' 44.. 4242 44% 22!!! '1' 4 'V - r ‘*-" ' . - .‘ v - o ——. . V 4»: .- .._- . _¢_.‘_: "'4 . -- 4. ‘ “‘7‘ A . - 4 -. -. a "' ..-. '2"‘ _ L. _-a~———— v " ' .— .2" _‘ _. --- .__._..._.' -‘_“"‘J7 - 2' w— -~—s— h.“— \.4‘ ‘ “"" w- 2.. ,- . I4' 4! ! , 2 4!... 2! 4,2,! ' ’2 222224 2 44 t '2 2 22 22‘!“ 2 22422222222 42 2! !H ‘4'!!! 242;! 22 424222! ! 4! 2222 22222 244 444 ! 2 r2; 2'2 '444 lW4 ! 224222 22224! 7 2 4 44. 2444 444! 4;]! 2,! 2l 42444122 ‘2'!‘42' .on — o w--— .~— 4. -_4 v w ._——‘- ' o _ ‘ .49 A... ‘.-— !!22!22, !2l!4 22!!!! H422!!! ‘! 42 22. " 24222 222' ll! ,0 I. A. . L 24,, ,4 4 4! 2!!!! '2!!! 24 I .4 II '2. 42,24!II‘, .444 2!! '4' . 422‘ , III :24III42II,I!,:I42!.12.4, 2 .me 222!!! IIIIIII . 2222222222222.1222!I" 2,I,I 424444 .42- ,II .4 2 2224 ,44I I24I442 I2!!!!! 4 .444. . 2.414 4442.442 ' 2:2[4 '22! H 1 2I 222 22222 H ‘2 "'2' 2'2 “H!!! 22 E! 4.44»: 22212.2'4. I 4'24}!!! .2 4,1,: 4:222 4 1"4 2'4 !2 2| 2 "41-2 !!.’ - 2' " 2! 22 II III I .4‘42222!2222! 4142*!!! III! )IIIIIII II 4! III L2,, . .232!!!“ 22; !L24 2222!!!!“ 2!! 92!!!! 2W2!!!~!4}g :.I . 2!!,.24!44,I!I21!, ! 4 4!!!! [2,!!! , 2:42:42; ,2!4!2!44!222 I2! 2222 J!!! !!!! :3 ‘ 4, 2“ H 2!! 442222,! 4!”! ‘42! I"!!! " 22!“? 42!: :4!4I!4 42! 22442.4 232222! I. 2 222112!!!HHHWH2HHN42222222 H!!! H 22222 24422 22' H2 22244 2 I 4 . 2"2224242 .12 44. ;,.424 2 '4 4 £1? ‘9 M*"_‘ WAKE: £4: .- -. L"'v-T'l-L ‘ .42.. _ . .4 3,4; 3: 4 4.2: 442444" 4 ,2" 24.42242 2222!!!! 1241!!!! 4. ' .' 42:?! I: .‘ 44!,422444 244422224444'V'5 "4* 2.444 .4;3, .‘ 3 4' 242242 2222‘ I!!! I‘.~~4:»2 ‘27.“.224 lllljllllfllllflljlllllllll"WNW”!!!llllHlIIllll 10392 462 1 4‘ u.- .--“ This is to certify that the dissertation entitled THE RELATIONSHIP OF SELF-ESTEEM AND LIFE SATISFACTION TO ATTACHMENT BETWEEN ELDERLY MOTHERS AND THEIR MI DDLE -AGED DAUGHTE RS presented by Cheryl Debra Steindel has been accepted towards fulfillment of the requirements for Ph . D. degree in Psychology 7,7 Major professor / // Date [‘2‘ [y (b / / MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 RETURNING MATERIALS: MSU Place 1‘ n book drop to LlBRARlES remove this checkout from n your record. FINES will be charged if book is returned after the date stamped below. M)?” w] 9M 2 7 1997 I THE RELATIONSHIP OF SELF-ESTEEM AND LIFE SATISFACTION TO ATTACHMENT BETWEEN ELDERLY MOTHERS AND THEIR MIDDLE—AGED DAUGHTERS By Cheryl Debra Steindel A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1981 ABSTRACT THE RELATIONSHIP OF SELF-ESTEEM AND LIFE SATISFACTION TO ATTACHMENT BETWEEN ELDERLY MOTHERS AND THEIR MIDDLE-AGED DAUGHTERS By Cheryl Debra Steindel This study explored affective aspects of the relationship between the elderly mother and her middle-aged daughter by examining the perspectives of both members of the attachment dyad. Congruence between each subject's perceived and preferred mother-daughter relationship and agreement between the mother's and her daughter's perceptions, preferences and ratings of the quality of their attachment were expected to be significant correlates of well-being. It was specifically hypothesized that 1) life satisfaction would be positively correlated with self-esteem and 2) life satisfaction and self-esteem would be negatively correlated with the extent of discrepancy a) between each subject's perceived-ideal attachment and b) between cross-generational comparisons of the mother's and her daughter's views and ratings of their relationship. The subjects were 25 mothers, 65 years or over, and their daughters, aged 40 to 64. Respondents were Caucasian, noninsti- tutionalized volunteers. Subjects independently completed the Rosenberg Self-esteem Inventory, Life Satisfaction Index and three Cheryl Debra Steindel questionnaires devised by the author, i.e., the Personal Inventory, Attachment Scale and Relationship Rating Semantic Differential. Each subject was then interviewed about her mother-daughter re- lationship. Internal consistency of the instruments had been established in a pilot study with middle-aged and elderly women. Three of the hypotheses were confirmed, but only for the daughters: Life satisfaction was positively associated with l) self-esteem and 2) negatively correlated with a) incongruent views of the perceived-ideal relationship and b) discrepant cross-genera- tional ratings of relationship quality. The only significant finding obtained for the mothers was against prediction. Mothers' self-esteem was positively associated with cross-generational dif— ferences in the mother's and her daughter's perceptions and preferences of their relationship. The differential findings between mothers and daughters were discussed in regards to the role demands, developmental issues, cohort effects, expectancies and degree of psychological differentiation which distinguishes women in middle age and senescence. Methodological limitations of the research design, instruments and sample selection were also examined. It was con- cluded that affectional ties between middle-aged daughters and their mothers are important, but further research must be con- ducted to better understand the specific significance of attachment to each dyad member. to Mom and Resa, who continue to discover and share with me, the trials and triumphs of being female in our family ACKNOWLEDGEMENTS As a teacher, Dr. Albert Rabin, chairperson of my committee, has contributed to my intellectual growth, supporting my explora- tion into research and clinical areas that were of interest to me. As a friend, he has always demonstrated his warmth and caring con- cern. I have valued his guidance throughout my professional and personal development. Dr. Gilbert DeRath continues to touch me by his Openness, his generosity and his ability to make me laugh, especially at myself. He has clearly entered my heart and I have cherished our professional association and friendship over the years. Dr. Lucy Ferguson's gentle guidance and encouragement helped me persevere when I felt most discouraged. However, I have appreciated not only her support, but also her contribution in instilling in me an interest in finding the questions, yet to be asked, as well as the answers. I feel fortunate to have had Dr. Ray Frankmann on my committee. He patiently provided statistical guidance, but more importantly, inspired a curiosity and comfbrtableness with methodological research issues. And finally, a special thanks goes to Linda Cohen who always seemed to provide a clear eye, a steady hand and a warm heart whenever I felt adrift and lost in uncertainty. For her professional colleagueship and cherished friendship over the years, I am deeply grateful. iv TABLE OF CONTENTS LIST OF TABLES. INTRODUCTION. REVIEW OF THE LITERATURE. Socialization and Dependency The Mother- -Daughter Bond. Dependency and Role Reversal. . Interdependency and Filial Maturity . Autonomy in Later Life. . Attachment and Family Relations . . Generational Perceptions and Preferences. STATEMENT OF THE PROBLEM. METHOD. Subjects. Instruments . Instrument Reliability. Sample Recruitment. . . Procedure with each Subject . Statistical Analysis. RESULTS . Additional Results. DISCUSSION. Re-examination of Attachment in Adulthood . Psychological Differentiation between Mother and Daughter . Similarities between Mothers and Daughters . . Methodological Issues . Implications for Future Research. Page vii 10 12 15 18 22 3O 33 33 38 47 48 49 50 52 55 66 66 71 77 80 83 SUMMARY. REFERENCES . APPENDICES . A Personal Inventory. B Attachment Scale. . C Relationship Rating Semantic Differential. . . D Rosenberg Self— esteem Inventory . E Life Satisfaction Index . F Interview Schedule. . . G Introductory Letter and Consent Forms . H Cross- Generational Statistical Compari- sons Using the Attachment Scale . I Results of Cluster Analysis of the Attachment Scale. . . J Additional Statistical Findings . K Glossary of Variable Abbreviations. vi Page 86 89 100 100 104 114 116 117 119 120 123 124 127 138 Table 10 11 12 LIST OF TABLES Comparison of Demographic Variables for the Research and National Sample of Women 55 Years and Over. Comparative Demographic Data for the Pilot and Research Samples. Chart of the Attachment Scale . Alpha Coefficients for Determining Reliabilities of the Instruments for the Pilot and Research Groups . Correlation of the Discrepancy between Perceived and Preferred Attachment with Self-esteem and Life Satisfaction . Correlations between Discrepancy of the Cross- Generational Views of Attachment and Self- esteem and Life Satisfaction. Correlations of Discrepant Mother-Daughter Ratings of their Relationship with Self-esteem and Life Satisfaction . Summary of the Results of the Hypothesized Relationships for Mothers and Daughters . Means, Standard Deviations and Two Tailed T-Test Comparisons of Research Variables between Mother and Daughter Samples . Two Tailed T-Test Comparisons between Mother and Daughter Groups regarding Components of the Attachment Scale. Cross-Generational Statistical Comparisons using the Attachment Scale. Results of the Cluster Analysis Performed on the Attachment Scale. vii Page 34 36 41 44 S3 53 54 56 58 59 123 124 Table 13 14 15 16 17 18 19 20 21 Statistically Significant T-Tests between Mothers and Daughters on Items of the Relationship Rating Semantic Differential. Statistically Significant T-Test Comparisons between Mothers and Daughters on the Attachment Scale . T-Tests of the Components of the Attachment Scale fer the Total Sample . Correlations between Age, Life Satisfaction, Self-esteem and Components of the Attach- ment Scale for Mothers . Correlations between Age, Life Satisfaction, Self-esteem and Components of the Attach- ment Scale for Daughters . Correlations for Mothers between Age, Life Satisfaction, Self-esteem and Attachment Components "To Seek, Get, Feel". Correlations for Daughters between Age, Life Satisfaction, Self-esteem and Attachment Components "To Seek, Get, Feel". Correlations between Mothers' and Daughters' Perceived Generational Similarity and Selected Variables . Correlations between Mothers' and Daughters' Discrepancy Scores on the Attachment Scale and Relationship Rating Scale viii Page 127 128 130 132 133 134 135 136 137 INTRODUCTION According to popular beliefs, the advent of contemporary, urbanized society has disrupted intergenerational relations, leading to the dissolution of extended family ties and the aban- donment of elderly relatives at a time when the older person's need for support is especially great because of the decline associated with aging (Bengtson G Treas, 1980; Stehouwer, 1968; Troll, 1971; Weishaus, 1979). Population demographics are interpreted as con- tributing to this bleak characterization of the family of later life. Statistics reveal an increase in the life span of parents and a decline in the birth rate, resulting in fewer children being available as resources for their aged parents (Donahue, 1969; Moss, 1969; Streib 8 Shanas, 1965; Troll, Miller G Atchley, 1979). Laslett (1979) has referred to this commonly held dis- paraging view of the contemporary family, and the corresponding glorification of the aging family of the past as the "world we have lost syndrome." Popular belief to the contrary, an extensive number of studies have not borne out this negative picture of intergenera- tional relations in middle and later life (Depner G Ingersoll, 1980; Shanas, 1968; Spark G Brody, 1970; Sussman, 1965; Townsend, 1965, 1968; Turner, 1975). Research demonstrates that the modified extended family is the prevalent kinship unit in modern Western society. Furthermore, residential propinquity, regular and fre- quent contact, acknowledgment of reciprocal obligations and an exchange of services across generations is characteristic of interactions between adult children and their elderly parents (Hill, 1965; Riley 8 Foner, 1968; Shanas, 1968). The responsibility for the physical and psychological maintenance of the family rests primarily with female lineage members (Hess G Waring, 1978; Troll, 1971; Troll 6 Smith, 1976). By virtue of their positions as designated kinkeepers (Wilen, 1979), mothers and daughters are assumed to have a strong affec- tive attachment with continuing contact and expectations of reliance on the other for aid, advice and emotional support across the life course (Cohler G Grunebaum, 1981). Clinical evidence suggests that the changes associated with aging necessitate important adjustments in parent-child relations in later life (Silverstone, 1979). These renegotiations of the parent-child relationship can, but do not inevitably, lead to conflict between the generations (Steinman, 1979). The older parent is faced with the possibility of increased financial, physical and social losses which, in turn, can threaten her self-concept. The ways in which the elderly deal with these developmental issues can likewise influence their relationship with their children. Correspondingly, the daughter, often middle-aged or post-retirement age herself, is at that point in her life course where she is coping with her own transition from middle to old age, as well as with the problems facing her elderly parents (Boyd, 1969; Brody, 1977; Cohen 8 Gans, 1978; Neugarten, 1969, 1970). Consequently, issues of attachment, especially centering around dependency, autonomy and interdependency in intergenerational relations, have drawn the attention of social theorists and aging family members (Bengtson, 1979; Clark 6 Anderson, 1969; Field, 1972; Hirschfield 6 Dennis, 1979; Spark & Brody, 1970). However, while the literature clearly documents the instrumental aspects of family fUnctioning in adulthood such as flow of aid, services, etc., comparatively little research is available regarding the affective nature of mother-daughter bonds in middle and later life (Streib, 1980; Troll, 1971; Troll, Miller G Atchley, 1979). The investigations that have been conducted on emotional ties past mid-life are scanty in number and have led to contradictory theorizing and findings (Cohler G Grunebaum, 1981; Schwartz, 1979; Spark & Brody, 1970). These inconsistent results have been attributed to the lack of integrated formulations of personality development in the middle and later stages of the life course (Cohler, 1980; Neugarten, 1968) and to the existence of culturally biased models of family adjustment which implicity value autonomy (Cohler G Grunebaum, 1981). In addition, researchers have tended to con- ceptualize attachment primarily as a characteristic of the indi- vidual, but not of the parent-child linkage itself (Troll 8 Smith, 1976). Typically, subjects are drawn from either the middle or older generation and are asked to comment on aspects of their familial relations. Another commonly used methodology is a cross-sectional design which compares the views of subjects from different generations who are not members of the same family. This practice of collecting data from only one member of the parent—child dyad or from unrelated members of different generations has limited the applicability of findings on affective family relations in later life (Troll, Miller 8 Atchley, 1979). The following study was an attempt to explore affective aspects of the relationship between the elderly mother and her middle-aged daughter by examining the perspectives of both members of the attachment dyad. Therefore, this research differed from the typical investigations in this area by comparing both the mother's and the daughter's experiences of their relationship. Our fbcus was on 1) the components of attachment in adulthood that can be labeled "emotional supplies" (e.g., seeking and getting con- tact, approval, advice, affection), 2) ratings of the quality of the mother-daughter relationship, and 3) the association between these emotional factors, self—esteem and life satisfaction. More specifically, mothers and daughters were asked to indicate what they a) perceive the pattern of exchanges of emotional supplies between them to be and b) what they would ideally prefer the pattern to be. It was proposed that the individual's self-es- teem and satisfaction with life would be negatively related to l) the amount of disparity between the subject's perceptions and preferences of her mother-daughter attachment and 2) the cross—gen- erational discrepancy between the mother's and her daughter's a) views and b) ratings of the quality of their relationship. REVIEW OF THE LITERATURE Socialization and Dependency The paucity of research on the affective aspects of inter— generational interactions in adulthood is especially striking given the presumed importance of family relations to personal and social adjustment across the life course (Bengtson 8 Treas, 1980; Herr 8 Weakland, 1979; Meissner, 1964; Treas, 1975). However, theoretical discussions of the evolution of dependent and inde- pendent relations within the family context can be found in the socialization literature (Lindgren, 1969). The infant's reliance upon a caretaker for satisfaction of biological and emotional needs is believed to impel the child to be attentive to and thus, learn from his/her providers (Cohler 6 Grunebaum, 1981). In other words, the ability of parents to influence their child is based upon the offspring's need for his/her parents. Therefore, the child's dependency upon the parent enables socialization to occur, providing the means by which the individual acquires culturally valued beliefs and standards and is integrated into the family and larger societal group (Mussen, Conger G Kagan, 1974). Some social learning theorists propose that dependency is a multidimensional construct and consequently, should not be treated in a global, undifferentiated way (Yarrow, 1972). Instead, 6 dependent behaviors can be distinguished according to the specific functions they serve. In this regard, dependency tends to be broken into two components--the need fOr love and security and the need for information and advice. These categories have been re- ferred to in the literature as "emotional" versus "instrumental" dependency (Sears, 1972) or "emotional" versus "cognitive" dependency (Kastenbaum 8 Cameron, 1969). Parens and Saul (1971), working from a psychoanalytic framework, draw similar parallel formulations. They distinguish between "libidinal" or "affectional" dependency based on the need for protection and security, and "ego-developmental" dependency defined as the reliance upon others for advice and interpretation of external reality in the aid of mastery. Maccoby and Masters (1970) contend that while theoretical distinctions are frequently made between classes of dependent behaviors, the two types are often difficult to differentiate in the research setting. Regardless of the specific theoretical approach or terminology being used, it is agreed that the initial helplessness of the child leads to a psychological dependence on the parent. As the child matures, his/her dependency decreases. Still, it is theorized that elements of that bond persist well beyond the period of actual infantile helplessness (Bengtson 8 Black, 1973). The mother, in particular, is represented as the earliest, most significant figure in the child's life, establishing her as an important source of cognitive and emotional supplies in the years to follow (Cohler, 1980; Goldfarb, 1969). The Mother-Daughter Bond The emotional bond between generations rather than material or instrumental assistance is regarded as central to the support of family ties (Brody, 1966). Women have been recognized as the persons most responsible for maintenance of these affective bonds in this and other societies (Troll 6 Smith, 1976). By virtue of this responsibility, they have been referred to as the ”designated kinkeepers" (Wilen, 1979). This linkage between females across generations is explained, in part, as a result of the socialization of women into traditional "affective, passive and dependent types of sex roles (Seelbach, 1977)." Cohler and Grunebaum (1981) argue that socialization experiences of females, unlike that of males, promote the daughter's dependence upon her mother--an affectional orientation which per- sists in adulthood. Social learning research corroborates that dependent behavior in childhood is discouraged in boys, but encouraged in girls, and is related to adult dependency in women, but not in men (Mischel, 1966). Furthermore, children learn to identify with the same sex parent. This identification process and the continuing dependence of the woman upon her mother is believed to contribute to the strengthening of the mother-daughter bond. As Cohler and Grunebaum (1981) articulate, the adoption of a gender identity like that of her mother fosters this dependence, for it increases the congruence in interests and life-style between mother and daughter. While the son must work hard at learning to become a man like his father and must renounce passive wishes earlier learned in his relationship with his mother, the little girl is encouraged to become even more like the original anaclitic object in her life (p. 21). Following a similar line of reasoning, Chodorow (1978) speculates that mother's and daughter's affectional and relational orienta- tion leads to a closeness and lack of separateness that does not characterize the relationship between sons and their fathers or mothers. In other words, women are taught to be sensitive to and respond to the emotional needs of other family members. Males, on the other hand, interact in a more autonomous and instrumentally oriented manner which promotes greater individuation. As was mentioned, the affectional ties between mother and daughter are expected to continue across the life course. Mutual socialization continues (Bell, 1968) with the mother helping the daughter's transition into the roles of wife and parent and the daughter introducing the mother into the grandparental role (Robertson, 1975). With advancing age, shared kinship responsi- bilities further contribute to the maintenance of the emotional bond and mutual reliance between women across the generations. Moreover, demographic analysis reveals that older men are likely to be married, while older women are more likely to be widowed (Gillaspy, 1979). This differential statistic has implica- tions for the mother-daughter relationship. After the loss of the spouse, patterns of mutual aid between children and parents are often altered such that older widows grow closer to their daughters (Troll, Miller G Atchley, 1979). In commenting 10 upon the saliency of the mother-daughter bond in adulthood, Cohler and Grunebaum (1981) declare, Most important of all, the continuing contact between adult women and their own mothers, further enhanced by adult socialization into the parental role, suggests a degree of closeness that has important psychological consequences for each generation. The relationship between adult women and their own mothers is perhaps the most complex and emotionally charged of all the relationships in the family (p. 24). Therefore, to summarize, the mother-daughter bond is con- sidered to be one of strong affective intensity. Rooted in early reliance of the young daughter upon the mother, this relation is regarded as persisting across the life course. It is explained as the result of 1) learning experiences reinforcing dependency in women, 2) cultural norms socializing women into positions of shared family responsibilities, 3) identification processes resulting in less psychological differentiation and 4) the statistical proba- bility that the mother will be the surviving parent in the later years. Dependency and Role Reversal While dependency is considered a normal, adaptive mode of interacting between mother and infant, it is often regarded criti- cally as inappropriate, anxiety provoking, immature and/or patho— logical when manifested within the context of adult relationships (Bowlby, 1969; Bumagin 8 Hirn, 1979; Schwartz, 1979). As Field (1972) explains, it is generally accepted in our culture that dependence of the adult child upon his parents, 11 as well as the dependence of the parent on his grown children, is conducive to an unwholesome situation (p. 57). Despite the negative associations to dependency in adults, dependency—based relations are considered powerful determinants of attachment across the life span (Bowlby, 1969). Goldfarb (1965) goes so far as to conclude that reliance upon others for cognitive and emotional resources and for confirmation of self-worth is the primary motivational basis underlying interactions between most adults in our society. A dependency based relationship labeled "role reversal" was the model of aging family functioning that dominated the literature in the 19405 and 19505 (Gelfand, Olsen 6 Block, 1978). Role reversal was founded on a more negative portrayal of aging as a degenerative, regressive phenomenon leading to greater depend- ence of the parent upon the child. The deterioration of the aged necessitated a switching of generational roles such that adult offspring became the caretakers of their needy, now child-like parent. This process was regarded l) as the inevitable, irrevers- ible deve10pmental task facing all families (Blenkner, 1965) and 2) as a disruptive force to intergenerational relations. Rautman (1962) epitomizes this view when he comments, . as the years of life pass, the ravages of time shift the once clear distinction between the roles of child and parent until those lines cross, so that they resemble nothing so much as fate's gigantic hourglass fallen on its side . . . communication between these two generations becomes increasingly hampered, if not completely inhibited by the inability of either of the two groups to accept this reversal of role, for the 12 reversal does, after all, require an almost com- plete reorganization of their entire 'Weltansu- hauung' (pp. 117-118). Today, case study and anecdotal material attests that aging families continue to interpret the parent-child relationship as one in which generational roles are reversed (Hirschfield 8 Dennis, 1979). However, scientists have challenged the notion of role reversal as a normal and inevitable consequence of aging (Spark 6 Brody, 1970). Two alternative proposals of Optimal parent-child relations in the later years are l) interdependency and 2) mutual autonomy. It would be beneficial to examine these models and then discuss attachment as a construct incorporating different theoretical perspectives of aging family adjustment. Interdependency and Filial Maturity Within this framework, the "healthy" family is described as a unit of interacting personalities that adjusts to the fluc- tuating needs and social demands of its members (Blenkner, 1965; Spark 6 Brody, 1970). Although the declining capacities of older relatives are seen as precipitating shifts in the family homeo- statis, the generations are not regarded as reversing their posi— tions as parent and child (Bumagin 8 Hirn, 1979; Goldberg 8 Deutsch, 1977). Blenkner (1965) has proposed that changes in family rela- tions in later life can be understood as the result of the adult child's attainment of a psychological state of "filial maturity" vis-a-vis his/her parents. Elaborating upon psychodynamic 13 personality formulations, filial maturity is conceptualized as a transitional stage between genital maturity and senescence. In describing this process, Blenkner writes, the filial crisis may be conceived to occur in most individuals in their forties or fifties, when the individual's parents can no longer be looked to as a rock of support in times of emotional trouble or economic stress, but may themselves need their off- spring's comfort and support. Successful accomplish- ment of the filial tasks, or performance of the filial role, promotes filial maturity which has its own gratification (p. 57). Therefore, successful resolution of the filial crisis is dependent upon the child relinquishing infantile views of the parent as omni- potent and immortal, and adopting a more realistic outlook of the parent's functioning (Verwoerdt, 1976). Blenkner's theoretical stance focuses on the child's revi- sion of expectations and perceptions. However, the achievement of filial maturity presupposes that parents reappraise their own capabilities, plus personal and relational expectations. In other words, a child can attain filial maturity, only if the parent is willing and/or able to depend more upon his/her offspring. Case study material indicates that this is not always the case (Cohen 6 Gans, 1978). An older parent can rigidly adhere to an independ- ent life-style to the point of denying increasing infirmities and the need for support from others. Empirical investigations of interdependency in the adult child-aged parent relationship have resulted in paradoxical findings (Weishaus, 1979). Reciprocity has been well documented as the established flow of assistance between generations (Hill, 1965). 14 However, continuing family relations are not necessarily associated with adaptive emotional functioning. Lowered morale has been cor- related with continued family responsibility among older persons, especially women (Hess 6 Waring, 1978; Kerckhoff, 1966; Maas 8 Kuypers, 1974), and among middle generation offspring (Silverstone, 1979). Lowered morale has also been linked to greater family contact (Kerckhoff, 1969; Mancini, 1979), and to higher filial ex- pectations, i.e., adult's obligations to meet parent's needs, among low socioeconomic elderly (Seelbach & Sauer, 1977) and also among upper status aged (Brown, 1960). These findings were not entirely supported by Mancini (1979) who reported higher morale with those aging parents who felt they contributed financially to their children and generally understood them. Moreover, family contact has been fOund to be both positively related to strengthened generational bonds and unrelated to measures of adjustment (Adams, 1967; Brown, 1974; Troll 8 Smith, 1976). Therefore, while mutual dependency is the prevalent type of exchange of services, its relationship to affective aspects of the parent-child bond is not clear-cut. There is some evidence that autonomous fUnctioning with limited intergenerational interactions is more directly related to positive morale and satisfaction. Still, this association is not corroborated in all studies in this area. In his review, Mancini (1979) reported the contradictory results that older parents have been found to have higher morale when they are more satisfied with family life in general and feel 15 closer to their children as well as when they feel more distant from their offspring, both geographically and emotionally. Mancini attributes these conflicting findings to the rudimentary level of our knowledge about aging family relations. Autonomy in Later Life The relationship of autonomy to family functioning in middle and later life has received attention in the gerontology literature. Autonomy is highly valued in this society (Clark, 1969). It is regarded as an indication of maturity in adulthood (Goldfarb, 1969), as a component of life satisfaction and self-es- teem in the elderly (Clark 8 Anderson, 1967; Cohen 8 Gans, 1978) and a correlate of positive intergenerational relations (Donahue, 1957; Field, 1972). As Clark and Anserson (1967) explain, A good relationship with children in old age depends, to large extent, on the graces and autonomy of the aged parent--in short, on his ability to manage gracefully by himself. It would appear that, in our culture, there simply cannot be any happy role reversals between the generations, neither an increasing dependency of parent upon child nor a continuing reliance of child upon parent (p. 275). Their conclusion is consistent with research in which older persons report preferences to live separately and function autonomously (Troll, Miller & Atchley, 1979), but expect continuing contact and support from their children when needed (Britton, Mather G Lansing, 1961; Hawkinson, 1965; Streib, 1975; Streib 6 Thompson, 1960; Treas, 1979). 16 Kastenbaum and Cameron (1969) compared the perceptions of three generations (young adult, parent, grandparent) regarding cognitive and emotional dependency in their family. The oldest generation saw themselves making fewer dependency demands on pe0ple upon whom they make any demands at all compared with other genera- tions. The difference is appreciable (p. 47). In comparison, other generations reported a higher number of dependency relations. Middle-generation respondents perceived themselves almost equally depending upon others and having others depending upon them. This outcome is congruent with the portrayal of the middle-aged as the "sandwich generation" or the "generation between" (Hirschfield 8 Dennis, 1979). The youngest group re- ported being in a slightly smaller number of dependency relation- ships than their parents, but endorsed similar reciprocal patterns of interactions. These generational differences were manifested in percep- tions of affectional exchange within the family as well. The parental and child group felt the flow of affection was balanced across the generations. The older respondents indicated they gave more emotional support than they received. These results led the authors to speculate that the elderly they sampled were either highly independent or, more probably, had a greater preference to picture themselves as autonomous than their younger kin. Johnson and Bursk (1977) substantiated positive affective correlates on parent-child relations and independence in adulthood. In another one of the few studies tapping the views of both 17 generations, they discovered that older parents and their children rated their relationship higher when parents were viewed as phys- ically and financially self-reliant and as having positive atti- tudes toward aging. Most of the appraisals of the relationship were positive and shared by dyad members. However, some observers of American culture have questioned what they perceive as an almost obsessive concern with the ideal of autonomy (Cohler 8 Grunebaum, 1981; Spark 8 Brody, 1970). The equation of individualistic self—reliance with morality (Clark, 1969) contributes to perceptions of increasing dependence producing tension, leading to denial or self-degradation among the adult child (Kuypers G Trute, 1978) and the aged parent (Kalish, 1969). Cohler (1980) has proposed a re-evaluation of assumed normative relations in middle and later life since models of parent-child relations do not adequately explain the high degree of interdependency in intergenerational relations, despite cultural prescriptions and some research findings which suggest a correla- tion between autonomy and adjustment. Cohler and Grunebaum (1981) write that much of the confusion surrounding the study of the relationship between mothers and their daughters is a result of the fact that this relationship is often viewed in terms of some ideal of individ- nation and autonomy which differs from the reality of family life in contemporary society (p. 25). 18 Attachment and Family Relations The emphasis on a progression from childhood dependency to increasing self-reliance in adulthood is also evident in attach— ment theories of personality development that use the mother-child bond in early years as the prototype for personal relations across the life course (Antonucci, 1976; Bowlby, 1969; Kalish 6 Knudtson, 1976). Stated broadly, attachment refers to the intense affec- tional ties one person forms for another specific individual. The class of behaviors identified as being representative of attachment are similar to those associated with dependency (Maccoby 8 Masters, 1970). However, unlike the negative connotations associated with the notion of dependent relations, attachment behaviors are not regarded as pathological when manifested past childhood (Bowlby, 1975). The term dependency is also criticized for being impre- cise and nonperson specific, i.e., it makes no reference to the individual to whom the dependent behavior is directed. Therefore, attachment encompasses behaviors deemed impor- tant to mother-child bonding, such as dependency and autonomy, without evoking the criticisms associated with the concept of dependency. Consequently, attachment is being proposed as an alternative conceptualization of human relations which incorporates and integrates different theoretical models (e.g., role reversal, mutual dependency, autonomy) (Kalish, 1976). Bowlby (1969), representing the ethological viewpoint, identifiedtflmeunderlying objectives of attachment behavior. These objectives are maintenance of proximity to, reduction of distance 19 from, and social engagement with, the attachment figure fer the biologically determined purpose of seeking protection. Sroufe and Waters (1977) recommend a revision of Bowlby's principles by stressing that the critical mediator of behavior is the individ- ual's subjective experience of security in relation to the attach- ment figure. Ethological formulations assume that the innate need for care and support facilitates the infant's attachment upon the mother. Subsequent well-being and adjustment in later years is traced to the quality of the primary attachment bond (Antonucci, 1976). Although the organization of attachment behaviors may differ when manifested in mother-infant as compared to aged mother-adult child interactions, the same goals of infantile attachment are believed to be operative in later years (Knudtson, 1976). Hartup and Lempers (1973) suggest that literature on family relations in later life has focused on parameters of interaction (e.g., proximity, frequency/forms of aid, contact, support and affective patterns) that are analogous to components of attachment deemed as crucial to mother-infant bonding. Troll and Smith (1976) applied a dyadic attachment analysis to family functioning. These researchers generated three independ- ent family attachment scales--proximity (contact frequency and residential contiguity), family integration (e.g., family reunions, possession of mementos) and affect. The affective scale encom- passed perceived intraorganismic factors (e.g., feelings of obliga- tion, responsibility, indebtedness, approval, acceptance and 20 admiration of Other) and interpersonal variables (e.g., quality and strength of relation, amount of influence). In their preliminary testing of the scales on a small, non- representative sample of graduate students, they found that affec— tive ties contributed more to feelings of attachment to family members than proximity factors. They tentatively concluded that dyadic affect overrides separation and distance in the maintenance of family attachment. It is interesting to note that the affective attachment scales used by Troll and Smith (1976) are similar to the opera- tional measures of cognitive and emotional dependency employed in Kastenbaum and Cameron's (1969) study. In the latter work, respondents were asked to indicate 1) which people confide in, seek approval from and ask advice from them, 2) how much, and 3) reciprocally, who the subjects then go to for a confidante, approval and advice. Moreover, the similarity can be extended to Knudtson's (1976) suggested representation of the dimensions of attachment as emotional components (love, like); an evaluative- emotional one (approve, accept, admire, want to be like); a desire for proximity (would seek out); and one relating to issues of power and interpersonal emotional effectiveness (expect support and get approval from). It would be interesting to greatly extend the latter power dimension not only in terms of reciprocity (would give support and approval to), but also in terms of amount and quality of perceived influence, trust, mutual self-disclosure and other forms of bidirectional risk-taking (p. 189). Therefbre, the concept of attachment, initially applied to mother-infant interactions, has evolved from its original 21 formulation of proximity seeking in early childhood, to a broadened conceptualization of interpersonal relations across the life course (Antonucci, 1976). Kalish and Knudtson (1976) concur that attach- ment has greater explanatory power than adherence to a particular, optimal model of aging. Stated more simply, the attachment con- struct enables a variety of behaviors—-autonomy, dependency or interdependency--to be represented within a relational context. Borszormenyi-Nagy and Spark (1973) have emphasized that individual actions cannot be analyzed apart from the family unit. They observe that one of the most important dimensions of the family is a hierarchy of obligations and expectations to which all members of the system are indebted and committed. This inter- generational ledger of debts and injustices, referred to as "invisible loyalties," constitutes the motivational underpinnings of individual and relational behaviors. As they declare, Our concept of relational autonomy pictures the individual as retaining a modified, yet fully re- sponsible and sensitively concerned dialogue with the original family members. In this sense, the individual can be liberated to engage in full, wholly personal relationships only to the extent that he has become capable of responding to parental devotion with concern on his part, and with the realization that receiving is intrinsi- cally connected with owing in return (p. 105). With this reasoning, it follows that dependent and independent behaviors are both the result of readjustments of expectancies and commitments within and across generations. 22 Generational Perceptions and Preferences The question arises how these family preferences and per- ceived exchanges affect the mother-daughter attachment in later years. As has been mentioned, the relationship between filial behavior, family contact and morale is not necessarily a direct one. Parent and child can perceive their interactions as being inconsistent with personal preferences or standards. Furthermore, relationships might be expected to differ as a function of the amount of intergenerational agreement between perceptions and idealized standards or desires of both the parent and the child. Hawkinson (1965) examined the relationship between expect- ancies, preferences and practices of contact, financial support and living arrangements between parent and adult child, as reported by aged parents. He found a discrepancy between what the elderly wanted to transpire in their intergenerational relations and what they reported as actually happening. Subjects wished things were different in their relationship with their children (e.g., more intimate contact), but they had low expectations that interactions with their offspring would change in the desired direction. How- ever, this investigation was not designed to evaluate how these differences between personal standards and perceived practices affected his respondents' feelings about themselves or their relationship, nor how the parents' views might correspond to those of their children. The confusion and equivocal results surrounding modes of intergenerational relations in later life might reflect, to a 23 degree, researchers' failures to consider the perspectives of both the parent and the child when attempting to understand their interactions. For example, Simos (1970, 1973) focused exclusively on the adult child's views of his/her familial relations. Her subjects were most distressed by the psychological disturbances they perceived in their aged parents, especially their depression and isolation. However, the offspring also indicated that their parents did not necessarily echo their own beliefs that the restricted social activities of the older generation was problemat- ical. Advocates of a phenomenological approach claim that per- ceptions of family functioning in later life are more important determinants of generational interactions than actual differences in people's behavior due to age (Ahammer G Baltes, 1972; Bengtson 8 Black, 1973; Cryns G Monk, 1972). Thomae (1970) postulated a cognitive theory in which adjustment to aging is a function of intrapersonal and relational perceptions that are consonant with the dominant concerns of the individual. For Thomae, adaptation is assumed to involve the continuation of or attempts at restructuring the balance between cognitive and motivational structures within the individual. Relationship quality depends upon the differing cognitions and motivations of the interacting generation members. Extending this line of thought, adaptive family relationships might be characterized by the greatest congruence or balance of the concerns and preferences within the individual and across 24 generations. Reflecting upon the advantages of his model, Thomae declared that . by defining adjustment to aging by the principle of the balance between situation as perceived on the one hand and motivational state or structure on the other hand, we neither have to superimpose an 'ideal' or 'normal' pattern of aging, nor a classificatory system on the dif- ferent varieties (p. 8). Streib and Thompson (1960), applying a sociological frame- work, propose similar principles. In their review of the older person's relations, they discuss the affectional functioning of the family in terms of role expectations and reciprocity. Roles are defined by cultural norms and personal definition. Affec- tional relations are satisfying when expectations are understood and reciprocated by other family members. Reciprocation is regarded as enhancing The person's feeling of identity and thus, is meaningfu1 to the individual involved and also increases stable social relations in general. Thus, even bickering and quarreling, when analyzed from this standpoint, may be func- tionally meaningful and may not be an index of family disunity (p. 472). Implicit in Streib and Thompson's formulations is the idea that reciprocity is a function of mutual recognition. The indi- vidual must perceive that his/her expectations or desires are being met. Secord and Backman (1961) have applied the term "interpersonal congruency" to the condition where a person (P) perceives another individual behaving in a manner that confirms P's self-concept. Congruency is believed to enhance P's liking of the Other and bolster P's own self-regard. 25 The principle of "perceived similarity" in needs, which is independent of congruency, has also been linked to dyadic attrac- tion. Research substantiates that perceived similarity and con- gruency are significant correlates of interpersonal attraction in friendship (Secord G Backman, 1961) and associated with the satis- faction and adjustment of couples married for 50 or more years (Sporakowski G Hughston, 1978). In studying perceptions of change over time in the relationship between middle-aged children and their parents, Wilen (1979) demonstrated that the children, especially the daughters, were more likely to report changes, in either a positive or nega- tive direction, while the older respondents stressed stability or enhanced closeness. Wilen interpreted her findings in terms of Bengtson and Kuyper's (1971) concept of the "developmental stake." Stated briefly, this postulate assumes that interactions between generations are influenced by individual perceptions which, in turn, reflect the differential investment or intentionality of one generation in regards to the other. In other words, a person selectively attends to those elements of interactions which justify their own generational position as determined by develOpmental and cohort factors. Older persons emphasize affective stability in inter- generational relations because they expect they might need their daughters more as they age. Thus, they are invested in viewing that relationship as an available source of support. Another hypothesis is that cultural values of individualism and autonomy 26 impel the parent to minimize changes in personal and relational functioning, as a way of denying the possibility of increasing dependency. Daughters, on the other hand, assume more of the kin- keeper obligations as the family ages. This position of responsi- bility requires them to be sensitive to alterations in the require- ments of individual relatives and in the adjustment of the family as a whole. Peterson (1979) has also collected evidence that percep- tions of mothers and daughters change qualitatively over time. Although contact and fulfillment of family responsibilities did not fluctuate appreciably, daughters' affective ties for their parents declined gradually past mid-life. Daughters tended to feel better about their mothers when they perceived them as showing interest, demonstrating approval and openly communicating. The mothers continued to value their daughters highly over time. Steinman (1979) speculated that conflicts reactivated between aging parents and their children in later life are the outcome of discrepant generational preferences. From her stand- point, family disturbances frequently center about concerns over independence and dependency. The problematical family situation is one in which the parents' wishes to l) depend upon or not depend upon their children, and 2) to have their children rely upon or be autonomous to them, do not match their children's wishes. In this respect, Johnson and Bursk (1977) found that an independent rater judged parent-child relations to be poorer when generational views of the bond were more discrepant. 27 Discrepancy between parent-adult child perceptions and preferences has been proposed as a correlate of poorer psychologi- cal adjustment (Gottesman, Quarterman G Cohn, 1973), conflictual interactions (Stern 8 Ross, 1952) and lowered satisfaction with life (Weinberg, 1969). However, these hypothesized relationships have not been empirically tested. For the purposes of this investigation, preferences and perceptions were specifically examined in relation to the exchanges of emotional supplies that are assumed to be components of attach- ment. The attachment literature suggests that interactions rele- vant to the mother-daughter bond in adulthood involve 1) the woman seeking, i.e., initiating,and getting, i.e., receiving, such factors as attention, affection, contact, reassurance, advice, approval, etc. from the other member of the dyad and 2) reciprocally, having the Other seek and get attention, affection, etc. from her. Attach- ment was also conceptualized as being associated with such feelings as security, influence (power) and interest (involvement with) in relation to the other attachment figure. These patterns of emotional exchanges can be analyzed on two levels--the individual's preferences and perceptions. Pre- ferences refer to what the mother or daughter wishes the pattern would be; perceptions refer to what she actually sees the pattern to be. Presumably, persons who perceived that their desires for contact, reassurance, etc. were being realized would report greater satisfaction with themselves and life, in general, than persons whose preferences were not recognized as being fulfilled. 28 Wilen (1979) provides some examples from her study which clarify this point. In reaction to the tasks of "parent-caring" assumed by children, one daughter described her relationship with her mother as 'Our roles have completely reversed . . . Our relationship was excellent: loving, warm, easy, supportive. Now, it's problem producing with anxiety, hostility, and anger on my part. I think her relation to me is no less loving; dependence is the key word here . . . The big change was physical-~ending in unfortunate dependency (p. 7).' Another daughter, commenting on the situation with her mother said, 'Mom has always taken care of me, and now I feel that I can have a chance to take care of her. She likes that and so do I (p. 7).' Clearly, these two statements depict contrasting reactions to the daughter's changing relations with her aging mother. The daughters hold similar perceptions, but different evaluations of their mothers' growing dependencies. The first woman quoted has a negative outlook on her mother's dependency. This contrasts with the second woman who not only accepted her mother's increasing reliance, but implied she welcomed and enjoyed it. Although specu- lation, it seems that the situation of greater discrepancy is potentially more conflictual and associated with a decrease in the daughter's self-regard and ratings of her overall satisfaction with life. Therefore, if perceptions do not generally confirm ideal- ized standards, the individual may develop a more negative view of her parent-child relationship, which could be mirrored in a less 29 desirable self-image. That is, if personal preferences vary considerably from her perceptions of her interactions, the indi- vidual will also tend to experience greater dissatisfaction and a corresponding lowering of her personal esteem and life satisfaction. Reasoning further, lower self-esteem and life satisfaction would be correlated with the amount of disparity between the mother's and her daughter's l) preferences and perceptions of their mutual attachment and 2) their views of the quality of their relation- ship. STATEMENT OF THE PROBLEM Physical, economic, social and psychological changes associated with the later years can affect the relationship between elderly mothers and their adult daughters who are coping with their own issues of aging. This phase of life has been popularly con- ceived as one of increasing dependency of the parent upon the child. However, popular beliefs to the contrary, affective family rela- tions in middle and later life are complex and not clearly under- stood or adequately researched. Theoretical formulations of individual and family func- tioning in adulthood do not seem to account for the variability of intergenerational relations represented in the literature. Attempts to identify normative and adaptive modes of parent-child bonding in adulthood have been biased by implicit value judgments. In addi- tion, generalizations about intergenerational interaction have been made on the basis of findings gathered from either the older or the middle generation, but rarely from both members of the adult parent-child dyad. The concept of attachment in later adulthood provides a possible framework for incorporating the different models of aging family relations, i.e., role reversal, independence and inter- dependency. Some of the relevant components of attachment which are emerging from the life-span and gerontological literature 30 31 seem to be confiding in, and seeking contact, approval, reassurance or advice from the significant other dyad member and reciprocation of these factors. The present research proposed that there is not a single Optimal pattern of attachment. Instead, self-esteem and satisfac- tion with life were expected to be negatively correlated with the amount of discrepancy between each individual's preferences and perceptions Of the exchanges of emotional supplies in her mother-daughter interactions. Self-regard and life satisfaction were also expected to be negatively correlated with cross-genera- tional discrepancies between the mother's and her daughter's views of their relationship. The specific hypotheses are stated as follows: I. The extent Of discrepancy between each participant's individual preferences and perceptions Of the exchanges of the emotional supplies in her mother-daughter relationship will be negatively correlated with A: her self-esteem. B: her life satisfaction. 11. Self-esteem will be positively correlated with life satisfaction. III. The extent of cross-generational discrepancy between the mother's and her daughter's preferences and perceptions of their mutual attachment (pattern Of exchanges of emotional sup- plies) will be negatively associated with each woman's 32 A: self-esteem. B: life satisfaction. IV. The extent of discrepancy between the mother's and her daughter's ratings Of the quality of their relationship will be negatively associated with each woman's A: self-esteem. B: life satisfaction. METHOD Subjects Research sample.--The research sample of mother-daughter pairs were 25 elderly women, aged 65 to 91 years old, and their daughters, aged 40 to 64. The mean age for the mothers was 77.1 years and 49.4 years for the daughters. All participants were volunteer, mentally competent Caucasians who resided independently in Eaton and Ingham counties in Michigan. Table 1 provides a com- parison between the demographic characteristics Of the present research sample and a national sample Of Caucasian, noninstitu- tionalized women taken in the United States in 1978. Examination Of the table reveals that the research group had a higher mean in- come, mean years of education and greater representation in white collar occupational positions. The present sample also comprised more Protestants, which is representative of the communities from which subjects were recruited. Marital status, mean number of children and living arrangements seem comparable across the two groups. Pilot sample.--A pilot study was conducted 1) to determine the ease and clarity of administration of the test materials, 2) to compute the reliability, that is, the internal consistency Of each instrument, and 3) to refine tests, when necessary, until an over- all alpha coefficient of .80 or higher was achieved, meeting the 33 34 TABLE l.--Comparison Of Demographic Variables for the Research and National Sample of Women 55 Years and Over.* Class of Groups1 Information Research (1981) National (1978) Marital Status: Single 0% 2% Married 48% 50% Separated 4% 1% Divorced 12% 7% Widowed 36% 40% Mean Number of Children: 2.5 2.4 Religion: Catholic 14% 35% Jewish 20% 26% Protestant 58% 39% Other 8% 0% Mean Income: $28,320 $6,175 Mean Years Of Education: 14.3 11.0 Living Arrangements: Live Alone 38% 41% With Spouse 46% 42% With Relative 16% 16% With Nonrelative 0% 1% Occupation: Homemaker: 48% 30% White Collar: Professional 26% 13% Managerial/Administrative 12% 8% Sales 3% 8% Clerical 10% 29% Blue Collar: 1% 12% *Taken from Current Population Reports, Special Studies, Series P-28, #85, US Department of Commerce, Bureau of Census, August 1978. 1Research group based on combined sample Of mothers and daughters (ages 40-91) and national sample based on Caucasian, noninstitutionalized women (ages 55 and over). 35 minimum criteria for scale reliability suggested by Crano and Brewer (1973). Therefore, data Obtained from the pilot sample were used to improve the internal consistency of the test materials before they were administered to the research subjects. It was assumed that with this procedure, the reliabilities Obtained with the pilot group would be less than or equal to the reliabilities Obtained on the refined instruments given to the research subjects. Further discussion Of reliability will be presented in the sections on instrumentation. The pilot sample consisted Of a nonrandom group Of Caucasian, noninstitutionalized, mentally competent women. For this preliminary study, it was not required that the volunteers be paired members Of a mother-daughter dyad. However, two mother-daughter pairs did participate in the pilot project. Inclusion in the pilot was based on the following criteria: 1) Daughters are 40 to 64 years Of age with a mother who is a) 65 years of age or over and b) noninstitutionalized. 2) Mothers are 65 or over with a daughter who is a) between the ages Of 40 and 64, inclusive, and b) independently residing. The participants in the pilot were 30 elderly women with a mean age of 73.2 years and 28 middle-aged women with a mean age Of 49.1 years. The majority Of the Older females were widowed, 64%, compared to 7% of the middle-aged group and 46% of the elderly women were married compared tO 68% Of the younger group. The average years of education were 13.1 for the mothers and 16.7 for 36 the daughters. Table 2 summarizes the comparative demographic characteristics of mothers and daughters from the pilot and the research samples. TABLE 2.--Comparison of Demographic Variables for the Research and National Sample Of Women 55 Years and Over. Groups1 Pilot Research Class of Information M D M D T N_ 30 28 25 25 50 Age (in years) Range 65-87 40-63 65-91 40-63 40-91 Mean 73.2 49.1 77.1 49.4 63.2 Marital Status: Single 0% 0% 0% 0% 0% Married 46% 68% 28% 68% 48% Separated 0% 0% 0% 8% 4o Divorced 0% 25% 8% 16% 12% Widowed 54% 7% 64% 8% 36% Mean number Of Children: 2.4 2.1 2.3 2.7 2.5 Mean number Of Siblings: 5.2 3.8 1.3 1.3 1.3 Religion: Catholic 13% 7% 16% 120 140 Jewish 27% 32% 20% 20% 20% Protestant 57% 54% 56% 60% 58% Other 3% 7% 8% % 8° Years Of Education: Range 9-19 12-20 9-21 12-18 9-21 Mean 13.1 16.7 13.3 15.7 14.3 Years at Job: Range 0 1-27 0 1-30 Mean 8.7 8.4 Years Retired: Range 0-49 0-4 1.36 0 Mean 9.6 .69 13.7 Table 2.—-Continued 37 Pilot Research Class of Information M T Primary Source Of Income: 85 53% 0% 64% 32% Pension 17% 4% 8% % ADC 0% 4% 0% 0% Investments 17% 0% 0% 0% Own Salary 0% 25% 4% 18% Husband's Salary 3% 29% 16% 26% Combined Salary 3% 27% 8% 20% Other 10% 11% 0% % Income (in thousands): Range 5-45 7.2-80 6-60 Mean 19.35 28.32 Type of Residence: Own home/apartment 60% 4 % 70% Rent home/apartment 37% 5 % 26% Other 3% % 4% Living Arrangements: Live alone 53% 68% 38% With spouse 40% 28% 46% With relative 7% 4% 16% With nonrelative 0% 0% 0% Distance from Other: Different country 0% 0% 0% Different city 37% 16% 16% Same city 27% 56% 54% Same neighborhood 23% 20% 22% Same street 10% 4% % Same house 3% 4% % Amount Contact with Other: Less than once a year 13% 0% 0% Several times a year % 0% 0% Once a month 23% 12% 10% Several times a week 40% 40% 40% Daily 17% 48% 50% 38 Table 2.--Continued Groups1 Pilot Research Class of Information M D M D T Changes in mother-daughter relationship in: Past Year Yes 17% 32% 10% 48% 34% NO 83% 68% 90% 52% 66% Currently Yes 7% 21% 12% 28% 20% NO 93% 79% 88% 72% 80% Future Yes 7% 29% 0% 8% 4% NO 93% 71% 100% 92% 96% Years at Current Address: Range 1-64 3-30 1-42 3-33 1-42 Mean 22.4 17.0 16.6 13.9 15.2 Years in Community: Range 1-71 3-48 2-69 3-58 2-69 Mean 27.8 27.7 36.8 30.4 33.6 Self-rated Health: Poor 0% 0% 0% % 0% Fair 20% 11% 8% 16% 12% Good 53% 29% 68% 28% 48% Very Good 27% 60% 24% 56% 40% l M=Mothers D=Daughters T=Combined research sample Instruments Personal Inventory.--Two forms of the Personal Inventory (P1) devised by the author consisted of 31 questions for the mothers and 28 questions for the daughters (see Appendix A). The items on both forms covered demographic information (e.g., age, marital status, family size, living arrangements, income, education), 39 family relations (e.g., type and frequency Of contact), self-ratings Of health, and recent or anticipated changes in the mother-daughter relationship. The three additional items on the mothers' PI asked about the residential propinquity, type and frequency of contact with the respondent's own mother if she was still living. Attachment Scale.--The Attachment Scale (AS) (see Appendix B) devised for this study by the author was based on the interview format used by Kastenbaum and Cameron (1969) and the family attach- ment scales Of Troll and Smith (1976). The A5 was compiled by generating a number Of items represented in the literature as pre- sumed components of attachment. These items were submitted to five psychologists who independently judged if the items represented attachment variables. Only those items that a majority Of the raters selected as indicators of attachment were included in the scale. The A5 required that respondents rate on a Likert-type scale 1) how much they perceive certain emotional exchanges actually occurring in their mother-daughter relationship and 2) how much they would ideally prefer these transactions to happen. More specifically, the subject indicated how much 1) I seek contact, attention, reassurance, admiration, approval, acceptance, emotional support, tO confide in, advice, affection, understanding and comfort from the Other. 2) the Other seeks contact, attention, reassurance (see list above), etc. from me. 4O 3) I get contact, attention, reassurance, etc. from the Other. 4) the Other gets contact, attention, reassurance, etc. from me. 5) I feel secure, responsible for, I have influence over, interest in, I want to be like, obligated to and care for the Other. 6) the Other feels secure, responsible for, etc. for me. Choices on each item range from this happens "Not at All," 1, to "A Great Deal," 5. Subjects answered the 62 items comprising the six sections presented above two times--first, according to what they see actually occurring in their mother-daughter relationship and second, according to what they would wish or prefer to have happen in their mother-daughter relationship. Therefore, to summarize, there are a total of 124 items indicating what the individual perceives and prefers in reference to attachment to the Other. Scores totaled for each half of the AS can range from 62 to 310. Two forms Of the AS were generated, one for the mothers and one for daughters. The questions were identical in content and order of presentation, but references to "the Other" were speci- fically labeled "to mother" or "to daughter" in an attempt tO clarify test instructions. Table 3 provides a representation Of the type Of response categories being elicited from subjects and the comparisons between A5 sections which can be made. 41 TABLE 3.--Chart of the Attachment Scale I Other I Other I Other Seek Seeks Get Gets Feel Feels From From From From Toward Toward Other Me Other Me Other Me I actually Perceive I ideally Prefer For the purpose of hypothesis testing, only discrepancy scores were calculated for the AS. One discrepancy score, labeled PREFACT, represents the sum of the absolute value of the differences between what the subject actually perceives and what she would pre- fer tO seek, get and feel in her mother-daughter relationship. In other words, PREFACT signifies the amount of fit between the per- ceived and ideal relationship--the greater the incongruity, the larger the PREFACT score. Since discrepancy on each item ranges from 0 to 4, the total difference score can range from 0 to 248. The other discrepancy score, referred to as D-ATTACH, is the sum Of the absolute value Of the differences between the mother's views and her daughter's views Of what she a) perceives to be happening and b) prefers or wishes to have happen in her mutual relationship. Restated, D-ATTACH reflects the extent of cross-generational agreement. A higher D-ATTACH score is associated with greater disagreement between mother-daughter pairs. For a description of the specific comparisons made on the AS using cross-generational difference scores, the reader is referred to Table 11 in Appendix H. 42 Reliability coefficients computed for the AS on the pilot sample data was .97 for the mothers and .98 for the daughters. This is comparable to the alphas Of .97 and .98 obtained in the research sample for mothers and daughters, respectively. Table 4 provides a comparative summary of the reliability coefficients of all the test instruments Obtained with the pilot and research subjects. Because the alpha for the AS was high, a cluster analysis was performed on the combined data for mothers and daughters from the research sample to identify a possible underlying factorial structure. The results Of this analysis can be found in Table 12 in Appendix I. As Table 12 displays, 12 factors were discovered. The standardized coefficient alpha for the factors ranged from .41 to .98 except for Factor 12 which had an alpha of 1.00 because it was comprised Of a single item. Interfactor correlations ranged from -.13 to .78. Relationship Rating;Semantic Differentia1.--The semantic differential (RR) (see Appendix C) was used to assess attitudes toward the mother-daughter relationship. For the pilot study, a list of 40 bipolar adjective pairs was generated by using items identified by Osgood, Suci and Tannebaum (1957) as primarily sampling the dominant evaluative factor typically found in semantic differentials. For the research study, a revised form was made up of 20 bipolar adjectives selected from the original list used in the pilot. These scales were chosen because they were the 20 pairs which accounted for the most variance in the mothers' and daughters' 43 pilot groups. The reliability coefficient obtained for the final version of the RR was .98. This compares with the alpha of .96 Obtained for the pilot mothers and .97 for the pilot daughters (see Table 4). On the RR, subjects were asked to describe their current, overall relationship with the Other as being "extremely," ”moderately," "slightly," or "equally" characterized by such adjectives as relaxed or tense, secure or insecure, ugly or beau- tiful, etc. Scoring was based on assigning the number 7 to the favorable end of each bipolar scale and l to the unfavorable end. Positive and negative leading stem words were alternated to mini- mize possible effects of response bias. If the stem word to the left was positive, scoring preceded from 7 to 1 points on that scale; if the stem word was negative, scoring preceded from 1 to 7. For each individual, scale scores were summed to provide an overall rating of the relationship which can range from 20 to 140, with higher scores signifying more positively characterized mother-daughter relationships. A discrepancy score was also cal- culated (RRDIFF) and based on the sum of the absolute value of the differences between the mother's scale scores and the scale scores generated by her daughter. Each difference scale score can range from 0 to 6 with the composite discrepancy score ranging from 0 to 120. Greater differences between the mother's and her daughter's rating of the quality Of their attachment are reflected in larger discrepancy scores on the RR. 44 TABLE 4.--Alpha Coefficients for Determining Reliabilities Of the Instruments for the Pilot and Research Groups.1 Pilot 2 Research Instrument M Q :1: M D _T_ Attachment Scale .97 .98 .98 .97 .98 .98 Relationship Rating Semantic Differential .96 .97 .97 .98 .98 .98 Rosenberg Self-esteem Inventory .79 .82 .80 .88 .75 .82 Life Satisfaction Index-A 3 15 items .80 .79 .80 NC .70 .67 item 6 deleted .68 .68 .68 1 For a discussion Of Cronbach's alpha coefficient, the reader is referred to Nunnally, J., Psychometric theory. NY: McGraw-Hill, Inc., 1978. 2M=Mothers;{DeDaughtersprTOtal sample. 3NC=reliability could not be computed because responses to item 6 had zero variance. Rosenberg Self-esteem Inventory.--The inventory (RSEI) consists Of 10 statements, five dealing with a favorable attitude about oneself and five with an unfavorable self-attitude (Appendix D). Positive and negative statements are interspersed in an attempt to reduce the effects Of response set. The subject was asked to indicate her amount of agreement with the item, ranging from 1, "Strongly Disagree," tO 4, "Strongly Agree." Rosenberg (1965), the RSEI's originator, tabulated re- sponses by using a Guttman technique in which each item's score 45 was based on its relation to other items on the scale. Following Monroe-Cook's (1979) revised procedure, for this study, items were scored individually and then summed to obtain a single composite score which reflected general level of self-esteem. Scores can range from 10 to 40--the higher the score, the higher the individ- ual's self-esteem. The RSEI was originally intended for use with adolescents. However, Monroe-Cook (1979) administered the instrument to college aged males and their parents who were primarily middle-aged. He found that the reliabilities, as measured by alpha coefficient, were .80 or greater in both the student and parent samples. Similar reliabilities were Obtained in the present research study with a = .88 for mothers and .75 for daughters, averaging to .82 for the combined research sample. Life Satisfaction Index.--Life satisfaction was measured by employing items from the Life Satisfaction Index-A (LSI) originally devised by Neugarten, Havighurst and Tobin (1961) and then refined and modified by Adams (1969) (Appendix E). The LS1 required subjects to indicate whether they agreed, disagreed or were undecided (uncertain) about statements representing more positive or more negative attitudes toward one's life. Following Adams' suggestions, items endorsed in a positive direction were counted as 2, negative responses counted as 0, and uncertain responses were counted as 1. Scores were summed, ranging from 0, low satisfaction, to 30, high life satisfaction. 46 The 15 items included in this study were only those that Adams identified through factor analysis as specific dimensions of life satisfaction. He labeled these factors as Mood Tone (items 2 to 6, l4), Zest for Life (items 1, 7, 8, 9, 12) and Congruence between achieved and desired goals (items 10, ll, 15). A cluster analysis performed on the current research data combined for mothers and daughters revealed three factors which differed in composition from those identified by Adams. The present combina- tions included Factor 1, Optimism (items 3, 5, 7, 8, l3), Factor 2, Past Reflections (items 1, 6, 9, 10, ll, 15) and Factor 3, Affective Orientation (items 2, 4, 12, 14). The standard score coefficient alphas for Factors 1 through 3 were .65, .64 and .57, respectively. The correlations between factors ranged from .16 to .29. The alpha coefficient obtained for the research daughters was .70, but alpha could not be computed for the mothers because item 6 had zero variance. All the mothers disagreed that "Most of the things I do are boring or monotonous." When this item was deleted, a reliability coefficient of .68 was obtained for both the research mothers and daughters. Family lnterview.-—The interview (see Appendix F) was intended to supplement findings Obtained on the questionnaires. It provided subjects with an opportunity to comment upon their relationship in their own words. The interview schedule was based on a structured fOrmat of nine primarily open-ended questions. TOpic areas included perceived changes in the mother-daughter 47 relationship over time, positive and negative aspects of the dyadic bond, and areas of importance in the relationship that had not been addressed. While the interview provided mainly qualitative data, two questions contributed quantitative information. Respondents esti- mated 1) how similar their answers on the test measures were to their mothers' or daughters' (item 7) and 2) how alike their current mother-daughter relationship was to the emotional relation- ship that existed between the elderly subject and her own mother (item 9). Choices ranged from 1, not similar or alike at all, to 5, very similar or alike. Instrument Reliability The RR was the only instrument from which items were deleted for the research group on the basis of the responses given in the pilot study. The A5 and RSEI had an overall alpha of better than .80 satisfying the criteria fOr scale reliability (Crano G Brewer, 1973). In terms of the LSI, since the alpha obtained for the total pilot sample was .80 and deletion of item 6 did not appreciably alter the internal consistency computed for the re- search sample, the 15 item LSI was used in the statistical analyses of the predictions. Examination of Table 4 demonstrates that the alphas for the combined research sample of mothers and daughters were equal to or higher than the reliability coefficients Obtained for the pilot study with the notable exception of the LSI. 48 Sample Recruitment Subjects were obtained by the author 1) getting permission from the appropriate authorities to 2) attend senior citizen centers and women's church groups. At those gatherings, a brief description of the research, a statement of the criteria for in- clusion in the study, and a request for volunteers was presented. All interested persons placed their name, address and telephone number on cards. In addition, the volunteer was asked to give the name(s) and telephone number(s) of 1) her daughter or her mother (whichever was appropriate) and 2) other women they felt might be interested in participating. Within a week, the researcher recontacted the initial volunteer and her mother or daughter, confirming their intent to be subjects and arranging an appointment for testing. Subjects were phoned the day before the scheduled meeting to confirm the time and place. Two of the mother-daughter pairs who originally volunteered did not participate because of unexpected illnesses resulting in the sudden hospitalization of the elderly mothers. Four individuals whose names were supplied by a subject declined to participate in the study when telephoned. This represents a refusal rate of 3.7% for the entire pilot and research study. When either the mother or daughter declined to participate, it was explained to the person who was interested in volunteering that the study required the involvement of both dyad members. It was communicated to that person though that her concern and cooperation were appreciated. In addition, research results were made 49 available to those persons who expressed an interest, but were excluded from the study. Procedure with each Subject The purpose, the design and an outline of what was expected of the participants was explained to each subject at the initiation of testing. Respondents were also informed of their rights as volunteers (e.g., access to results, confidentiality) and were asked to sign a standard consent form for participation in research conducted under the auspices of the Department of Psychology, Michigan State University (Appendix G). The actual testing session consisted of two parts: self-re- port instruments and a brief interview. Written instruments were presented first, followed by the family interview schedule. All written materials were combined in a booklet and presented in the following order: 1. Introductory letter (Appendix G) 2. Research consent form 3. Personal Inventory 4. Attachment Scale 5. Relationship Rating Semantic Differential 6. Rosenberg Self-esteem Inventory 7. Life Satisfaction Index Following presentation of these measures, subjects were interviewed by the researcher using the schedule already discussed. Subjects were asked for a written consent to tape record the interview. One 50 subject refused consent to be tape recorded. With her agreement, her interview responses were written down by hand by the examiner. Data was collected from individuals in their residences or a place of their choosing. Three of the daughters were tested at their place of employment; the remainder of the subjects were seen in their homes. Mothers and daughters filled out the forms separately and were interviewed independently. The researcher was present throughout test administration. The test was admin- istered orally to those subjects who had physical limitations (e.g., arthritis, poor eyesight) and who requested questions be verbally presented. Testing and interviewing was completed in a single session. The average time to fill out the forms was 45 minutes with an average time of 20 minutes for the interview. No subject took longer than one and one-half hours to participate. At the end of each session, subjects were thanked for their cooperation and involvement. It was recognized that by partici- pating in this project, the subject might have personal issues raised about herself and/or her relationship with her mother or her daughter. One daughter in the pilot study indicated she desired assistance in discussing her evoked concerns. The author helped her contact the appropriate community and professional 4 resources . Statistical Analysis This research was designed as a correlational study involving continuous variables. Hypotheses were tested using the 51 Pearson product-moment correlation coefficient. In addition, t-tests of the difference between means was computed for selected variables to assess possible cohort differences between the middle-aged and elderly samples. RESULTS The correlations between self—esteem, life satisfaction and the discrepancy between the individual's perceived and ideal mother-daughter relationship yielded inconsistent findings. Hypothesis I stated that discrepancy between each individual's perceptions and preferences (PREFACT)1 of her mother-daughter bond will be negatively correlated with her self-esteem (S-ESTEEM) and life satisfaction (LIFE-S). This prediction was confirmed, but only with the daughters' life satisfaction (r = -.44, p < .01). Table 5 reveals that the hypothesized inverse relationship between the extent of discrepancy and the other measures was in the pre- dicted direction with the daughters' self-esteem and the mothers' life satisfaction. However, these correlations are low and not statistically significant. Hypothesis II postulated a positive relationship between self-esteem and life satisfaction. This prediction was supported, but for the daughters only (r = .43, p < .02). For the mothers, the correlation between self-esteem and life satisfaction approached, but did not reach statistical significance (r = .28, p < .09). Hypothesis III stating that cross-generational discrepancy (D-ATTACH) between the mother's and her daughter's perceptions and 1Appendix K provides a glossary of variable abbreviations. 52 53 preferences of their attachment will be negatively associated with each woman's self-esteem and life satisfaction was not confirmed. As Table 6 demonstrates, there was a predicted negative, but statistically nonsignificant correlation between the daughter's life satisfaction and discrepant views of the mother-daughter relationship. This correlation of -.28 approached a level Of significance with p < .09. TABLE 5.-—Correlation of the Discrepancy between Perceived and Preferred Attachment with Self-Esteem and Life Satis- faction. Group/PREFACT S-ESTEEM LIFE-S Mothers .12 -.17 Daughters -.25 -,44* *p < .01. TABLE 6.--Correlations between Discrepancy of the Cross-Generational Views of Attachment and Self-Esteem and Life Satisfaction. Group/D-ATTACH S-ESTEEM LIFE-S Mothers .54* .05 Daughters -.03 -.28 * p < .003. Further inspection of Table 6 reveals a significant posi- tive correlation between the mother's self-esteem and the 54 cross-generational discrepancy score (r = .54, p < .003). This strong relationship is against prediction. The correlations between discrepancy and mothers' life satisfaction and daughters' self-esteem were negligible. The prediction of Hypothesis IV was that the discrepancy between the mother's and her daughter's ratings of the quality of their relationship (RRDIFF) will be negatively correlated with each woman's self-esteem and life satisfaction. The hypothesis was confirmed with the daughters' life satisfaction (r==-a45, p < .01). As Table 7 shows, the correlations are low, but in the expected directions for the daughters with self-esteem and for the mothers' group with life satisfaction. The association between the mothers' self-esteem and discrepant ratings of the relationship is very weak. TABLE 7.--Correlations Of Discrepant Mother-Daughter Ratings of their Relationship with Self-Esteem and Life Satisfaction. Group/RRDIFF S-ESTEEM LIFE-S Mothers .03 -.12 Daughters -.17 -.45* 'k p < .01. The findings for the predicted relationships for mothers and daughters are summarized in Table 8. As the table shows, three of the hypotheses were confirmed, but only for the daughters. With the middle-aged group, life satisfaction 1) was inversely correlated 55 with an incongruent perceived-ideal mother-daughter relationship and discrepant cross-generational ratings of the relationship quality, and 2) was positively associated with self-esteem. Two trends in the hypothesized directions were noted. As expected, mothers' self—esteem was positively correlated with life satisfaction; daughters' life satisfaction was inversely correlated with dis- crepant cross-generational views (D-ATTACH). These findings approached (p < .09), but did not reach the minimum criteria Of significance of .05. One statistically significant finding was obtained for the mothers, but this was against prediction. Mothers' self-esteem was found to be positively correlated with cross-generational differences in the mother's and her daughter's perceptions and preferences of their attachment. Additional Findings During the course of this investigation, certain additional results emerged which are worth examining. They are presented below. Cohort Differences.--Two tailed t-test comparisons of independent means were computed to determine if there were signif- icant differences between the mothers' and daughters' groups. This analysis was performed with 1) demographic variables, 2) composite scale scores for each of the instruments, 3) section scores on the AS (see Table 3) and 4) an item by item comparison on each test measure. 56 TABLE 8.--Summary Of the Results of the Hypothesized Relationships for Mothers and Daughters. Hypothesis Results Direction Of Number Predicted Correlation Variables Mothers Daughters Ia Negative Self-esteem G PREFACT .12 -.25 lb Negative Life Satisfac— * tion 8 PREFACT -.17 -.44 11 Positive Self-esteem a Life Satisfaction .28 .43+ IIIa Negative Self-esteem 8 ** D-ATTACH .54 -.03 IIIb Negative Life Satisfaction G D-ATTACH .05 -.28 IVa Negative Self-esteem G RRDIFF .03 -.17 IVb Negative Life Satisfaction G * RRDIFF -.12 -.45 +p < .02. * p < .01. ** p < .003. The demographic characteristics of mothers and daughters, as measured by the PI, were compared. The following statistically significant differences emerged between the middle and older generations: 1) Mothers tended to be widowed, while daughters were married (t = 4.47, df = 24, p < .001). 2) Mothers lived alone, while daughters tended to live with a relative, most Often a spouse (t =-5.63, df = 24, p < .001). 57 3) The primary source of income was social security for the elderly sample and the husband's salary for the middle-aged subjects (t=- 7.77, df = 24, p < .001). 4) The daughters' mean income was significantly higher than the mothers' (t =-4.38, df = 17, p < .001). 5) Mothers were less likely than daughters to indicate that within the past year, something had happened to themselves or another member of the family which influenced their mother-daughter relationship (t =-3.06, df = 24, p < .005). When composite scale scores were examined, the only signif- icant finding was on the RR. Using the total overall rating of the quality of the relationship as the unit of comparison, mothers tended to characterize their relationship at a more positive level than their daughters (t = 2.82, df = 24, p < .01). Table 9 con- tains the means, standard deviations and t-tests for the RR, AS, RSEI and LSI. NO other comparison revealed statistically signifi- cant results. However, when the AS is examined in terms of its component sections (refer to Table 3), several significant differences were discovered between the generations. A summary of these results can be found in Table 10 and are described as follows: 1) Mothers see themselves getting more from their daugh- ters than daughters indicate they get from their mothers (t = -2.24, df = 24, p < .01). 2) Mothers see their daughters getting more from them than daughters indicate their mothers get from them (t = 2.63, df = 24, p < .02). 3) Mothers prefer or wish to get from their daughters more than daughters prefer to get from them (t = 2.30, df = 24, p < .03). 4) Mothers believe their daughters feel towards them more than daughters believe their mothers feel in regards to their daughters (t = 7.07, df = 24, p < .001). 58 5) Mothers prefer their daughters feel more for them than the daughters prefer their mothers feel (t = 2.89, df = 24, p < .01). TABLE 9.--Means, Standard Deviations and Two Tailed T-Test Com- parisons of Research Variables between Mother and Daughter Samples. Variable Mean S.D.1 df t p 2 Attachment Scale 4 Perceived relation- M:224.8 40.0 ship D:216.0 41.3 24 1.13 .269 Preferred relation- M:23l.0 37.3 ship D:224.0 39.9 24 .71 .485 PREFACT M:49.0 28.1 D:48.8 21.3 24 .05 .961 Self-Esteem (RSEI) M:33.4 4.1 D:33.8 3.5 24 -.46 .648 Life Satisfaction M:24.1 4.1 (LSI) D:23.8 4.1 24 .30 .764 Total Relationshi M:129.2 17.1 * Rating (RR) Score D:121.8 21.4 24 2.82 .009 * Meets minimum criteria for level of statistical signifi- cance, p < .05. 1S.D. = Standard deviation; df = Degrees of freedom. sum of items 1-62 on AS; sum Of items 63-124 on A5. 2Perceived relationship Preferred relationship 3Total overall rating of quality of relationship. 4M = Mothers; D = Daughters 59 TABLE 10.--Two Tailed T-Test Comparisons between Mother and Daughter Groups Regarding Components of the Attachment Scale. Component of the 1 Attachment Scale Mean S.D. t p I seek from Others M:36.4 14.3 D:34.4 8.1 .73 .475 Other seeks from me M:36.2 11.3 D:40.6 10.0 -l.68 .106 I get from Other M:49.2 9.4 * D:43.3 10.4 2.84 .009 Other gets from me M:51.6 7.6 * D:45.8 11.3 2.63 .015 I feel for Other M:25.0 5.0 D:27.4 5.9 -l.97 .061 Other feels for me M:26.4 6.3 * D:20.0 4.6 7.07 .001 I prefer to seek from Other M:39.8 13.5 D:39.5 11.1 .10 .923 I prefer Other seeks from me M:38.5 11.8 D:38.6 10.6 -.04 .966 I prefer to get from Other M:49.0 10.1 * D:43.8 9.1 2.30 .031 I prefer Other gets from me M:51.l 7.3 D:48.5 9.5 1.37 .184 I prefer to feel for Other M:26.l 4.4 D:26.0 5.3 .11 .911 I prefer Other feels for me M:26.6 4.6 * D:24.l 4.6 2.89 .008 Actually seek1 M:72.6 23.9 D:75.0 16.4 -.49 .628 * Meets minimum criteria for level of statistical signifi— cance of .05. 1Degrees of freedom =24. 60 TABLE 10.--Continued Component of the Attachment Scale Mean S.D. t p Actually get2 M:100.7 16.4 * D:89.0 20.3 3.06 .005 Actually £6613 M 51.4 10 6 * 0 47.4 9 8 2.46 .021 Ideally seek4 M:78.3 23.3 D:78.3 20.6 .03 .977 Ideally get5 M:100.0 16.7 0 92.2 18.1 1.98 .060 Ideally feel6 M:52.7 8.9 0 50.0 9.5 1.62 .119 Seek7 M:150.9 40.7 0 153.1 33.0 -.25 .805 Get8 M:200.8 27.5 * D:181.3 34.9 3.80 .001 Feelg M:104.1 18.7 * 0 97.5 18.1 2.41 .024 * Meets minimum criteria for level of statistical signifi- cance of .05. 1Actually seek = I seek + Other seeks. 2Actually get = I get + Other gets. 3Actually feel = I feel + Other feels. 4Ideally seek = I prefer to seek + I prefer Other to seek. 5Ideally get = I prefer to get + I prefer Other to get. 6Ideally feel = I prefer to feel + I prefer Other feels. 7Seek = Actually seek + Ideally seek. 8Get = Actually get + Ideally get. 9Feel = Actually feel + Ideally feel. 61 To summarize, the elderly generation tended to report getting and feeling more than the middle-aged group indicated. There were no significant differences between mothers and daughters on attach- ment variables involving seeking. For a more thorough investigation of possible cohort dif- ferences, t-tests were computed between the age groups for each item of all the measures. With the RR, the mothers' endorsements were significantly more positive than the daughters on half the scale items. These adjective pairs are presented in Table 13 in Appendix J. There were no statistically significant differences between the age groups on the RSEI items. When the LSI items were examined, three comparisons proved to be statistically significant. These results are 1) Daughters, more than mothers, feel their "life could be happier" (t = 2.14, df = 24, p < .04). 2) Mothers were more likely to indicate they have "gotten pretty much what they expected out Of life" (t = 2.30, df = 24, p < .03). 3) Mothers were more likely to "feel old and somewhat tired" than the daughters (t = -2.21, df = 24, p < .04). A t-test analysis of the AS items yielded significant findings on 27 of the 124 comparisons which were made. Table 14 in Appendix J presents these results. Components of Attachment.--Because the concept of attach- ment is only beginning to be applied to relationships in middle and later life, further statistical analyses were conducted on the components and correlates of attachment as measured by the AS. When the data from the mothers and daughters were combined, t-test 62 comparisons of the sections of the AS revealed certain patterns. These findings are displayed in Table 15, Appendix J. To sum- marize the significant results, there was a tendency for the sub- ject to indicate 1) I get less than the Other. 2) I prefer to get less than the Other. 3) Other gets more than she seeks from me. 4) I get more than I seek from the Other. 5) I prefer the Other gets more than she seeks from me. Pearson product-moment correlations were computed between the variables of age, life satisfaction, self-esteem and com- ponents of the AS. These findings are represented in Tables 16 to 19, Appendix J. The tables identify several statistically significant correlations at the .05 level or better. Positive relationships were found between I seek and Other seeks, I get and Other gets, I feel and Other feels, I wish (prefer) to seek and wish Other seeks, I wish to get and wish Other gets, I wish to feel and wish Other feels. Moreover,composite ratings of what the individual perceives in her mother-daughter relationship (ACTREL) is positively associated with subject's totaled pre- ferences (PREFREL). Put more simply, if the person sees more things occurring in her mother-daughter transactions, she also wishes to have more in her ideal mother-daughter relationship. The correlates of age yielded significant results, but primarily for the mothers. Mothers' self-esteem was inversely related to age; mother's age was positively associated with what 63 she indicated on the AS she gets, feels, wishes to get, wishes Other gets and wishes Other feels. Interestingly, the only signif- icant and positive correlation with the daughter's age was with the attachment variable "I feel toward mother." This corresponds with the finding that mother's age is positively associated with the mother's view on how the daughter "feels toward me." Ap- parently, for this sample, as the daughter ages, she acknowledges more feelings toward her mother--an event which is recognized by the older mothers. For daughters, life satisfaction was positively associated with the Other seeks, I get and Other gets, I feel and Other feels, wish I get, wish I feel, ACTREL and PREFREL. When the mothers' data are examined, significant positive relationships are revealed between satisfaction and I seek, wish I seek, wish Other seeks and PREFREL. The only significant correlate of self—esteem was found with the daughters' group and the attachment component of the Other seeks. Overall, higher satisfaction with life is associated with higher rates Of "seeking." Moreover, "getting" tends to be directly linked with "feeling." In addition, for the middle-aged group, life satisfaction was positively related to "getting" and "feeling." Perceived Similarity.--In the interview, subjects esti- mated the amount of similarity they expected between themselves and the Others' answers to the questionnaires. A two tailed t-test determined that mothers anticipate greater similarity 64 with their daughters than the daughters expected. This difference is significant at the .04 level (t = 2.21, df = 22). Participants also evaluated how alike their current mother-daughter bond was to the attachment that existed between the elderly mothers and their own mothers. NO significant dif- ferences were found between the mothers' and daughters' estima- tions when a t-test was computed. To further assess possible correlates of perceived genera- tional similarity, Pearson product-moment correlations were calculated between selected variables. These results, found in Appendix J, Table 20, demonstrate that mothers' and daughters' ratings of assumed similarity on the testing were positively related. Also, mothers' estimates of similarity were directly associated with both the mothers' and the daughters' life satis- faction. Likewise, daughters who rated themselves as more simi- lar to their mothers had higher life satisfaction. In this re- spect, mothers' and daughters' satisfaction with life were posi- tively correlated. There was also a significant positive correlation between mothers' and daughters' evaluations of the amount of "alikeness” between their current attachment and the ties between the mothers and their own parents. Greater perceived likeness of the mothers corresponded to them having higher self-esteem. Discrepancy Variables.--The interrelationships between the various discrepancy or difference measures (PREFACT, D-ATTACH, RRDIFF) were investigated (refer to Table 21, Appendix J). 65 Mothers' and daughters' disparity between each's perceived-ideal attachment (PREFACT) were positively related--greater discrepancy for the mothers was linked with greater discrepancy in their daughters' incongruity between actual and wished for ties. A direct correlation was discovered between PREFACT and RRDIFF, that is, the higher the perceived-ideal incongruence, the more mothers' and daughters' ratings of the quality of their relation- ship differed. This finding was significant for both age groups. Moreover, for the entire sample, a significant positive correla- tion existed between PREFACT and D-ATTACH. In other words, if there is a greater difference between what the individual sees and what she wants with the Other, there is also more cross-genera- tional discrepancy between the individual's and other attachment figure's views of their mutual bond. DISCUSSION While the results suggest that life satisfaction and self—esteem are positively associated for women in middle and later life, generalizations about the relationship of life satisfaction and self-esteem to mother's and daughter's perceptions, preferences and ratings of their mutual bond are harder to formulate. Overall, the findings reveal that attachment ties between mothers and daugh- ters are generally unrelated to how elderly women feel about them- selves and their lives, but significantly correlated with the daughters' satisfaction with life. A notable and unexpected finding is that mothers have higher personal regard when they and their daughters adhere to more disparate perceptions and pre— ferences of their attachment. Possible explanations for these generational differences in attachment will be discussed. In addition, methodological and other theoretical issues relevant to this research will be ex- plored. Re-examination of Attachment in Adulthood An implicit assumption underlying this present work was that the attachment bond past mid-life was equally meaningful for both members of the dyad. The effects of the relationship were treated as bidirectional, influencing both mother and daughter in 66 67 an identical manner and to a comparable extent. Put more simply, it was inferred that attachment had the same type and degree of impact on middle-aged daughters and their elderly mothers. The findings of this study clearly challenge the validity of this assumption, questioning the possible role of other factors mediating the attachment relationship in adulthood. Bowlby (1973) posited that affectional bonds in early childhood influence individual functioning throughout the life course. Yet, he also maintained that the attachment behavioral system is not always activated nor manifested in a similar fashion in each interpersonal situation (Bowlby, 1979). Stress, illness, anticipated or actual loss and separation are some of the vari- ables which contribute to the patterning and intensity of the attachment relationship. Thus, contextual factors are believed to influence the form and consequently, the experience of the mother-child bond in adulthood. Life stage theorists have long been interested in the identification of the developmentally related tasks which char— acterize each phase of life (Bischof, 1976). There is consider- able consensus that the issues of the middle-aged differ from those with which the older individual must contend (Cavan, 1965; Rogers, 1979; Sussman, 1965). As has been mentioned, the middle generation is involved in a multiplicity of roles and commitments vis-a-vis younger and older age groups, with women having the primary responsibility for maintaining family solidarity and continuity. These role demands have led researchers to view 68 middle-agers as the most overcommitted (Hess G Waring, 1978) and pivotal age group, serving as a mediator of generational trans- actions (Robertson, 1975). Erikson (1963) has theorized that the central psychologi- cal issue of the middle-aged individual is the achievement of "generativity." Generativity involves guiding succeeding genera- tions so that the middle-aged person makes a productive contribu- tion to humanity, thereby confirming the significance of her own, limited self-existence. Failure to attain generativity results in "stagnation," a narrow, narcissistic orientation about oneself and the world. Applying Erikson's principles further, the psychological task of significance to the older person is attainment of "ego integrity" versus "despair." The integrated individual reflects upon and comes to accept the inevitability of her death and the direction in which her life has unfolded. Those who have failed to establish a meaningful existence approach life's end with fear and dread. A comparison of ego integrity to generativity suggests that the older individual has had her opportunity to affirm her place in the social order and now, in the later years, is less dependent, on an intrapsychic level, on interactions with others to confirm self-worth. In contrast, a large component of the developmental functions and challenges of the middle-aged person is relational in nature, that is, integrally connected to success- ful negotiation of interpersonal commitments. 69 Reformulated in sociological terms, the older woman has fulfilled, with varying degrees of success, many of the role requirements of spouse and parent associated with her younger years (Hess 8 Waring, 1978). Conceivably, the previously shared function of kinkeeper may be increasingly delegated to younger lineage members. A shift in kinship duties may parallel changes in the mother's identification as primary provider of her family's emotional needs. Interestingly, some evidence has been gathered to suggest aged women who continue to provide a major source of support to younger generations have lower morale than older females with less familial duties (Cohler, 1980; Eckels, 1981). A differential sense of family responsibility between mother and daughter was clearly manifested by women in this study. The daughters consistently demonstrated they felt more respon- sible for and obligated to their parents than mothers felt in relation to them. Moreover, daughters preferred to feel respon- sible for their mothers--a desire tending to match the mothers' preference for their daughters to feel a sense of duty towards them. As one elderly respondent articulated, I don't feel responsible for my daughter at all. I paid my dues when I was raising the children, but M. is grown now and she has her husband and a family of her own to care for. As for myself, I'm pretty independent and don't need much. But if things should change, M would take care of me. That's only right. Daughters should feel responsible for their parents when they are not in a position to care for themselves. Therefore, in contrast to their mothers, middle-aged women's positions of responsibility promote greater sensitivity 70 (or vulnerability) to possible incongruences in the relational functioning of family members. In analyzing the impact of inter- generational perceptions on behavior, Bengtson and Kuypers (1971) declare, Difference is indicative of 1055. Loss is feared, hence the demand to counteract the feared loss (p. 257). Consequently, discrepancy, whether real or anticipated, may be a source of conflict or tension for the daughter who experiences a concomitant lowering of her life satisfaction. In general, satisfaction with life may be more susceptible to transitory changes in current functioning than self—esteem and thus, demonstrates a significant association with the discrepancy measures. Correspondingly, self-esteem represents a more global, stable index of self-concept (Robinson 8 Shaver, 1972; Tippett 8 Silber, 1965) and may be less likely to fluctuate with changes in relational ties. Overall, participants in this research tended to regard themselves highly, suggesting that subjects had more autonomous, integrated and adaptively functioning personalities (Wells 8 Maxwell, 1976). Self-systems characterized by such high self-es- teem are believed to act as "conceptual buffers for evaluative stimuli (Ziller, Hagey, Smith 8 Long, 1969, p. 84)." In other words, persons who tend to consistently think positively of them- selves can better tolerate negative, diverse or ambiguous feedback arising from social interactions without experiencing a deprecia- tion in personal worth. Given this reasoning, it would be more 71 likely that self-esteem would be negatively correlated with dis- crepant views of attachment in those persons 1) who, like middle-aged women, are potentially more vigilant of sources of interpersonal conflict (such as discrepancy) and 2) who generally experience lower levels of self-regard. To review, attachment ties appear to be differentially associated with mothers' and daughters' life satisfaction and self-esteem. These findings may, in part, reflect the role demands, developmental issues and life circumstances which distinguish middle age and senescence. In addition, salient aspects and cor- relates of well-being may vary depending upon the age or other particular characteristics of the individuals being measured (Bultena, 1969; Knapp, 1976; Tibbett G Silber, 1965). Therefore, while attachment may be important to individual adjustment past mid-life, the psychological significance of the attachment rela- tionship and of the various dimensions of morale may not be the same for mothers and their daughters. Psycholggical Differentiation Between Mother and Daughter So far, the discussion has focused on why discrepant relational views appear to be salient to the life satisfaction of middle-aged daughters, but it has not addressed why the mothers' self-esteem and incongruent cross-generational perceptions and preferences of attachment were highly correlated. Perhaps, this unanticipated finding can be understood by considering issues of 72 identity and psychological differentiation relevant to women in the middle and later years. Chodorow (1978) contends that the relationship between mothers and daughters in adulthood is not exclusively a function of early socialization experiences and explicit role training. Chodorow relates the lack of (or limited) psychological differen- tiation between adult mother-daughter dyads to internalization Of the primary attachment and thus, to a continuation, at least symbolically, of what Mahler, Pine and Bergman (1973) refer to as the "primary symbiosis." Women who reach maturity and become mothers perpetuate the preoedipal relationship with mother so that, women have an investment in mothering in order to make reparation to their own mother (or to get back at her). Throughout their development, moreover, women have been building layers of identity with their mothers upon the primary internalized mother-child relationship (Chodorow, 1978, p. 204). If individuation from mother threatens the daughter's internal representation of the primary attachment figure, generat- ing anxiety, it could be argued that discrepant views of mother-daughter relations are experienced as threatening for the daughters and thus, are associated with less satisfaction with one's self and life. However, while the correlations of self-es- teem and life satisfaction with incongruent views of mother-daugh— ter bonds were generally in the predicted direction for daughters, the relationships were not significant, making generalizations highly speculative. 73 Although the quanitative data do not provide confirmatory evidence regarding this individuation hypothesis, the interview material does suggest that, at least, for some of the daughters, issues of psychological differentiation from mother were mani- fested. As illustrated by a 52 year old daughter explaining as she filled out the Attachment Scale, I wasn't clear who I was answering for. I became aware that my mother would want me to want these things from her and that would give her pleasure. But I don't really want more for myself now, although I did when I was younger. Sometimes, it's difficult to sort out what I'm doing for me and what I'm doing because my mother wants me to or I sense she wants it from me. It would follow that if incongruent cross-generational relational perceptions and ideals are equated with psychological differentiation, the mother's higher self-esteem in conditions of greater discrepancy could be interpreted as a sign of the Older woman's more complete individuation and separateness from her daughter. Interestingly, Neugarten (1973) found consistent dif- ferences in perceptions of self as a function of the respondent's age and sex. Germane to the present discussion is that women's preoccupation with inner resources and needs increased with advanced age, suggesting a transition from a relational to more egocentric orientation. Neugarten labeled this developmental shift as a move to greater "interiority." Moreover, the older women were more accepting of aggressive and narcissistic impulses when compared to younger females tested. In addition, on the basis of their longitudinal study of female generational ties in 74 adulthood, Cohler and Grunebaum (1981) speculated that a shifting of emotional investments may facilitate greater psychological differentiation and adaptiveness in senescence than is exper- ienced by younger women. Nevertheless, while the individuation argument is intri- guing and deserving of further inquiry, it is highly speculative and questionable when applied to the current research. Discrepancy between mothers' and daughters' evaluations of their relationships is not necessarily an index of psychological differentiation between parent and child. Furthermore, the difference scores used in the computations do not represent the women's conscious acknow- ledgment of disagreement. When mother's and daughter's ratings of perceived similarity on the questionnaires are considered, a dif- ferent picture emerges altogether. The elderly mothers anticipated greater similarity with their daughters' test answers than the daughters' expected in relation to their mothers' responses. MOthers who believe their views are consonant with their daughters' perspectives report greater satisfaction with life and have daughters with higher life satisfaction. Likewise, daughters who expect similarity have higher life satisfaction than those who anticipate differences between themselves and their older parent. This relationship between perceived generational similarity and well-being extended to the elderly mother's attachment with her own mother. Mothers and daughters tended to agree on how much their attachment compared to the bond between the mother and 75 her own parent. If the elderly respondent felt her present relationship with her daughter matched that which she experienced with her own mother, the older subject benefited from greater self-esteem. These results caution against over-interpreting and generalizing when studying intergenerational perceptions and self-concept measures. As other investigators have claimed, per- ceived and objective differences between age groups do not neces- sarily correspond nor have similar effects on intergenerational relations (Ahammer 6 Baltes, 1972; Bengtson G Kuypers, 1971; Nardi, 1973; Thomae, 1970). Perhaps, the most reasonable inter- pretation that can be made of the present findings is that mothers and daughters are more satisfied with their lives when they believe they share similar views of their attachment, even though their actual evaluations may be more discrepant than realized. Attach- ment, life satisfaction and self-esteem are multidimensional con- structs. The structure and meaning of relational ties and variables of well-being seem to reflect the respective types of individual and family life adjustment of middle-aged daughters and their older mothers. However, the relative effects on morale of discrepant views of attachment and acknowledgment of similarity to others are not yet clearly understood. Before continuing, it should be mentioned that the mother-daughter bond does not constitute the only significant affectional relationship in a woman's life. The subjects in this study reported networks of husbands, children, other kin and 76 friends who most probably facilitate individual adaptation and thus, contribute to well-being. Consequently, investigation of the correlates of morale by focusing exclusively on the mother-daughter bond may be a too simplistic approach to under- standing interpersonal functioning in adulthood. A more meaningful research question would be to ask who mothers and daughters 1) go to among their social support system and 2) for what kinds of emotional supplies. As a middle-aged sub— ject reflected upon the complexity of her emotional network, I suppose things will change in the future if I or my mother become widows. Right now we go to each other without restraint, but I know for me, my husband is the first person I turn to for ad- vice or if I want to be comforted. The same is true for her [mother]. Dad is her best friend. If something happens to us, well, I know it'll be different than it is now. Therefore, as Wood and Robertson (1978) speculate, the function Of an individual's ongoing relationship with another and the relative saliency of types of relationship (e.g., kin, friend, organiza- tional affiliations) may change with age. However, a cautionary warning must be offered regarding interpretation of the discussion up to this point. Although generational differences between mothers and daughters have been discussed in reference to the attachment, life satisfaction and self-esteem variables, it is inaccurate to automatically attribute these group distinctions to age. The concept of "generation" is treated in three distinct ways in the literature (Troll, 1970); that is, l) as an age stratum with individual ranked according to 77 years of life, 2) as a position within the family life cycle which carries with it identifiable developmental tasks and role responsibilities (Cohler G Grunebaum, 1981), and 3) as a cohort, i.e., a group of persons, who by virtue of their common ages, experience major social, economic or historical events which are presumed to affect them in ways which are different than individ- uals of other Cohorts (Riley, 1973). The cross-sectional design of this study does not permit the relative contributions of age, family life cycle position or cohort effects to the current findings to be ascertained. The attachment patterns which were identified may be multiply determined. Moreover, until now, this discussion has stressed the discontinuities between mothers and daughters, but additional findings speak to the similarity between the generations. Similarities Between Mothers and Daughters Bowlby (1969, 1970) maintained that physical proximity to the attachment figure was not the sole determinant of the quality of affectional bonding. The emotional availability and responsive- ness of the Other were equally crucial to the form and persistence of emotional ties. In the present work, clear distinctions emerged between the presumed components of attachment, most notably, "seeking (initiating)" and "getting (receiving)" behaviors. Individuals more readily recognized the Other trying to obtain emotional support, but they did not see themselves seeking similar 78 affective supplies. One of the daughters seemed to typify the attitude of the sample when she expressed, I don't really have to ask for these things. I'm able to get it from Mom without having to let her know what I want. I guess I must do something even though I'm not aware of it, but it doesn't go on much. For the most part, it's kind of naturally giving between us. Therefore, women see themselves giving and getting more in affective exchanges than they acknowledge seeking. This self-per- ception is consistent with socialization of the woman primarily into the role Of provider and secondarily, in the role of recipient of affective transactions within the family. To recognize one's own initiative in securing emotional contact, the individual is acknowledging the presence of a personal need. This realization may contradict the woman's self-image as kinkeeper. Another explanation is that the women sampled may not have valued or felt the need to seek from the Other to any great extent. This interpretation is in agreement with the noted tendency for intergenerational interactions in this culture to be characterized by a high degree of autonomy (Clark, 1969). Subjects commonly vocalized that while emotional support, comfort, reassurance, etc. was important to them, it was not always wanted, sought for, nor appreciated. As one mother declared, I'm showered with emotional support--sometimes too much. Still, I know I can count on it if I ever do need it. This quote illustrates the high value subjects placed on having an emotional reserve, that is, a ready and accessible 79 source of support in the Other when it is needed. The expecta- tion that kin will offer assistance in times of crisis has been demonstrated in other studies Of the aging family (Atchley, 1977; Brody, 1966; Treas, 1975). The impact of expectancies on the affective bond between mother and child in adulthood has not been clearly elucidated by this or other research in the other. Expectancies of emotional exchanges may serve a mediational function, ameliorating or intensifying the effect of incongruent views of attachment. Con- sequently, both the individual's level Of expectancy and con- gruency in her perceived-ideal mother-daughter relationship must be considered. Well-being may be less influenced by highly dis- crepant perceptions and preferences of attachment if the individual has low or moderate expectations that personal needs will be met in her mother-daughter interactions. Contrastingly, high expect- ancy of the Other may lead to disappointment, feelings of anger, etc., which compound the impact of incongruency between how mother and daughter regard their relationship. In this respect, Seelbach and Sauer (1977) demonstrated that higher filial expectations in elderly women were linked with lowered morale. The importance of expectancies to affective ties gains credence from the comment of one elder studied in this project. She revealed, Of course I'd want more of these [emotional support] from J, but I don't expect it. She has her own life and I've got mine. We don't expect more. If there's a week where I see her more 80 often, all the better for me, but I don't go looking for it. I'd say I'm pretty satisfied with whatever goes on between us. Moreover, problems inherent in the use of dyadic dif- ference scores based on actual-ideal comparisons could have further obscured the intelligibility of the results (Cronbach, 1958). Discrepancy between preferred and perceived attachment does not provide any standard to estimate the individual's initial level of satisfaction with her "actual" mother-daughter relation- ship (Calvin 8 Holtzman, 1953; Hanlon, Hofstaetter 8 O'Conner, 1954). If the individual's overall evaluation of her current mother-daughter relationship is high (as was obtained with this sample), incongruent perceptions and preferences may be a less significant correlate of morale than the individual who is basically dissatisfied and thus, more unaccepting of the dis- crepancy she experiences between what she has and what she wants with the Other. Methodolggical Issues The foregoing discussion of the results is premised on the assumption that the validity of the test materials is estab- lished. Previous applications of the life satisfaction and self-esteem measures (George G Bearon, 1980) and the semantic differential (Osgood, Suci G Tannebaum, 1957) suggest that the assumption of validity is warranted for these instruments. Yet, the validity of the Attachment Scale is still under question. 81 While the face validity of the AS was achieved, it has not been determined that the construct under investigation, namely attachment, was actually being measured by the AS. The issue of validity becomes even more crucial given the inconsistent findings obtained in this research. Varying responses between mothers and daughters may not be representative of differential attachment ties, but instead signify the tendency fOr the AS to measure dif- ferent variables as a function of the age, cohort membership, etc. of the respondent. Although internal consistency reflected a high level of homogeneity among the items, in the absence of validation external to the test itself, generalizations and interpretations of the results are tentative (Anastasi, 1968). Furthermore, the temporal stability of the instrument has not been assessed. Attachment is conceptualized as a persistent, relatively stable bond. Even accounting for the effects of situational modifiers on the affectional bond, a person's responses might not be expected to vary considerably on successive testings of the AS within a limited time period. However, the design of this study did not enable estimates of this type of reliability to be obtained. Moreover, the AS has limited applicability. It is a self-report inventory requesting the subject to discriminate personal preferences and perceptions on a variety of emotional dimensions. In this regard, the instrument requires a certain degree of psychological sophistication on the part of the respond- ent. Administration of the A5 to larger, more heterogeneous 82 samples would add clarity to these questions of validation and reliability. The sample in this study appeared to be a relatively advantaged one. A majority of the volunteers were married, educated persons who have been living in privately owned resi- dences for several years. The women tended to regard themselves as being in good health and indicated, in conversation, that they were involved in a variety of social activities and volunteer organizations. The factors of self-rated health and activity have been linked to higher life satisfaction over time in persons in middle and later adulthood (Palmore & Kivett, 1977), suggesting that this sample has experienced rewarding life events. Both mothers and daughters were characterized by high levels of self-esteem and life satisfaction. Even an examination of the mean difference scores reveals that the sizes and variances of discrepancy in the participant's views and ratings of attach- ment are not great. This possible lack of sufficient variation in the spread of scores on the attachment and well-being measures might have attenuated the magnitude of the correlations which were obtained, limiting the ability to detect significant relation- ships between the variables studied. In addition, the sampling procedure was probably biased because of a self-selecting factor. The very fact that both a mother and her daughter volunteer in a study examining their affective ties presupposes the existence of a degree of trust, intimacy and comert with the relationship on the part of the 83 dyad members. If attachment is sensitive to stress and discrepant perceptions and preferences, the examination of relationships marked by greater conflict and/or ambivalence then characterized the present sample might yield interesting results. Implications for Future Research This study was regarded as an exploratory attempt to quantify the dimensions of the attachment relationship in adulthood by examining the views of the middle-aged daughter and her elderly mother. Interpretations of attachment based on the current findings are complicated because of the Attachment Scale's lack of estab- lished validity. Also, the homogeneity of the sample, self—select- ing sampling bias and restriction of range of scores on the test instruments further limit the generalizability of the results. Attachment, as measured in this research, constituted an evaluation of an affective state taken at one point in the respond- ent's life. However, attachment is also a process, and as such, should be measured over time (Spanier, 1976). Most of the research on attachment in adulthood is based on cross-sectional data. Longitudinal designs enable the investigation of stability, not only of attachment behaviors, but of subjective measures of self and life satisfaction. This methodological format can better assess the relative influence of age, developmental and cohort differences on mother-daughter ties past mid-life. This research confirms that perceived and actual dis- crepancies are not necessarily associated with the same correlates 84 of well-being. Therefore, investigations employing both self-re- port and Objective behavioral or observational measures increase the potential explanatory power of the findings. As George and Bearon (1980) contend, Life quality includes both the conditions of life and the experience of life (p. 2). Generalizability of findings can be increased by testing larger, more representative samples of women including persons undergoing known periods of transition such as retirement or bereavement. Furthermore, work on attachment in middle and later life has primarily been conducted with community residents. Our knowledge about how affective bonds between an aged parent and her adult child are influenced by the intervention of an "insti- tutional" attachment figure, such as a hospital or nursing home (Kalish G Knudtson, 1976) is even more limited. Moreover, while the relational ties between mother and daughter were the focus of this investigation, adjustment may be better understood as a function of the entire fabric of the social network available to the individual in middle and later adulthood. Comparatively little work has been conducted on other types of familial relationships such as with siblings or father-daughter and mother-son bonding past mid-life. In review, affectional ties were found to be significant correlates of certain measures of well-being, but these relation- ships were not consistent across the generations. The assertion that discrepant views of attachment, in general, are related to 85 morale in adulthood may obscure the complex interplay between family members across the generations. The equivocal findings of this and other research suggests that further investigations are needed which are based on longitudinal designs and use combined subjective and objective behavioral measures with larger, more heterogeneous samples. Thus, while it seems safe to conclude that bonding between mothers and daughters in adulthood seems important to selected aspects of individual functioning, the specific significance of this attachment for each dyad member is not clearly understood. Further inquiry is needed so that new meaning can be brought to the old adage voiced by an elderly mother A son is a son until he takes a wife, but a daughter is a daughter for the rest of her life. SUMMARY Comparatively little research is available regarding the affective nature of mother-daughter bonds in the middle and later years. Attempts to identify normative and adaptive modes of parent-child bonding in adulthood have been biased by implicit value, judgments and are primarily based on findings gathered from either the older or the middle generation, but rarely both members of the adult mother-daughter dyad. Consequently, theoret- ical formulations and contradictory research findings regarding individual and family functioning past mid-life do not seem to account for the variability of intergenerational relations repre- sented in the literature. This study is an attempt to explore affective aspects of the relationship between the elderly mother and her middle-aged daughter by examining the perspectives of both members of the attachment dyad. It was assumed that no single optimal pattern of attachment would be associated with life satisfaction and self-esteem in adult women. Instead, congruency between each individual's perceived and preferred (ideal) mother-daughter relationship and cross-generational agreement between the mother's and her daughter's views and ratings of their mutual relationship were expected to be significant correlates of morale. More specifically, it was hypothesized that 1) life satisfaction would 86 87 be positively related to self-esteem and 2) life satisfaction and self-esteem would be negatively correlated with a) the extent of discrepancy between each subject's individual perceptions and preferences of her attachment, b) the extent of cross-generational discrepancy between the mother's and her daughter's preferences and perceptions of their mutual relationship and c) the extent Of cross-generational discrepancy between the mother's and her daugh- ter's ratings of the quality of their bond. Pearson product-moment correlation coefficients were computed to test the hypotheses. The subjects were 25 mothers, 65 years and older, and their daughters, aged 40 to 64. All participants were Caucasian, mentally competent, noninstitutionalized volunteers. Each respondent independently completed a series of written questionnaires and a brief structured interview. The instruments consisted of the Rosenberg Self-esteem Inventory, the Life Satisfaction Index and three measures devised by the author for this study, i.e., the Personal Inventory, the Attachment Scale and the Relationship Rating Semantic Differential. A pilot study conducted with 30 elderly and 28 middle-aged women was used to evaluate the internal consistency of the test materials and to make needed revisions before administering the questionnaires to the research sample. Three of the hypotheses were confirmed, but only for the daughters: Life Satisfaction was 1) positively associated with self-esteem and 2) negatively correlated with an incongruent perceived-ideal mother-daughter relationship and discrepant cross-generational ratings of relationship quality. The only 88 significant finding obtained for the mothers was against pre- diction. Mothers' self-esteem was positively associated with cross-generational differences in the mother's and her daughter's perceptions and preferences of their relationship. The differential findings between mothers and daughters were discussed in regards to the role demands, developmental issues, cohort effects, expectancies and degree of psychological differentiation which distinguishes women in middle age and senescence. Methodological limitations of the research design, instruments and sample selection were also examined in reference to the results. It was concluded that affectional ties between middle-aged daughters and their mothers are important to well-being, but the particular significance of attachment to each dyad member needs to be better understood. In this respect, specific areas of future research were proposed. REFERENCES REFERENCES Adams, B.N., Interaction theory and the social network. Socio- metry, 1967, 393 64-78. Adams, D.L., Analysis of a life satisfaction index. Journal of Gerontology, 1969, 24(4), 470-474. Ahammer, J.M. and Baltes, P.B., Objective versus perceived age differences in personality: How do adolescents, adults and Older people view themselves and each other? Journal of Gerontolggy, 1972, 21(1), 46-61. Ainsworth, M., Attachment and dependency: A comparison. In Gewirtz (Ed.), Attachment and dependency. Washington, DC:V.H. Winston, 1972, 97-137. Ainsworth, M., Object relations, dependency and attachment: A theoretical review of the infant-mother relationship. Child Development, 1969, 59, 969-1025. Anastasi, A., Psychological testing, 3rd Edition. Toronto: Macmillan Company, 1968. Antonucci, T., Attachment: A life-span concept. Human Development, 1976, 12, 135-142. Atchley, R.C., The social forces in later life. Belmont, CA: Wadsworth Publishing Company, Inc., 1977. Bell, R., A reinterpretation of the direction of effects in studies of socialization. Psychological Review, 1968, 25(2), 81-95. Bengtson, V. and Black, K., Intergenerational relations and con- tinuities in socialization. In Baltes, P. and Schaie, K (Eds.), Life-span develppmental_psychology: Personalipy and socialization. NY: Academic Press, 1973, 207-234. Bengtson, V. and Kuypers, J., Generational differences and the developmental stake. Aging and Human Development, 1971, 23 249-260. 89 90 Bengtson, V. and Treas, J., The changing family context of mental health and aging. In Birren, J. and Sloane, R. (Eds.), Handbook of mental health and aging. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1980, 400-428. Bischof, L., Adult psychology, NY: Harper and Row, 1976. Blenkner, M., Social work and family relationships in later life with some thoughts on filial maturity. In Shanas, E. and Streib, G. (Eds.), Social structure and the family: Generational relations. Englewood Cliffs, NJ: Prentice- Hall, Inc., 1965. Boszormeny-Nagy, I. and Spark, 6., Invisible loyalties: Reciprocity in intergenerational family therapy. NY: Harper 8 Row, Publishers, Inc., 1973. Bowlby, J., Attachment and Loss: Volume 1, Attachment. NY: Basic Books, 1969. Bowlby, J., Attachment and Loss: Volume 2, Separation. NY: Basic Books, 1973. Bowlby, J., Attachment theory, separation anxiety and mourning. In Arieti, 5. (Ed.), American handbook of psychiatry, Second edition. NY: Basic Books, 1975, 292-309. Bowlby,J The makingand breaking of affectional bonds. London: Tavistock Publications, 1979. Boyd, R. The valued grandparent. A changing social role. In Donahue, W. (Ed. ), Living in the multigeneration family. Ann Arbor: University of Michigan Press, 1969, 90- 106. Britton, J., Mather, W. and Lansing, A., Expectations for older persons in a rural community: Living arrangements and family relationships. Journal of Gerontology, 1961, 19(2), 156-162. Brody, E., The aging family. The Gerontologist, 1966, 9(4), 201-206. Brown, A., Satisfying relationships for the elderly and their patterns of disengagement. The Gerontologist, 1974, 11, 258-262. Bultena, 6., Life continuity and morale in old age. The Geronto- logist, 1969, 9(4), 251-253. Bumagen, V. and Hirn, K., Aging as a family affair. NY: Thomas Y. Crowell, Publisher, 1979. 91 Calvin, A. and Holtzman, W., Adjustment and the discrepancy between self-concept and inferred self. Journal of Con- sulting Psychology, 1953, 11, 39-44. Cavan, R., Family tensions between the Old and the middle-aged. In Vedder, C. (Ed.), Problems of the middle-aged. Springfield, ILL.: Charles C. Thomas, 1965, 82-91. Chodorow, N., The reproduction of mothering: Psychoanalysis and the sociology ofygender. Berkeley: University of California Press, 1978. Clark, M., Cultural values and dependency in later life. In Kalish, R. (Ed.), The dependencies of Oldypeople. Ann Arbor: University of Michigan Press, 1969, 59-72. Clark, M. and Anderson, B., Culture and aging. Springfield, 111.: Charles C. Thomas, Publisher, 1967. Cohen, S., and Gans, B., The other generation gap: The middle-aged and their aging parents. Chicago: Follett Publishing Company, 1978. Cohler, B., Autonomy and interdependence in the family of adult- hood: A psychological perspective. Paper presented at the Annual Meeting of the Gerontological Society of America, San Diego, CA., 1980. Cohler, B. and Grunebaum, H., Mothers, grandmothers, and daughters. NY: John Wiley and Sons, 1981. Crano, W. and Brewer, M., Principles of research in social psychology. NY: McGraw-Hill, Inc., 1973. Cronbach, L., Proposals leading to analytic treatment of social perception scores. In Taguiri, R. and Petrullo, L. (Eds.), Person perception and inteppersonal behavior. Stanford, CA.: Stanford Press, 1958, 353-379. Cryns, A. and Monk, A., Attitudes of the aged toward the young: A multivariate study of intergenerational perception. Journal of Gerontolggy, 1972, 22(1), 107-112. Depner, C., and Ingersoll, Social support in the family context. Paper presented at the Annual Meeting of the Gerontological Society of America, San Diego, CA, 1980. Donahue, W., Emerging principles and concepts: A summary. In Donahue, W. and Tibbitts, C. The new frontiers of aging. Ann Arbor: University of Michigan Press, 1957, 198-206. 92 Donahue, W., Living in the four-generation family. In Donahue, W. (Ed.), Living in the multigeneration family. Ann Arbor: University of Michigan Press, 1969, 37-51. Eckels, E., Negative aspects of family relationships fOr older women. Paper presented at the 34th Annual Meeting of the Gerontological Society of America, Toronto, Canada, 1981. Edwards, J. and Klemmack, D., Correlates of life satisfaction: A re-examination. Journal of Gerontology, 1973, 22(4), 497—502. Erikson, E., Childhood and society. NY: Norton Books, 1963. Field, M., The aged, the familyyand the community. NY: Columbia University Press, 1972. Gelfand, D., Olsen, J. and Block, M., Two generations of elderly in the changing American family: Implications for family services. The Family Coordinator, 1978, 22(4), 395-403. George, L,, and Bearon, L., Quality of life in Older persons. NY: Human Sciences Press, Inc., 1980. Gillaspy, R., The older population: Considerations for family ties. In Ragan, P. (Ed.), Aging parents. LA: University of Southern California Press, 1979, 11-26. Goldberg, S. and Deutsch, F., Life-gpan individual and family development. Monterrey, CA.: Brooks Cole Publishing CO., Inc., 1977. Goldfarb, A., Psychodynamics and the three-generation family. In Shanas, E. and Streib, G. (Eds.), Social structure and the family: Generational relations. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1965, 10-45. Goldfarb, A., The psychodynamics of dependency and the search for ' aid. In Kalish, R. (Ed.), The dependencies of old pgople. Ann Arbor: University of Michigan Press, 1969, 1-16. Gottesman, ., Quarterman, C. and Cohn, G., Psychosocial treatment of the aged. In Lawton, M. and Eisdorfer, C. (Eds.), Psychology of adult development and aging. Washington, DC: American Psychological Association Press, 1973, 378-427. Hanlon, R., Hofstaetter, P. and O'Conner, J., Congruence of self and ideal self in relation to personality adjustment. Journal of ConsultingyFsychology, 1954, 12, 215-218. 93 Hartup, W. and Lempers, J., A problem in life-span development: The interactional analysis of family attachments. In Baltes, P. and Schaie, K. (Eds.), Life-span developmental psychology; Personality and socialization. NY: Academic Press, 1973, 235-252. Hawkinson, W., Wish, expectancy and practice in the interaction of generations. In Rose, A. and Peterson, W. (Eds.), Older people and their social worlds. Philadelphia: F.A. Davis Co., 1965, 181—192. Herr, J. and Weakland, J., Counseling elders and their families. NY: Springer Publishing Company, Inc., 1979. Hess, B. and Waring, J., Changing patterns of aging and family bonds in later life. The Family Coordinator, 1978, 22, 303-314. Hill, R. Decision making and the family life cycle. In Shanas, E. and Streib, G. (Eds.), Social structure and the family: Generational relations. Englewood Cliffs, NJ: Prentice- Hall, 1965, 113-139. Hirschfield, I. and Dennis, H., Perspectives. In Ragan, P. (Ed.), Agipg parents. LA: University of Southern California Press, 1979, 1-10. Johnson, E.S. and Bursk, B.J., Relationships between the elderly and their adult children. The Gerontologist, 1977, 22(1), 90-96. Kalish, R., Of children and grandfathers: A speculative essay on dependency. In Kalish, R. (Ed.), The dependencies of old people. Ann Arbor: University of Michigan Press, 1969, 73-83. Kalish, R. and Knudtson, R., Attachment versus disengagement: A life-span conceptualization. Human Development, 1976, 22, 171-181. Kastenbaum, R. and Cameron, P., Cognitive and emotional dependency in later life. In Kalish, R. (Ed.), The dependencies of Old people. Ann Arbor: University of Michigan Press, 1969, 39-58. Kerckhoff, A., Family patterns and morale in retirement. In Simpson, I. and McKinney, J. (Eds.), Social aspects of aging. Durham, NC: Duke University Press, 1966. Knapp, M., Predicting the dimensions of life satisfaction. Journal Of Gerontolo y, 1975, 21(5), 595-604. 94 Knudtson, F., Life-span attachment: Complexities, questions, con- siderations. Human Development, 1976, 22, 182-196. Kozma, A. and Stones, M., Some research issues and findings in the study of the psychological well-being in the aged. Canadian Psychological Review, 1978, 22(3), 241-249. Kuypers, J. and Trute, B., The older family as the locus of crisis intervention. The Family Coordinator, 1978, 22(4), 405-411. Laslett, P., Societal development and aging. In Binstock, R. and Shanas, E. (Eds.), Handbook of aging and the social sciences. NY: Van Nostrand Reinhold, 1976, 87-116. Lindgren, H., An introduction to socialypsychology. NY: John Wiley 8 Sons, Inc., 1969. Maas, H. and Kuypers, J., From thirty to seventy. San Francisco: Jossey-Bass, 1974. Maccoby, E. and Masters, J., Attachment and dependency. In Mussen, P. (Ed.), Carmichael's manual of child psychology, 3rd Edition, Volume 11. NY: John Wiley, 1970, 73-158. Mahler, M., Pine, F., and Bergman, A., The psychological birth of the human infant. NY: Basic Books, 1975. Mancini, J., Family relationships and morale among people 65 years of age and older. American Journal of Orthopsychiatpy, 1979, 32(2), 292-300. Meissner, W., Thinking about the family--psychiatric aspects. Family Process, 1964, 2(1), 1-40. Mischel, W., A social-learning view of sex differences in behavior. In Maccoby, E. (Ed.), The development of sex differences. StanfOrd, Calif: Stanford University Press, 1966, 56-81. Moss, B., Caring for the aged. NY: Doubleday and Company, Inc., 1966. Mussen, P., Conger, J., and Kagan, J., Child development and personality, fourth edition. NY: Harper 8 Row, Inc., 1974. Nardi, A., Person-perception research and the perception of life- span development. In Baltes, P. and Schaie, K. (Eds.), Life-span developmental psychology: Personality and social- ization. NY: Academic Press, 1973, 285-301. 95 Neugarten, B., Adult personality: Towards a psychology of the life cycle. In Neugarten, B. (Ed.), Middle agg and aging. Chicago: University of Chicago Press, 1968, 137-141. Neugarten, B., Continuities and discontinuities of psychological issues into adult life. Human Development, 1969, 223 121-130. Neugarten, B., Dynamics of transition of middle age to old age: Adaptation to the life cycle. Journal of Geriatric Psy- chiatry, 1970, 9(1), 71-87. Neugarten, B., Personality change in late life: A developmental perspective. In Eisdorfer, C. and Lawton, M., 299_ psychology of adult deve10pment and agipg. Washington, D.C.: American Psychological Association, 1973, 311-338. Neugarten, B., Havighurst, R., and Tobin, S., The measurement of life satisfaction, Journal of Gerontology, 1961, 29, 134-143. Nunnally, J., P§ychometric theory, Second edition. NY: McGraw-Hill, Inc., 1978. Palmore, E., and Kivett, V., Change in life satisfaction: A longitudinal study of persons aged 46-70. Journal of Gerontology, 1975, 92(3), 311-316. Parens, H. and Saul, L., pgpendence in man: A psychoanalytic study. NY: International Universities Press, Inc., 1971. Peterson, J., The relationships of middle-aged children and their parents. In Ragan, P. (Ed.), Aging parents. LA: University of Southern California Press, 1979, 27-36. Osgood, C., Suci, G. and Tannebaum, F., The measurement of meaning. Chicago: University of Illinois Press, 1957. Rautman, A., Role reversal in geriatrics. Mental Hygiene, 1962, 99, 116-120. Riley, M., Aging and cohort succession: Interpretations and mis- interpretations. Public gpinion Quarterly, 1973, 92, 35-49. Riley, M. and Foner, A., Aging and society: Volume 1, An inventory of research findings. NY: Russell Sage Foundation, 1968. Robertson, J., Interaction in three generation families, parents as mediators: Toward a theoretical perspective. International Journal of Aging and Human Development, 1975, 9(2), 103-109. 96 Robinson, J. and Shaver, P., Measures of socialpsychological attitudes. Ann Arbor, MI: Social Research Institute, 1972. Rogers, D., The adultyyears: An introduction to aging. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1979. Rosenberg, M., Society and the adolescent self-image. Princeton, NJ: Princeton University Press, 1965. Rosow, 1., Social integration of the aged. NY: The Free Press, 1967. Schwartz, A., Psychological dependency: An emphasis on the later years. In Ragan, P. (Ed.), Aging parents. LA: University of Southern California Press, 1979, 116-125. Sears, R., Attachment, dependency and frustration. In Gerwitz, J. (Ed.), Attachment and dependengy. Washington, DC: V.H. Winston Books, 1972, 1-27. Secord, P. and Backman, C., Personality theory and the problems of stability and change in individual behavior: An inter- personal approach. Psychological Review, 1961, 92(1), 21-32. Seelbach, W., Gender differences in expectations for filial responsibility. The Gerontologist, 1977, 22(5), 421-425. Seelbach, W. and Sauer, W., Filial responsibility, expectations and morale among aged parents. The Gerontologist, 1977, 22(6), 492-499. Shanas, E., Family help patterns and social class in three countries. In Neugarten, B. (Ed.), Middle gge and aging. Chicago: University of Chicago Press, 1968, 296-305. Silverstone, B., Issues for the middle generation: Responsibility, adjustment and growth. In Ragan, P. (Ed.), Aging parents. LA: University Of Southern California Press, 1979, 107-115. Simos, B., Adult children and their aging parents. Social Work, 1973, 22(3), 78-85. Simos, B., Relations of adults with aging parents. The Gerontolo- gist, 1970, 29, 135-139. Spanier, G., Measuring dyadic adjustment: New scales fOr assessing the quality of marriage and similar dyads, Journal of Marriage and the Family, 1976, 15-28. 97 Spark, B., and Brody, E., The aged are family members. Family Process, 1970, 9(2), 195-210. Sporakowski, J. and Hughston, G., Prescriptions for happy mar- riage: Adjustments and satisfactions of couples married for 50 or more years. The Family Coordinator, 1978, 27(4), 321-327. '— Stehouwer, J., The household and family relations of old people. In Shanas, E., Townsend, P., Wedderburn, D., Friis, H., Milh¢g, P. and Stehouwer”]_(EdS.), Old people in three industrial societies. NY: Atherton Press, 1968, 117-226. Steinman, L., Reactivated conflict with aging parents. In Ragan, P. (Ed.), ggingyparents. LA: University of Southern California Press, 1979, 126-143. Stern, E. and Ross, M., You and your aging parents. NY: A.A. Wyn, Inc., 1952. Streib, G., Older families and their troubles: Familial and social responses. In Sze, W. (Ed.), Human life cycle. NY: Jason Aronson, Inc., 1975, 671-690. Streib, G., Untitled address presented at the Annual Meeting of the Gerontological Society of America, San Diego, CA, 1980. Streib, G. and Shanas, E., Social structure and the family: Generational relations. Englewood Cliffs, NJ: Prentice— Hall, Inc., 1965, 2-8. Streib, G. and Thompson, W., The older person in a family context. In Tibbitts, C. (Ed.), Handbook of socialygerontology. Chicago: University of Chicago Press, 1960, 447-488. Sroufe, L. and Waters, E., Attachment as an organizational con- struct. Child Development, 1977, 993 1184-1199. Sussman, M., Intergenerational family relationships and social role. In Vedder, C. (Ed.), Problems of the middle-aged. Springfield, ILL.: Charles C. Thomas, 1965. Sussman, M., Relation of adult children with their children in the United States. In Streib, G. and Shanas, E., (Eds.), Social structure and the family: Generational relations. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1965, 62-92. Thomae, H., Theory of aging and cognitive theory of personality. Human Development, 1970, 293 1-16. 98 Tippett, J. and Silber, E., Self-image stability: The problem of validation, Psychological Reports, 1965, 223 323-329. Townsend, F., The effects of family structure on the likelihood of admission to an institution in old age: The application of a general theory. In Streib, G. and Shanas, E., (Eds.), Social structure and the family: Generational relations. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1965, 163-187. Townsend, F., The emergence of the four generation family in industrial society. In Neugarten, B., (Ed.), Middle age and aging. Chicago: University Of Chicago Press, 1968, 255-257. Treas, J., Aging and the family. In Woodruff and Birren (Eds.), Aging scientific perspectives and social issues, 1975, 92-108. Treas, J., Intergenerational families and social change. In Ragan, P. (Ed.), éging parents. LA: University of Southern CalifOrnia Press, 1979, 58-65. Troll, L.E., The family of later life: A decade review. Journal of Marrioge and the Family, 1971, 99(2), 1971, 263-290. Troll, L., Issues in the study of generations. International Journal of Aging and Human Developggnt, 1970, 2, 199-218. Troll, L.E., Miller, S. and Atchley, R., Families in later life. Belmont, Calif: Wadsworth Publishing Company, Inc., 1979. Troll, L. and Smith, J., Attachment through the life span: Some questions about dyadic bonds among adults. Human Develop- ment, 1976, 29, 156-170. Turner, J., Patterns of intergenerational exchange: A develop- mental approach. International Journal of Aging and Human Development, 1975, 9(2), 111-115. Verwoerdt, A., Clinical geropsychiatry. Baltimore: Waverly Press, Inc., 1976. Weinberg, J., Interpersonal relationships in multigeneration families. In Donahue (Ed.), Livipg in the multiggnera- tion family, Ann Arbor: University of Michigan Press, 1969. Wells, L. and Marwell, G., Self-esteem: It's conceptualization and measurement. Beverly Hills, CA: Sage Publications, Inc., 1976. 99 Wood, V. and Robertson, J., Friendship and kinship interaction: Yarrow, Ziller, Differential effect on the morale of the elderly. Journal of Marriagg and the Family, 1978, 99(2), 367-375. L., Attachment and dependency: A developmental perspective. In Gewirtz, J. (Ed.), Attachment and dependency. Washington, DC: V.H. Winston, 1972, 81-95. R., Hagey, J., Smith, M. and Long, B., Self-esteem: A self-social construct. Journal of Consulting and Clinical Psychology, 1969, 99(1), 84-95. APPENDIX A 10. 11. 12. 13. 14. 15. APPENDIX A PERSONAL INVENTORY MOTHER Name: Age: Marital Status: Single Married Separated Divorced Widowed How long have you been married, separated, divorced or widowed? Race: Caucasian Black Other Religion: Catholic Jewish Protestant Other Years of education completed: Specialized training and/or degrees: Main occupation: How long have you been at your present job? If retired, for how long? Present source of income: Yearly household income (in thousands) Living quarters: Own home/apartment Rent home/apartment Other Living arrangements: Live alone Live with spouse Live with relative (please specify) Live with nonrelative (please specify) 100 16. 17. 18. 19. 20. 21. Sex 22. 101 MOTHER How long have you lived at your current address? How long have you lived in your community? How is your health? Poor Fair Good Very Good Present age of parents: (Put D in space if parent is deceased) Mother Father How Old were you when your parent died? When your mother died? When your father died? How many children do you have? Amount of Contact (M=male, Marital Number of (e.g., letters, visits) F=female) Age Status Children With Your Child* (Use back of page if more room is needed.) *a. Indicate if contact is one of the following: Less than once a year. Several times a year. About once a month. Once a week or more. Daily. mthr—I b. Next to the child you feel closest to, put an "X." How far does your daughter who is in this study live from you? Lives in a different state Lives in same state, but different city Lives in same city, but different neighborhood Lives in same neighborhood Lives on same street Lives in same house 23. 24. 25. Sex 26. 102 MOTHER How much contact do you have with this daughter? Less than once a year Several times a year About once a month Once a week or more Daily Kind of contact with this daughter: rank order these alter- natives with "1" being the most frequent type of contact and "3" being the least frequent. Letter writing Telephone conversations Face-to-face visits Do you have any brothers or sisters? (Indicate if person is deceased). (M=male, Marital Number of Amount of Contact F=female) Age Status Children With Sibling* (If more room is needed use back of page.) *a. Indicate if contact is one of the following: Less than once a year. Several times a year. About once a month. Once a week or more. Daily. m-hMNv—l b. Next to the sibling you feel closest to, put an "X." Within the past year, has anything happened to you or any other member of your family which you feel has influenced your relationship with your daughter? Please explain briefly. 27. 28. 29. 30. 31. 103 MOTHER Is anything currently happening to you or any other member of your family which you feel has influenced your relationship with your daughter? (Explain) Do you expect anything to happen to you or any other family member in the near future which you feel might influence your relationship with your daughter? (Explain) If your mother is still alive: How far does she live from you? Lives in a different country Lives in a different state Lives in a different city, same state Lives in same city, different neighborhood Lives in same neighborhood Lives on same street Lives in same house How much contact (e.g., letter writing, phoning, face-to-face visits) do you have with your mother? Less than once a year Several times a year About once a month Once a week or more Daily Kind of contact with your mother: rank order these alterna- tives with "1" being the most frequent type of contact and "3" being the least frequent. Letter writing Telephone conversations Face-to-face visits APPENDIX B APPENDIX B MOTHER'S FORM Listed below are a number of things you may SEEK from your daughter. SEEK means to "ask for," "to search for" or "to try to get.” There- fore, SEEKING means that you initiate getting these things. Please indicate how much you generally see yourself SEEKING these things. There are five choices ranging from happens "Not at All (1)" to happens "Very Much (5)." Circle the single number, 1 to 5, which best describes what you feel generally happens in your current relationship with your daughter. Make each item a separate and independent decision. Do not worry over individual items. Just put down your first impressions. NOT AT A GREAT I ACTUALLY SEEK __féi;. 2 3 4 DEAL CONTACT (phone, write, visit) with my daughter. . . . . . . . l 2 3 4 5 ATTENTION from my daughter. . . 1 2 3 4 5 my daughter's REASSURANCE . . . 1 2 3 4 5 my daughter's ADMIRATION. . . . 1 2 3 4 5 my daughter's APPROVAL. . . . . l 2 3 4 5 my daughter's ACCEPTANCE. . . . l 2 3 4 5 my daughter's EMOTIONAL SUPPORT l 2 3 4 5 to CONFIDE IN my daughter . . . l 2 3 4 5 my daughter's ADVICE. . . . . . 1 2 3 4 5 my daughter's AFFECTION . . . . l 2 3 4 5 my daughter's UNDERSTANDING . . 1 2 3 4 5 my daughter's COMFORT . . . . . 1 2 3 4 5 104 105 MOTHER'S FORM Now please indicate how much you feel your daughter SEEKS these things from you. Do not refer to the previous page or wonder about your answers. MY DAUGHTER SEEKS CONTACT with me. my ATTENTION . my REASSURANCE . my ADMIRATION. my APPROVAL. my ACCEPTANCE. my EMOTIONAL SUPPORT . to CONFIDE IN me . my ADVICE. my AFFECTION . my UNDERSTANDING . my COMFORT . NOT AT ALL 2 l l A GREAT DEAL S 5 106 MOTHER'S FORM We are now interested in what you feel you GET from your daughter. Circle the number, 1 to 5, which best describes how much you see these things happening. Once again, answer each item. I ACTUALLY GET CONTACT with my daughter. ATTENTION from my daughter. my my my my my to my my my my daughter's daughter's daughter's daughter's daughter's CONFIDE IN daughter's daughter's daughter's daughter's REASSURANCE . ADMIRATION. APPROVAL. ACCEPTANCE. EMOTIONAL SUPPORT my daughter . ADVICE. AFFECTION . UNDERSTANDING . COMFORT . NOT AT ALL 4 l 1 DO not refer back to other pages. A GREAT DEAL 5 5 What do you feel your daughter GETS FROM YOU? MY DAUGHTER ACTUALLY GETS CONTACT with me. my my my my my my to my my my my ATTENTION . REASSURANCE . ADMIRATION. APPROVAL. ACCEPTANCE. EMOTIONAL SUPPORT . CONFIDE IN me . ADVICE. AFFECTION . UNDERSTANDING . COMFORT . 107 NOT AT ALL 1 l MOTHER'S FORM A GREAT 3%.. 108 MOTHER' S FORM Below are some FEELINGS you may experience in relation to your daughter and your daughter in relation to you. As on the other pages, circle the number which best describes how much YOU FEEL or you see YOUR DAUGHTER FEELS. I ACTUALLY FEEL SECURE when with my daughter. RESPONSIBLE FOR my daughter . I INFLUENCE my daughter . INTERESTED IN my daughter . I WANT TO BE LIKE my daughter . OBLIGATED TO my daughter. I CARE FOR my daughter. MY DAUGHTER ACTUALLY FEELS SECURE when with me . RESPONSIBLE fOr me she INFLUENCES me . INTERESTED IN me she WANTS TO BE LIKE me . OBLIGATED TO me . she CARES FOR me. NOT AT ALL 1 1 b A GREAT DEAL 109 MOTHER'S FORM Now we are interested in what you IDEALLY WISH or PREFER YOU WOULD SEEK from your daughter. Please put down your first impressions of what you WISH WOULD IDEALLY HAPPEN. I WISH I WOULD SEEK CONTACT with my daughter. ATTENTION from my daughter my my my my my to my my my my daughter's daughter's daughter's daughter's daughter's CONFIDE IN daughter's daughter's daughter's daughter's REASSURANCE . ADMIRATION APPROVAL. ACCEPTANCE. EMOTIONAL SUPPORT . my daughter . ADVICE AFFECTION . UNDERSTANDING . COMFORT . NOT AT ALL A GREAT 22.4.1.1... What do you WISH your daughter WOULD SEEK FROM YOU? I WISH MY DAUGHTER WOULD SEEK 110 MOTHER'S FORM CONTACT with me. my my my my my my to my my my my ATTENTION . REASSURANCE . ADMIRATION. APPROVAL ACCEPTANCE. EMOTIONAL SUPPORT . CONFIDE IN me . ADVICE. AFFECTION . UNDERSTANDING . COMFORT . NOT AT ALL 1 1 A GREAT DEAL 111 MOTHER'S FORM Go through these lists and indicate what you WOULD WISH TO GET FROM your daughter. NOT AT A GREAT I WISH I WOULD GET ALL 2 3 4 DEAL CONTACT with my daughter. . . . . 1 2 3 4 5 ATTENTION from my daughter. . . . l 2 3 4 5 my daughter's REASSURANCE . . . . l 2 3 4 5 my daughter's ADMIRATION . . . . l 2 3 4 5 my daughter's APPROVAL . . . . . l 2 3 4 5 my daughter's ACCEPTANCE. . . . . 1 2 3 4 5 my daughter's EMOTIONAL SUPPORT . l 2 3 4 5 to CONFIDE IN my daughter . . . . l 2 3 4 5 my daughter's ADVICE. . . . . . . l 2 3 4 5 my daughter's AFFECTION . . . . . l 2 3 4 5 my daughter's UNDERSTANDING . . . 1 2 3 4 5 my daughter's COMFORT . . . . . . l 2 3 4 5 What do you WISH your daughter WOULD GET FROM YOU? I WISH MY DAUGHTER WOULD GET 112 CONTACT with me. my my my my my my to my my my my ATTENTION . REASSURANCE . ADMIRATION APPROVAL ACCEPTANCE. EMOTIONAL SUPPORT . CONFIDE IN me . ADVICE. AFFECTION . UNDERSTANDING . COMFORT . MOTHER'S FORM NOT AT ALL 2 4 1 2 4 1 2 4 l 2 4 l 2 4 1 2 4 1 2 4 l 2 4 l 2 4 1 2 4 l 2 4 l 2 4 1 2 4 A GREAT DEAL 113 MOTHER'S FORM A; Now please indicate what you PREFER or WISH YOU WOULD FEEL in relation to your daughter and WISH your daughter WOULD FEEL ABOUT YOU. I WISH I WOULD FEEL SECURE when with my daughter. RESPONSIBLE FOR my daughter . I INFLUENCE my daughter . INTERESTED IN my daughter . I WANT TO BE LIKE MY daughter . OBLIGATED TO my daughter. I CARE FOR my daughter I WISH MY DAUGHTER WOULD FEEL SECURE when with me . RESPONSIBLE FOR me she INFLUENCES me . INTERESTED IN me she WANTS TO BE LIKE me . OBLIGATED TO me . she CARES FOR me NOT AT ALL A GREAT DEAL APPENDI X C APPENDIX C RELATIONSHIP RATING SEMANTIC DIFFERENTIAL FOR MOTHER On the next page, there are a list of adjectives arranged in pairs that are opposite in meaning. These word pairs can each be used to describe your current relationship with your daughter. For example, if you feel that your relationship with your daughter is vepyrstropgly characterized (described) by one or the other end of the line, put an X as follows: friendly X : : : : : : unfriendly OR friendly : : : : : : IX unfriendly If you feel that your relationship with your daughter is only slight2y described by one or the other end of the pair, put an X as follows: friendly : X : : : : : unfriendly OR friendly : : : : : X : unfriendly If you feel your relationship is equally associated or characterized by either adjective, place your X in the middle slot. friendly : : : X : : : unfriendly 114 115 FOR MOTHER Therefore, the place you put your X depends upon: 1) which of the two adjectives best describes your relationship; 2) and to what extent (very strongly, moderately, slightly, equally). Please put only one X on a single line and answer every line. Make sure your X is clearly in a slot. Do not worry about your answers; just put your first impressions. relaxed tense secure insecure ugly beautiful positive negative mature immature cruel kind stable unstable good bad generous selfish meaningless meaningful important unimportant stale fresh undependable dependable pleasant unpleasant strong weak certain uncertain unfair fair satisfying dissatisfying successful unsuccessful cool warm APPENDIX D APPENDIX D ROSENBERG SELF-ESTEEM INVENTORY Using the following scale, circle the number which best represents your feelings about each of the statements below: Strongly Strongly Disagree Disagree Agree Agree 1. I feel that I'm a person of worth, at least on an equal basis with others. 1 2 3 4 2. I feel that I have a number of good qualities. 1 2 3 4 3. All in all, I am inclined to feel that I am a failure. 1 2 3 4 4. I am able to do things as well as most people. 1 2 3 4 5. I feel I do not have much to be proud of. l 2 3 4 6. I take a positive attitude toward myself. 1 2 3 4 7. On the whole, I am satis- fied with myself. 1 2 3 4 8. I wish I could have more respect for myself. 1 2 3 4 9. I certainly feel useless at times. 1 2 3 4 10. At times I think I am no good. 1 2 3 4 116 APPENDIX E APPENDIX E LIFE SATISFACTION INDEX Here are some statements about life in general that people feel differently about. Would you read each statement on the list, and if you agree with it, put a check mark in the space under "AGREE." If you do not agree with a statement, put a check mark in the space under "DISAGREE." If you are not sure one way or the other, put a check mark in the space under "?" PLEASE BE SURE TO ANSWER EVERY QUESTION ON THE LIST. DIS- AGREE AGREE ? 1. As I grow older, things seem better than I thought they would be. 2. This is the dreariest time of my life. 3. I am just as happy as when I was younger . 4. My life could be happier than it is now . . . . . . . . . . . . 5. These are the best years of my life . 6. Most of the things I do are boring or monotonous . . . . . 7. I expect some interesting and pleasant things to happen to me in the future. 8. The things I do are as interesting to me as they ever were. 9. I feel old and somewhat tired . 10. As I look back on my life, I am fairly well satisfied . . 117 11. 12. 13. 14. 15. 118 AGREE AGREE I would not change my past life even if I could. Compared to other people my age, I make a good appearance . I have made plans for things I'll be doing a month or a year from now. Compared to other people, I get down in the dumps too often. I've gotten pretty much what I expected out of life . . . APPENDIX F APPENDIX F INTERVIEW SCHEDULE In what ways has your relationship with your mother/daughter changed over the past 5 years? How do you feel about these changes? How do you expect your relationship with your mother/daughter to be in the future? What do you like or value about your relationship with your mother/daughter? What kind of problems do you see in your relationship with your mother/daughter? How do you deal with these problems? How much similarity do you feel there is between your answers on the questionnaires you just filled out and your mother's/daughter's answers? Please indicate by giving me a number 1 to 5 with 1 being the least similar and 5 being the most similar. If you expect differences, please describe briefly what they might be. What feelings about your relationship with your mother/daugh- ter do you believe are important, but were not mentioned in the questionnaires or this interview? How does your current relationship with your mother/daughter compare to the type Of emotional relationship you have (your mother had) with your (her) own mother? Please indicate how alike you feel they were by giving me a number 1 to 5 with 1 being the least alike and 5 being the most alike. 119 APPENDIX G APPENDIX C INTRODUCTORY LETTER AND CONSENT FORMS To the Volunteer, Our knowledge Of the emotional relationship between mothers and their middle-aged daughters is comparatively limited. Coopera- tion, from individuals, like yourself, is helping us find out more about the feelings of mothers and daughters in the middle and later years of life. Therefore, your interest in this study is greatly appreciated. This booklet contains a variety of different materials. Complete each section in this booklet in the order in which they appear. 00 not move ahead to the next section until you have finished answering all the items before you. Please answer items as truthfully as possible. Since we are asking for your own ideas, there are no right or wrong answers. All responses will be kept confidential. If you have any questions while going through this booklet, please feel free to ask them. Once again, thank you for participating in this project. Cheryl Steindel Project Coordinator 120 MICHIGAN STATE UNIVERSITY Department Of Psychology DEPARTMENTAL RESEARCH CONSENT FORM 1 have freely consented to take part in a scientific study being conducted by: Cheryl Steindel under the supervision of: Dr. A.I. Rabin Academic Title: Professor of Psychology The study has been explained to me and I understand the explana- tion that has been given and what my participation will involve. I understand that I am free to discontinue my participation in the study at any time without penalty. I understand that the results of the study will be treated in strict confidence and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. I understand that my participation in the study does not guar- antee any beneficial results to me. I understand that, at my request, I can receive additional explanation of the study after my participation is completed. Signed: Date: TITLE OF RESEARCH PROJECT: The emotional relationship between elderly mothers and their middle-gged dapghters 121 CONSENT TO TAPE RECORD FOR RESEARCH PURPOSES In the research project on the "emotional relationship between middle-aged daughters and their elderly mothers," conducted by Cheryl Steindel under the supervision of Dr. A.I. Rabin, Psychology Department, Michigan State University, 1. I am being asked to participate in an interview to understand my views about my mother-daughter relationship as stated in my own words. I understand that it is being asked that a tape be made of this interview for the convenience of recording my responses. I understand that my refusal to be tape recorded will not in any way affect my or any other person's participation in this study. I understand that tapes will be kept for six months from today and then will be erased. I understand that I am free to discontinue and/or erase the tape any time during the interview itself. I understand that I am free to request that the researcher erase the tape at any time during the next six months while the tapes are being held. I understand that the recording will be treated in strict confidence. I will remain anonymous. Within this restriction, the results of the study will be made available to me at my request. Please sign appropriate line below: I agree to tape record the interview. I do not agree to tape record the interview. Date 122 APPENDIX H 123 .m< 6:» Op momcommop m.poucw:mp paw m.no;uoE may «0 mnemwnmmEoo Hmcowumhocomlmmouo mucomonmon xm .Hozpoz mucomonmon 2N .uouzmsma monomOHHOH OH x mammm z x mbmu 2 x mxmmm N2 x ammm H SMHIA SsHHlHDflVO x Hmo H x xmmm H 4mmm H Emu H xmmm H wammm o whmu a wxmmeo mZmH> m.mm:Hoz .onom pcoenomuu< may wchD mcomwnwmsou Hmowumwumum HHCOHumhocoonmmonu-u.HH mgm vouuoHom van quuaHHan HacoHuauucou vo>Houuoa .nuoususua can .eeogue: eooxuon neefiue~eeueu--.e~ msn