaller differences between men and women, both the women being more affected by maternal employment than were the men. Furthermore, the effect of maternal employment was to raise the estimation of one's own sex--that is, each sex added positive traits usually associated with the opposite sex: daughters of working mothers saw women as competent 40 and effective while sons of working mothers saw men as warm and expressive. So the father, at least among middle-class males whose wives work, may be seen as a nurturant figure possibly because of his taking over some of the child care role. V Several investigators of adolescent adjustment to maternal employment have not analyzed their data separately for boys and girls. These studies have found few differ— ences related to maternal employment, though Nye (1963) did find delinquency to be more common, if mother worked, among middle-class children living in intact homes. Nye used a self report measure of delinquency. This finding was also supported by Gold (1961), who used police contact as a meas- ure. In both studies, the relationship was obtained for the middle class and did not hold for the lower class. Glueck and Glueck (1957), in their study of lower-class boys, found that regularly employed mothers were no more likely to have delinquent sons than were non-employed mothers. Sons of the occasionally employed women in their study were found to be more delinquent, but the group as a whole was clearly more unstable than the groups where mother worked regularly or not at all. In summary,the mother's working generally appears to have no harmful effects on adolescent adjustment, although lower-class boys and girls may experience some difficufties in adjusting to full-time maternal employment. The effects of maternal employment are differentiated between adolescent 41 boys and girls, especially as regards boys' perceptions of fathers in lower-class working mother families. So maternal employment seems to affect adolescents' conceptions of men, father and his role, as well as of women, mother and her role. Again, the importance of analyzing data separately by sex of child is stressed, and the importance of social class differences becomes evident. Summary From all of the above studies on the effects of mater- nal employment on children, it appears that the effect of mother's absence is dependent on many factors. These include substitute child care, age of child, sex of child, social class of family, mother's satisfaction, quality and quantity of mother-child interaction, and social custom. These results provided research underpinnings in setting up the design of the present study. They also provided impor- tant theoretical considerations and speculations in discuss- ing the results of this study. Theory of Perceptual Psychology The theory of perceptual psychology grew out of human- istic psychology and the phenomenological approach. This maintains that a person's experiencing of his environment is most meaningful to his personality development. Behavior is understood from the point of View of the behaver. The per— ceptions, then, of the behaver become a frame of reference 42 for studying "whole" people, their behavior, and their development. Perceptions from this perceptual framework are defined as "any differentiation a person is capable of mak- ing in his perceptual field whether or not an objectively observable stimulus is present" (Combs 1976). The process of making differentiations in the perceptual field is con- sidered similar for the perceptions of seeing, hearing, smelling, feeling as well as those made by conceiving, know- ing, and understanding. Each involves some measure of per- sonal meaning on the part of the perceiver. Taguiri more specifically defines person perceptions as the process by which man comes to know and think about characteristics, qualities, and inner states (Combs 1976; Taguiri 1974). In the same vein, perceptions will differ between per- sons in the same situation. Each individual will interact or respond to the situation in terms of what it means to him. Behavior is in turn determined not necessarily by objective facts but by the perceiver's own perceptions. Factors effective in determining behavior of an individual are only those experienced by the individual at the time of his behavior. The concept of complete determination of behavior by perceptions or the entire field of the percep- tions called the "perceptual field" is the basic postulate of this approach. Combs states that "all behavior without exception is completely determined by and pertinent to the perceptual field of the behaving organism" (Combs 1976). 43 Since no two people ever have identical perceptual fields, no two people can have identical meanings. Subse- quently, no two people can have identical behavior. However, people who have similar experiences tend to have common characteristics in their phenomenal fields and thus see things the same way. As a result, they show common ten— dencies in their behavior. Many factors go into determining the perceptual field and are identified by different theorists in different ways. It is important to know what some of these are since all behavior is determined by the perceptual field, and to change behavior,it is necessary to change the perceptual field. There are many factors controlling and limiting the process of perceiving and the functioning of the perceptual field. Combs categorizes these variables into seven groups: (1) physical organism, (2) need, (3) time, (4) opportunity, (5) phenomenal self, (6) goals and values, and (7) organiza- tion of the field. Taguiri discusses some of these factors as age, developmental trends, time, sex, roles, dyadic interplay, cognitive processes, and affective processes. Others have postulated an interpersonal perceptual style (Shrauger 1964). For the purpose of this study, it is assumed that a number of these factors may be altered in working mother homes as compared to non-working mother homes (time spent, roles, child rearing behaviors) which would in turn alter children's perceptions by altering the perceptual field. It is then assumed that there would be 44 differences in behavior between children from working mother homes and those from non-working mother homes. The implica- tions of these differing perceptions leading to differing behaviors are important as well in interpersonal relations because of dyadic interplay. This is the process whereby perceptions of parents' behavior affect children's behavior which in turn affects parents' perceptions and subsequent behavior. To further extend the implications of children's perceptions of their mother's and father's caregiving behav- iors, Jersild (1968) also notes that perceptions of mother and father retained as the child grows older are likely to influence perceptions of others who may be mother and father figures. So children's perceptions as mediating factors between what parents do and how children perceive and pro- cess what they do have implications for behavior and inter- personal relationships both present and future. Children's Perceptions of Parental Behavior Recognition of the importance of children's perceptions of parental behavior is fairly recent. Most of the research on personality development has focused on the effects of child training methods. Assumptions about what the parents reported they did or were observed to do and subsequent per- sonality factors in children were made. Only a few studies have paid attention to children's perceptions of parental behavior or children's feelings about and reaction to par- ents' child rearing methods. The theory of perceptual 45 psychology provides a framework for studying this mediating process between parental behavior and child response. Research in this area has made a connection between chil- dren's perceptions of parental behavior and child adjustment as well. One of the first studies taking this position was done in 1954 by Ausubel, et al. This analysis of perceived par- ental attitudes and children's ego structure led to the con- clusion that although parental behavior is an objective event in the real world, it affects the child's ego develop- ment only to the extent that and in the form in which the child perceives it. Hence, the child's perception of paren- tal behavior is in reality a more direct and relevant deter- minant of personality than the actual stimulus content to which it refers. This study also pointed out that it is less relevant to establish the nature of the actual environ- ment to which the individual is exposed than to ascertain the distinguishing features of his perceived world. The researchers also assumed that children's perceptions of par- ent behavior and attitudes could be measured more validly than the observed behavior itself or parents' report of it. It was assumed that both interview and observation would be contaminated by the parents' understandably strong motiva- tion to perceive their role behavior in a favorable light and to similarly impress others. Though this was true to a certain extent, difficulty in measuring perceptions validly and reliably because of the lack of an instrument with 46 statistical properties was encountered. The researchers ruled out using projectives because of reliability problems and ended up analyzing the data from a parent attitude rat- ing scale which they created. Serot and Teevan (1961) formulated and found signifi- cance for three hypotheses concerning parents' and chil- dren's perception of the parent-child relationship and its relation to child adjustment. These findings were that: (a) a child's adjustment is related to his perception of his relationship with his family; the well adjusted child will perceive his parent-child relationship as relatively happy and close to the theoretical ideal, whereas the mal— adjusted child's perception of his relationship will be far from ideal, (b) the child's perception of the relationship is unrelated to his parents' perception of the same, and (c) the parents' perception of the relationship is unrelated to their offspring's adjustment. This study established the fact that a child's view may be very different from what the parents have in mind and may be a more valid predictor of his/her adjustment than parents' perceptions. Jersild (1968) stated that if parents had the courage and interest to inquire and children the freedom to respond candidly, most parents would probably be surprised by the pictures their children paint of them. These studies began to speak to the significant portion of the variance of child behavior not related to observer-reported parental management and 47 rearing practices but related to the child's phenomenologi- cal perceptions of persons and situations (Goldin 1969). As previously mentioned, one of the problems of meas- uring children's perceptions has been the need for accurate and consistent instrumentation. In a review of the litera- ture on children's report of parents' behavior, Goldin (1969) reports that though there are various instruments used in measuring children's perceptions (most commonly Schaefer's or Bronfenbrenner's scales) there is consistency in reporting of three main behaviors in the universe of children's reports of parental behavior; these are affec- tion, dominance, and punishment. Contemporary factor analy- tic work has supported the validity of these variables and extended their understanding (Schaefer 1956; Seigelman 1965). Both of these investigators evolved systems in which these three orthogonal factors of children's reports of par- ent behavior explained the reliable variance. By using these variables--loving, punishing, and controlling--as a framework for comparing studies on children's perceptions of parental caregiving behaviors, understanding of trends and finding commonalities in results has been facilitated. Studies have also found correlations between negative social and psychological adjustment on the one hand and both (a) adolescents' perceptions of their parents as rejecting and (b) adolescents' perceptions of their parents' view of them as negative (Dubin and Dubin 1965). In addition, Goldin (1969) found delinquents perceiving parents as more 48 rejecting, neglecting, less loving, lower in authority, and lax in parental discipline when compared to "normals." In summarizing a number of studies using a variety of tech- niques to identify maladjusted normals, Goldin reports mal— adjusted normals as perceiving parents as less accepting, unloving, and more rejecting than do matched controls. There also was a tendency for maladjusted normals to report parents as excessively psychologically controlling, exces- sively disciplining, dominant, nagging, and restrictive. In child guidance clinic patients, correlations were also found between acting out symptomatology and child perceptions of parents as undercontrolling and indulgent, and withdrawn symptomatology and child perceptions of parents as insuffi- ciently loving and highly controlling. In summary, percep- tion of parents is clearly related to behavior in several ways for these deviant groups of children. All groups are more likely to report parents as rejecting. While delin- quents perceive parents as undercontrolling, maladjusted normals and clinic children are more apt to report parents as overcontrolling. It is apparent, however, that greater homogeneity exists in the latter two groups. Differences between them require further exploration with greater speci- fication of subgroups within these two populations. It would seem useful to determine whether the groups are at all differentiable along the acceptance dimension in terms of, for example, extent of rejection. It might be hypothesized 49 as well that these groups differ in the situations they report as leading to parental rejection. Extending research in children's perceptions of paren- tal caregiving behavior and its relationship to child adjustment, Stollak, Michaels and Messe (1977) examined a normal population of school-aged children and included par- ents' perceptions of caregiving behavior as well. Their sample included eighty seven—year old children and their parents who completed Bronfenbrenner's questionnaire. Chil- dren filled out the inventory for their perceptions of their parents' caregiving behavior; parents rated their percep- tions of their own caregiving behaviors and then filled out the questionnaire as they thought their children would. Ratings of children's social behavior were made from video- tapes of structured interactions with an undergraduate col- lege student in a playroom setting. Three separate analyses were done. The first found significant differences between parents' self perceptions and children's perceptions along the dimensions of loving, punishing, and demanding caregiv- ing behaviors and on the same dimensions between each of these measures and parents' inferences about how their child perceived them. In the second analysis, each perception measure--1oving, punishing, and demanding--was found to be independently related to adaptive and non-adaptive child behavior. The third finding demonstrated that the accuracy of parent inferences and similarity between parents' self perceptions and children's perceptions were found to be 50 independently related to adaptive and non-adaptive child behavior in the playroom. Conclusions stress the complexity and importance of person perception processes in family relationships and for the child's adjustment in a normal population. The study also demonstrated the independent relationship of each of the original and derived perception measures to social functioning of the child. All five per— ception measures and each composite measure within a measure were found to be related to some of the playroom behavior categories. Also, confirmation of the hypotheses from the three experiments can be used to support predictive validity for the perception measures studied. This is important since a major problem in studying the children's perceptions has been validity of the measures used. A modification of the Bronfenbrenner scale will be used in the present study. Besides the establishment of a relationship between child behavior and adjustment and children's perceptions of parental behavior, several other findings are supported in the literature. These have implications for the present study in examining sex of children effects, sex of parent effects, age effects, and socioeconomic status effects. In general, there is support for the hypothesis that children perceive both parents favorably but perceive»mothers more in terms of loving and father more in terms of punish- ing, with indication that either parent may be seen in terms of controlling. That mothers are classified as accepting received general support from Harris and Tseng (1957) and 51 from DuValle (1937), who reported more positive attitudes were expressed about mothers than fathers. Droppleman and Schaefer (1963) also reported that mother is perceived as more loving, less ignoring and neglecting than father. Clear differences between mother and father as to puni— tiveness emerge with children judging fathers as more puni— tive (Kagan, et a1. 1961). The results are conflicting as regards children's reports of controlling parental behavior. In some studies, fathers are seen as more demanding and powerful (Kagan 1961), but others report mother as more powerful and dominant (Morgan and Gaier 1956). These studies did not consider age and sex of child differences which may explain the contradictions and the lack of clear relationships. At the behavioral level, Emmerich (1959) found that mother was perceived as more facilitating of the child's action than father, while father was viewed as interfering more than mother. Dubin and Dubin (1965) explain this find- ing in terms of a model on continuity/discontinuity of interaction between child and parent. The usual more or less continuous interaction between mother and child from birth leads the child to perceive her as facilitating. The episodic and disjointed interaction between father and child could lead to the child's perception that when father is encountered he usually interferes with activity. The child's differential perceptions might be explained by the continuity or discontinuity of his interactions with them. 52 If this is the case, then there may be different perceptions in working mother families where the pattern of mother being the continuous caregiver and father being the discontinuous caregiver are altered. If father takes part in more child care and domestic functions as reported in the literature, then children's perceptions may be different from this sup- ported finding in working mother homes. It is plausible to predict that this difference of perceiving mother as more loving and father as more punishing may not exist, and par- ents may be perceived as more equally loving and punishing or facilitating and interfering. When the sex of the child is considered, there is gen- eral support for the following trends in children's report of parental behavior. Boys perceive parents as less accept- ing and loving, more psychologically controlling, and more punitive than do girls. In a study by Droppleman and Schaefer (1963), boys saw fathers as less accepting than did girls in that they are seen by boys as lower in love, affec- tion, and nurturance. As regards psychological control, results are similar. Boys compared to girls reported mothers as using both more overt and covert controls. As regards reported controlling behavior in father, boys when compared to girls also saw father as using more overt and covert controls. These findings are supported by Siegelman (1965). An additional finding from Siegelman is that boys note greater physical punishment by mother than do girls and more physical punishment and isolation from father as 53 compared to girls' reports. These studies clearly indicate that sex of child affects perceptions. The generalized picture is that girls tend to be more favorably oriented than boys toward parents. Speculation as to why these dif- ferences occur includes possible different child rearing techniques used by parents for girls than boys. Or perhaps there is some psychological reason for girls being more favorably oriented towards adults. Again, this finding may be different in working mother families because of differ- ence in child rearing patterns so that there may not be the split between boys and girls. Age effects are also supported with the general conclu- sion being that with increasing age children's perceptions of adults become more realistic--meaning that they corres— pond more accurately with objective characteristics of par- ents. They also become grounded in more subtle perceptual characteristics. However, mother still persists as the pre- ferred parent (Dubin and Dubin 1965). These effects are also complicated by the interaction between age and sex, though Goldin (1969) reports in general a lessened feeling of acceptance during adolescence than during early child- hood. Theoretically, perceptions would change and become more subtle and abstract and less concrete as children move from the concrete to operational levels of cognition (Piaget 1950). In research regarding the effect of socioeconomic status on children's perceptions of parental behavior, it 54 appears that children of lower socioeconomic status per- ceive parents as less accepting and possibly as more psycho- logically controlling. Siegelman (1965) found social class was positively correlated with love. It has also been found that middle-class boys perceive parents as more accepting and interested (Rosen 1964) and that children of laborers, compared to children of professionals, report more family disharmony (Hayward 1935). However, no recent study report- ing this finding was found. The results are suggestive of a relationship between children's report of psychological con- trol and class whereby parents in lower-class homes are seen as overcontrolling. However, studies are somewhat outdated. No clear socioeconomic status differences can be drawn regarding children's report of parental behavior and charac- teristics. In summary, then, children's perceptions of parents are related to child behavior and adjustment, sex of child, age of child, and possibly socioeconomic status of child. These findings provide a theoretical justification for the design of the present study. Children's Perceptions of Parental Behavior and Maternal Employment In general, maternal employment has been found to be largely unrelated to children's perceptions of maternal behavior. However, the studies in this area have confounded maternal working status (combining full and part-time work- ing mothers) and have not clearly defined working mothers. 55 These studies also failed to subgroup along theoretically relevant dimensions. Problems with instrument validity and reliability have also affected outcomes. The research that has studied effects of maternal employment on children's perceptions of both parents has suggested that there may be less favorable perceptions of fathers when mother works among lower-class boys. However, these studies were not studying perceptions per se. Finkelman (1967), from the results of questionnaires administered to ninety-six fifth and sixth-grade children, found no differences in perceived nurturance or authority of parents as a function of mother's working status (undefined), sex of child, or social class. Though he examined chil- dren's perceptions of fathers, he did not find any differ- ences as a function of mother's work status and class as expected. Franke (1972) again found no significant differ- ences in the child's perceptions of mother's behavior which could be attributed to mother's undefined working status, though she found one significant interaction between maternal employment and sex of child: punishment by working mothers was perceived as more severe by girls than by boys whereas punishment by non-working mothers was perceived as more severe by boys than by girls. She concluded that work- ing mothers and non-working mothers relate to their children in the same ways. She did not account for the above finding theoretically. 56 Woods (1972) administered a modified parent behavior inventory to black lower-class ten-year old children of working mothers. Mothers who were employed full time and part time were perceived similarly on all variables but one: mothers employed full time were more consistent in their discipline. Coupled with the above finding regarding dif— ferences in perceived punishment as a function of mother's working status suggests a possible trend. Perhaps working mothers do change their disciplinary function and technique as compared to non-working mothers. In Propper's study (1972) of lower-class high school students, she reports that incidences of parent-child dis- agreement over a wide range of issues were more common among both boys and girls when the mother was employed. However, perceptions of parental interest, help with school and personal problems, and degree of closeness to parents were generally unrelated to mother's employment status for children of both sexes. The one exception to this finding was that sons of working women were less likely to choose their fathers as the man they most admired than were sons of non—working women. Other studies of lower-class boys have produced strikingly similar results. Douvan (1963) reported that sons of full-time working women less frequently chose their fathers as an adult ideal while McCord et a1. (1963) found that sons of working mothers showed more disapproval of their fathers. 57 It can be speculated, then, that when children's per- ceptions of parental caregiving behaviors are specifically studied with methodological improvements in subgrouping and instrumentation, differences in working mother families ver- sus non-working mother families may be detected as a func- tion of sex of child, class, and sex of parent. These differences are hypothesized because of the changes in 2233 mother's and father's behavior as a result of mother's work- ing. It is assumed that these role and behavior changes alter the traditional model of mother's being the continuous and most consistent caregiver to a more shared model. These behavior changes alter the child's perceptual field and thus his/her perceptions. It is speculated that in working mother families there may not be as great a difference between the perception of mother as the more loving parent and father as the more punishing. Parents in working mother families may be viewed by their children as more similar in caregiving behaviors than in non—working mother families. Research already indicates mother may be seen as more pun- ishing when she works and father as more nurturing when mother works when comparing perceptions of children from working mother families and non-working mother families. It is speculated that changes in perceptions may be more pronounced for boys' perceptions of fathers. However, in a recent study, these findings were not supported. Rosenthal and Hansen (1981) assessed the impact of maternal employment on young adolescents' self concept, school achievement, 58 vocational development, and the perceptions of their par- ents. No significant differences in the subjects' percep— tions of either parent were found. Subjects were seventh, eighth, and ninth-grade students from New York from both two-parent and one-parent homes. Perceptions were measured by the parent-child Relations Questionnaire. Categories of perceived parental behavior characteristics were loving, rejecting, casual, demanding of attention. Mother's work status was undefined. They concluded that even in nonintact homes in which one parent is absent, mother's employment had no significant impact on children's self concept, school achievement, occupational aspirations, emotional maturity, or their perceptions of their parents. They speculated that data in their study did not support previously held conclu- sions that: (a) children with working mothers view their fathers with more disapproval than children from non-working mother families (McCord, et al. 1963), and (b) children with working mothers are either emotionally or cognitively deprived or perceive her absence as rejection (Hoffman and Nye 1974). They did not account for social desirability factors, socioeconomic status, instrumentation, and did not specify their definition of working mother versus non- working mother. Summary To summarize, the "deprivation" model concluding that mother's absence may be rejecting and harmful does not hold 59 up under the research conditions presented here where mother's absence is defined as working. A perceptual model focusing on what a child perceives about his/her parents rather than on what a parent does may add more information and a new dimension to this area of maternal absence through work. Very few studies have demonstrated significant dif- ferences in children's perceptions of parents as a function of mother's work status. Those differences that were dis- cusSed were inferences about changes in boys' and girls' perceptions of fathers along the punishing/nurturing dimen- sion. Most of these studies were undertaken ten years ago with methodological problems. The present study, using improved methodology with the perceptual model as a theoret- ical base, attempts to update and reexamine the results from previous studies. CHAPTER III METHODOLOGY The purpose of this research was to explore children's perceptions of their parents' caregiving behavior as a func— tion of family work status (working mother family or non- working mother family), sex of child (male or female), sex of parent (mother or father), and role satisfaction of par- ents (satisfied or dissatisfied with mother's role). This area has been relatively unexplored or explored, but with methodological weaknesses. The following research design and analysis was created to more adequately research this area. This chapter is subdivided into nine sections cover- ing the following topics: design of the research, popula- tion, description of the sample, procedures for selection of the sample, data collection, training of interviewers, instrumentation, statistical hypotheses, and methods of data analysis. Design The design of the study is causal comparative as defined by Borg (1977). This design is aimed at the discov- ery of "possible causes for a behavior pattern by comparing Ss in whom this pattern is present with similar 85 in whom 60 61 it is absent." It is also called ex post facto research, since causes are studied after they have exerted their effect on another variable. This design is chosen as the nature of the relationships to be studied does not permit experimental manipulation. The observed effects (children's perceptions) are present in this study. The antecedents of these effects will try to be determined. The limitation of this design is that causality cannot be determined, rather only that a relationship exists between variables. This study was a 2x2 (2x5) design with the following three independent variables: sex of child (male or female), family work status (working mother family or non-working mother family), and sex of parent (mother or father). Role satisfaction of mother and father's satisfaction with mother's role could not be used as independent variables as not enough parents endorsed being dissatisfied. Therefore, all parents in this study endorsed being satisfied: mothers with their roles and fathers with their spouses' roles. These variables were chosen because of suggestions in the literature that they may be related to children's percep— tions and may account for differential effects of working mothers on children. Role satisfaction of mother and work- ing status of mother were both determined from the Mother's Questionnaire described below. Father's satisfaction with mother's work status was determined from the Father's Ques- tionnaire. 62 Three dependent variables were obtained from children's perceptions of their parents' caregiving behavior as meas— ured by the Cornell Parent Behavior Inventory (PBI), also described below. The three dependent variables are: loving, controlling, and punishing parent behaviors. Each factor has been determined to be independent through factor analy- sis and is to account for twenty-eight percent of the vari- ance (Rogers 1966). These variables are used repeatedly in the literature as the three main categories of children's perceptions of parental caregiving behavior (Goldin 1969). The 55 were divided into four groups according to the independent variables with between eleven and twenty—one Ss in each group. The four groups were male child working mother families, female child working mother families, male child non-working mother families, and female child non- working mother families. Borg suggests a minimum of fifteen Ss in each cell to be compared. Because of the possibility of not being able to control certain variables, e.g., class and educational level of parents, and because only small differences on the dependent variables were expected, twenty 55 per group rather than fifteen 55 per group was the N desired. The female working mother family group was the only group that did not approximate or surpass this number. It had eleven families in it. Raising the N in this manner also increased confidence in any findings and increased the likelihood of detecting subtle differences. A model of the design is presented in Figure 3.1. 63 .Amusmmms cmpmmaomv uofl>mnmn mcfl>figmumo m.uonumw 0cm m.umcuos mo mcowumoouoa m.:oupaflno am am Ha Ha mamHo om om ma ma msom mHOEHOZ whmfiflmm WHOLflOZ thr—HMM— emammaumm omammaumm nmammaumm emammaumm mmHaHzaa mmmeoz oszmozizoz mmHaHzaa mmmaoz oszmoz .Hmpoz cgflmma mmusmmoz poummaom Amxmv Nxm .H.m ousoflm 64 Population The population consisted of students in the third, fourth, and fifth grades of the East Lansing, Michigan, public school systems and their parents and third and fourth graders and their parents from Bath, Michigan, public school system. This age group was selected as it is a young school age population. Young children may not be as influenced by social desirability factors as older children in reporting parental behavior. This can be a problem in a self report inventory. They also may be more directly dependent on their parents for caregiving than older children. Third graders were chosen as the youngest age group because of reading ability and attention span factors in taking a paper and pencil inventory. Description of the Sample The sample consisted of seventy third, fourth, and fifth-grade students and their parents from the East Lansing, Michigan, public school system and the Bath, Mich- igan, public school system. A few families from other area schools were also included to increase the N and broaden the sample. The sample was a volunteer sample. All families in the sample were two-parent families with completed data. Excluded families included single-parent families, step families, separated families, and families who did not com- plete all interview tasks. 65 Selection of the Sample From the population, a volunteer sample was obtained. To obtain subjects, a letter describing the intent of the research and requesting participation was sent home with all of the children in grades 3 and 4 in Bath, Michigan, public schools and grades 3, 4, and 5 in East Lansing, Michigan, public schools (see Appendix A for letter). Interested fam- ilies were requested to fill out an informational card to return to the researcher indicating their interest in more information and/or their willingness to participate. Also requested were parents' names, their children's names and ages, their address, and telephone number. Returned cards were then distributed to trained student interviewers, and families were contacted by phone. More details of the project were explained to the parents, and if they were still interested in participating, an appointment was set up to meet with the family in the home. All 172 volunteer families were interviewed. During this meeting, an explanation of the study was again presented to the parents as well as the confidential nature of the interview (see Appendix C). The parents' con- sent to participate for themselves and their children was obtained. Questionnaires to parents were distributed and explained, and the age appropriate child was interviewed privately by the interviewer. Twenty-five other families were also asked to partici- pate through word-of-mouth referral. This was done to 66 increase the number of families in the study and to tap fam— ilies of a different type than those who responded to the letter. It was hoped that these twenty-five families would broaden the sample and possibly contain more dissatisfied families. It was speculated that the nature of the letter responders may be "satisfied" and thus willing to give some time to a project of this type. However, this group of twenty-five also consistently endorsed being satisfied: mother with her role and father with his wife's role. Data Collection Obtaining Children's Perceptions Children were interviewed in the home in a private room. Children were tested individually. They were asked to mark items of the Cornell Parent Behavior Inventory (PBI) in a forced choice scale format. The items were orally administered by the interviewer with the children being asked to read along. The inventory was administered twice, once for perceptions of mother's caregiving behaviors and once for perceptions of father's caregiving behaviors. The test was alternated as to which parent form was administered first because of the possibility of confounding results due to fatigue. Between inventory administrations, brief play was offered to give the children a rest and time to clear their minds for the next parent administration. Response format and item order was counterbalanced to control for response set and fatigue factors. 67 The interviewers were thirteen junior and senior psy— chology students who were specially trained to interview children. The interviewers consisted of eight female and five male students who were offered psychology credits for their training and interviewing. All interviewers had had several psychology courses and at least some previous for- mal or informal field contact with the age group in the study. Male and female interviewers alternated administra- tions of the inventory from boy to girl to avoid consis- tently interviewing the child of the same sex. This would control for sex effects of the interviewer. The inventory was administered with uniform directions (see Appendix B), and the interviewer was careful to make sure the children understood the directions for each part of the interview before they answered. Any words that a child did not understand were defined. Otherwise the interviewer did not elaborate on any of the items. If a child did not understand or could not answer a question, the interviewer proceeded to the next question. To make sure that the chil- dren understood the questions, the interviewer asked them to repeat the response format to the questions before he/she began. The responses were repeated again after each item was orally administered. Training of Interviewers Student interviewers received approximately ten hours of training consisting of the initial "how to interview 68 children" session, lecturettes on the research topic, group discussion, scoring and coding interview data, and biweekly checks on interview progress (Appendix B). The initial training session covered the purpose of the research, pilot data, goals of the interview, confidential- ity considerations, review of materials, definition of terms, review of tape of actual interview with ten-year old child, role plays, and general procedural issues. Though there was no formal interviewer evaluation, the biweekly sessions pro- vided an opportunity to discuss interviewer progress and detect interviewer problems. Instrumentation The Cornell Parent Behavior Inventory (PBI) was devel- oped by Uri Bronfenbrenner, et al., at Cornell Human Devel- opment Center in 1961 for the purpose of determining the antecedents and consequences of children's perceptions of the behavior of their parents toward them. The instrument was developed to help alleviate the problems with observa- tion of parental behavior and parental self report of behav- ior (Rogers 1966). Problems with these methods were observer bias, parental defense, and low reliability. Observation studies also cover only a sampling of a small part of behavior. Other types of instruments, such as pro- jectives and homemade inventories, resulted in inconsistent results and are not standardized for comparability. By 69 creating a standardized instrument with standard scores, results could be quantifiable. The PBI is a group-administered questionnaire in the form of rating scales appropriate for children from third to sixth grade. It has value from the phenomenological point of view which states only children's perceptions of parental behavior towards them affects personality development (Test and Measurements Handbook 1971). The original scale of one hundred items was divided among twenty variables with five items representing each variable thought to cover different and important aspects of parental behavior. This original scale has gone through a number of modifications. In 1965, in response to an article by Marvin Siegelman, the scale was reduced to forty-five items with three items representing each of fifteen parent behavior variables. This form was used in a study compar- ing 600 American and German children (Devereaux, Bronfen- brenner, Suci 1961). With a few further modifications in response format and wording, this form was also used in a study comparing children's perceptions of parental behavior, parental self perception, and inference of their children's perceptions (Michaels, Messe, Stollak 1977). A third ver- sion of the PBI was created by the Cornell group from the results of the American German study and was administered to approximately 900 English and 1,000 American children in 1964 and 1965 (Devereaux, Bronfenbrenner, Rogers 1966). Out of this study grew the fourth version of the instrument, 70 which has been worked into both a short and a long form. The short form has been used in two further studies in Israel comparing socialization practices of parents, teach- ers, and peers in the kibbutz, the moshav, and the city (Avgar, Bronfenbrenner, Henderson 1977; Devereaux, Shouval, Bronfenbrenner, Rogers, Kav-Venaki, Kiely, Karson 1974). In this short form, a single item is used to represent each of eighteen different dimensions of behavior measured by clus- ters of items in the longer form. The child is asked to respond to the eighteen-item questionnaire separately for different socializing agents--mother, father, teacher. Work with the short form has shown that each of the dimensions included in the questionnaire represents a different and relevant dimension in the behavior of various socializing agents. Factor analysis of the items in the short form and intflmaSiegelman, Michaels, et al.; Devereaux and Bronfen- brenner studies consistently resulted in three clearly identifiable independent factors: Support, Control, and Punishment. These factors contain twenty-eight percent of the total variance (Rogers 1966). Reliability estimates, calculated by the Spearman Brown method, a split half reliability method derived from the average correlation among the instrument items, show the instrument consistently measures what it purports to measure (Table 3.1) (Rogers 1966). It is this short form that was used in the present study. This form was chosen because of its high reliability 71 .guocoH CH mEouH x Ho some .mummu 03H cow3uwn :oHumHmuuoo 0:» Ho mmumEHumo =mhozvm ooHoom:o .poumHH mEouH may mcosm coHumHomhhoo momuo>m on» EOHH ©m>Hh®© moumEHumo czoumlcmEhmmam .0>Humucwm0haou ance muo3 who; pmumHH mEmuH on» :oan Ho houomm pmumuOH 0:» >3 now coucsoonm . mpcwHuH >2 «~qu on on we ocfisoH Hm. was .3 9: $5355 in . mH .oE pm wow: Iii .HH .oE xomEm no we DH: HHquii m>Mm II: .mxHH u.cmoop III acqumEOm 00 H HH .wH .oE QOHm III .mH .umuumn o>mzon “Icon H HH mac—swam m 9: m>Hv HHHS ...i m>mm Ii .VH .wE mxcmam in: .MH vb. mm. mp. mp. m AmcHschsmv acmszchsm .HHH .cmthHno .355 :93 umuumn 06 on we Houmm mammo— iii .NH 6238 0» me 3:03 in 30: can 9: Ho wuooaxo i... no.3 socx H . HH .hopuo coca :H mmcHnu >5 apex ou we muomaxo iii .oH .Hoozom :H HHoz op 09 we houmm maoox in: .m . on H uo>oumn3 .: when .2. on on me 05:95 maoox Ii .m .uH usonm DUHuum >uo> mH Iii .oE Ho pouomaxm mH Doc: 00 u.:oc H HH .h we. mm. mm. o». m locaHHouucoov Hoeucoo .HH $23 mchHme in .mcHnumEom on CD we 3.83 Ii :95 .0 .ocH£u>um>m usonm II... 4.33 ...—Hop coo H Hmom we moo—ME III . m .moHnsouu w>mn H :on3 ms mQHw: cam muuoHEoo III .v .cumoH on Dan: H mocHnu we monomou III .m .uzo we QHG: ou i... :0 £500 :mo H .EmHnoua «0 0:3. mEom 96: H HH .N .11: com: H HH when» mH III mew we mome III .H «m. mm. mp. Hm. HH loca>oav uuoaasm .H mucohmali . H0909 umnuoz Monumm .mum> acoucou EouH can mHanHm> c853 ammumsflmm 333.6391 158... ocoHumHmhuoo mmumeHumm >uHHHanHmm Show uuonm I >uoucm>cH uoH>mgmm ucmhmm HHmcuoo .HAHoHnme 72 which is comparable to the long form. It is more current, as it has been used in the last two large research projects done by the Cornell group, and it is more applicable to the population being studied (younger children and two adminis- trations of the same instrument). Its items are also broken down into the subscales of loving, controlling, and punish- ing. The instrument has construct validity as demonstrated from the factor analysis. All items are shown to be related for each cluster. The items are similar in content and can be clumped under one heading. The factors of loving, con- trolling, and punishing have also been repeatedly used in studies on parental caregiving behaviors. These factors not only have professional agreement but have been consistently used over time. This is also supportive of their construct validity (Goldin 1969). Various response scales have been used with the PBI including "never, hardly ever, sometimes, fairly often, very often" (Bronfenbrenner 1962, 1965). Other response scales have been "in every case, in most cases, sometimes, seldom, never," "almost every day, almost once a week, about once a month, only once or twice a year, never" (Siegelman 1965), and "definitely yes, probably yes, definitely no, probably no" (Michaels, et al. 1966). The first response format of "never, hardly ever, some- times, fairly often, very often" was chosen for the present study because reliability estimates were calculated using 73 this format. Scoring was done by giving "never" a score of one, "hardly ever" a score of two, "sometimes" a score of three, "fairly often" a score of four, and "very often" a score of five. Means were then calculated for the subscales of loving, controlling, and punishing for each child inter- viewed. Mother's Questionnaire The questionnaire that mothers of $5 filled out while the PBI was being administered to their children is a col- lection of questions to elicit demographic data and to determine mother's work status and satisfaction with her role. This information was also used in the data analysis to determine if detected differences in perceptions were possibly attributable to other factors (see Appendix D for copy of questionnaire). Mother's working status was determined from her report of numbers of hours worked per week and number of months worked prior to the study. The definition of full-time working status was determined from the literature. Full- time working status consistently is defined as over thirty hours per week. Length of employment prior to the study varies in the research on the effects of maternal employment on children from two months to the life of the 85. Two years was decided upon for this study because it establishes a pattern while allowing for inclusion several mothers who may have returned to work when their children entered school. 74 The other significant piece of information gathered by the questionnaire was mother's report of role satisfaction. This was determined by mother's answering yes or no to the following questions. If working, are you on the whole happy and satisfied in your role as a working mother? If not working, are you on the whole happy and satisfied in your role as homemaker? Mothers were then asked to explain their answers in one or two sentences. Mothers were also asked to rate their degree of satisfaction with their role on a scale from one to ten with one representing dissatisfaction and ten representing satisfaction. Mothers were also requested to prioritize activities as to satisfaction. The above questions were chosen as they are similar to the questions asked by Hoffman (1965) in her study finding significant differences between satisfied and dissatisfied working mothers and satisfied and dissatisfied non-working mothers. They are also similar to questions asked by Yarrow, who also found differences between working mothers when subgrouped according to role satisfaction. By request— ing an explanation of why a parent is satisfied or not, it was hoped to discover different reasons and motivations for mother's working or not. In turn, it was hoped this would aid in interpreting findings. The addition of the rating scale is to see if the questions on satisfaction are differ- entiating satisfaction to any degree and to help further explain results. 75 85 were differentiated by class according to Hollings- head's two-factor Index of Social Position classification system. This system classifies families into socioeconomic status by using educational level and type of occupation in a weighted formula. The classes are I-V, with I being the highest level and V the lowest socioeconomic status. Classes I-III are considered middle class. Classes IV and V are considered lower or working class. This system has been traditionally used in studies of this type. Father's SES has been consistently used to determine family SES. The use of this system thus enables generalizability of results of this study to other studies across socioeconomic status and promotes comparison with other studies. Both mother's and father's SES were computed. Final report of data used father's SES to classify families. Father's Questionnaire The questionnaire that fathers filled out while the PBI was being administered to their children is a collection of questions to elicit demographic data to use in computing father's socioeconomic status and to use in data analysis and discussion. Father's work status and satisfaction with mother's role were also determined from this questionnaire. Full-time working fathers were defined as fathers who worked over thirty hours per week for paid employment outside of the home for two years prior to data collection. Only full- time working fathers were included in the study. 76 Father's satisfaction with mother's role was determined by father's answering yes or no to the following questions depending upon mother's work status. "On the whole are you happy and satisfied with your wife being a working mother?" or "On the whole are you happy and satisfied with your wife being a homemaker?" Fathers were asked to briefly explain their answers and to rate their degree of satisfaction with mother's role on a scale from one to ten, with one repre- senting dissatisfaction and ten representing satisfaction. Fathers were also requested to prioritize activities as to satisfaction. These questions were asked to differentiate families by satisfaction/dissatisfaction and to aid in interpreting findings. The above instruments, as well as the whole data col- lection procedure and subsequent data analysis, were piloted on sixteen Ss to anticipate problems in design, data collection, and analysis of data. Statistical Hypotheses The hypotheses examined in this study are presented below. These hypotheses were tested at the .05 level of significance. Hypothesis 1 Null Hypothesis: There is no significant difference in children's perceptions of loving parent behavior as a 77 function of family work status, sex of the child, and sex of the parent. Alternative Hypothesis: Children's perception of lov- ing parent behavior significantly differs as a function of family work status, sex of the child, and sex of the parent. Hypothesis 1 was broken down into the following corol- lary hypotheses: Hola: There is no significant difference in children's per- ception of loving parent behavior as a function of family work status. Hala: Children's perception of loving parent behavior sig- nificantly differs as a function of family work status. Ho There is no significant difference in children's per- lb' ceptions of loving parent behavior as a function of sex of child. Halb: Children's perception of loving parent behavior sig- nificantly differs as a function of sex of child. Hole: There is no significant difference in children's per- ceptions of loving parent behavior as a function of sex of parent. Halc: Children's perception of loving parent behavior sig- nificantly differs as a function of sex of parent. Hold: There is no significant difference in children's per- ceptions of loving parent behavior as a function of family work status and sex of child. Hald: Children's perception of loving parent behavior sig- nificantly differs as a function of family work status 78 and sex of child. Hole: There is no significant difference in children's per- ceptions of loving parent behavior as a function of family work status and sex of parent. Hale: Children's perception of loving parent behavior sig— nificantly differs as a function of family work status and sex of parent. Holf: There is no significant difference in children's per- ceptions of loving parent behavior as a function of sex of child and sex of parent. Half: Children's perception of loving parent behavior sig- nificantly differs as a function of sex of child and sex of parent. Hypothesis 2 Null Hypothesis: There is no significant difference in children's perception of controlling parent behavior as a function of family work status, sex of the child, and sex of the parent. Alternative Hypothesis: Children's perceptions of con- trolling parent behavior significantly differs as a function of family work status, sex of the child, and sex of the parent. Hypothesis 2 was broken down into the following corol— lary hypotheses: 79 Ho2a There is no significant difference in children's per- ceptions of controlling parent behavior as a function of family work status. HaZa: Children's perception of controlling parent behavior significantly differs as a function of family work status. HoZb: There is no significant difference in children's per- ceptions of controlling parent behavior as a function of sex of child. HaZb: Children's perception of controlling parent behavior significantly differs as a function of sex of child. HoZC: There is no significant difference in children's per- ceptions of controlling parent behavior as a function of sex of parent. Hazcz Children's perception of controlling parent behavior significantly differs as a function of sex of parent. HoZd: There is no significant difference in children's per- ceptions of controlling parent behavior as a funCtion of family work status and sex of child. HaZd: Children's perception of controlling parent behavior significantly differs as a function of family work status and sex of child. Hoze: There is no significant difference in children's per- ceptions of controlling parent behavior as a function of family work status and sex of parent. Haze: Children's perception of controlling parent behavior 80 significantly differs as a function of family work status and sex of parent. Ho There is no significant difference in children's per- 2f: ceptions of controlling parent behavior as a function of sex of child and sex of parent. Hazf: Children's perception of controlling parent behavior significantly differs as a function of sex of child and sex of parent. Hypothesis 3 Null Hypothesis: There is no significant difference in children's perceptions of punishing parent behavior as a function of family work status, sex of the child, and sex of the parent. Alternative Hypothesis: Children's perception of punishing parent behavior significantly differs as a func— tion of family work status, sex of the child, and sex of the parent. Hypothesis 3 was broken down into the following corol- lary hypotheses: Ho3a: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of family work status. Ha3a: Children's perception of punishing parent behavior significantly differs as a function of family work Status . 81 Ho3b: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of sex of child. Ha3b: Children's perception of punishing parent behavior significantly differs as a function of sex of child. Ho3C: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of sex of parent. Ha3C: Children's perception of punishing parent behavior significantly differs as a function of sex of parent. HoBd: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of family work status and sex of child. Ha3d: Children's perceptions of punishing parent behavior significantly differs as a function of family work status and sex of child. Ho3e: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of family work status and sex of parent. Ha3e: Children's perception of punishing parent behavior significantly differs as a function of family work status and sex of parent. Ho3f: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of sex of child and sex of parent. Ha3f: Children's perception of punishing parent behavior 82 significantly differs as a function of sex of child and sex of parent. Methods of Data Analysis The basic method of data analysis was an unweighted means repeated measures analysis of variance performed by program BALANOVA on Michigan State University's Computer. All factors were fixed. There were two non-repeated fac- tors: sex of child and family work status. The one repeated factor was the scale used, whether for mothers or for fathers. A normal distribution of population scores was assumed as was equal population variances. The sample was not random in that it was a volunteer sample. However, it is representative in that it is not fixed. Other similar studies (Bronfenbrenner 1969; Finke 1969; Franke 1971; Avager, et a1. 1977) have consisted of volunteer samples and have used parametric statistics as well because of their strength in comparison to non-parametric statistics. Hypoth- eses were tested at the .05 level of significance. When statistically significant main effects were found, pot hoc analyses to determine the source of the interaction were computed. For the non-repeated and repeated factors, an analysis of the simple two-way interaction effects was performed (Kepel 1973). Again, significance was determined at the .05 level. Pearson correlations (PPMCC) were run between six pos- sible confounding variables and the three dependent measures 83 of loving, controlling, and punishing to determine the pos- sibility of covariates. These variables were the socio- economic status of mother and father, the child's age, the birth order of the child, the number of children in the family, and the age of the child when mother became absent from the home by going to work. These correlations were run on the sample as a whole and on individual groups. Descriptive data was statistically analyzed by using the Statistical Package for the Social Sciences computer program at Michigan State University. Frequencies were run for all collected sample characteristics for each group. Means and standard deviations were also determined for the three dependent measures. These were determined separately for mothers and fathers on the sample as a whole and on indiVidual groups. Pearson correlations were also run to see if any of the dependent measures of loving, controlling, punishing were correlated with each other. Data on mother's satisfaction with her role and father's satisfaction with mother's role as well as mother's perception of father's satisfaction with her role were also collected. However, not enough mothers or fathers endorsed the dissatisfaction items on the questionnaire to enable a statistical analysis with satisfaction/dissatisfaction as an independent variable. Nevertheless, this outcome has sig— nificance in understanding the results of this study and will be discussed in Chapter 5. 84 Summary Seventy third, fourth, or fifth-grade children and their parents who met specific criteria were selected for the study. These criteria included: (a) being a two-parent family, (b) having a full-time working mother (one who has worked outside the home for paid employment for over thirty hours per week or more for two years prior to the study) or a non-working mother in them, and (c) having completed all interview tasks. These families were interviewed in their homes by trained interviewers to determine if children's perceptions of the parental behaviors of loving, controlling and punishing differed as a function of the sex of child, the sex of parent, and the family work status. Children were interviewed using the Cornell Parent Behavior Inven— tory. Mothers were requested to complete a Mother's Ques- tionnaire, and fathers completed a Father's Questionnaire. These were completed independently. Families were divided into four groups according to the sex of the child and family work status: male child working mother families, female child working mother families, male child non-working mother families, and female child non-working mother families. A causal comparative experimental design was used to determine possible relationships among the dependent and independent variables. Data met the assumptions for using parametric status. An unweighted means repeated measures ANOVA was performed with all factors fixed. Hypotheses were 85 tested at the .05 level of significance. Post hoc analyses were performed when a significant interaction was found to locate the source of the interaction. The test for this was an analysis of the simple two-way interaction effects. CHAPTER IV ANALYSIS OF RESULTS The research data are analyzed in this chapter. A restatement of the research problem, pilot data, and a description of the sample are included. Each hypothesis is restated, and the data and analyses relevant to each hypothesis are presented and discussed. Restatement of the Problem The study was designed to examine third, fourth, and fifth graders' perceptions of parental caregiving behaviors as a function of family work status, sex of the child, sex of the parent, mother's satisfaction with her role, and father's satisfaction with mother's role. Children's per- ceptions of loving, controlling, and punishing parental behaviors were gathered, scored, and analyzed. A causal comparative design was used to explore possi- ble relationships between the dependent and independent variables. Data was analyzed using parametric statistics. An unweighted means repeated measure analysis of variance was performed by the computer program BALANOVA to determine if there were statistically significant differences. 86 87 Pilot Study Prior to the main study, a pilot study was carried out on sixteen volunteer families from the Bath, Michigan, public school system. Families were solicited through a letter requesting help with the project which was distributed to third and fourth graders (see Appendix A). The volunteer families were interviewed by four (two male and two female) trained junior and senior psychology students. Data from interviews was analyzed with an unweighted means repeated measures analysis of variance to determine significant dif- ferences in children's perceptions as a function of family work status, sex of the child, and sex of the parent. Post hoc analyses were not run on the pilot data because of the low N. Pilot Sample The pilot sample consisted of all Caucasian families with a mean age of 9.3 years for children, thirty-five years for mothers, and forty years for fathers. The mean number of children per family was 2.4, and most of the chil- dren in the study were first children. The mean socio— economic status (SES) level of both mothers and fathers was SES level 2. All parents reported satisfaction: mothers with their role and fathers with their wife's role. The sample contained two step families and two part-time working mother families which were categorized as working mother families. 88 Pilot Results Prior to testing the main hypotheses, Pearson correla— tions were run between the dependent measures of loving, controlling, and punishing parental behaviors and six vari- ables the literature described as relevant to maternal employment and perception studies. Since no significant correlations were detected, it was determined that no covar- iate analysis of variance was necessary. The main hypothe- ses tested at the .05 level of significance were: Hol: There is no significant difference between children's perceptions of loving parent behavior as a function of family work status, sex of child, and sex of parent. (H02: There is no significant difference between children's perceptions of controlling parent behavior as a funcg tion of family work status, sex of child, and sex of parent. H03: There is no significant difference between children's perceptions of punishing parent behavior as a function of family work status, sex of child, and sex of parent. Corollary hypotheses were also tested and are fully deline— ated in Chapter III. Results for the first hypothesis indicate that no sig- nificant differences were found between means of children's perceptions of loving parent behavior as a function of the three independent variables (sex of child, sex of parent, family work status) or their interaction. Since p = .86 for Hol and is not significant, the decision is not to reject 89 Hol. Data presented in Table 4.1 show that none of the hypotheses tested meet the requirement for statistical sig- nificance. Results for the second hypothesis indicate a signifi- cant difference in children's perceptions of controlling parent behavior as a function of the interaction between family work status and sex of parent, where p = .032 and is less than the predetermined .05 significance level. The decision is not to reject the main Ho2 but to reject corol- lary hypothesis HoZb. There is no significant difference in children's perceptions of controlling parent behavior as a function of family work status and sex of parent. Data presented in Table 4.2 show this significant interaction. Results for the third hypothesis indicate a signifi- cant difference in children's perceptions of punishing par- ent behavior as a function of family work status, where p = .005. Since p = .005 is less than the previously estab— lished significance level of c =.05, the decision is to reject corollary hypothesis Ho3a. There is no significant difference in children's perceptions of punishing parent behavior as a function of family work status. Significance results were also obtained for the inter- action effect of family work status and sex of the child. Children's perceptions of punishing parent behavior were significantly different as a function of the interaction between the sex of child and the family work status, where p = .029. The decision was made to reject corollary Ho3d 90 Table 4.1. Pilot Analysis of Variance Summary: Scores for Loving Scale. Sources of ss df MS F p Variation A .05 l .05 .14 .713 NS B .45 l .45 1.21 .194 NS AB 1.11 l 1.11 3.01 .111 NS S/AB 4.05 11 .37 C .04 l .04 .25 .628 NS AC .05 1 .05 .33 .580 NS BC .02 l .02 .12 .732 NS ABC .005 l .005 .03 .856 NS S/CXAB 1.56 11 .14 A = Family work status (working mother family or non-working mother family) Sex of child (male or female) Sex of parent (mother scale or father scale) [D II II 91 Table 4.2. Pilot Analysis of Variance Summary: Scores for Controlling Scale. Sources of SS df MS F p Variation A 1.59 1 1.59 2.63 .133 NS B 1.53 1 1.53 2.54 .140 NS AB .18 1 .18 .30 .594 NS S/AB 6.65 11 .60 C .002 l .002 .02 .902 NS AC .65 1 .65 6.06 .032 * BC .16 l .16 1.45 .253 NS ABC .38 l .38 3.56 .086 NS S/CXAB 1.18 11 .11 A = Family work status (working mother family or non-working mother family) Sex of child (male or female) Sex of parent (mother scale or father scale) (D II ll *¢= .05 92 since this value is less than the pre-established c =.05 level of significance. The main hypothesis and all other corollary hypotheses failed to produce significant results (see Table 3.3). Pilot Discussion Since pilot data were analyzed with a very small (16 families) biased (included part-time working mother fami- lies and step families) N, no conclusions were drawn from significant results. Procedures of data collection and data analysis of the pilot project study were replicated in the main study. Except for the four families that did not meet sample selec- tion criteria (two step families, two part-time working mother families), pilot sample characteristics were compara- ble to main data sample characteristics. Therefore, a decision was made to include the eleven pilot families fit- ting selection criteria in the data analysis of the main study. Subjects The sample for the main study consisted of 172 families drawn from a population of 1,174 families. Only seventy of these families fit the criteria of the study and were used in the final analysis. These seventy families were two- parent families who completed all parts of the interview procedure and whose mothers fit the criteria of working full 93 Table 4.3. Pilot Analysis of Variance Summary: Scores for Punishing Scale. Sources of SS df MS F p Variation A 4.46 1 4.46 12.05 .005 * B .69 1 .69 1.88 .198 NS AB 2.32 l 2.32 6.25 .029 * S/AB 4.07 11 .37 C .004 l .004 .05 .830 NS AC .17 l .17 2.05 .180 NS BC .01 l .01 .14 .712 NS ABC .00006 1 .00006 .001 .980 NS S/CXAB .90 11 .08 A = Family work status (working mother family or non-working mother family) B = Sex of child (male or female) C = Sex of parent (mother scale or father scale) *“= .05 94 time or not. Working mothers were defined as mothers who had been working outside the home for paid employment for at least thirty hours per week for at least two years prior to the data collection. Non-working mothers were defined as mothers who had not worked at all outside the home for paid employment for two years prior to the data collection. Part-time working mother families were excluded. One- parent families, separated families, step families, and families with incomplete data were eliminated from the anal- ysis to avoid additional confounding variables. Of the 102 eliminated families, only ten were not included because of incomplete data. These families were investigated, and in five out of the ten families, fathers were in another coun— try or out of town on business for an extended period and could not return their questionnaires. Five fathers failed to return their questionnaires, though they had agreed to upon follow-up contact. Ten non-responder families were also investigated. It was determined that their unwilling- ness to participate was based on time factors and/or a deci— sion not to respond to new volunteer projects because of previous volunteer commitments. However, there did not seem to be any trend in type of family failing to respond, as many had volunteered for similar projects previously. The final seventy families consisted of eighteen male child working mother families, eleven female child working mother families, twenty male child non-working mother fami- lies, and twenty-one female child non-working mother 95 families. Families could not be further broken down by mother's and father's satisfaction/dissatisfa¢tion scores due to too few subjects per group. Mothers and fathers con- sistently endorsed being satisfied regardless of working status. Therefore, this sample consists solely of mothers who reported satisfaction with their roles and fathers who reported satisfaction with mother's roles. The children in the study were third, fourth, and fifth graders ranging in age from eight to eleven years old. Each group had all ages represented with the mean age of children in each group being 9.5 years old. Seventy-five percent of the children in the sample were from the East Lansing school system, and twenty—five percent of the chil- dren in the sample were from Bath, Haslett, Lansing, and Okemos school systems. Most children were first or second children in their families, with first children ranging from forty percent of the subjects in the male child non-working mother families to 72.7% of the subjects in the famale child working mother families. Second children ranged from 18.2% of subjects in the female child working mother families to forty-five percent of subjects in the male child non-working mother families (see Table 4.4). The race of the families in the study was primarily Caucasian. Ninety percent of the families in the sample were Caucasian, and ten percent were minorities. These fam- ilies were black, Hispanic and Oriental. In general, this 96 Table 4.4. Frequency Distribution of Children's Ages. 6:33P WORKING MOTHER FAMILY NON-WORKING MOTHER FAMILY Male Female Male Female No. % No. % No. % No. % 8 3 16.7 1 9.1 l 5 6 28.6 9 6 33.3 7 63.6 8 40 10 47.6 10 5 27.8 2 18.2 8 40 3 14.3 11 4 22.2 1 9.1 2 10 2 9.5 Blank l 5 sample approximates the percentage of minorities in the pop— ulation at large. Examining mother's and father's characteristics, it is noted that most of the parents in the sample were between the age ranges of thirty-one to thirty-five years old and thirty-six to forty years old. The mean age of mothers was thirty-three, and the mean age of fathers was thirty—eight. In examining the frequency table on socioeconomic status of mothers, it can be seen that the SES of mothers was very similar regardless of her present work status during the last two years. The non-working mothers had attained similar educational levels and, when working, similar kinds of occu- pations as the current working mothers. The non—working mother groups, larger in this study, had more unskilled and less educated mothers in them as well. Most mothers fell into SES II or III, middle-class classifications (see Table 4.5). Father's SES appeared quite similar across groups also, with most fathers falling into SES I or II. Seventy-four percent of all fathers in the sample fell into SES I and II. 97 Table 4.5. Frequency Distribution of Socioeconomic Status (SES) of Mothers. 1.25:1 WORKING MOTHER FAMILY NON-WORKING MOTHER FAMILY Male Female Male Female No . 76 No . % No . % No . % I 4 22.2 2 18.2 1 5 2 9.5 II 9 50.0 5 45.5‘ 9 45 8 38.1 III 3 16.7 3 27.3 5 25 4 19 IV 2 11.1 1 9.1 4 20 6 28.6 V l 5 I—V = Hollingshead's levels as defined by weighted formula. I-III = middle class; IV-V = lower class. Female working mother families was the group with most diversity of father's SES, with fifty-four percent in SE8 I and II and 27.3% of fathers in the group in SE8 IV. These findings indicate that the sample was weighted towards fami- lies with college-educated fathers in professional and mana- gerial positions. The sample was also biased towards middle-class mothers, as categories I, II, and III have tra- ditionally been considered middle class, and categories IV and V have been traditionally considered lower class (Table 4.6). When income is examined, it appears that in the working mother families, father's income ranged around $31-40,000. In the non-working mother families, father's income was higher in the female child non-working mother families and more diverse in the male child non-working mother families when compared to the working mother groups. 98 Table 4.6. Frequency Distribution of Socioeconomic Status (SES) of Fathers. Lzsgl WORKING MOTHER FAMILY NON-WORKING MOTHER FAMILY Male Female Male Female No. % No. % No. % No. % I 9 50.0 4 36.4 9 45 8 38.1 II 5 27.8 2 18.2 5 25 10 47.6 III 3 16.7 2 18.2 4 20 l 4.8 IV 1 5.6 3 27.3 1 5 2 9.5 V ' 1 5 I-V = Hollingshead's levels as defined by weighted formula. I-III = middle class; IV-V = lower class. In summary, all groups in the sample for the study appeared quite similar as regards children's ages, birth order, race, mother's SES, father's SES, and mother's and father's ages. When compared to the population at large, the sample had a lower percentage of minorities and a higher percentage of educated professional and managerial fathers and middle-class mothers. This may be because of the voluntary natureof the sample and the composition of the main city from which the sample was drawn. East Lansing, Michigan, is a highly educated, professional com— munity. These characteristics limit generalizability of results. Pearson Correlations Prior to the actual examination of research hypotheses, Pearson correlations (PPMCC) were performed between the dependent variables of loving, controlling, and punishing 99 parental behaviors and six variables which the literature indicated are important considerations in studies on mater- nal employment and perception. These were the socioeconomic status of mother and father, the child's age, the birth order of the child in the study, the number of children in the family, and the age of the child when mother became absent from the home by going to work. Correlations were run for all seventy families across groups. They were also run for each subgroup: male child working mother families, female child working mother families, male child non-working mother families, and female child non-working mother fami- lies. No significant correlations were consistently found across groups between the dependent variables and the above factors where ¢ = .05. This result indicated that there was no need to use any of the above factors as covariates in the analysis of variance of the research hypotheses. There did appear to be a trend towards father's socioeconomic status being correlated with the dependent measures of children's perceptions of loving, controlling, and punishing behaviors of fathers. A few significant correlations were found within subgroups as well. These findings are elaborated upon in the discussion section (see Appendix F). No significant correlations resulted when Pearson cor- relations were run to see if any of the dependent means, loving, controlling, and punishing parent behaviors were correlated with each other. This supports previous research 100 finding these three factors as clearly identifiable and independent (see Appendix F). Hypothesis 1 H01: There is no significant difference in children's per- ceptions of loving parent behavior as a function of family work status, sex of child, and sex of parent. Hal: Children's perception of loving parent behavior sig- nificantly differs as a function of family work status sex of child, and sex of parent. These hypotheses were formulated to examine whether differences in children's perception of loving parent behav- ior were related to family work status (working mother or non-working mother), the child's sex, and/or the sex of the parent. Data could not be statistically analyzed according to the variables of mother/father satisfaction/dissatisfac— tion with mother's role as not enough parents endorsed being dissatisfied. Mean data presented in Table 4.7 show that means for all groups are very similar. An unweighted means repeated measures ANOVA indicated no significant results. Specifically, F = .10 on main hypothesis 1. Since p = .75 is much greater than C = .05, the pre-established significance level, Ho is not signifi- l cant and the decision not to reject Ho is made. All other 1 corollary hypotheses were also found to be nonsignificant. Table 4.8 shows F values and probability estimates for each of the variables and interactions tested. 101 Table 4.7. Means and Standard Deviations for Loving Score. WORKING MOTHER FAMILY NON-WORKING MOTHER FAMILY Mother Father Mother Father MALE 4.39 4.36 4.52 4.56 .43 .63 .59 .48 FEMALE 4.40 4.30 4.48 4.34 .66 .50 .60 .64 The conclusion is that the family work status, sex of the child, and/or the sex of the parent does not signifi- cantly affect children's perceptions of loving parent behavior. Hypothesis 2 H02: There is no significant difference in children's per- ceptions of controlling parent behavior as a function of family work status, sex of child, and sex of parent. Haz: Children's perception of controlling parent behavior significantly differs as a function of family work status, sex of child, and sex of parent. These hypotheses were developed to determine if mean differences in children's perceptions of controlling paren- tal behavior were related to family work status, sex of the child, and/or sex of the parent. Data could not be statis- tically analyzed according to the variables of mother/father 102 Table 4.8. Analysis of Variance Summary: Scores for Loving Scale. Sources of SS df MS F p Variation A .28 1 .28 .53 .470 NS B .20 l .20 .37 .548 NS AB .10 l .10 .19 .665 NS S/AB 35.32 66 .54 C .06 l .06 .44 .511 NS AC .02 l .02 .15 .696 NS BC .10 l .10 .73 .396 NS ABC .01 l .01 .10 .751 NS S/CxAB 8.71 66 .13 A = Family work status (working mother family or non—working mother family) B = Sex of child (male or female) = Sex of parent (mother scale or father scale) 103 satisfaction/dissatisfaction with mother's role as not enough parents endorsed being dissatisfied. The means and standard deviations of controlling perceptions are presented in Table 4.9. The largest mean difference is between males and females in working mother families. Table 4.9. Means and Standard Deviations for Controlling Score. WORKING MOTHER FAMILY NON-WORKING MOTHER FAMILY Mother Father Mother Father MALE 3.96 3.93 3.81 3.47 .66 .69 .71 .64 FEMALE 3.61 3.86 3.96 3.76 .71 .78 .74 .87 An unweighted means repeated measures ANOVA indicated significant results for H02. A triple interaction was indi- cated to be significant at the .045 level of significance which fits the predetermined level of significance, ¢ = .05. All other tested corollary hypotheses were determined to be insignificant (see Table 4.10). In an effort to look further into the data and find the locus of interaction leading to the triple interaction effect, an analysis was made of the simple two—way interac- tion effects (Kepel 1973). Results of this analysis are presented in Table 4.11. 104 Table 4.10. Analysis of Variance Summary: Scores for Controlling Scale. Sources of ss df MS F p Variation A .21 l .21 .24 .628 NS B .90 l .90 1.03 .315 NS AB .08 l .08 .10 .758 NS S/AB 57.86 66 .88 C .04 1 .04 24 .624 NS AC .16 l .16 .86 .357 NS BC .004 1 .004 .02 .877 NS ABC .76 l .76 4.18 .045 * S/CxAB 12.00 66 .18 A = Family work status (working mother family or non-working mother family) Sex of child (male or female) Sex of parent (mother scale or father scale) a- A ll .05 Table 4.11. 105 Scores for Controlling Scale. Summary of Simple Two-Way Interaction Effects: Sources of SS df MS F Variation AxB at C1 .66 1 .66 .23 NS AxB at c2 0 l 0 0 NS BxC at al .16 l .16 .89 NS BxC at a2 .16 l .16 .89 NS AxC at b1 .08 1 .08 .46 NS AxC at b2 .04 l .04 .22 NS A = Family work status a1 = working mother family a2 = non-working mother family B = Sex of child bl = male b2 = female C = Sex of parent c1 mother scale father scale C2 106 As can be seen from the F values, no significant simple two—way interaction effects were detected. This is not a surprising finding as the F probability of the triple inter- action was only p = .045. As a result of the post hoc analyses, a decision was made to reject H02. However, since the post hoc analyses of simple two-way interactions effects did not locate the source of interaction, it is unclear how to interpret the triple interaction effect. The conclusion is that there is a trend towards the interaction of family work status, sex of child, and sex of parent significantly affecting chil- dren's perceptions of controlling parent behavior (see Figure 4.1). In observing the interaction effect in Figure 4.1 and the mean data in Table 4.9, it appears that in work— ing mother families boys see mothers as more controlling than girls, and that in non-working mother families boys see fathers as more controlling than girls. Hypothesis 3 H03: There is no significant difference in children's per- ceptions of punishing parent behavior as a function of family work status, sex of child, and sex of parent. Ha3: Children's perception of punishing parent behavior sig— nificantly differs as a function of family work status, sex of child, and sex of parent. These hypotheses were created to see if there is a relationship between children's perception of punishing 107 Figure 4.1. Three-Way Interaction of Factors A, B, and C for Controlling Parent Behavior. 33 8 4'0 "' 3.96 m 0 3'9 ... 3'93 3.86 Father 5 3.8 4- P4 6' u 3.73% 4H) 8 3.6 i- 3.61 Mother U c 3.54— m m z t t Male Female WORKING MOTHER FAMILY 8 8 4.0 v— 3.97 m 3.96 Mother 0') 3.9 4* F 3.81 S 3.81- '3 3.76 Father 3 3.7-P c 8 3.5.- S 3.5 «F o 2 Male Female NON-WORKING MOTHER FAMILY A = Family work status (working mother family or non-working mother family) B = Sex of child (male or female) C = Sex of parent (mother or father scale) 108 parent behavior and the variables of family work status, sex of child, and sex of parent. Data could not be statisti- cally analyzed according to the variables of mother/father satisfaction/dissatisfaction with mother's role as not enough parents endorsed being dissatisfied. These hypotheses were tested using the unweighted means repeated measure analysis of variance. Mean and standard deviation data are presented in Table 4.12 showing that largest mean differences are between children's perceptions of parents punishing behavior in working mother families versus non-working mother families. Table 4.12. Means and Standard Deviations for Punishing Score. WORKING MOTHER FAMILY NON—WORKING MOTHER FAMILY Mother Father Mother Father MALE 1.82 1.84 2.14 1.82 .55 .50 .86 .51 FEMALE 1.76 1.74 2.03 2.12 .51 .66 .70 .83 Results indicate no significant F values for the main Ho3 or any of the corollary hypotheses tested. Specifically, F = 1.36 with a probability of p = .25 for Ho3 This is insignificant at the .05 level of significance. Thus the decision not to reject Ho is made. 3 109 In examining Table 4.13, it is noted that there is a family work status effect for the children's perception of punishing parent behavior. Children from working mother families may see parents as less punishing than children in non-working mother families. However, this is significant only at the .10 level, which does not fit the criterion for rejecting the null hypothesis. The conclusion is that the variables of family work status, sex of child, and sex of parent do not significantly effect children's perceptions of punishing parental behavior. Summary Three major hypotheses and eighteen corollary hypothe- ses were tested to examine the effect of family work status, sex of child, and sex of parent on children's perceptions of loving,‘controlling, and punishing parent behaviors. Hypotheses were examined in a pilot study prior to main data collection and analysis. Alpha level was set at the .05 level for each test. An unweighted means repeated measures analysis of variance was performed to examine the possible relationships between dependent and independent variables. Prior to the examination of the hypotheses, Pearson correla- tions (PPMCC) were performed to determine possible covari- ates. None were found. The results basically did not establish differences between the children's perceptions and the explored independent variables. Though a triple 110 Table 4.13. Analysis of Variance Summary: Scores for Punishing Scale. Sources of ss df MS F p Variation A 1.85 1 1.85 2.73 .103 NS B .0002 1 .0002 .00 .985 NS AB .22 1 .22 .32 .571 NS S/AB 44.86 66 .68 C .0008 l .0008 .009 .927 NS AC .0005 l .0005 .006 .940 NS BC .06 1 .06 .66 .419 NS ABC .13 l .13 1.36 .257 NS S/CxAB 6.32 66 .09577 :1:- ll Family work status (working mother family or non-working mother family) B = Sex of child (male or female) = Sex of parent (mother scale or father scale) 111 interaction between family work status, sex of child, and sex of parent was found for the children's perception of controlling parent behavior, the results did not indicate any specific locus of interaction upon post hoc analysis of simply two-way interactions. Though no further analyses were significant, the triple interaction suggests a trend. A summary of the hypotheses and results are presented in Table 4.14. 112 Table 4.14. Summary of Hypotheses and Results. Actual Level Hypothesis Test “ of Decision Significance . Ho1 RM .05 .781 Failed to Reject Hola RM .05 .470 Failed to Reject Ho1b RM .05 .548 Failed to Reject Ho1C RM .05 .665 Failed to Reject Ho1d RM .05 .551 Failed to Reject Ho1e RM .05 .696 Failed to Reject Ho1f RM .05 .396 Failed to Reject H02 RM .05 .045 Rejected Ho2a RM .05 .628 Failed to Reject Ho2b RM .05 .315 Failed to Reject Ho2C RM .05 .758 Failed to Reject Ho2d RM .05 .624 Failed to Reject Ho2e RM .05 .357 Failed to Reject Ho2f RM .05 .877 Failed to Reject Ho3 RM .05 .257 Failed to Reject Ho3a RM .05 .103 Failed to Reject Ho3b RM .05 .985 Failed to Reject Ho3C RM .05 .571 Failed to Reject Ho3d RM .05 .927 Failed to Reject Ho3e RM .05 .940 Failed to Reject Ho3f RM .05 .419 Failed to Reject RM = Repeated measures analysis of variance. CHAPTER V SUMMARY, CONCLUSIONS, DISCUSSION AND IMPLICATIONS FOR FUTURE RESEARCH This chapter is divided into four major sections. The first section is a summary of this research project. Next, conclusions are stated. The third section is a discussion of the findings with respect to the theoretical concepts upon which the study was based. The final section suggests implications for future research. Summary The purpose of this study was to explore whether or not children's perceptions of their parents' caregiving behav— iors were affected by the family work status (working mother family or non-working mother family), the sex of the child (male or female), and the sex of the parent (mother scale or father scale). Caregiving behaviors were defined as loving, controlling, and punishing parent behaviors as measured by the three subscales of the Cornell Parent Behavior Inventory. The need for this study arose from the awareness that more and more mothers work without conclusive or recent evi— dence as to the effect of their work absence on their 113 114 children. According to the 1980 census, over forty-three percent of women with children under six years work for paid employment outside the home, and over fifty-three percent of women with children under the age of eighteen work. Par- ents, especially mothers, need to have adequate knowledge of the effects of their absence on their children's behavior and development. Without adequate information on the effect of maternal employment on children, parents remain guilt ridden and conflicted. Career decisions and lifestyle deci- sions are made without conclusive answers. A survey of the literature indicates that much of the research on the effects of maternal employment on children has been conflicting, methodologically problematic, and devoid of theoretical underpinnings. The '60's found more juvenile delinquents in working mother families and chil- dren, especially boys, with sexual identity problems. Prob- lems with bonding were also speculated. Other research, especially survey studies in the early 1970's, found few significant differences between children from working mother families and children from non-working mother families. However, most of these studies did not control for confound- ing variables such as sex of child, mother's feelings about work, father's feelings about her work, supervision, age of children when mother began working, birth order, number of children in family, etc. Hoffman (1974) suggested that maternal employment is a complicated issue and that it is too simplistic an approach to merely look at one child 115 characteristic and tie it to maternal employment. She sug- gested that studying interaction effects and the intervening process between maternal employment and a child's character- istics would perhaps clarify effects. ,Children's percep- tions was proposed as an area needing research since it is a mediating process between maternal employment and any child characteristic or behavior. Ego psychology theory, supported by studies on the effects of maternal deprivation on children, has been used as a theoretical base for many of the maternal employment studies. This theory maintains that children with absent mothers are either cognitively or emotionally deprived and perceive her absence as rejection (Hoffman and Nye 1974). The theory of perceptual psychology maintains that any change in perceptual field alters perceptions which, in turn, alter behavior. According to Combs (Combs and Snygg 1974), behavior is a function of the perceptions of the indi- vidual. Research has shown that mothers and fathers both inter- act with their children in different ways when mother works. Thus, it was assumed that children's perceptions of their parents would be different depending upon mother's work status in interaction with other variables. It was pre- dicted under the perceptual model that there would be dif- ferences in perceptions between working mother children and non-working mother children. Under the "deprivation" model, 116 it was predicted that mother's working absence would be per- ceived as rejection. The sample for the study consisted of seventy families with third, fourth, or fifth graders in them. Eighteen families were male child working mother families, eleven were female child working mother families, twenty were male child non-working mother families, and twenty-one were female child non-working mother families. Both parents in the sample endorsed being satisfied: mothers with their role either as working mother or homemaker and fathers with their spouses' role. All families were volunteer families from the Lansing-East Lansing area solicited by letter through the schools. Twenty-five families also volunteered upon direct request from the researcher. These were word- of-mouth referrals. All families were two-parent and con- tained a working mother, a mother who had worked continu— ously for thirty hours per week or more for two years prior to the study for paid employment outside the home or a non- working mother, a mother who had not worked for paid employ- ment outside the home for two years prior to the study. No part-time working mothers, one-parent families, step fami— lies, or families with incomplete data were included in the sample. The sample was mainly composed of middle-class families, as defined by Hollingshead, and was ninety percent Caucasian, ten percent minorities. The mean age of the children was 9.5 years. 117 A causal comparative design was used to determine the effects of existing conditions: family work status, sex of child, and sex of parent on children's perceptions of lov— ing, controlling, and punishing parent behaviors. Trained student interviewers interviewed children pri- vately in their homes using the Cornell Parent Behavior Inventory. Parents were requested by the interviewers to fill out the Mother's Questionnaire and the Father's Ques- tionnaire while the child was being interviewed. Confiden— tiality of responses was assured. Twenty-one hypotheses were tested at the .05 level of significance using an unweighted means repeated measures analysis of variance performed by the computer program BALANOVA at Michigan State University. No covariates were found when Pearson correlations (PPMCC) were run between variables and dependent variables prior to the study. Conclusions Effect of family work status on loving perception. No evidence was found to indicate that the family work status significantly affects children's perceptions of loving par— ent behavior. Effect of sex of child on loving perception. No evi- dence was found to indicate that the sex of the child signif- icantly affects children's perceptions of loving parent behavior. 118 Effect of sex of parent on loving perception. No evi- dence was found to indicate that the sex of the parent sig- nificantly affects children's perceptions of loving parent behavior. Effect of family work status and sex of child on loving perception. No evidence was found to indicate that the fam- ily work status and sex of the child significantly affect children's perceptions of loving parent behavior. Effect of sex of child and sex of parent on loving per- ception. No evidence was found to indicate that the sex of the child and sex of the parent significantly affect chil- dren's perceptions of loving parent behavior. Effect of family work status and sex of parent on lov- ing perception. No evidence was found to indicate that the family work status and sex of the parent significantly affect children's perceptions of loving parent behavior. Effect of family work status, sex of child, and sex of parent on loving perception. No evidence was found to indi- cate that the family work status, sex of the child, and sex of the parent significantly affect children's perceptions of loving parent behavior. Effect of family work status on controlling perception. No evidence was found to indicate that the family work status significantly affects children's perceptions of con- trolling parent behavior. Effect of sex of child on controlling perception. No evidence was found to indicate that the sex of the child 119 significantly affects children's perceptions of controlling parent behavior. Effect of sex of parent on controllinggperception. No evidence was found to indicate that the sex of the parent significantly affects children's perceptions of controlling parent behavior. Effect of family work status and sex of child on con— trollingperception. No evidence was found to indicate that the family work status and sex of the child significantly affect children's perceptions of controlling parent behavior. Effect of sex of child and sex of parent on controlling perception. No evidence was found to indicate that the sex of the child and sex of the parent significantly affect children's perceptions of controlling parent behavior. Effect of family work status and sex of parent on con- trollingperception. No evidence was found to indicate that the family work status and sex of the parent significantly affect children's perceptions of controlling parent behavior. Effect of family work status, sex of child, and sex of parent on controllingperception. Evidence was found to indicate that the interaction of family work status, sex of the child, and sex of the parent did significantly affect children's perceptions of controlling parent behavior. Under post hoc analyses, the source of the interaction between family work status, sex of the child, and sex of the 120 parent could not be determined. No simply two-way interac- tion effects were found significant. Observation of the data indicates that the finding may be that boys in working mother families see mothers as more controlling than girls in working mother families and that boys in non-working mother families see fathers as more controlling than girls in non—working mother families. Thus, the interaction is a trend which warrents replication. Effect of family work status on punishing perception. No evidence was found to indicate that the family work status significantly affects children's perceptions of pun- ishing parent behavior. Effect of sex of child on punishing perception. No evidence was found to indicate that the sex of the child significantly affects children's perceptions of punishing parent behavior. Effect of sex of parent on punishing perception. No evidence was found to indicate that the sex of the parent significantly affects children's perceptions of punishing parent behavior. Effect of family work status and sex of child on pun- ishing‘perception. No evidence was found to indicate that family work status and sex of the child significantly affect children's perceptions of punishing parent behavior. Effect of sex of child and sex of parent on punishing perception. No evidence was found to indicate that the sex 121 of the child and sex of the parent significantly affect children's perceptions of punishing parent behavior. Effect of family work status and sex of parent on pun- ishing perception. No evidence was found to indicate that the family work status and sex of the parent significantly affect children's perceptions of punishing parent behavior. Effect of family work status, sex of child, and sex of parent on punishing perception. No evidence was found to indicate that the family work status, sex of the child, and sex of the parent significantly affect children's percep- tions of punishing parent behavior. Discussion According to the "deprivation" model of mother's absence, the perceptual theory of psychology,and the research in these areas, children's perceptions of parental behavior were expected to be altered as a function of mother's work status in interaction with the sex of the child, the sex of the parents, mother's satisfaction with her role, and father's satisfaction with mother's role. It was assumed that family interaction and structure were altered from the traditional in working mother families and would, in turn, affect children's perceptions of parent's behavior. This section is a discussion of the results of the study with reference to these theoretical constructs. The limitations of the study are also discussed. 122 Under the deprivation model, mother's absence is con- sidered detrimental to a child's normal adjustment, especi- ally if occurring at an early age. It is hypothesized that without the stable consistent and continuous care of mother, a child will develop abnormally, without adequate object constancy, with cognitive deficits, with poor self- esteem, and with feelings of inadequacy (Hoffman and Nye 1974). It is also speculated that when mother's absence is conceptualized as "deprivation," mothers and fathers both feel guilty and as a result overcompensate in their behavior and interactions with their children (Hoffman 1974). Research has been mixed in bearing out this deprivation explanation, and it has begun to be more widely advocated that mother's working absence may not be deprivation at all (Rosenthal and Hansen 1971). The present study adds credence to this position that middle-class working mothers, when satisfied with their role and when their husbands are satisfied with mother's role, may not be depriving their children from the age group examined at all. If a child were indeed deprived or felt deprived by mother's work absence, it is speculated that children would perceive working mothers as more unloving and/or more punishing when compared to children in non- working mother families. It is unclear how fathers would be perceived, an area relatively unexplored with respect to mother's employment status. 123 In the present study, since no significant differences were detected between loving scores and punishing scores as a function of family work status, sex of child, or sex of parent, it appears that children in satisfied middle-class families may not perceive mother's working absences as deprivation by either parent. This finding holds up regard- less of child's age when mother returned to work. It is interesting to note that in other studies of children's per— ceptions of parents, sex differences were found as well as parent differences where girls saw parents as, in general, more loving than boys and where fathers were seen as more punishing than mothers. This finding did not hold up in this study. This can perhaps be accounted for by cultural factors where role expectations between men and women have become less differentiated and parents are assuming more similar roles, at least in this satisfied middle-class sub— group. Though no significant simply two-way interaction main effects were found upon post hoc analysis, a significant three-way interaction did occur between family work status, sex of child, and sex of parent when examining the control- ling parent behavior. This indicates that the interaction of these factors may significantly affect children's percep— tions of controlling parent behavior. Visual examination of the data indicates that the source of this interaction may be that in non-working mother families, boys sees fathers as more controlling than girls do, and in working mother 124 families, boys see mothers as more controlling than girls do. This possible finding is interesting and possibly sug- gests support for Hoffman's theory, one that grows out of the deprivation model. She hypothesizes that when parents conceptualize mother's work status as deprivation then they feel guilty and overcompensate. This may be experienced by children as controlling behavior. So mother, out of guilt, may overcompensate and overcontrol. In non-working mother families, according to traditional stereotypes, father may be more controlling. Boys may be more sensitive to control because of cultural stereotypes that boys must be indepen- dent and in control of themselves, not controlled (Hartley 1964). This finding could also reflect that in traditional families, father may raise, take responsibility, or control sons, and mother may raise, take responsibility, or control daughters. In working mother families, these roles may be changed, and mothers may take more responsibility for rais—I ing sons or both parents take more responsibility for rais- ing all children regardless of sex. Again, there may be -less role differentiation. These role changes show up in boys' perceptions because of boys' sensitivity to control issues and girls' expectations to be controlled. The find- ing could also be conceptualized in terms of power issues where traditionally the one who works (father) is perceived as having the power and is thus perceived as more control- ling. Mother's employment may alter this perception so that she is viewed as more controlling in working mother families 125 since she has the power of the traditional role plus now the power of working. However, these are speculations on a trend which need to be futher explored in future studies. It is also possible that the three-way interaction was a spurious finding or a statistical artifact. Yet it is sug- gested that when children's perceptions of parents' control- ling behavior are examined, the family work status, the sex of the child, and the sex of the parent must be taken into consideration. The results of this study do not support the depriva— tion model of conceptualizing mother's work absence but instead support the idea that working mothers and fathers from satisfied middle-class families can and do relate to their children in the same basic pattern as do non—working mothers and fathers from satisfied middle-class families. It is suggested that deprivation or abandonment may not be an issue at all in working mother families for elementary school children. Guilt over working and depriving her chil- dren may play a role in mother's overcompensating behavior which the child perceives as controlling. This finding may also be viewed as reflecting more role differentiation in working mother families where mothers take more responsibil- ity for raising sons and fathers for daughters when compared to the traditional family. Support for this conclusion also comes from perceptual psychology theory. According to this theory, perception is affected by a number of factors including need, time, goals, 126 values, and roles. Though no two people have identical per- ceptual fields, people who have similar experiences tend to have common characteristics in their phenomenal fields and thus see things the same way. Since behavior is a function of perceptions, people with common perceptions show common tendencies in their behavior (Combs 1974). In this study, it was assumed that children's perceptual field would be different in working mother families when compared to non- working mother families since a number of factors control- ling and limiting the perceptual field would be affected by mother's work status. More specifically, these would be time and roles. However, since no differences were obtained in the perceptions measured, it can be speculated that the perceptual field of children from working mother fmailies and non-working mother families may not be significantly different at all, and mother's employment status may not be as influential a factor in child rearing or children's per- ceptions for this age group as has been assumed and hypothe- sized. It is suggested again that parents in this sample from satisfied middle-class non~working mother families and working mother families relate to their children in basi- cally the same way and with the same patterns. Other fac- tors in families may affect patterns of relating between parents and children more significantly than mother's work status. In visual examinations of family characteristics, of the ten children (six females and four males) reporting 127 their parents as most loving (I = 4.8—5 on loving scale) and least punishing <§'= <2 on punishing scale), some interest- ing patterns emerged suggesting what these other factors might be. These ten families consisted of six non-working mother families and four working mother families. SES of fathers was 1 or 2, and all were first or second-born chil- dren. The average age of the children was nine years old. In all cases, both parents rated their degree of satisfac- tion with mother's role above 7 on a lO-point scale. It is assumed this scale was differentiating satisfaction to some extent as degree of satisfaction scores were not consis— tently high in the sample as a whole. This possibly means that these parents were not overly concerned or conflicted by mother's role but highly satisfied with it whether mother works or not. The other interesting trend which warrants speculation is that in eight out of the ten families, both parents, when prioritizing activities that give them great- est satisfaction in the family, endorsed being a husband or wife first and being a father or a mother second. The other two families reversed this pattern. This trend could indi- cate that these parents, perceived as highly loving and low in punishment, may haVe very happy and satisfying marriages and that this is an extremely important factor in affecting parents' behavior and children's perceptions of parent behavior towards them. Parents in satisfying marriages of these kinds of families may relate in a loving and non- punitive manner towards their children. 128 To further investigate the above findings, the data from the ten families where children perceived their parents as least loving and most punishing were examined. Though demographic characteristics were similar, it was discovered that five of these families were from the eliminated data of dissatisfied parents (not enough parents endorsed being dis- satisfied with working mother's role to compose groups large enough for statistical analysis). In seven out of the ten families, at least one parent reported dissatisfaction with mother's role. Degree of satisfaction scores widely varied on the 1-10 scale. In all but one family, getting satisfac- tion from being a wife or husband was a lower priority than in the ten loving, non-punishing families. Usually there was disagreement between parents as to satisfying activities where, for example, mother endorsed getting greatest satis- faction from mothering first and friends second, and where father endorsed getting greatest satisfaction from job with leisure activities second. These findings support the sug- gestions above that satisfaction with mother's role and par- ent relationship issues may be more influential factors in how parents relate to their children and are thus perceived than mother's work status, sex of the child, or sex of the parent. These findings warrant further investigation. Alternative explanations of the findings are also possi- ble, the most significant being the possibility that the behaviors measured are too crude to get at the sensitive process of relating between parents and children. 129 Differences may be too subtle to be detected in a pencil and paper inventory, or the variables measured in this study may not be the variables affected by mother's work status. Since the Cornell Parent Behavior Inventory has discrimin— ated differences between children's perceptions in the past, it is assumed that the instrument has adequate sensitivity. More complex ways to examine the parent-child interactive process may need to be developed, however. Other factors, such as mother's self-esteem, father's self-esteem, mother's ego strength, father's ego strength, and parents' satisfacw tion/dissatisfaction with their relationship in combination with maternal employment may be variables that more signifi- cantly affect parent/child interaction than ones identified in this study. The administration of the CPI was different in this study than from the administration in the majority of other studies. The instrument was created to be administered to groups of children. In this study, the questions were admin- istered individually and read aloud to control for misunder- standing and reading ability. This may have increased the likelihood of social desirability factors. Children in group administrations may feel more anonymity whereas chil- dren in this study may have felt that the trainer would see their results. Since parents were in the home at the time of administration, children may have felt their parents would see the results. Confidentiality was assured and questions coded in an attempt to control for this. Michaels 130 Stollak and Messe (1977) also administered the same instru— ment in the same way to a comparable age group and detected significant differences in perceptions. However, social desirability may be an alternative explanation for results. Though activity level between parents and children was not measured in the present study, it has been found to be less in working mother families in previous research. It is possible, however, that in these satisfied middle-class fam- ilies with children of these ages, activity level between parents and children did not differ. Thus, there is no "deprivation" or "abandonment" in working mother families. Similar activity levels and amounts of "time spent" between parents and children may account for the findings as well. Future research may find differences in perceptions between working mother and non-working mother families if activity level is computed and examined. Another factor accounting for the lack of significant results could be the low N. Though twenty families per group were proposed, one group had eleven families in it meeting the specified criteria. A larger N would increase the likelihood of detecting differences. In summary, then, possible methodological problems could have accounted for the lack of results, the most likely of these being a small N and social desirability fac- tors. However, many confounding variables were carefully controlled for: mother's and father's socioeconomic status, age of child, birth order, number of children in family, age 131 of child when mother returned to work, in a clearly defined sample with standardized data collection procedures. These factors lend credibility to the following generalizations: (1) Children in "satisfied" middle-class families may not perceive mother's work absences as deprivation by either parent. Deprivation may not be an issue at all for this population. (2) Fathers and mothers from satisfied middle—class working mother families and satisfied middle-class non-working mother families may relate to their children in similar ways and in the same basic pattern. (3) Parental satisfaction with mother's role and with their marriage may be a more relevant and influential factor on children's perceptions of parental behavior than maternal employment. (4) Mothers from working mother families may feel guilty about their absence and overcompensate or take more responsibility for their children which shows up as controlling parent behavior. Generalizability of this study is limited to the popu- lation described. Implications for Future Research If research studies are continued in this area, several design considerations might be examined. Sampling in ear— lier studies was a problem as mother's work status was not clearly defined or controlled for, especially in the survey 132 studies of the '60's and early '70's. Though this factor was carefully controlled in this study, the N decreased con- siderably as a result. Future studies should concentrate on getting a larger N of clearly defined working and non- working mother families. It may also be fruitful to examine part-time working mother families to see if this status affects children's perceptions when compared to the other two groups. Examining working class families may also pro- duce different results and broader generalizations. Since there were not any dissatisfied families in this study enabling statistical examination of perceptions along the parental satisfaction/dissatisfaction dimension, it is suggested that greater effort be made to find dissatisfied working mother and non-working mother families. Satisfac- tion questions could be refined to elicit more specific sources of satisfaction such as: In what ways are you satisfied/dissatisfied with your role as working mother or non-working mother? In what ways are you satisfied/ dissatisfied with your wife's role as working or non-working mother? This variable may possibly be more significant in affecting children's perceptions than mother's work status. Other research suggests various variables be accounted for in studies on maternal employment. These variables include the age of the child, the socioeconomic status of both mother and father, the birth order of the child, the number of children in the family, and the age of the child when the mother returned to work. Though no correlations 133 were discovered between these variables and the three depen- dent measures of loving, controlling, and punishing parent behaviors, certain correlations did exist within subgroups. These variables are suggested for future research explora- tions as identified covariates within subgroups. A separate analysis of covariance could be performed for each group. The identified covariates in this study were birth order of subject, age of subject, and age of the child when mother returned to work in examining the punishing parent behavior for female child non-working mother families; the sex of the child when examining the punishing parent behavior and father's socioeconomic status when examining loving and con- trolling perceptions for male child non—working mother families; the number of children and mother's socioeconomic status for female child working mother families; and birth order and age of child when mother returned to work when examining loving parent behavior for male child working mother families. Significant differences in perception may be found within further subgroupings with identified covari- ates. Refinement in data collection may also improve studies of this nature. To control for social desirability factors, it may be helpful to interview children individually away from the home with more anonymity. It is also suggested to administer mother's and father's scales at different times to avoid having the child's responses on the first parent 134 affect and interfere with his/her responses on the second parent. Though the Cornell Parent Behavior Inventory has increased reliability and validity when compared to similar instruments used in studies of this type, available instru- ments still might not possess the desired degree of sensi- tivity in measuring perception. Instruments might need to be developed and/or refined in order to obtain a more accu- rate assessment of children's perceptions of parent's behavior. Longitudinal studies with data collected over time and analyzed with respect to cognitive and perceptual develop- ment could provide helpful information with regards to the process between parents and children in working and non- working mother families over time. It also may add informa- tion on the effect of mother's working on the behavior of children at earlier and later age periods. APPENDIX A LETTERS TO PARENTS 135 7513 Stoll Road Bath, Michigan 48808 June 8, 1981 Dear Parent(s): My name is Jaye Hamilton, and I am completing a doctoral degree in Counseling Psychology at Michigan State University. First, I would like to thank the Bath Public School adminis— tration and staff for their support and help so that your child was able to bring home this letter. I am currently engaged in research under the supervision of Drs. Gary Stollak, Norman Kagan, Al Aniskiewicz, and Sam Plyer concerning children's perceptions of different aspects of family life. I would like to administer a brief question- naire to a large number of third and fourth graders over the next few months. I anticipate that it will take a child approximately fifteen minutes to complete his/her question— naire which will be individually administered in your hometnr a trained interviewer. I will also be asking for parents to complete a brief questionnaire which will take approximately fifteen minutes to complete. All collected information will be anonymous and identifiable only by code. If you are interested in participating or interested in learning more about the research before reaching a final decision whether to participate or not, please complete and return the enclosed postcard and I will contact you. If, after I answer your questions, you and your family are‘milling to participate, I will set up a meeting time convenient for you and your child. Your child will be interviewed in your home at that time, and I will deliver the parent questionnaire also at that time. When you have completed the questionnaire, an addressed, stamped envelope will be provided for the ques- tionnaire's return. If desired I will provide you with speci- fic details and results of the study after the data has been completely collected and analyzed. If you have any questions, please call me at 641-4582 (home) or 351-3451 (office). I sincerely thank you for your help in advance. Sincerely, Jaye Hamilton Enclosure 136 MICHIGAN STATE UNIVERSITY DEPARTMENT OF PSYCHOLOGY EAST LANSING ' MICHIGAN ' 48824 PSYCHOLOGY RESEARCH BUILDING Dear Parent(3)= "y name is Jaye Hamilton, and I an completinr a doctoral dearee in Counseling Psychology at Hichiqan State University. First, I would like to thank the East Lansine public School administration and staff for their support and help so that your child was able to bring home this letter. I am currently engaged in research under the supervision of Drs. Cary ”tollak, Norman Kagan, A1 Aniskiewicz, and Sam Plyer concernine children’s perceptions of different aspects of family life. I would like to administer a brief questionnaire to a large number of third and fourth graders over the next few months. I anticipate that it will take a child approximately fifteen minutes to couplete his/her questionnaire which will be individually administered in your home by a trained interviewer. I will also be askine for parents to complete a brief questionnaire which will take approximately fifteen minutes to complete. All collected information will be anonymous and identifiable only hy code. If you are interested in participating or interested in learning more about the research before reaching a final decision whether to participate or not, please complete and return the enclosed postcard and I will contact you. If, after I answer your questions, you and your family are willine to participate. I will set up a meetine time convenient for you and your child. Your child will then he interviewed, and I will deliver the parent duestion- naire at that time. "hen you haye completed the nuestionnaire, it should be returned in the addressed, stamped envelope provided. If desired I will provide you with specific details and results of the study after the data has been completely collected and analyzed. If you have any questions, please call me at 641-4582 (home) or 351-3451 (office). I sincerely thank you for your help in advance. Sincerely, 957“ "MW Jaye Hamilton 137 MICHIGAN STATE UNIVERSITY IJI-J’AR'I'MI'ZN'I' OI‘ PSYCHOLOGY EAST LANSING ' MICHIGAN ° 48824 PSYCHOLOGY RESEARCH BUILDING Dear Parent(s): My name is Jaye Hamilton, and I am completing a doctoral degree in Counseling Psychology at Michigan State University. First, I would like to thank the East Lansing Public School administration and staff for their support and help so that your child was able to bring home this letter. I am currently engaged in research under the supervision of Drs. Gary Stollak, Norman Kagan, Al Aniskiewicz, and Sam Plyer concerning children's perceptions of different aspects of family life. I would like to administer a brief questionnaire to a large number of fifth graders over the next few months. I anticipate that it will take a child approximately fifteen minutes to complete his/her questionnaire which will be individually administered in your home by a trained interviewer. I will also be asking for parents to complete a brief questionnaire which will take approximately fifteen minutes. All collected information will be anonymous and identifable only by code. If you are interested in participating or interested in.learning more about the research before reaching a final decision whether to participate or not, please complete and return the enclosed postcard and I will contact you. If, after I answer your questions, you and your family are willing to participate, I will set up a meeting time convenient for you and your child. Your child will then be interviewed, and I will deliver the parent question- naire at that time. When you have completed the questionnaire, it should be returned in the addressed, stamped envelOpe provided. If desired, I will provide you with specific details and results of the study after the data has been completely collected and analyzed. If you have any questions, please call me at 641-4582 (home) or 351-3451 (office). I sincerely thank you for your help in advance. Sincerely, 91x 7 L QZ/x Ind-Z [cw/ex Jaye Hamilton MSU is an Al/irnum'm' Action/Equal Opportunity Institution APPENDIX B TRAINING MATERIALS (A) (B) (C) (D) (E) (F) (G) 138 OUTLINE OF TRAINING SESSION Introduce selves. State field experience with children in this age group. State interest in this study. Lecturette on purpose of this research, theory, and background of study. Presentation of pilot data. Review "How to Do Interview." Review consent forms, mother's and father's question- naires, Cornell Parent Behavior Inventory for mothers and fathers. Listen to tape of interview with 10 year old boy. Role play interviewing children with a partner. 139 HOW TO DO INTERVIEW Goals of Interview: (1) To obtain informed consent to participate in the study from parents for themselves and for their child. To obtain child's assent to participate. (2) To interview child and obtain his/her perceptions of his parent's behaviors. (3) To have parents fill out their questionnaires. (4) To answer questions about the questionnaires. 1. Call parents--names of family members and phone numbers on post card. Identify self-—associate of mine, MSU student helping me collect data for my doctoral dissertation. They have returned the post cards I passed out at East Lansing Elementary Schools indicating willingness to partic— ipate in this study. Set up a time convenient to them to interview their child (3rd, 4th, or 5th grader). Get direc- tions to their home. 2. Make sure code on questionnaires. Take packet of materials and training packet to home--will need training packet to read from. 140 3. At the home--introduce self and go over informed consent papers. Both mother and father need to sign for themselves--only one parent need sign for child. Explain the study as research exploring children's perceptions of different aspects of family life. Don't tell them that I am looking at working versus non-working mothers, etc., because this information could bias their responses. Explain that you will interview the child separately for about half an hour. During this time you would like them to fill out their questionnaires so that you can answer questions that may arise when you are finished with the child interview. Tell them that group results will be made available to them--most likely next spring--but that a child's indi- vidual responses are not available to them. If they want to know specifically what child is being asked then give examples like-—we'll be asking the child questions about the differences between mothers and fathers—-How often does your mother comfort you? Do fathers comfort you? How often? Does your mother want you to do well in school? Your father? If parents want to see the Cornell Parent Behavior Inventory, show them. Ask if they have any questions. 4. Parents' Questionnaires. Again try to get parents to fill out their questionnaires while you're in the inter— view with the child. Otherwise have them look questionnaire over and leave stamped, addressed envelope for them to return it. When giving parents their questionnaires, 141 remind them to not share their answers with their spouse until each has completed his/her questionnaire. If there is not an answer applicable to them, have them write their own answer in. Parents need not disclose salary. All participants have the right not to answer a question. If step family, have the step father/mother fill out the father's/mother's questionnaire if child lives with step parent most of the time. Make a note of this on questionnaire. 5. Interview with child. Need to have pencil and something for child to write on. Go to private place to interview child--bedroom, family room, outside. Establish rapport. Explain what doing--why interviewing. Example: "I'm helping a person with a big paper so she can graduate from MSU. She is writing a book about children. She's interested in what kids think about families. I'd like to ask you some questions about families. Is that okay?" (child's assent). Tell the child that all of his/her answers are private--no one will see them including you or parents and that they can be kept private because he/she has been given a number so no name appears on the questionnaire. Encourage child to ask questions if does not understand word or meaning of a question. 142 Read directions and if child seems to understand begin reading the questions. Read every question and every response format as response format changes, and some chil- dren do not read very well. Also, if child does not under- stand response format explain as: Very often means most of the time Fairly often means more than half the time Sometimes means half the time Hardly ever means less than half the time Never means never Define any unknown words. Alternate whether give mother or father question- naire first. Indicate on questionnaire which is given first. Offer brief play between parent administrations. Also indicate on the questionnaire whether ygg, the interviewer, are male or female. If child lives with a step parent most of the time, have them fell out the questionnaire for the step parent and indicate this on the questionnaire. Find out how often child sees absent parent. Also in single-parent families, have kids fill out the forms for both parents even though only lives with one. Again, find out how frequently child sees absent parent. 6. Answer any questions parents may have. APPENDIX C INFORMED CONSENT l. 144 RESEARCH CONSENT FORM I have freely consented to have my child take part in a scientific study being conducted by: under the supervision of: Academic Title: The study has been explainedtx>me and I understand the expla- nation that has been given and what my child's participation will involve. I understand that my child is free to discontinue his/her par- ticipation in the study at any time without penalty. I understand that the results of the study will be treated in strict confidence and that my child will remain anonymous. Within these restrictions, results of the study deJ. be made available to my family at my request. I understand that my child's participation in the study does not guarantee any beneficial results to him/her. I understand that, at my request, I can receive additional ex- planation of the study after my child's participationis com- pleted. Signed: Dated: TITLE OF RESEARCH PROJECT: APPENDIX D CORNELL PARENT BEHAVIOR INVENTORY CODE: AGE: 145 Male or female MOTHER Here are some descriptions of things mothers do. Read each statement below along with the instructor and check the answer which best describes your mother as she has been during the past year. Be sure to answer every statement. Please do not leave any out. Remember, you are describing your mother as she has been just during the past year. 1. She comforts and helps ma when 1 have troubles. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIREY OFTEN VERY OFTEN 2. She keeps after me to do better than other children. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES RARDEY EVER NEVER 3. She says she will give me a spanking if I don't behave better. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 6. If I have some kind of problem, I can count on her to help ma out. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES EARDLY EVER NEVER 5. She keeps after me to do well in school. I I I I I I I I I I NEVER RARDLY EVER SOMETIMES FAIREY OFTEN VERY OFTEN 6. She punishes me by not allowing me to be with my friends. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES EARDLY EVER NEVER 7. She makes me feel I can talk with her about everything. I I I I I I I I I I NEVER RARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 146 8. If I don't do what is expected of me, she is very strict about it. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVER NEVER 9. If I do something she doesn't like, she says she will hit me or smack me. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 10. She teaches me things I want to learn. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVEN NEVER 11. She keeps pushing me to do by best at whatever I do. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VENT OFTEN 12. She spanks me. I I I I I I I I I I VERx OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVER NEVER 13. When she wants me to do something, she explains why. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VENT OFTEN 16. She expects me to keep my things in good order. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVER NEVER 15. She nags at me. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 16. She makes me feel she is there if I need her. I I I I I I I I I I VERY OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVER NEVER 147 17. I know what she expects of me and how she wants me to behave. I I I I I I I I I I NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 18. She slaps me. I I I I ~ I I I I I VERY OFTEN FAIRLY OFTEN SIMETIHES EARDLY EVER NEVER I CODE: AGE: 148 Male or female FATHER Here are some descriptions of things fathers do. Read each statement below along with the instructor and check the answer which best describes your father as he has been during the past year. do not leave any out. during the past year. 1. He comforts and helps me when I have troubles. Be sure to answer every statement. Remember, you are describing your father as he has been Please (I I) I) I) I) NEVER RARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 2. He keeps after me to do better than other children. I) I) I) I) I) VERY OFTEN FAIRLY OFTEN SOMETIMES NARDLY EVER NEVER 3. He says he will give me a spanking if I don't behave better. I) I) (I I) I) NEVER RARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN I. If I have some kind of problem, I can count on him to help me out. I) I) I) II I) VERY OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVER NEVER 5. He keeps after me to do well in school. I) I) (I I) I) NEVER HARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 6. He punishes me by not allowing me to be with my friends. I) I) I) I) (I VERY OFTEN FAIRLY OFTEN SOMETIMES HARDLY EVER NEVER 7. He makes me feel I can talk with him about everything. (I I) I) I) I) NEVER. HARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 149 8. If I don't do what is expected of me, he is very strict about it. ( ) ( ) ( ) ( ) ( ) van 0mm um! 0mm sown-ms HARDLY mm mm 9. If I do something he doesn't like, he says he will hit me or smack me. ( ) < ) ( ) ( ) ( ) maven mom mm soum'ms mm! am VERY 0mm 10. He teaches me things I want to learn. ( ) ( ) ( ) ( ) ( ) my om um! um smms mu m m 11. He keeps pushing me to do my best at whatever I do. ( ) < ) ( ) ( ) ( ) mm HARDLY m sown-ms um! om my om 12. He spanks me. ( ) ( ) ( ) ( ) ( ) VERY OFTEN FAIRLY OFTEN SCMETIMES NARDLY EVER NEVER 13. When he wants me to do something, he explains why. ( ) ( ) ( ) ( ) ( ) NEVER . EARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 15. He expects me to keep my things in good order. ( ) ( ) ( ) ( ) ( ) VERY OFTEN FAIRLY OFTEN SOMETIMES EARDLY EVER NEVER 15. He nags at me. ( ) ( ) ( ) ( ) ( ) NEVER RARDLY EVER SOMETIMES FAIRLY OFTEN VERY OFTEN 16. He makes me feel he is there if I need him. ( ) ( ) ( ) ( ) ( ) VERY om FAIRLY 01mm SOMETIMES mu mm: mm 150 17. I know what he expects of me and how he wants me to behave. ( ) . ( ) ( ) ( ) ( ) NEVER HARDLY EVER SGIETIMES FAIRLY OFTEN VERY OFTEN 18. Re slaps me. ( ) ( ) ( ) ( ) ( ) VERY OFTEN FAIRLY OFTEN ‘ SOMETIMES NARDLY EVER NEVER APPENDIX E PARENT QUESTIONNAIRES CODE NUMBER: DATE: 151 MOTHER'S QUESTIONNAIRE Please read the instructions below, then proceed. INSTRUCTIONS: On the questionnaire itself, circle the letter of the response which is most applicable to you. Try to answer every question. Please do not share your responses with your husband until the questionnaire is returned. If you have questions, feel free to call Ms. Hamilton (641-4582). 1. What is your age? 21-25 years old 26-30 years old 31-35 years old 36-40 years old 41-45 years old (1100373’ 2. What is your marital status? A. Married B. Single C. Divorced D Remarried 3. What is your race? A. Caucasian B. Black C. Hispanic D. Other 4- _How-many children do you have? One child TWO children Three children Four children Five children Over five children "flt‘flUOWJ’ 5. What is the birth order of the child in this study? First Second Third Fourth Fifth Other TWP??? 152 If you are now engaged in paid employment outside the home, please answer the following set of questions. the home, please proceed to question 17. 10. What is your present occupation according to the following classification system? ammo Executive or upper level professional Manager or lower level professional Administrator in a large concern; owner of a medium-sized independent business Owner of a small business; clerical or sales worker; technician Skilled worker Semi-skilled worker Unskilled worker What is your present yearly income? IDVWDOW> 0 - 5,000 6,000 - 10,000 11,000 - 15,000 16,000 - 20,000 21,000 - 25,000 26,000 - 30,000 31,000 - 35,000 Above 36,000 What is your level of educational attainment? ommonw> How onw> How onw> Graduate training with a graduate degree Standard college or university graduate Partial college training High school graduate Partial high school (tenth or eleventh grade) Junior high school (seventh - ninth grade) Less than seven years of school many hours per week do you work (for paid employment outside the home)? 1 - 10 hours per week 11 - 20 hours per week 21 - 30 hours per week Over 30 hours per week long have you worked for paid employment outside the home? 6 months - 1 year 1 - 3 years 4 - 6 years Above 6 years If you are not now engaged in paid employment outside 11. 12. 13. 153 How long have you worked continuously over 30 hours per week prior to this study? A 1 year B. 2 years C. 3 years D 4 years E 5 years F For the life of child in study What was the age of the child in this study when you went back to work over 30 hours per week (paid employment outside the home)? Below 6 months Below 1 year 1 - 2 years old 3 - 4 years old 4 - 5 years old Older than 6 years old mmonoa> On the whole, are you happy and satisfied with being a working mother? A. Yes B. No In one or two sentences, please explain your answer. 14. 15. 16. Please rate your degree of happiness and satisfaction with being a working mother. 1 2 3 4 5 6 7 8 9 10 Dissatisfied Satisfied What activities give you greatest satisfaction? Please rank order the following list, with 1 being the activity that gives you greatest satisfaction. ( ) Being a wife ( ) Being a mother ( ) Doing leisure activities ( ) Your job ( ) Being a volunteer ( ) Being a friend ( ) Doing housework ( ) Cooking ( ) Other Do you think that on the whole your husband is happy and satisfied with your being a working mother? A. Yes B. No Please explain your answer. 154 If you are not now engaged in paid employment outside the home, please answer the following set of questions. If you are now engaged in paid employment outside the home, then you are finished with the questionnaire. Thank you. 17. What is your level of educational attainment? Graduate training with a graduate degree Standard college or university graduate Partial college training High school graduate Partial high School (tenth or eleventh grade) Junior high school (seventh - ninth grade) Less than seven years of school nmmonw> 18. If you have previously worked (for paid employment outside the home), what was your occupation according to the following classification system? Executive or upper level professional Manager or lower level professional Administrator in a large concern; owner of a medium-sized independent business Owner of a small business; clerical or sales worker; technician Skilled worker Semi-skilled worker Unskilled worker nw> ammo 19. Have you ever worked ( for paid employment outside of the home) during the lifetime of the child in this study? A. Yes B. No (If you answer no to this question, go on to question 20.) a. If yes, how old was your child when you began work? A. Below 6 months B. Below 1 year C. l - 2 years D. 3 - 4 years E. 4 - 5 years F. Older than 6 years b. For how long did you work during the lifetime of the child in this study? 6 months 1 year 2 - 3 years 4 - 5 years Over 6 years monm> c. For approximately how many hours per week did you work? 0 - 10 hours 11 - 20 hours 21 - 29 hours 30 or over onw> 20. 155 d. When was the last time you worked over 30 hours per week for paid employment outside the home? Within the past 6 months Within the last year Within the last 2 years Within the last 3 years Within the last 5 years mcow> If not presently working for paid employment outside the home, on the whole, are you happy and satisfied with being a homemaker? A. Yes B. No In one or two sentences, please explain your answer. 21. 22. 23. Please rate your degree of satisfaction with being a homemaker (non-working mother). 1 2 3 4 S 6 7 8 9 10 Dissatisfied Satisfied What activities give you greatest satisfaction? Please rank order the following list, with 1 being the activity that gives you greatest satisfaction. ( ) Being a wife ( ) Being a mother ( ) Doing leisure activities ( ) Your job ( ) Being a volunteer ( ) Being a friend ( ) Doing housework ( ) Cooking ( ) Other Do you think that on the whole your husband is happy and satisfied with your being a homemaker? A. Yes B. No Please explain your answer. Thank you very much for your time and consideration. CODE NUMBER: DATE: 156 FATHER'S QUESTIONNAIRE Please read the instructions below, then proceed. INSTRUCTIONS: On the questionnaire itself, circle the letter of the response which is most applicable to you. Try to answer every question. Please do not share your rexponses with your wife until the questionnaire is returned. If you have questions, feel free to call Ms. Hamilton (641-4582). 1. What is your age? 21-25 years old 26-30 years old 31-35 years old 36-40 years old‘ 41-45 years old MUOW> 2. What is your race? A. Caucasian B. Black C. Hispanic D Other 3. What is your occupation according to the following classification system? A. Executive or upper level professional B. Manager or lower level professional C. Administrator in a large concern; owner of a medium-sized independent business Owner of a small business; clerical or sales worker; technician Skilled worker Semi-skilled worker Unskilled worker Unemployed USD'UITIU 4. What is your level of educational attainment? Graduate training with a graduate degree Standard college or university graduate Partial college training High school graduate Partial high school (tenth or eleventh grade) Junior high school (seventh - ninth grade) Less than seven years of school CD'TJITIDOW> 157 5. How many hours per week do you work (for paid employment outside the home)? A. 0 - 10 hours per week B. 11 - 20 hours per week C. 21 - 30 hours per week D Over 30 hours per week 6. How long have you worked continuously over 30 hours per week prior to this study? 1 year 2 years 3 years 4 years 5 years For the life of child in study mmcnw> If your wife works for paid employment outside the home, please answer questions 7 through 10. If your wife does not work for paid employment outside the home, please proceed to question 11. 7. On the whole, are you happy and satisfied with your wife being a working mother? A. Yes B. No In one or two sentences, please explain your answer. 8. Please rate your degree of happiness and satisfaction with your wife being a working mother. 1 2 3 4 S 6 7 8 9 10 Dissatisfied Satisfied 9. Do you think your wife is happy and satisfied with being a working mother? A. Yes B. No 10. What activities give you greatest satisfaction? Please rank order the following list with 1 being the activity that gives you greatest satisfaction. ( ) Being a husband ( ) Being a father ( ) Doing leisure activities ( ) Your job ( ) Being a volunteer ( ) Being a friend ( ) Doing housework ( ) Cooking ( ) Other 158 If your wife does not now work for paid employment outside the home, please answer questions 11 through 14. If your wife does work for paid employment outside the home, please go to question 15. 11. On the whole, are you happy and satisfied with your wife being a homemaker? A. Yes B. No In one or two sentences, please explain your answer. 12. Please rate your degree of happiness and satisfaction with your wife being a homemaker. l 2 3 4 5 6 7 8 9 10 Dissatisfied Satisfied 13. Do you think your wife is happy and satisfied with being a homemaker? A. Yes B. No 14. What activities give you greatest satisfaction? Please rank order the following list with 1 being the activity that gives you greatest satisfaction. ( ) Being a husband ( ) Being a father ( ) Doing leisure activities ( ) Your job ( ) Being a volunteer ( ) Being a friend ( ) Doing housework ( ) Cooking ( ) Other 15. What is your present yearly income? A. 0 - 10,000 B. 11,000 ~ 20,000 C. 21,000 - 30,000 D. 31,000 - 40,000 E. 41,000 - 50,000 P. 51,000 - 60,000 G. 61,000 - 69,000 H. Over 70,000 16. If you are presently unemployed, what was your last yearly income? A. 0 - 10,000 B. 11,000 - 20,000 C. 21,000 - 30,000 D. 31,000 - 40,000 E. 41,000 - 50,000 F. 51,000 - 60,000 G. 61,000 - 69,000 H. Over 70,000 Thank you very much for your time and consideration. APPENDIX F PEARSON CORRELATIONS (PPMCC) 159 xuoz om. u mo. u a me. u a mo. u a mo. u a He. u a mwsmwmuuwmm woe n.6HHno .No. u me. u a .mo. u a NH. u a me. n a mo. u a cmumanwwmww .Ho. u mo. u a so. u H mm. u a om. u H mm. u a “mommhmwuwm «mo. u .vo. u a «So. u a mo. u a no. u a «A. u a mo< n.6Heno mo. u mo. u a .mo. u a .mo. u a «me. u a mm. u d uwspma mam .mo. n me. u a mo. u a on. u H mm. u a cu. u a umnboz mum nosumm emcee: Manama Macao: nonumm monuoz oszmstm quqqomezou ozH>oq .masouo HH< How mouoom ceaummuuom 0cm ouaumumufiq on» ca UmflmflucovH moumflum>oo xflm coozaom Auozmm. mcofiumHouuou comumom .H.< magma 160 xuoz . n . u . u . n . u . u ow vmcuzuom me u a so a am a we a .mo 6 we a macho: can: woe m.paaso . >Hasmm ca " Q o " Q 0 .III Q o " Q o " Q o III Q . . 3 mm cm 2. am 3 $363.10 more . n . u . u . u . u . u cameo no me u d ow n a mm I H mm u a OH I a .Ho I a macho spasm mm. u a me. u a me. u a Hm. u a me. u a no. u a woe m.oeeco mo. u m cm. u a ca. u a ca. u a av. u a mm. u a umsumm mmm Hm. u a Hv. n a om. u a me. u a me. u a mm. u a macho: mam umnumm nacho: umnbmm nacho: umnume nacho: eszmstm equqomezoo ezH>oq .mmfiaesma Honuoz mcfixuoz UHfiLU mam: mow mouoom coauaoouom mam musumuouflq one . :fi omflMfiucmnH maumflum>oo xwm cmmzpmm Aouzmmv mcoflpmflouuoo comummm .~.< wanna 161 xuoz be. u a me. u a .mo. u a be. u e um. u a mm. u a mwnmwuumwmm woe m.oaanu .Ho. u a .Ho. u a me. u a He. u a me. u a mm. u a .cmuwwwmwquM me. u a Na. u a be. u a as. u a om. u a He. u a umnmmemwumm no. u m cm. u a mm. u a «N. u a .mo. u a we. u a one m.6~eno em. u H mm. u a mm. u a mm. u m cm. u a ov. u a umcbmm mam «moo. u a «we. u a as. u m cm. a d me. u a me. u a nacho: mmm monumm “mayo: Honumm Hague: Monumm “mayo: eszmstm equqomazou ezH>oa .mofiaasmm mozuoz mcfixuoz GHHLU mamemh new museum cofiuaouuom mam ouzumHOHHa one ca voflmHucomH moumfium>ou xwm comzpom Auuzmm. mcowumaouuou comumom .m.< magma 162 ago: «a. u a «a. u a av. u a ow. u a me. u a «a. u mwnmwmumwmm woe m.oaaeu 2. u a ..m. ... a .3. u a 2. u a Z. n a om. u cmuwmmwwocw ma. u a Na. u a no. u a av. u a om. u a am. u umoWWamwhwm ow. u a mm. u a ha. u a mo. u a me. u a ma. u wee m.eaano ea. u a pm. u a «ac. u a mm. u a .mo. u a mo. u umsama mum mm. u a ma. n a we. u a me. u 6 ca. u a mm. u uwzaoz mum umnama emcee: assume Congo: Canaan umnaoz ezammazsa ozaqqomezou 62a>oa .mmaaaema nmnaoz mcaxuozncoz UHfizu mam: How monoum coHaQoOuom new muaumumuaa on“ :w OmwmaucooH moamaum>oo xwm coozumm Auuzmmv mcofiumaouuou comumom .v.< magma 163 xuoz «mo. u a mo. u a ev. u a mv. u a mN. u d mv. u a mwnmwmumwmm mm< m.maano oN. u a mN. u a ma. u a oN. u a mN. u a aa. u a cmuwwmmwwmww .mo. u d NN. u a mN. u a mm. u a vN. u a av. u a umoWMamwnwm no. u a .mo. u a mN. u a mm. u a oa. u a MN. u a woe n.6aano «ao. u a mo. u a vm. u a ma. u e «vs. n 6 av. u a umcuma mum no. u a me. u a NN. u a aN. u a aa. u a mm. u a umcuoz mum umnamm nacho: panama nacho: umsume umnuoz ezammazam 02aaqomezoo 62a>oq .mmaaaemm umnaoz mcaxuozlcoz maanu mamsmm mom mouoom cowaaouumm mam ousumumuaq may Ca UONHflucmvH mannaum>oo xwm coozuom auozmmv macaumaouuou comummm .m.< manma 164 d 0 .aoo. u a .aoo. n ma. u m cm. u a ma. u a Nv. u a .852 m S H .aoo. u a «aoo. u om. ... a N. u a Em. u a mo. u a uofioz m D m ma. u m cm. u .aoo. u a .aoo. u a ma. u a N. u a 353 m 0 ..1 om. u a N. u ..aoo. u a .aoo. n a N. u d av. u a 350: m w n1 ma. u a 3. n ea. u a N. u a ..aoo. u d ..aoo. u a ..Bfimm m I N Nv. u a mo. u N. u a ev. n a ..aoo. u a ..aoo. u a .550: 9 monumm nosuoz umnumm umcuoz Manama Honuoz oszmHZDm OquqomBzou 02H>OJ .mmaouo aa< How monoum :oHvamuumm cmmzuom auuzmmv mCONHMHOHHOU comummm .o.< wanes BIBLIOGRAPHY BIBLIOGRAPHY Ainsworth, M.D. The development of mother-infant interac- action among the Ganda. In Determinants of Infant Behavior, 1963, 2, 67-112. Edited by B.M. Foss. New York: Wiley. Ausubel, D.P.; Balthager, E.E.; Rosenthal, I.; Blackman, L.S.; Schpoont, S.N.; and Welkowitz, J. Perceived parent attitudes as determinants of children's ego structure. Child Development, 1954, 25, 173-183. Augar, A.; Bronfenbrenner, V.; and Henderson, C.R., Jr. Socialization practices of parents, teachers, and peers in Israel: kibbutz, moshav and city. Child Development, 1977, 48, 1219-1227. Bauducri, R. The effect of mother's employment on the achievement, aspirations and expectations of the child. Personnel and Guidance Journal, 1967, 46, 263-267. - Baruch, G.K. Maternal influences upon college women's attitude toward women and work. Developmental Psy— chology, 1972, 6, 32-37. Birnbaum, J.A. Life patterns, personality style, and self esteem in gifted family oriented and career committed women. Ph.D. dissertation, University of Michigan, 1971. Bommarito, J., ed. Tests and measurements in child devel- opment--A handbook. San Francisco: Jossey-Boss, Inc., 1971. Borg, W., and Gall, M.S. Educational research: An intro— duction. New York: Longman, 1977. Bowlby, J.A. Attachment. New York: Basic Books, 1969. Bronfenbrenner, V. Some familial antecedents of responsi- bility and leadership in adolescents. In Leadership and Interpersonal Behavior. Edited by L. Petrullo and B.M. Bass. New York: Holt, Rinehart and Winston, 1961. 165 166 Burchinal, L.G. Personality characteristics of children. In The Employed Mother in America. Edited by F.I. Nye and L.W. Hoffman. Chicago: Rand McNally, 1963. Callahan, S.C. The workingymother. New York: The Macmil- lan Co., 1971. Campbell, J.D., and Radke-Yarrow, M. Person perception of children. Merrill-Palmer Quarterly, 1963, 2, 57-72. Combs, A.W.; Richards, A.C.; and Richards, F. Perceptual psychology. New York: Harper and Row, 1976. Curtis, J. Working mothers. New York: Doubleday, 1976. Devereaux, E.C., Jr.; Bronfenbrenner, V.; and Suci, G.S. Patterns of parent behavior in the United States of America and the Federal Republic of West Germany, a cross-national comparison. International Social Science Journal, 1962, 13, 488-506. Devereaux, E.C.; Shouval, R.; Bronfenbrenner, V.; Rodgers, R.R.; Kav-Venaki, 5.; Kiely, E.; and Karson, E. Socialization practices of parents, teachers, and peers in Israel: The kibbutz vs. the city. Child Development, 1974, 25, 269—281. Douvan, E. Employment and the adolescent. In The Employed Mother in America. Edited by F.I. Nye and L.W. Hoffman. Chicago: Rand McNally, 1963. Droppleman, L.F., and Shaefer, E.S. Boys' and girls' reports of maternal and paternal behavior. Journal of Abnormal and Social Psychology, 1963, El, 648-654. Dubin, R., and Dubin E.R. Children's social perceptions: A review of research. Child Development, 1965, fig, 809-838. DuValle, E.W. Child-parent social distance. Sociology Society Research, 1937, 31, 458-463. Emerich, W. Family role concepts of children ages six to ten. Child Development, 1961, 32, 609-624. Etaugh, C. Effects of maternal employment on children: A review of recent research. Merrill-Palmer Quarterly, 1974, 39(2), 71-98. Ferguson, L.R.; Partyka, L.B.; and Lester, B.M. Patterns of parent perceptions differentiating clinic from non- clinic children. Journal of Abnormal Child Psychology, 1974, 2 169-181 I 167 Finkelman, J.V. Maternal employment, family relationships, and parental role perception. Ph.D. dissertation, Yeshiva University, 1966. Franke, H.B. A comparison of perceived parental behavior characteristics of eighth grade children of working and non-working mothers. Ph.D. dissertation, Univer- sity Of Florida, 1972. Glueck, S., and Glueck, E. Working mothers and delinquency. Mental Hygiene, 1957, 21, 327-352. Gold, D., and Andus, D. Developmental comparisons between ten-year old children with employed and non-employed mothers. Child Development, 1978, 12, 75-84. Goldin, P.C. A review of children's reports of parent behavior. Psychological Bulletin, 1969, 11, 222-236. Harris, B.D., and Tseng, S.C. Children's attitudes toward peers and parents as revealed by sentence completions. Child Development, 1957, 2Q, 401-411. Hartly, R.E. Children's concepts of male and female roles. Merrill-Palmer Quarterly, 1959-1960, 2, 83-91. . Current patterns of sex role: Children's perspec- tives. Journal of the National Association of Women. Deans and Counselors, 1961, 25, 3—13. . A developmental view of female sex-role defini- tion and identification. Merrill-Palmer Quarterly, 1964, i2, 3—16. Hastorf, A.H.; Schneider, D.J.; and Polefka, J. Person perception. Reading, Mass.: Addison-Wesley, 1970. Hayward, R.S. The child's report of psychological factors in the family. Archives of Psychology, New York, 1935, No. 189. Hoffman, L.W. Mother's enjoyment of work and effects on the child. In The Employed Mother in America, 95-105. Edited by F.I. Nye and L.W. Hoffman. Chicago: Rand- McNally, 1963. Hoffman, L.W., and Nye, F.I. Working mothers. San Francisco: Jossey—Bass Publishers, 1974. Hollingshead, A.B. Class differences in family stability. In Class, Status and Power. A Reader in Social Stratification, 184-191. Edited by R. Bendix and M. Lipset. New York: Free Press of Glencoe, 1953. 168 Howell, M.C. Effects of maternal employment on the child- (11). Pediatrics, 1973, 52(3), 327-343. Ince, R. Children's and parents' interpersonal perceptual style and children's adjustment. Ph.D. dissertation, Michigan State University, 1979. Jersild, A.T. Child psychology. Englewood Cliffs, N.J.: Prentice Hall, Inc., 1968. Kagan, J. The child's perception of the parent. Journal of Abnormal and Social Psycholggy, 1956, 53, 157-158. Kagan, J., and Lemkin, J. The child's differential percep- tion of parental attributes. Journgl of Abnormal and Social Psychology, 1960, 91, 440-447. Keppel, G. Design and analysis: A researcher's handbook. Englewood Cliffs, N.J.: Prentice Hall, Inc., 1973. Kirk, R. Experimental design: Procedures for the behav- ioral sciences. Belmont, Calif.: Brooks/Cole, 1968. Livesley, W.J., and.IBrom1eyg D.B. Person perception in childhood and adolescence. New York: John Wiley and Sons, 1973. Maccoby, E.E., and Feldman, S.S. Mother attachment and stranger-reactions in the third year of life. Morrographs of the Society for Research in Child Development, 1972, 3111) (Serial No. 146). Mahler, M. The psychological birth of the human infant. New York: Basic Books, Inc., 1975. Marantz, S.A., and Mansfield, A.F. Maternal employment and the development of sex-role stereotyping in five to eleven year old girls. Child Development, 1977, fig, 668-683. McCord, J.; McCord, W.; and Thurber, E. Effects of mater- nal employment on lower class boys. Journal of Abnormal Psychology, 1963, El, 177-182. Mead, M. Working mothers and their children. Childhood Education, 1954, 21, 66-71. Michaels, G.Y.; Messe, L.A.; and Stollak, G. Seeing paren— tal behavior through different eyes: Exploring the importance of person perception processes in parents and children. Paper presented to American Psychologi- cal Association Meeting, San Francisco, California, 1977. 169 . Relationships among children's perceptions of parent behavior, parents' inferences of their chil- dren's perceptions and parents' self perceptions. Paper presented to Midwestern Psychological ASS meet- ing, Chicago, Illinois, 1977. Morgan, P.K., and Gaier, E.D. The direction of aggression in the mother-child punishment situation. Child Development, 1956, 21, 446-457. Mott, S.M. Concept of mother: A study of four and five year old children. Child Development, 1954, £2, 99-106. Moore, T.W. Effects on the children. In Working Mothers and Their Children, 105-124. Edited by S. Yudkin and A. Holme. London: Michael Joseph, 1963. . Stress in normal childhood. Human Relations, 1969, 2;, 235-250. . Children of full-time and part-time mothers. International Journal of Social Psychiatry, 1964, Special Congress Issue No. 2. Nolan, F.L. Effects on rural children. In The Employed Mother in America, 122-124. Edited by F.I. Nye and L.W. Hoffman. Chicago: Rand-McNally, 1963. Nye, F.I. The adjustment of adolescent children. In The Employed Mother in America, 133-141. Edited by F.I. Nye and L.W. Hoffman. Chicago: Rand-McNally, 1963. Nye, F.I., and Hoffman, L.W. (Eds). The employed mother in America. Chicago: Rand-McNally, 1963. Piaget, J. Theypsychology of intelligence. London: Rontledge and Kegan, 1950. Propper, ARM. The relationship of maternal employment to adolescent roles, activities and parental relation- ships. Journal of Marriage and the Famil , 1972, 33, 417-421. Rabin, A.I. Growing up in the kibbutz. New York: Springer, 1965. Roe, A., and Siegelman, M. Parent-child relations ques- tionnaire. Child Development, 1963, fig, Roland, A., and Harris, B. Career and motherhood. New York: Human Sciences Press, 1979. 170 Rosenthal, D., and Hansen, J. The impact of maternal employment on children's perceptions of parents and personal development. Sex Roles, 1981, 1,593-598. Schranger, S., and Altrocchi, J. The personality of the perceiver as a factor in person perception. Psycholo- gical Bulletin, 1964, fig, 289-308. Schaefer, E.S. Children's reports of parental behavior: An inventory. Child Development, 1965, §§, 413-424. Serot, N.M., and Teevan, R.C. Perception of the parent- child relationship and its relation to child adjust- ment. Child Development, 1961, 23, 373-378. Siegel, A.B., and Haas, M.B. The working mother: A review of research. Child Development, 1963, 21, 513-542. Siegelman, M. Evaluation of Bronfenbrenner's questionnaire for children concerning parental behavior. Child Development, 1965, fig, 164-174. Spitz, R.A. Hospitalism: An inquiry into the genesis of psychiatric conditions in early childhood. Psycho- analytic Studies of the Child, 1945, 1, 53-74. Stolz, L.M. Effects of maternal employment on children: Evidence from research. Child Development, 1960, 3;, 749-782. Taguiri, R. Person perception. In The Handbook of Social Psychology, 3. Edited by G. Lindzey and E. Aronson. Reading, Mass.: Addison-Wesley, 1969. Vogel, S.R.; Broverman, I.K.; Broverman, D.M.; Clarkson, F.E.; and Rosenkrantz, P.S. Maternal employment and perception of sex roles among college students. Development Psychology, 1970, , 384-391. Von Mering, F.H. Professional and non-professional women as mothers. Journal of Social Psychology, 1955, 2;, 21-34. Warr, P.B., and Knapper, C. The perception of people and events. London: John Wiley and Sons, 1968. Woods, M.B. The unsupervised child of the working mother. Development Psychology, 1972, g, 14-25. 171 Yarrow, M.R.; Scott, P.; DeLeeuw, .; and Heinig, C. Child-rearing in families of working and nonworking mothers. Sociometry, 1962, 15, 122-140. Yudkin, S., and Holme, A. Working mothers and their children. London: Michael Joseph, 1963. "111111111111111.1111“