Adult Development and Aging

Psychology and Aging 1996. Vol. 11. No. 4. 591-606

Copyright 1996 by the American Psychological Association. Inc. 0882-7974/96/13.00

Sources of Tension in the Aging Mother and Adult Daughter Relationship

Karen L. Fingerman Pennsylvania State University

This study explored sources of tension in the aging mother-adult daughter relationship. Forty-eight

dyads of healthy mothers over the age of 70 years (mean age = 76 years) and their adult daughters

(mean age = 44 years) were interviewed individually and then together about their relationship.

Responses to questions about tension were coded as referring to intrusiveness, exclusion, inappro-

priate care of self or other, eras referring to general habits or traits. The term developmental schism

is introduced to explain possible sources of tension in this relationship. Aging mothers and middle-

aged daughters are at different points in their adult development; developmental discrepancies may

foster interpersonal tension in their relationship. Mothers and daughters who described sources of

difficulty that were not related to developmental differences had more positive regard for the relationship.

Psychologists regard difficulties between offspring and par- ents as integral to child and adolescent development, but pat- terns of tension between aging parents and offspring have been all but ignored in the psychological literature. Intergenerational strife in early life has been linked to the child's developmental tasks of acquiring a sense of autonomy and identity (Ainsworth, 1979; Bettleheim, 1965; Erikson, 1963; Paikoff & Brooks- Gunn, 1991). Adult offspring are thought to have acquired identities and to have matured to a point where they no longer need to struggle against their parents (Baruch & Barnett, 1983; Blenkner, 1963). The establishment of identity does not pre- clude tension stemming from other issues, however. In fact, ten- sion between parents and offspring earlier in life may have less to do with the child's individual need to establish autonomy than with the differences between the child's and the parents' stages of development. Parent-child tension may stem from the difficulties that individuals who are at discrepant points in de- velopment confront when trying to maintain an intimate rela- tionship. Such discrepancies in developmental stage may also underlie intergenerational difficulties in adulthood.

Association between older parents and their offspring is not

This research was originally part of a doctoral dissertation conducted

at the University of Michigan, Department of Psychology, chaired by Eric Bermann and Joseph Veroff. A draft of this article was written

while the author was on the faculty at the University of San Francisco.

Lillian Troll and Diana Engle provided invaluable comments on drafts

of this article. Rosemary Blieszner offered advice concerning the fram-

ing of the problem.

This study was partially supported by the following awards: Univer-

sity of Michigan Rackham Graduate School Dissertation Award, Amer- ican Psychological Association Dissertation Grant, Society for the Study of Social Issues Grants-in-Aid Award, and a grant from Sigma Xi.

Correspondence concerning this article should be addressed to Karen

L. Fingerman, Department of Human Development and Family Stud-

ies, 110 Henderson Building South, Pennsylvania State University, Uni-

versity Park, Pennsylvania 16802. Electronic mail may be sent via In-

ternet to [email protected].

mandatory, and it could presumably be discontinued when dis- satisfying. Accordingly, theorists have suggested that parents and offspring who experience difficulties in adulthood maintain more distant relationships (Bedford & Blieszner, in press). Some psychologists have argued, however, that interpersonal tension is not antithetical to close relationships in later life (e.g., Rook, 1984, 1990). Tension may play an important role in any intimate relationship, and sources of tension may reflect differ- ences in the ways in which individuals approach the relation- ship. In fact, understanding sources of tension may illuminate the roles and perspectives each individual brings to a given rela- tionship. Thus, investigating what causes difficulties between aging parents and their offspring may lend insight into the na- ture of intergenerational relationships in old age.

Aging Mothers and Adult Daughters

The purpose of this study was to examine sources of tension in aging mother-adult daughter relationships when mothers are healthy, active, and independent. Although there is considerable individual variation in parents' and children's interactions, the quality of parent-child relationships in adulthood tends to vary as a function of gender. Yet, a lack of differentiation in previous research between older fathers and older mothers and among adult daughters, sons, and spouses has obscured an understand- ing of unique emotional characteristics of particular parent- offspring combinations (Thompson & Walker, 1984; Troll, 1988). The present study focused specifically on the mother- daughter dyad because this relationship tends to be distinct from other parent-child combinations in later life.

Mother-daughter bonds tend to be the strongest and most enduring of filial connections (Rossi & Rossi, 1990; Rossi, 1993). Women generally exhibit greater activity and more con- sistent interactions with members of generations above and be- low them than do men (Fischer, 1986; Hagestad, 1984; Walker & Thompson, 1983). In addition, Troll and Smith (1976) ar- gued that mothers and daughters communicate with one an- other directly, whereas other parent-offspring combinations

591

592 FINGERMAN

tend to engage in triadic interactions, usually involving a spouse. Fathers often interact with their offspring through the mother, and sons often relate to their own parents through their wives. Thus, negative interactions may have a greater impact on the mother-daughter relationship than they would on relation- ships that are less close and less insular. In fact, previous re- search has found that mothers and daughters claim to experi- ence greater tension, as well as greater intimacy, in their rela- tionships (Lehr, 1984; Troll, 1987).

Not only may the experience of tension play a more impor- tant role in mother-daughter relationships than in relationships involving sons or fathers, but the sources of difficulties in the relationship appear to vary as a function of gender. Lehr (1984) analyzed conflict situations from biographies of the life histories of 141 women and 185 men. She found that women's conflicts tended to be over family matters. Men tended to have less fre- quent conflicts, which were related to occupation, leisure time, or political events. Hagestad (1984, 1987) examined sources of intergenerational conflict among 148 three-generation families in Chicago and found similar gender differences. In her study, relatives reported that family matters were principal causes of trouble with daughters, mothers, and grandmothers. Political and social issues tended to underlie difficulties with male re- lations. Lehr's and Hagestad's findings suggest that tension be- tween female relatives in adulthood may be related to relation- ships in the family. However, neither study examined genera- tional differences in the experience of interpersonal tension. The question arises, do mothers and daughters cite the same issues as sources of difficulties in their relationships?

Both the sources of tension and the bonds mothers and daughters maintain appear to center around their shared invest- ment in familial relationships, and particularly their own rela- tionship. In this respect, mother-daughter relationships appear to share the dyadic feature of marital relationships. Spousal ten- sion also often revolves around the marital relationship itself. In fact, difficulties in such relationships may stem from the de- mands of maintaining a dyadic relationship. Moreover, there appear to be systematic differences related to role and back- ground that influence perceptions of difficulties. For example, researchers have found that gender differences in communica- tion styles appear to carry over into marital conflict; the im- plication of this research is that people in different social roles may communicate their feelings differently when they are upset (Acitelli, Douvan, & Veroff, 1993; Crohan, 1992; Gottman, 1979; Gottman & Krokoff, 1989). Older mothers and daugh- ters occupy different roles in the relationship and are also con- fronting different developmental changes; sources of tension may reflect differences in their developmental positions coupled with these role differences.

Building on Lehr's (1984) and Hagestad's (1984, 1987) findings, in the present study I sought to further differentiate sources of tension between aging mothers and adult daughters. Existent research on intergenerational conflict has tended to be atheoretical in nature. Although the present study was also in- tended to be descriptive, mothers' and daughters' responses to open-ended questions asked in individual and joint interviews were coded into content categories on the basis of a theoretical understanding of adult development.

The Developmental Schism

Tension between older mothers and adult daughters may re- flect issues pertaining to recent changes in their lives and their present point in development, or to more continuous or unre- solved early problems that have endured over the course of a lifetime (Havighurst, 1972; Steinman, 1979; Weishaus, 1978). Continuous or reactivated problems are often assessed in the context of present developmental changes. For example, physi- cal or mental decline that leads to an aging parent's new depen- dencies on offspring may evoke early feelings of rivalry with sib- lings (Brody, 1985). Although negative feelings may be associ- ated with earlier patterns, the source of difficulty is most readily attributed to current changes in the relationship, in particular the parent's decline. For this reason, in the present study I fo- cused on understanding the role of current developmental differences in generating mother-daughter tension.

The field of gerontology is based on a premise of continuing development across the course of adulthood. Accordingly, in many aspects of functioning, older adults are expected to be at a different point in their development than are middle-aged adults. Two people from different cultures may experience problems in their relationship as a function of their discrepant backgrounds. So, too, may mothers and daughters experience problems as a function of being at different points of develop- ment. The premise of this study is that interpersonal tension between mothers and daughters may stem from what I will refer to as a "developmental schism." A developmental schism in- volves a gap created by differing developmental processes expe- rienced by parents and offspring at any age. Developmental differences may exist across a wide range of developmental is- sues and may result in interpersonal tension, regardless of the quality of attachment or the strength of the relationship. The source of such tensions may change as mothers and daughters seek to maintain an ongoing intimate relationship across a lifetime.

The developmental schism is related to sociological concep- tions of parent-child differences in perceptions of their relation- ship. Bengtson and Kuypers (1971) introduced the term "de- velopmental stake" to describe intergenerational differences in perceptions of compatibility. They argued that parents are in- vested in offspring as a symbol of their own future, and thus, perceive greater compatibility with offspring. Offspring wish to establish their own mark on the world, and thus perceive less similarity. The developmental stake can be viewed as an exam- ple of a developmental schism, in which intergenerational dis- parities result from a difference in developmental tasks. How- ever, the premise of the developmental stake is linked to parent- child roles, without consideration of changes that occur within these roles over the course of the lifespan. The investment is assumed to be constant across time. The principle of a develop- mental schism subsumes the concept of developmental stake. Parents and offspring may experience tension not only as a function of the positions they occupy in their relationship at the sociological level, but as a function of discrepancies in specific developmental issues at a more psychological level. The devel- opmental issues an individual faces change over the course of the lifespan, from the two-year-old's increasing needs for auton- omy to an older adult's possible increasing needs for care and

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 593

support in the face of debilitating illness. Thus, although par- ents and children may experience tension stemming from de- velopmental differences across the lifespan, the specific mani- festation of developmental schisms in the relationship would be expected to change as individuals mature and confront new de- velopmental changes.

Developmental schisms may be found throughout the par- ent-child life cycle; individuals confront new developmental tasks beginning at birth and ending at death (Erikson, 1963; Havighurst, 1972). Not all developmental tasks infringe equally on the parent-child relationship, however. For example, the middle childhood years are often considered ones of relative tranquillity in the parent-child relationship. Yet, children are clearly undergoing developmental tasks during this period, as are their parents. In fact, certain changes may be viewed as pos- itive for the relationship by one or both parties. The child's sec- ond year of life is marked by gains in language ability, as well as by efforts to establish autonomy. These linguistic advances are often viewed favorably by parents, who find it easier to commu- nicate with a child whose communication style begins to match their own more closely. Offspring may appreciate parents* more relaxed parenting styles as they mature in their roles over time. Not all developmental changes generate parent-child tension, just as not all gender differences generate marital tension. Rather, where tension exists in parent-child relationships, un- derstanding the interplay of developmental patterns may ex- plain the tension. Developmental issues that generate friction are likely to involve changes that infringe on the other person in some manner.

In other words, the key aspects of understanding tension lie in each individual's perception of the source of tension and in their interactions together. Parents who perceive their two-year- old's increasing autonomy as a sign of budding genius are un- likely to frame her exploratory behavior as a source of friction, and are also unlikely to interact with her in a manner that gen- erates tension around her exploration. In a parallel situation, a middle-aged daughter who views her mother's desire to travel to India as a sign of her continuing vitality is less likely to consider this issue a source of tension and less likely to discuss this issue as a problem with her mother. A middle-aged daughter may be- gin to worry about her mother's aging and mortality as she con- fronts her own transition to mid-life (Cicirelli, 1988; Neugar- ten, 1968). Such a daughter might worry about her mother and view her desire to travel to India as too risky. She may hint to her mother that a trip to India is inappropriate at her age, and mother-daughter tension may ensue from such interactions.

Developmental Issues in This Study

The issues examined in this study reflect disparities stem- ming from differences in the lives of middle-aged and older women that might generate tension in the mother-daughter re- lationship. To understand tension in relationships between younger mothers and daughters, different developmental issues would need to be examined. Discrepancies in many aspects of development might have an impact on relationships in later life between older parents and their middle-aged offspring. For ex- ample, there is a lot of literature emphasizing the stresses that middle-aged daughters and their mothers experience when the

mother's physical changes necessitate changes in patterns of de- pendency (Brody, 1985, 1990; Litvin, 1992; Semple, 1992; Stull, Kosloski, & Kercher, 1994; Zarit, Reever, & Bach-Pe- terson, 1980). There has been little attention to tension mothers and daughters experience when both are relatively healthy. The present study focused on a limited examination of two general areas where developmental differences might create tension be- tween middle-aged daughters and older mothers. The areas ex- amined included perceptions of the parameters of the relation- ship and perceptions of the other's self care.

The manifestation of developmental concerns may vary as a function of culture, ethnicity and social class. Although the pro- posed sources of disparity are steeped in theory of adult devel- opment that may cut across groups, the ways in which individ- uals interpret the developmental differences they have with their parents or offspring may vary as a function of their own cul- tural, educational, or other past experiences. The issues exam- ined in this study are thought to be most applicable to the well- educated, healthy American sample used. Other areas of difficulties might be more salient amongst other groups. The fundamental question is how human development contributes to difficulties between parents and offspring rather than the par- ticular issues examined.

Importance of the Relationship

The first area of tension explored in this study involved differences in perceptions of the salience of the relationship. Mothers and daughters may experience tension in later life stemming from differences in the primacy they give to their re- lationship. In recent years, there has been considerable research investigating and substantiating a model originally proposed by Carstensen(1987, 1991, 1992), referred to as socioemotional selectivity theory. Carstensen posited that as people age, they narrow the focus of their social network to maximize emotional gains from fewer relationships. According to the socioemotional selectivity model, older adults might view their relationships with offspring as more important than offspring do. For exam- ple, middle-aged daughters would be expected to have a greater number of ongoing social relationships than would mothers. Thus, relationships with mothers might be less salient and meaningful to daughters. This possible discrepancy in the im- portance of the relationship is an example of a developmental schism. Tension may result from the mother's increasing social selectivity if mothers and daughters perceive the other's behav- iors as either demanding of. or inattentive to. their relationship. The individual descriptions of tension might involve instances where a daughter perceives her mother as intrusive or where a mother perceives her daughter as excluding her. Yet, the un- derlying cause of this tension may reflect levels of investment in the relationship.

Generational differences in perceptions of the parameters of the relationship were anticipated. Hypotheses were generated in a manner reflecting the descriptive nature of the data. Daugh- ters were expected to be more likely to cite their mothers' intru- siveness as sources of difficulty, particularly when the mothers rated the relationship as more salient. Mothers were expected to be more likely to describe feelings of exclusion from their daughters' lives, again particularly when they were more in-

594 FINGERMAN

vested in the relationship. In the joint interview, dyads were ex- pected to discuss disagreements over contact.

Care of Self or Other

Changes pertaining to the mother's aging may also evoke ten- sion in the mother-daughter relationship. Offspring may expe- rience stress in anticipation of problems in the aging process, even when the parent is still healthy and independent. Midlife is a period in which individuals begin to think about how much time they have left; middle-aged adults ponder growing old and dying (Kastenbaum&Aisenberg, 1972; Neugarten, 1968). Ac- cordingly, middle-aged adults may become increasingly aware of their parents' aging. In fact, as a middle-aged daughter be- comes aware of her mother's aging, she may see it as a greater contributor to problems than her mother does. Cicirelli (1988) coined the term "filial anxiety" to describe tension generated when an adult child worries about care an elderly parent may need someday. As a middle-aged daughter worries about her mother's aging, an older mother may resent her daughter's con- cern about her future decline while she is still healthy. It is also possible that a mother would consider her own aging as a source of tension with her daughter, particularly if health declines pre- vented the mother from engaging in the relationship as actively as she had in previous times. In the present study, awareness of the mother's aging was examined as a potential source of ten- sion. It was anticipated that healthier older mothers would be less likely to describe their aging as an issue in their relationships with their daughters than would their daughters.

A complementary category of tension was examined pertain- ing to the daughter. A mother's complaints about her daughter's self-care may be related to enduring aspects of the maternal role or to developmental changes in the daughter's life. Across the lifespan, a mother might be expected to act "motherly" by ex- pressing concern over her daughter's well-being. However, there may be shifts in this pattern as the mother and daughter grow older. The mother's solicitude for the daughter in general may be seen as an ongoing source of tension in the relationship, par- ticularly if she has always been a mother who worries. Yet. changes in the nature of the concerns may be manifested as the daughter grows older. Concerns may begin to center around the daughter's care for her own spouse or her own children; these concerns are related to the daughter's current position in the lifecycle. Thus, the mother may continue to function as a mother in her solicitude for daughter. At the same time, the mother's concerns over the daughter's ability as a wife or mother may reflect changes in their relationship related to individual development.

Nondevelopmental Issues

The mother-daughter relationships examined in this study have endured over the course of a third to half a century. In addition to issues related to present developmental status, mothers and daughters might also experience more continuous issues as sources of present tension. Although all mothers and daughters differ in age, these differences may be more important or less important across relationships at different points in time. Sources of tension unrelated to developmental differences were also considered in this study.

Mothers' and daughters' complaints about the others' habits, traits, or general behavior were examined. These sources of ten- sion were not considered to represent developmental schisms. Such problems may be inherent to interpersonal relationships. Regardless of gender, age, or cultural differences, people en- gaged in intimate relationships may lake note of faults in the other person. In the context of mother-daughter relationships, a mother may be bothered by her daughter's housekeeping or lack of promptness at any age. A daughter may feel that her mother has had embarrassing habits across the lifespan. Such difficulties seem to be less a function of the mother's aging, the daughter's adult status, or the quality of their relationship than a matter of perceiving something undesirable in an otherwise loved other. Elsewhere, researchers interested in friendships re- port that older adults who maintain a problematic friendship in later life tend to see the source of difficulties in that relationship as something pertaining to their friend's personality attributes (BIieszner& Adams, 1995).

The distinguishing aspect of perceiving a habit or trait in a loved one as the thing that annoys one most lies in the ability to see this trait as an aspect of the other person unrelated to the relationship. It is possible to see something annoying about the other person that infringes on the relationship, in which case the problem becomes something the other "is doing to you." References to habits or traits here refer to issues perceived as a part of a loved one that may be annoying but are not experi- enced as an affront. The ability to perceive faults in a parent without taking these faults personally may be part of what Blenkner (1963) referred to as filial maturity. She argued that offspring may undergo filial maturity during their thirties, al- lowing adult children to perceive their parents as human beings with weaknesses and vulnerabilities. An aspect of this matura- tion may involve refraining tensions so that they are no longer perceived as being based in the relationship but, rather, are per- ceived as being based in the parent. The ability to perceive un- desirable attributes in a loved parent suggests an absence of the types of developmental differences that get played out in parent- child tensions. Such perceptions on the part of offspring may contribute to a better parent-offspring relationship in later life.

By contrast, an aging mother who sees her child's flaw as a source of tension may feel worse about her relationship with that offspring. Previously, researchers have found that parents' perceptions of how their grown children turned out has an effect on their own well-being (Ryff, Lee, Essex, & Schmutte, 1994). In the early parenting years, parents may derive a sense of com- fort from the thought that their children will outgrow whatever current developmental issues seem to underlie tensions in their relationship. By old age, parents may feel more comfortable with developmental discrepancies than do offspring. For exam- ple, an aging mother might be able to accept the idea that her working daughter has many competing demands on her time, and she might use this to explain the differences in their levels of investment in the relationship. A mother's sense that the daughter is lazy, however, might be related to her greater dissat- isfaction with the relationship.

Assessing Tension in Aging Parent-Child Relationships

Finally, this study was designed to gather information about sources of difficulty by using dyadic analyses obtained across

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 595

two interviews. Previous research has generally concluded that

parent-child tension is minimal in later life because older par-

ents often report that they have no problems in their relation-

ships with their offspring (Mancini, 1989; Streib & Beck,

1980). However, the failure to obtain data may be an artifact of

the methods used to assess such tension rather than an actual

lack of tension. Much previous research on parent-child con-

flict in later life has relied on survey items involving closed-

ended ratings (Aldous, 1987; Cicirelli, 1983a, 1983b; Thomp-

son & Walker, 1984; Whitbeck, Hoyt, & Huck, 1994). In their

review of a decade of literature on aging parents, Mancini and

Blieszner (1989) suggested that a better understanding of con-

flict and tension might be obtained with open-ended interviews,

allowing time for participants to build rapport with the in-

terviewer. Talbott (1990) obtained information pertaining to

older widows' difficulties with adult children indirectly, using

open-ended interview discussions. She did not systematically

seek information concerning tension, but rather coded sponta-

neous mentions of problems from lengthy interviews.

The present study employed an open-ended interview tech-

nique, designed to more systematically assess descriptions of

problems in the relationship. A broad understanding of diffi-

culties was used. An emphasis was placed on gathering infor-

mation about the causes of problems rather than about the com-

munication of such problems. Aging mothers and adult daugh-

ters may not openly confront one another with things that

bother them in their relationship (Fingerman, 1995; Wodak &

Schulz, 1986). Thus, the term tension was operationalized to

emphasize internal negative experiences as well as overt con-

frontation, including interpersonal difficulties, negative feel-

ings, disparities of interest, miscommunications, and open

confrontation.

To understand the role that each individual's perceptions of

the relationship play in interpersonal tension, both parties'

points of views must be considered (Thompson & Walker,

1982). Previous studies addressing intergenerational conflict

have been limited to only the parents' (Aldous, 1987; Talbott,

1990) or the offsprings' (Cicirelli, 1983a; Troll, 1987) reports

of difficulties. In the present study, mother-daughter pairs were

examined using a dyadic approach. Furthermore, a joint assess-

ment of tension may lend insight not available from compari-

sons of individual interviews (Acitelli, Douvan, & Veroff,

1993). When individuals are interviewed alone, they may de-

scribe underlying problems in the relationship that may not be

recognized by the other person. When two people are inter-

viewed together about difficulties in their relationships, they

may be more likely to discuss issues that receive attention in the

relationship, problems of which both individuals are aware.

In the present study, individual and joint interviews were

conducted.

Although the principal focus of this study was on providing a

descriptive understanding of dyadic perceptions of sources of

tension, the impact of such tension was also considered. The

criterion variable for the study was the quality of the mother-

daughter relationship, operationalized as mothers' and daugh-

ters' regard for the relationship. Elsewhere, researchers have in-

dicated that the quality of the parent-child relationship in later

life is important to adults' general well-being and functioning

(Ryffetal., 1994).

Table 1

Demographic Characteristics of the Sample

Characteristic

Marital status Married Remarried Engaged or living with other Divorced Widowed Single

Religion Protestant Catholic Jewish Other

Years of education At least some high school High school Some college College Graduate school

Employment status Currently works for pay Has worked for pay Has never worked for pay

Mothers (n = 48)

18 1 0 2

27 0

18 10 14 6

5 11 15 8 9

3 41

4

Daughters (n = 48)

35 5 2 4 1 1

14

7 16 11

1 2 7

20 18

34 11 3

Sources of tension may be related to levels of regard for the

relationship. It was hypothesized that daughters who experi-

enced tension around their mothers' habits or traits would have

higher regard for their relationship than would those daughters

who described issues related to differences in their individual

development. Therefore, daughters' citing issues related to the

parameters of the relationship, the mother's aging, and the

mother's failure to care for self were expected to be associated

with less regard for the relationship. Daughters' describing is-

sues that seem to be less related to individual development, such

as taste in clothes, where to have lunch, or general attributes

were expected to be associated with higher regard for the rela-

tionship. For mothers, issues related to developmental differ-

ences were expected to be associated with higher regard for the

relationship, whereas perceptions of daughters' flaws were ex-

pected to be related to lower regard for the relationship.

Method

Participants

The data presented here are derived from a larger study of aging mothers and their adult daughters. The larger study was designed to assess positive and negative aspects of mother-daughter relationships.

The sample consisted of 48 dyads ( N = 96) of older mothers (mean age = 76.01 years. SD = 5.22 years) and their adult daughters (mean age = 44.02 years, SD = 7.03 years). Table I contains individual demographic

variables describing the mothers and daughters in this sample. Forty-six of these dyads were White, and the other 2 were African-American. Mothers and daughters were highly educated, though the daughters tended to be more highly educated than their mothers. All women claimed to be in good health at the time of the study. Mothers reported being ill an average of 13.60 days (SD = 25.30 days) in the past year, compared with daughters' reports of 6.45 days ( S D - I I .60 days). (For

596 FINGERMAN

further details pertaining to this sample and recruitment procedures,

see Fingerman, 1995.)

Mothers and daughters all resided in separate households, within an

average of l-hr's drive of one another. Mothers and daughters reported

frequent contact with one another. Ninety percent of mothers and

daughters reported at least biweekly visits and weekly telephone calls.

Thirty mothers were either divorced or widowed and resided alone, and

all but 6 daughters had either a spouse or a live-in boyfriend. Of those 6

daughters, all but 2 had serious on-going relationships with a man.

Mothers had an average of 3.06 children (SD = 1.83), with an average

of 1.96 residing in the area (including the participant daughter).

Daughters had an average of 2.45 children (SD = 1.33), with an average

of 0.88 children (SD = 1.00) residing in their homes. The women who participated in this study are not representative of

the general population. The mothers in this study were healthier, wealth-

ier, and more well-educated than most other women their age. However, the aging mothers in this study and their relationships with their daugh-

ters provided a forum for investigating sources of tension when many

factors associated with the downside of the aging process are absent, including economic dependency, educational disparities, poor health,

cognitive decline, and emotional isolation. The generalizability of the

findings may also be limited by the underrepresentation of ethnic mi-

nority elderly. Thus, sources of difficulties examined in this study are

thought to most closely reflect the experiences of healthy, economically,

and educationally advantaged women in the dominant American

culture.

Procedure

Mothers and daughters were first interviewed separately, were then

given questionnaires to complete on their own, and were finally inter-

viewed together 1-2 weeks later. The interviews lasted approximately 1

hr each. All interviewers were women in their mid-twenties, with at least

a bachelor of arts degree and two years of research experience in the

field of psychology. In addition to sources of tension, interviews and

questionnaires addressed family background, other aspects of conflict,

positive aspects of the relationship, demographic information, and well- being. I have reported data pertaining to mothers' and daughters1 rat-

ings of their behaviors during a recent past conflict in a previous article

published in this journal. These data were obtained during the joint

interview, after mothers and daughter had discussed their most recent tense situation (see Fingerman, 1995). The descriptions of tension pre-

sented here were obtained during the individual and joint interviews;

the ratings of the relationship were included in the questionnaires.

Salience of Relations hip

Information pertaining to the social context of the mother-daughter

relationship was obtained at the beginning of the individual interview.

Pilot testing revealed that older mothers and daughters had a difficult

time ranking the importance of the other in their social network when

asked to do so in a precise manner. In this study, mothers and daughters were asked to rate the importance of the other using four categories: the

most important person in their life, within the 3 most important, within

the 6 most important, and within the 12 most important people. Partic-

ipants were also asked to name the person with whom they get along best, in whom they confide most easily, and with whom they are an-

noyed most often.

Sources of Tension

To allow time to build rapport, participants first discussed positive aspects of their relationship during the individual interview. The first

question about sources of tension came in the middle of the interview. This question was worded to encourage discussion of discontent, even

in the absence of overt conflict. Participants were asked to "Tell me

about the last time you felt irritated, hurt, or annoyed during a visit with your daughter/mother."

In the joint interview, mother and daughter dyads were asked to select

their last tense situation. They were asked to determine the incident that

constituted the last time they had had a disagreement or that one of

them had felt hurt or annoyed by something the other person said or

did. This question focused on more overt conflict than the parallel ques-

tion in the individual interview. The purpose of the joint interview was

to have dyads select events that both mother and daughter had experi-

enced as tense, to examine how individual perceptions differed from

joint experiences of difficulties. Most mother-daughter dyads reached

a consensus on their situation in a matter of 2-5 min, with no dyad

taking more than 10 min. To facilitate mothers' and daughters' comfort

with speaking openly during the individual interviews, participants

were assured of their confidentiality in that interview. Thus, sources of

tension described during the individual interviews were only discussed

in the joint interview if participants themselves brought up those issues.

Content Codes

The content of the individual and joint interpersonal tension re-

sponses was coded by two independent coders from audio tapes and

transcriptions taken during each interview. Table 2 contains examples

of content codes from the individual interviews and interrater

agreement for each code, which was established on ratings of 24 ran-

domly selected dyads. The applicability of the content codes for the

individual and joint interviews was established post hoc on the basis

of the theoretical justification posited in the introduction. All content

categories were brought up in at least two mother and daughter in-

terviews. Details of content codes are discussed below. The general cod-

ing procedure was as follows: The theme of each respondent's individual

interview response was coded first. Responses were coded as including

people in addition to the mother-daughter dyad (e.g., grandchildren,

spouses, lovers, in-laws, friends). Finally, participants' responses were

coded for spontaneous discussion of the mother's aging and for inclu-

sion of positive material or praise for the other,

Importance of the relationship. Three codes were included to exam-

ine intergenerational differences in perceptions of the primacy of the

relationship. Discussions describing the other as being obtrusive, mak-

ing critical remarks, and making demands were coded under the con-

densed theme intrusion, on the basis of coders' inability in many in- stances to distinguish between these three subcategories. A respondent's

report that the other did not include her in activities or did not spend

enough time with her was coded as exclusion,

To more fully examine intergenerational differences in perceptions of

the parameters of the relationship, the people included in the responses

to questions about tension were examined. Each response was coded as

being limited to the mother-daughter dyad or focusing on outsiders, such as husbands, siblings, or children.

Care of self-other. Complaints that the other did not take appropri-

ate care of self or of others or took inappropriate risks were coded as

inappropriate care. In addition, mothers and daughters who described

physical risks the other was taking or allowing others to take were coded

as inappropriate care- The target of the inappropriate care, self or other,

was also coded.

No participant mentioned the mother's aging as a principal source of tension. However, many participants mentioned the mother's age in the

course of their discussion. For example, a daughter might preface her

discussion of what irritates her about her mother with a comment on

her mother's age, "She's growing older now, so it's harder for her to remember. Sometimes she forgets important things I've told her. That

bothers me." Spontaneous remarks related to the mother's aging were coded from the audiotapes.

Habits-traits. Finally, enduring individual attributes such as an-

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 597

Table 2

Content Codes and Examples

Content code Example

Parameters of the relationship Intrusion

Exclusion

People discussed Others included

Dyad only

Vulnerability Care of self-other

Mother's aging

Attributes Habits- traits

Compatibility Praise of other

Mother: "She sends people into the kitchen to help me." Daughter. "She tells me how to discipline my kids." Mother: "She has conversations with her father where I feel left out.'1

Daughter: "I came over to visit and she just watched TV."

Mother: "Her husband kept turning up the radio every time I turned it back down!" Daughter: "Dinner at mother's went past the kid's bedtime. They were overtired." Mother: "The way she diets, she obsesses over it. I hate having lunch with her." Daughter: "She kept criticizing my driving when we went shopping last week."

Mother: "She lets her girls go into the city alone. It's dangerous!" Daughter: "She won't use a wheelchair in the airport." Mother: "She is well intentioned, but she makes me feel my old age." Daughter: "She's older now. She can't remember things as well. It's hard."

Mother: "She's too hard on other people: her family, her co-workers." Daughter: "She has terrible taste in clothes."

Mother: "She rarely bothers me, she's so easy-going." Daughter: "She kept talking about people I don't even know . . . She really is sociable!"

.93

.91

.91

.91

.97

.79

.86

.83

noying behaviors, traits, and bad taste were grouped together under the

theme habits and traits.

Praise. Positive remarks made in response to the questions about interpersonal tension were coded from the audiotapes. Several mothers

and daughters either prefaced or interrupted their discussions of diffi- culties with comments about the high quality of their relationship or

positive attributes the other individual possessed.

The content category of each response was coded from the perspective

of the person interviewed. For example, if a mother complained that her daughter did not watch over the grandchildren enough, that response

was coded as pertaining to care of self or others. If the mother com- plained that the daughter made unwanted doctor's appointments for her (the mother) that situation was coded as intrusion. If a daughter

complained that her mother told her how to clean her house, that re-

sponse was coded as intrusion. If a mother complained that her daugh- ter was not a good housekeeper, that complaint was coded as a habit or

trait. The category codes were not mutually exclusive. A given response could fall under more than one category. Every mother and daughter response was classified under at least one content category. One mother

and 1 daughter, from different dyads, described political issues as sources of tension; the daughter went on to discuss other things that

bothered her about her mother, and these issues fell under habits and

traits. Given the low incidence rate, political issues were not addressed further as a cause of tension.

Joint interview scenarios. The joint interview responses were coded in a manner akin to that used for the individual interviews. There were

some differences in the codes, however, as mothers and daughters were often more tentative in their discussions of sources of tension in the

joint interview. Rather than discuss intrusiveness, mothers and daugh-

ters tended to refer to unwanted advice that was offered by one or the

other. Situations discussed in the joint interviews also fell under fewer,

larger themes. For example, participants did not mention each other's habits or traits as a source of tension during the joint interview and they

did not bring up the mother's age. Thus, coding of joint interviews was

limited to the following: unsolicited advice, inappropriate care, and in- clusion of outsiders. One additional level of coding was applied to the joint interview: The joint interviews were coded with regard to whose

domain the story fell under—the mother's life, the daughter's life, or both-unclear.

Assessment of Relationship Quality

Positive regard for the relationship was assessed in the questionnaire, using a modified version of Bengtson's (1973} Positive Affect Index. This instrument is comprised of 2 five-item scales assessing perceptions of feelings for the relationship held by self and by other with regard to trust, respect, understanding, fairness, and affection (Bengtson &

Schrader, 1982). Responses were rated on a 5-point Liken scale ranging from 1 (not well) to 5 (very well). The instrument was originally devel- oped and validated on individuals in three-generation families (N = 1,350) drawn from a list of members of a health plan. Ratings of self and other can be combined into a single index of perceptions of the

quality of the relationship. Cronbach's alphas for this measure in the present study were .90 for mothers and .86 for daughters.

Results

Salience of Other

Table 3 contains information pertaining to the salience of the

mother to the daughter and the daughter to the mother. Where

relevant, McNemar'stest(Conover, 1980) was conducted to ex-

amine intergenerational differences in mothers' and daughters'

ratings of salience. McNemar's test is a nonparametric statistic

allowing paired comparisons in a 2 X 2 contingency table. The

test allows calculation of exact p values for instances when fewer

than 30 cases have different values in the matched pair; signifi-

cant findings indicate lack of symmetry in the table.

Mothers and daughters tended to rate the other person as be-

ing important in their lives. Ninety percent of mothers and

daughters ranked the other as being amongst the six most im-

portant people in their lives. However, mothers tended to be

more likely to name the daughter as being the most important

598 FINGERMAN

Table 3

Salience of Mother to Daughter and Daughter to Mother

McNemar's test exact

Salience Ratings Mothers Daughters p value

Ranked importance of other Most important person in life Within the top 3 Within the top 6 Within the top 10

Who get along with best Target (mother or daughter)

Who can talk to when upset Target (mother or daughter)

Who annoyed with most often Target (mother or daughter)

1 1 25

T

5

23

26

4

4 24 15 5

6

8

15

.0963'

.0005

.0002

.0350

a McNemar's test calculated using 2 groups: I = top 1 or top 3 most important, 2 = amongst 6 or 10 most important.

person or amongst the three most important people in their

lives; 75% of mothers indicated the daughter fell amongst the

three most important, compared with 58% of daughters.

There were intergenerational differences in perceptions of the

other person's emotional role in the relationship. McNemar's

exact p values were calculated using a binomial code in which

naming the mother or target daughter in response to questions

regarding important social functions was coded as 1 and nam-

ing someone other than that individual was coded as 2. Forty-

seven percent of mothers (n = 23) named the participant

daughter when asked whom they got along with best compared

with only 1 3 % ( « = 6) of daughters. Mothers were also more

likely to name the target daughter when asked whom they spoke

to when upset; 52% (n - 26) of mothers named their daughters,

compared with 17% (n = 8) of daughters naming their mothers.

By contrast, daughters were more likely to cite their mothers as

the person who annoyed them most: 30% ( n = 15) of daughters

cited their mothers as the principal source of annoyance com-

pared with to 9% (n = 4) of mothers.

Point-biserial rs were calculated to examine the association

between size of family and whether the mother and daughter

named the other as preferred confidant or source of irritation.

The mother's family network seemed to influence her ratings

of the daughter and their relationship. There was a significant

correlation between number of children mothers had and how

important they rated the participant daughters in their lives ( r

= .27, p < .05). Mothers with more children were also more

likely to name a child other than the target daughter as their

preferred confidant. Being widowed was not related to mothers'

ratings of daughters; even married mothers were likely to nomi-

nate a child as a preferred confidant. The size of the daughter's

network of siblings and her own children seemed to be orthog-

onal to the importance she attributed to her relationship with

her mother.

Content oj Tension in the Individual Interviews

Table 4 shows the proportions of responses from mothers'

and daughters' individual interviews fitting each content code

and McNemar's test examining differences in these propor-

tions. McNemar's tests were calculated by devising two catego-

ries for each code: (a) gave response to which code applies and

(b) did not give response to which code applies.

Response Rates

Although all daughters in this study answered the interper-

sonal tension questions, 8 mothers claimed to be unable to

think of a recent time when they felt upset with their daughters.

It is unclear whether these mothers' failures to respond to ques-

tions about tense situations in the individual interview repre-

sents an actual lack of perceived difficulties in the relationship,

memory problems, social desirability considerations, or a com-

bination of factors. These mothers will be referred to as "no-

example mothers" in this study, as an indication of their not

providing a response to the question about tension in the indi-

vidual interview, regardless of the underlying reason.

All mothers participated in the joint interview. Thus, dyads

with no-example mothers in the individual interviews were not

excluded from the study; rather, care was taken to conduct anal-

yses of the individual interviews to deal with their "missing

data" conservatively. Analyses involving data from the individ-

ual interviews were conducted twice, once including all moth-

ers, and once excluding no-example mothers and their daugh-

ters. The pattern of results was parallel. The number of mothers

and daughters giving responses included under the category

codes can be found in Table 4. McNemar's exact p values in this

table reflect more conservative calculations than would result

from using the entire sample: these values were calculated ex-

cluding the eight dyads where the mothers did not provide an

example of difficulties.

Intrusion, Exclusion, and Salience of the Other Person

There were generational differences in the number of mothers

and daughters who described the other's intrusiveness as a

source of tension in the individual interviews. Nearly half of the

daughters described incidents where their mothers were critical,

demanding, or intrusive. Fewer mothers offered instances of

their daughters' being intrusive.

Table 4

Numbers of Mothers and Daughters Mentioning Each

Category of Content Coded in Individual Tension

Responses and McNemar's Test

Content code

Provided no example Intrusion Exclusion Discussed others Appropriate care Mother's aging Praise of other Habits-traits

Mother interview

8 11 7

22

6 7

27

14

Daughter interview

0 23

5 16 1 1 24

15

13

McNemar's test exact/? value3

.0039

.0500 ns

.0301 ns

.0009

.0453 m-

a McNemar's test was conducted in a conservative manner using only those mother-daughter dyads in which the mother responded.

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 599

There were no significant differences in the proportions of

mothers and daughters whose responses were coded as involv- ing complaints about feeling excluded. Only 7 mothers and 5 daughters described situations in which they felt excluded by the other.

Mothers were more likely to include other people in the inci- dents they described in the individual interviews. Mothers' re- sponses involved grandchildren, other offspring, sons-in-law, and their own husbands. More daughters confined their discus- sions to situations involving themselves and their mothers.

Salience of other and complaints about intrusion. It was posited that complaints about the parameters of the relation- ship might stem from a developmental schism. If this were the case, intergenerational differences in the emotional salience of the relationship would contribute to mother-daughter tension over the parameters of the relationship. Mothers' and daughters' descriptions of situations involving intrusiveness or exclusion were examined as a function of their ratings of the salience of the other by using point-biserial correlations. Dichotomous codes were derived by examining whether the mother or daugh- ter named the other in response to different questions about emotional functions ( 1 = naming the target as the person par- ticipant gets along with best, 0 = naming other person; 1 -nam- ing the target as the person participant confides in, 0 = naming other person; 1 = naming the target as the person participant is annoyed by most, 0 = naming other person) and whether the mother or daughter felt intruded upon (1 = yes, 0 = no).

Daughters' feeling intruded upon was related to mothers' rat- ings of the salience of their daughters. Daughters whose mothers named them as the person they prefer to confide in were more likely to describe tense situations in which their mothers were intrusive (r — .37, p < .01). Daughters' emotional investment in their mothers was not related to whether they felt intruded

upon by their mothers. Mothers' descriptions of intrusion were not predicted by

their own ratings of the salience of their daughters. Salience of other and reports of exclusion. Because only 7

mothers reported feeling excluded, McNemar's tests were used to examine the distributions of ratings of salience and the mother feeling excluded. Dichotomous codes were developed as described above. Mothers who value their daughters but do not turn to them for positive emotional functions appear to be more likely to feel excluded. Mothers who named someone other than the target daughter as the person they prefer to confide in and mothers who named someone other than the target daughter as the person they got along with best were more likely to feel excluded (McNemar's exact ps = .0019 and .0033, respec- tively). Mothers who said their daughters were amongst the top 3 people most important people in their lives were also more likely to feel excluded than mothers who said the target daughter was amongst the top 6 or 12 (McNemar's exact p = .0001).

Daughters who rated their mothers as less important in their own lives were more likely to have mothers who felt excluded (McNemar's exact p = .0005). In addition, daughters who de- scribed their mothers as intrusive had mothers who were more likely to describe feelings of exclusion. Of the 7 mothers who felt excluded, 5 had daughters who reported feeling intruded upon (McNemar's exact p = .0007).

Care for self-other. Eleven daughters and 3 mothers de-

scribed situations in which the other individual's failure to care for self lay at the root of tension. In addition, 3 mothers de- scribed situations in which a daughter's inappropriate care of her own offspring was the root of difficulties.

Daughters were more likely to cite their mother's aging in their responses than were mothers. Nearly half of the daughters mentioned that their mother was getting older, compared with only 7 mothers mentioning their own aging. Daughters' men- tion of their mothers' aging was not related to the mothers' age or health status, though it was related to their own age ( r = .24, p < .05), with older daughters more likely to mention their mothers' age.

Given the small number of mothers who complained about daughters' self-care, analyses were not conducted to examine the age or health of these mothers. The number of and ages of the daughters' children were not related to the mother's express- ing concern over the daughters' care of others, although, again, only 3 mothers described their daughters' care of offspring as a principal source of concern.

Habits-traits. There were no significant differences in de- scribing the other's habits-traits as sources of tension. Nearly a third of mothers and daughters cited the other's attributes as a source of tension.

Compatibility. The majority of mothers spontaneously praised their daughters or commented on positive aspects of their relationship when asked to discuss difficulties. For exam- ple, 1 mother responded to the question about feeling hurt, irri- tated, or annoyed with, "She's so generous, I just wish she'd save a little more money." The mothers who gave no example of tension in the individual interviews often praised their daugh- ters rather than indicate a time when they were upset with them. Daughters were less likely to discuss their mothers' positive attributes.

Daughters of no-example mothers. To more fully un- derstand the dyads involving no-example mothers, the descrip- tions of tension provided by the daughters whose mothers did not supply a response were examined separately. These daugh- ters were more likely to describe their mothers' habits and traits as a source of annoyance. Six of the 9 daughters who had no- example mothers described situations involving their mothers' annoying habits or traits.

Qualitative examples. The distribution of coding of re- sponses seems to be accounted for by the manner in which mothers or daughters framed their irritation. The types of issues described under any given content code might also have fallen under a different code had it been perceived differently. For ex- ample, 1 daughter, whose response fell under habit-traits ex- plained, "My mother has horrible taste in clothes. Sometimes she brings my girls clothes that are just hideous. We smile and are gracious, but you know, I just pack them up and send them to Good Will. Her taste has always been that bad." This daugh- ter's response might have fallen under the intrusion code, had she framed her complaint differently. For example, the daughter might have stated instead, "My mother tries to foist her taste in clothes on my kids. She's never had taste like mine. Yet, she insists on bringing my girls clothes she knows they aren't going to wear and that I don't think they should wear."

In another example, a mother and a daughter both cited differences in their religious practices as the principal source of

600 FINGERMAN

difficulty in the relationship (the daughter converted to a different religion when she married). The mother felt excluded from her daughter's family life, and the daughter felt impatient and guilty about her mother's insistence that the daughter had converted to another religion in order to hurt her. In 2 different mother-daughter dyads, the daughters' conversion to another religion at marriage was not mentioned as a source of tension, and 1 mother even stated that she enjoyed not having to fight with in-laws (who were not Jewish) for her daughter's attention during the Jewish holidays.

There were also intergenerational differences in the qualita- tive types of issues mothers and daughters raised when describ- ing the other as intrusive. When mothers claimed their daugh- ters were intrusive, their complaints tended to focus on un- wanted care, such as "She's always sending people into the kitchen to help me." When daughters complained about their mothers' intrusiveness, they tended to focus on excessive de- mands, criticism, and unsolicited advice.

Content of the Joint Responses

All dyads were able to establish an incident that constituted their last tense situation during the joint interview. One dyad insisted on using the debate they had had in responding to the question about their last tense situation to constitute their last tense situation. All other dyads selected incidents outside the context of the study.

Mothers' and daughters' relative participation in the joint in- terview discussion was coded from audiotapes of their conver- sation and written transcriptions obtained during the in- terviews by two independent raters. Cohen's kappas were calcu- lated on ratings of 21 of the 48 dyads coded to establish interrater agreement. Chi-square analyses revealed that there were no significant differences in mothers' and daughters' like- lihoods to instigate discussion (« = .87), choose the situation ( K = .87), or disagree with something the other brought up during that conversation ( K = .72). Daughters were more likely to de- scribe what had happened in the situation in question, after the incident had been selected, x 2 = ( 1 , N = 48) 12.84, p < .005 (* = .91). Chi-square analyses also revealed that the pattern of behavior did not vary as a function of whether the mother had been responsive during the individual interviews.

In parallel to the results of the individual interviews, sources of tension cited in the joint interviews tended to revolve around parameters of the relationship. As can be seen in Table 5, the majority of responses in the joint interviews focused on unso- licited advice. Although intrusion was a common source of ir- ritation discussed in the individual interviews, participants toned down their complaints in discussing a parallel source of tension in the other's presence. Topics of advice ranged from childrearing to business to health care.

Over half of the mother-daughter pairs included individuals outside of their dyad in their discussions of interpersonal tension.

Mothers and daughters tended to discuss issues pertaining to proper care of self or dependents more frequently during the joint interview than they had in the individual interviews. Twenty-one percent of the dyads (n = 10) discussed issues re- lated to appropriate care during the joint interview, 15% of the

Table 5 Number of Dyads Mentioning Each Category of Tension Coded From the Joint Interviews

Content code Number of dyads

Unsolicited advice Mother gives advice Daughter gives advice

Included others Inappropriate care

Mother's care Daughter's care

General disagreements Whose jurisdiction scenario falls under

Mother only Daughter only Both-unclear

19 12 28

7

3

11 21 16

dyads ( n = 7) discussed the mother's failure to care for self, and 6% (n = 3) discussed the daughter's care for self or others.

Associations Between Sources of Tension

in the Individual and Joint Interviews

and Qualitative Examples

Analyses were conducted to examine relationships between complaints in the joint interview and the analogous complaint from the individual interview. McNemar's tests were conducted to examine patterns of responses between the individual and joint interviews. The types of responses mothers and daughters gave in the individual and joint interviews did not appear to be related.

When a mother or daughter complained about intrusiveness during the individual interviews, the joint discussion was not more likely to revolve around unsolicited advice than when nei- ther had discussed intrusiveness alone. So few dyads discussed the daughters' care for self or other during the joint interview that the distributions of these dyads were not examined relative to source of complaint in the individual interviews. A dyad's discussing mothers' risk taking during the joint interviews was not related to daughters' individual complaints about mothers' self care.

In addition, analyses were conducted to examine whether re- sponse patterns in the individual and joint interviews might re- veal a lack of open communication within certain dyads. There appeared to be a dyadic pattern in the individual interviews sug- gesting less openness over problems. Mothers who could not think of a last problematic situation were more likely to have daughters who described their mothers' traits as problems. Such mothers and daughters might also be less likely to describe overt confrontation as a recent source of tension in the relationship in their joint discussions. McNemar's tests were conducted to examine the distribution of these two individual response types with discussion of a recent confrontation as a source of tension in the joint interview. The analyses were conducted coding no- example mothers as /, and mothers who provided an example as 2, compared with a dyad citing an open confrontation as their last source of difficulty coded as /, and not having an open con- frontation coded as 2. A similar analysis was conducted com-

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 601

paring daughters' description of their mothers' habits as a

source of tension to dyads' discussion of a confrontation. These

tests were not significant. As was mentioned previously, the overall tone of the joint in-

terview responses tended to be less negative than responses in

the individual interviews. Even when the same basic story came

up in the two interviews, the tone was less negative in the other

person's presence. For example, in the individual interview, 1

daughter described her mother's intrusiveness with regard to

her child rearing. She claimed that her mother had suggested

that she withhold ballet lessons from her own little girl as lever-

age to get the little girl to practice the violin.' The daughter then

proceeded to discuss her mother's tendency to be pushy in

matters concerning this daughter's own children and her resent-

ment that her mother felt free to reprimand the daughter's chil-

dren. In the joint interview, the same issue was presented, only

this time in terms of a disagreement over when children should

make their own decisions and the mother's offering advice that

did not seem to work well when applied to the daughter's

children. In another dyad, mother and daughter alike complained ve-

hemently of situations where they had felt hurt by the other

when answering the question about tension during the individ-

ual interviews. The mother lamented that her daughter pre-

ferred her cat over this mother. The mother explained that she

is allergic to the cat and the daughter would not lock the cat in

the basement when the mother visited. The daughter com-

plained about her mothers' intrusiveness and insensitivity to the

daughter's schedule. The daughter gave an example of a time

when the mother came and stayed for 2 hrs, expecting the

daughter to visit with her, although the daughter repeatedly

mentioned that she had an imminent deadline for a project at

work the next day. However, mother and daughter alike hesi-

tated to bring up these incidents when asked to discuss negative

feelings in each other's presence. During the joint interview,

each one proposed an incident concerning tastes in home dec-

orating that the other claimed not to remember. They finally

settled on a minor dispute over the daughters" house paint as

their last tense incident.

Relationship Quality

Mothers and daughters generally reported having high regard

for the relationship, according to the Bengtson Positive Affect

subscales. On a scale of 5-25, mothers rated their regard for the

relationship,!/=21.58(5/5 = 2.84), and daughters rated their

regard M = 20.52 (SD = 2.82). Although mothers expressed

more positive regard for the relationship than daughters, t(47)

= 2.63, p £ .05, this difference occurred in the context of very

high overall ratings for the relationship.

Relationship Quality and Interpersonal Tension Indices

A series of multiple regressions were calculated to examine

the impact of daughters' and of mothers' described sources of

tension on quality of relationship. Dummy variables were cre-

ated in which 1 = describing a given source of tension and 0 =

not describing that source of tension. Quality of relationship was

assessed from the mother's and the daughter's point of view.

Table 6

Regressions Predicting Relationship Quality From Mothers'

and Daughters' Descriptions of Sources of Tension

Description of tension 3 coefficient SE

Mothers' reported quality of relationship regressed on mothers' descriptions of tension"

Daughter as intrusive Daughter as excluding Daughters' care of self Habits-traits Constant

.26 -.42

.2!

.02 20.59

1.09 1.15 .23

1.01 2.49

-2.82 -0.24

1.29 0.12 5.89

m <.OI

ns ns

fi.Ol

Mothers' reported quality of relationship regressed on daughters' descriptions of tension1*

Mother as intrusive Mother as excluding Mothers' care of self Habits-traits Constant

-.44 -.09 -.03

.04 18.51

1.18 1.54 .47 .48

4.18

- 2.14 -0.50 -.23

-0.18 4.42

£.05 ns ns ns

<.OOI

Daughters' reported quality of relationship regressed on daughters' descriptions of tension0

Mother as intrusive Mother as excluding Mothers' care of self Habits-traits Constant

-.39 .05

-.15 .17

17.21

1.10 1.49 .47

1.16 3.91

-1.98 0.26

-1.02 0.90 4.39

<.05 ns ns ns

=i.ooi

*F(4,42) = 2.72,;) s .05, R2 = .24. rF(4,42) = 3.09, p ̂ .05, R- = .24.

4, 42) = 2.72,ps.Oi,R2 = .21.

Thus, four sets of regressions were calculated with mothers' and

with daughters' reports of relationship quality regressed on

mothers' and daughters' reported sources of tension separately.

Table 6 contains results from three of these regressions. The

fourth regression, involving daughters' reports of the quality

of the relationship regressed on mothers' described sources of

tension was not significant.

The sources of tension mothers described significantly pre-

dicted mothers' own reports of relationship quality, F(4,42) =

2.72, p < .05. The degree to which mothers reported feeling

excluded by their daughters was a significant predictor of rela-

tionship quality (/3 = —.42).

Daughters' descriptions of tension predicted their own and

their mothers' ratings of their relationship, F(4, 42) = 2.72, p

< .05, and F(4, 42) = 3.09, p < .05, respectively. Daughters'

reports of the mother as intrusive were a significant predictor of

decreased regard for the relationship in both equations (0 =

—.44 for daughters' ratings of the relationship and /3 = —.39 for

mothers' ratings of the relationship).

Additional regressions were estimated to examine the inde-

pendent contribution of mothers' and daughters' citing the oth-

ers' attributes as sources of tension on relationship quality.

Daughters' descriptions of the mothers' attributes as a source of

tension significantly predicted daughters' feeling better about

1 Because mothers and daughters who participated in the study might recognize their joint response, minor details have been changed to pro- tect mothers' and daughters' confidentiality.

602 FINGERMAN

the relationship (0 = . 41 ),/•'( 1,45)= 8.77,p<. 001. and moth- ers' feeling better about the relationship ((3 = . 3 3 ) , F ( 1 , 4 5 ) = 5.33, p < .05. Mothers' descriptions of the daughters' habits significantly predicted mothers' feeling better about the rela- tionship as well (0 = .29), F( 1,44) = 4.07, p <, .05. (Results are not shown here.)

In addition, quality of relationship was regressed separately on mothers' and daughters' citing of the mothers' aging, praising the other, and inclusion of others in their discussions. These re- gressions were not significant.

Finally, a point-biserial correlation was calculated to explore associations between failure to answer questions in the initial interview and regard for the relationship, with failure to re- spond coded as 1 and responding coded as 0. No-example mothers reported higher quality relationships; declining to pro- vide a tension response in the individual interview was associ- ated with mothers' higher regard for the relationship ( r — .33, p

Discussion

The women in this sample tended to be highly educated, healthy, and to lead independent lives. Tension in these mother- daughter relationships did not appear to be an artifact of many factors associated with parental aging, such as educational de- privation, financial difficulties, or stress induced from poor health in the later years of life (Troll & Stapley, 1985; Walker & Thompson, 1983). Rather, the sample in this study provided a forum for investigating negative aspects of later life parent- child relationships, controlling for some of the contextual vari- ables that have most readily been associated with stress in this relationship elsewhere. Although the select characteristics of the sample might have allowed for reduction of tension, mothers and daughters still described problems in their relationship ei- ther when interviewed alone or when interviewed with the other person. There were intergenerational differences in the types of tense situations and perspectives on those situations that moth- ers and daughters presented. In addition, the pattern of re- sponses was distinct in the individual and the joint interviews, with dyads giving a lower level of negativity in descriptions of problems in the relationship than had mothers and daughters when interviewed alone.

The Developmental Schism

The premise that developmental schisms underlie difficulties in parent-child relationships was partially supported by the findings in this study. Mothers' and daughters' reports of sources of tension often were related to predicted areas of de- velopmental differences.

Socioemotional Selectivity Theory

The first area of developmental difference was examined with regard to socioemotional selectivity theory (Carstensen, 1987, 1 99 1 ) or the salience of the relationship for mothers and daugh- ters. Sociological interpretations of parent-child relationships posit that tension between mothers and daughters may reflect general tendencies of mothers to have a greater investment in their relationship as a function of the parental role ( Bengtson &

Kuypers, 1971). Findings in the present study are in keeping with a premise that tension stems not only from the mothers' role, but from changes in the social meaning of relationships as mothers age. Mothers and daughters alike appeared to consider the other person as important in their lives; the majority of mothers and daughters rated the other person as amongst the top six most important people in their life. Yet, mothers ap- peared to give greater positive emotional value to their daugh- ters than daughters did to mothers. Mothers were more likely to name their daughters as their preferred confidant or the person with whom they most enjoy spending time. The greater emo- tional value mothers placed on their relationships with daugh- ters is in keeping with expected changes in the emotional sa- lience of important social relationships in later life (Carstensen, 1992).

Intergenerational differences in reports of tension may stem from these differences in emotional investment. Daughters were more likely to find their mothers intrusive, and these complaints appear to be related to their mothers' investment in their rela- tionship. Daughters' reports of feeling intruded upon were re- lated to their mothers' rating them as the person they prefer to confide in.

Another fundamental difference in the manner in which mothers and daughters view their relationship appears to bo re- lated to their conceptions of the parameters of their relation- ship. Daughters seem to view their own spouses and offspring as constituting a family unit unto itself, distinct from their rela- tionships with their mothers. Daughters tended to view the mother-daughter relationship dyadically, rarely bringing up is- sues involving their mothers in relationship to their own spouses, siblings, or children. Mothers more often included other people in their discussions of difficulties with their daugh- ters. Thus, older mothers' investment may lead them to go be- yond viewing their daughters as discrete individuals. Rather, mothers may perceive themselves as an integral part of a larger family, including their daughters, daughters' spouses, siblings, and children. As a result, mothers may feel free to offer advice or to direct affairs in the daughters' lives in a manner that daughters experience as intrusion.

Exclusion

Although there were no generational differences in reports of feeling excluded, for mothers these reports were more closely related to the differences in mothers' and daughters' ratings of the importance of the relationship. Mothers who reported feel- ing excluded from their daughters' lives also appeared to be less invested in their daughters' lives; mothers who did not confide in their daughters or rate them as someone they get along with best felt more excluded. Yet, 5 of the 7 mothers who felt ex- cluded claimed their daughter was amongst the three most im- portant people in their lives. Mothers also seem to report feeling excluded more often when their daughters value them less. Mothers were more likely to feel excluded when daughters rated their mothers as less important or as the person who annoys them most. It is also possible that when mothers feel excluded from their daughters' lives, they protect themselves by not con- fiding in these daughters and by finding other people, even an- other child, with whom they get along better. The findings in

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 603

this study may be spurious (only 7 mothers described situations

in which they felt excluded) or, alternately, these findings may reflect a mother-daughter typology. The mothers' sense of ex- clusion may stem from ongoing difficulties in the relationship rather than present developmental differences. Of particular note is the general value the mothers place on the relationship, though they do not report emotional functions that it serves.

In addition, dyadic processes were suggested by the findings of this study. Daughters' reports of feeling that their mother is intrusive related to the mother's feeling excluded from the daughter's life. The daughter's experience of tension may, in turn, have a negative impact on their treatment of their mothers. As she ages, a mother may value her relationship with her daughter more than the daughter does. As a result, the mother may engage in activities the daughter interprets as intrusive, and the mother may end up feeling excluded by her daughter.

Larger samples with a different selection criteria are needed to further understand the role of exclusion as a source of tension in mother-daughter relationships. The findings from this study reflect generally optimal relationships. In this study, all mothers rated their daughters as being amongst the 12 most important people in their lives, and mothers and daughters claimed that their relationships were generally positive. Studies involving sam- ples of mothers and daughters with more problematic relation- ships, such as those who are seeking therapy, may reveal a greater proportion of mothers who feel excluded from their daughters' lives and allow greater understanding of the patterns found here.

Although a premise of this article was that differences in the salience of the relationship are related to changes associated with socioemotional selectivity theory, the differences in invest- ment in the relationship might also be attributed to life context variables. For example, generational differences in investment might reflect differences in marital status; the mothers were more likely to be widowed, whereas daughters were more likely to be married or engaged. Yet, marital status does not fully ac- count for the mothers' preference for offspring; mothers who had husbands did not list their husbands as their preferred con- fidant. Cohort differences in men's roles as listeners may also partially account for generational differences in mothers' and daughters' preferences for one another as confidant. Daughters' preferences for their husbands as confidants may reflect societal changes in men and women's relationships. At the same time, the pattern of results in this study suggest that differences in investment, rather than life context, underlie tension mothers and daughters experience.

Care for Self-Other

The tension in these adult relationships may be distinctive from that experienced in younger parent and offspring relation- ships. Daughters appeared to be aware of their mothers' aging and also concerned about their mothers' well-being. As had been predicted, tension appeared to arise from discrepancies in moth- ers' and daughters' perceptions of the mothers' needs. Daughters cited tension related to concerns about their mother's self care, including a mother's refusal to get a flu shot or overexerting her- self unnecessarily. Daughters in this study were also more likely to mention their mothers' aging than were mothers.

Other aspects of the interview process also reveal the daugh-

ters' concerns for their mothers' well-being. For example, the discussion of tension in the joint interview rarely focused on the mother or something in her life; daughters seemed to seek to protect their mothers from criticism in such a context.

The premise of mothers and daughters experiencing a "role reversal" per se in later life has largely been discounted in the gerontological literature (Brody, 1990; Seltzer, 1990). Rather, daughters' concerns for their mothers' well-being in the present study may be related to their own present developmental issues. For example, there was evidence of filial anxiety in which daughters worry about their own abilities to care for their moth- ers should their mothers' health decline (Cicirelli, 1988). Daughters complained about their mothers' self-care, even in the self-defined healthy sample in the present study. As I have suggested elsewhere, daughters in mid-life may undergo a "role enhancement" in which they take on solicitude for their moth- ers' physical well-being, regardless of the mothers' actual health status (Fingerman, 1996).

At the same time, mothers seem to experience their daugh- ters' concerns and solicitousness for them as intrusive, suggest- ing that daughters' awareness of their mothers' aging may be experienced in a negative manner by their mothers, at least when mothers still define themselves as generally healthy. The results of the present study suggest that changes incumbent with the mother's aging may underlie some difficulties, but that these difficulties reflect discrepancies in perceptions of mothers' age- related needs rather than actual caregiving needs. Moreover, the experience of tension over the mothers' aging or health does not necessarily have a negative impact on the quality of the relationship.

Only three mothers expressed concern for the daughters' care of her own children, a smaller proportion than was originally anticipated. Although mothers actively included grandchildren and sons-in-law in the incidents they described as sources of difficulty with their daughters, they did not necessarily consider their daughters' treatment of these significant others as a source of difficulty.

Habits-Traits

Nearly a third of mothers and daughters described an unde- sirable attribute as the source of difficulty in the relationship. In this study, complaints about habits or traits were deemed to be less intrinsically related to stage of adult development. As was originally anticipated, reports of habits were associated with higher ratings of the relationship for daughters. Com- plaints over habits or traits may reflect the daughter's ability to view the mother as an individual, rather than as a member of the mother-daughter relationship only. Daughters who described their mothers' habits as a source of tension may consider aspects of the mother that they dislike, without regarding those aspects as a personal assault. Such daughters perceive the problem as "she is too talkative" rather than "she insists on talking to me when I have other things to do."

Dyadic processes pertaining to complaints or habits as sources of tension are also suggested by the data in this study. Daughters who described their mothers' traits as sources of difficulty were more likely to have mothers who provided no example of tension in their individual interviews. Mothers who

604 FINGERMAN

were unable to describe a tense situation during their individual interview reported higher relationship quality than did other mothers. Although mothers' failures to provide an example and higher ratings might both be an artifact of social desirability constraint, the pattern of results indicates a more complex pat- tern. Mothers and daughters who perceive problems with the other person as an aspect of that person rather than an aspect of the relationship may circumvent the discrepancies in their relationship stemming from differences in their emotional in- vestment or concerns about care. It is unclear whether this ap- proach to understanding problems in the relationship is unique to mother-daughter relationships in later life or whether it might be found at other points in time in the parent-child rela- tionship. Do parents who perceive difficulties with toddlers as an aspect of the toddler's temperament and emerging personal- ity feel more favorable about their relationship? Do teens who think their parents are "weird" rather than "mean to me" feel better about their relationships with these parents? Future re- search might examine this issue at other points in the lifespan.

These types of complaints also appeared to be related to a less direct communication style in this study. A party's com- plaints over the other's habits or traits as a source of difficulty appear to be less likely to be raised in overt confrontation in the relationship. The pattern of communication between mothers and daughters may be more complex than the portrait pre- sented here, however. The data presented here suggest that mothers and daughters who do not discuss experiences of ten- sion have higher quality relationships. Elsewhere, in findings from the same study of mothers and daughters, I reported that daughters who avoid telling their mothers when they are upset and who have mothers who do not perceive their efforts at avoid- ance feel worse about the way conflict situations turn out. How- ever, mothers who engaged in aggressive confrontational styles and whose daughters perceive their behaviors as such feel worse about the situation (Fingerman, 1995). Thus, experiencing sources of tension removed from the relationship or that appear to be uncommunicated are not necessarily indicative of positive outcomes. Rather, the congruence between each person's un- derstanding of the other's feelings may also play a role in how tense incidents are experienced.

Complaints about attributes might be more characteristic of friendships or professional relationships than in parent-child relationships. Future research might include examination of friendships, collegia! or work-related relationships, and mar- riages with large age and developmental task discrepancies to investigate whether patterns of tension found in the present study generalize across relationship types. At the same time, age differences alone do not account for developmental differences. Relationships with friends and colleagues tend to center on shared developmental position as parents of children of compa- rable ages, junior faculty members in the same department, equals at golf, or peers on the assembly line, regardless of age. The parent-child relationship is unique in that individuals maintain intimacy in the face of developmental discrepancies.

Praise

Although mothers and daughters reported generally strong positive feelings for the relationship, these feelings were stronger

for the mothers than for the daughters. Mothers praised daugh- ters while discussing their faults, or failed to discuss a fault at all. When mothers spoke up during the joint interview, they seemed to want to protect their daughters from having to do so. Mothers' focus on other people when discussing what bothered them about their daughters may also have been a means of shielding their daughters from being the sole target of complaint.

Limitations of the Study

There are several limitations to this study. The predomi- nantly White composition of the sample limits the generaliz- ability of this study. The central sources of tension reported amongst these women reflect dominant themes of their culture, including dependency versus autonomy, adult women's role in the family of origin, and perceptions of changes accompanying aging. Although the general pattern of developmental differ- ences might remain the same, descriptions of sources of tension related to intrusion-exclusion and regard for the relationship might be different in groups where familial connections are given different value. Future research might investigate devel- opmental differences between adults and manifestations of sources of intergenerational tension in other ethnic groups. Though the manifestation of content underlying intergenera- tional tension between women of different cultures may vary, the central theme may revolve around differences in adult development.

For example, future research in different cultural subgroups might reveal a greater proportion of mothers claiming to feel excluded as a function of differences in level of investment in the relationship as mothers age. In cultures where a greater value is placed on parental input throughout the life course, daughters' complaints about intrusive mothers might be less frequent. Rather, mothers' complaints of exclusion might be more frequent in cultures where they expect to play a greater part in their daughters' lives in adulthood. In such contexts, mothers' citing exclusion as a source of difficulty would be thought to reflect a manifestation of developmental differences in the salience of the relationship rather than ongoing rifts be- tween mother and daughter, as was suggested by the data for this sample.

Cohort differences, as well as developmental differences, may also account for some of the generational differences in this study. Women of the older generation may have been more reti- cent to discuss negative aspects of their lives with an interviewer. Yet, all analyses were conducted separately for mothers and daughters, as well as dyadically. There was variation in behavior within cohorts. Results for mothers and daughters suggest that issues pertaining to developmental differences have a more neg- ative impact than those issues related to individual attributes.

Although the women in this study may not be representative of the population of mothers over the age of 70 years and their daughters, this sample provided a forum for exploring intergen- erational tension in the absence of some of the contextual vari- ables that have been most readily associated with stress in this relationship in previous research. The focus of the study was on the contribution of developmental differences to intergenera- tional tension in later life. One might expect the pattern of re-

OLDER MOTHER AND DAUGHTER SOURCES OF TENSION 605

suits of the present study to be even more evident in a sample

with less healthy older mothers and a greater developmental gap

between generations.

This study poses questions about the nature of future devel-

opment in mother-daughter relationships. Do sources of ten-

sion shift again as the mother's health declines? How do mothers

and daughters evolve from the types of relationships described

in this study to those marked by caregiver stress described else-

where (Brody 1985, 1990; Litvin, 1992)? Do daughters who

complain about their mothers' habits while the mothers are

healthy assume the role of caregiver more readily than those

who are upset about their mothers' intrusiveness? Do they be-

come different types of caregivers? The answers to these ques-

tions may best be answered in the context of longitudinal

research.

References

Acitelli, L. K.., Douvan, E., & Veroff, J. (1993). Perceptions of conflict and similarity in the first year of marriage: How important are sim- ilarity and understanding? Journal of Social and Personal Relation- ships, 10, 5-19.

Ainsworth, M.S. (1979). Infant-mother attachment. American Psy- chologist, 34, 932-937.

Aldous, J. (1987). New views on the family life of the elderly and the near-elderly. Journal of Marriage and the Family, 49, 277-234.

Baruch, G., & Barnett, R. C. (1983). Adult daughters' relationships with their mothers. Journal of Marriage and the Family, 45, 601-

606.

Bedford, V. H.,&Blieszner, R.(in press). Persona] relationships in later life families. In S. Duck (Ed.), Handbook of personal relationships (2nd ed.). New York: Wiley.

Bengtson, V. L. (1973) Positive affect index. In D. J. Mangen & W. A. Peterson (Eds.), Research instruments in social gerontology: Vol. 1. Social roles and social participation (p. 154). Minneapolis: Univer- sity of Minnesota Press.

Bengtson, V. L., & Kuypers, J. A. ( 1 9 7 1 ) . Generational difference and

the developmental stake. Aging and Human Development, 2, 249- 260.

Bengtson, V. L., & Schrader, S. S. (1982). Parent-child relations. In D. J. Mangen & W. A. Peterson (Eds.), Social roles and social panic-

ipation(w- 115-129). Minneapolis: University of Minnesota Press. Bettleheim, B. (1965). The problem of generations. In E. Erikson

(Ed.), The challenge of youth (pp. 76-109). New York: Anchor. Blenkner, M. (1963). Social work and family relations in later life with

some thoughts on filial maturity. In E. Shanas & G. F. Streib (Eds.), Social structure and the family: Generational relations (pp. 46-59). Englewood Cliffs, NJ: Prentice-Hall.

Blieszner, R., & Adams, R. G. (1995, November). The causes and con- sequences of problems with friends in old age. In K. L. Fingerman (Chair), Negative aspects of social relations in later life. Symposium conducted at the meeting of the Gerontological Society of America, Los Angeles.

Brody, E. M. (1985). Parent care as normative family stress. Geronlol- ogisl,25, 19-29.

Brody, E. M. (1990). Role reversal: An inaccurate and destructive con- cept. Journal of Gerontological Social Work, 15, 15-23.

Carstensen, L. L. (1987). Age-related changes in social activity. In L. L. Carstensen & B. A. Edelstein (Eds.), Handbook of clinical gerontol- agy(pp. 222-237). New York: Pergamon Press.

Carstensen, L. L. (1991). Selectivity theory: Social activity in life-span context. In K. W. Schaie(Ed.), Annual review of'gerontology and ger- iatrics (pp. 195-217). New York: Springer.

Carstensen, L. L. (1992). Social and emotional patterns in adulthood:

Support for soicoemotional selectivity theory. Psychology and Aging, 7.331-338.

Cicirelli, V. G. (1983a). Adult children's attachment and helping be- havior to elderly parents: A path model. Journal of Marriage and the

Family. 45.815-825.

Cicirelli, V. G. (1983b). Personal strains and negative feelings in adult children's relationships with elderly parents. Academic Psychology Bulletin, 5, 31-36.

Cicirelli, V. G. (1988). A measure of filial anxiety regarding anticipated care of elderly parents. Gerontologist, 28,478-482.

Conover, W. J. (1980). Practical nonparametric statistics (2nd ed.). New York: Wiley.

Crohan, S. E. (1992). Marital happiness and spousal consensus on be- liefs about marital conflict: A longitudinal investigation. Journal of Social and Personal Relationships, 9, 89-102.

Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: Norton.

Fingerman, K. L. (1995). Aging mothers' and their adult daughters'

perceptions of conflict behaviors. Psychology and Aging. 10, 639- 649.

Fingerman, K. L. (1996, August). More than a daughter: Role enhance- ment in mid-life. In G. Labouvie-Vief (Chair), Daughters in mid-

life. Symposium conducted at the 104th Annual Convention of the American Psychological Association, Toronto, Ontario, Canada.

Fischer, L. R. (1986). Linked lives: Adult daughters and their mothers. Newbury Park, CA: Sage.

Gottman. J. M. (1979). Marital interaction: Experimental investiga- tions. New York: Academic Press.

Gottman. J. M., & Krokoff, L. J. (1989). Marital interaction and mar-

ital satisfaction: A longitudinal view. Journal of Consulting and Clin- ical Psychology, 57, 47-52.

Hagestad. G. O. (1984). Multi-generational families, socialization, sup-

port, and strain. In V. Garms-Homolova, E. M. Hoerning, & D. Schaeffer(Eds.), Intergenerational relationships^. 105-114). Got- tingen, Germany: Hogrefe.

Hagestad, G. O. (1987). Parent-child relations in later life: Trends and

gaps in past research. In J. B. Lancaster. J. Altmann, A. S. Rossi, &

L. R. Sherrod (Eds.), Parenting across the lifespan (pp. 405-433). New York: Aldine de Gruyter.

Havighurst, R. J . ( 1972). Developmental tasks and education (3rd ed.). New York: McKay.

Kastenbaum. R., & Aisenberg, R. (1972). The psychology of death.

New York: Springer.

Lehr, U. (1984). The role of women in the family generation context.

In V. Garms-Homolova, E. M. Hoerning, & D. Schaeffer (Eds.), In-

tergenerational relationships (pp. 125-132). Goltingen, Germany: Hogrefe.

Litvin, S. J. (1992). Status transitions and future outlook as determi- nants of conflict: The caregiver's and care receiver's perspective. Ger- ontologist. 32, 68-76.

Mancini, J. A. (1989). Family gerontology and the study of parent-

child relationships. In J. A. Mancini (Ed.), Aging parents and adult

children (pp. 3-12). Lexington, MA: Lexington Books.

Mancini, J. A., & Blieszner, R. (1989). Aging parents and adult chil-

dren: Research themes in intergenerational relations. Journal of Mar- riage and the Family, 51, 275-290.

Neugarten, B. L. (1968). The awareness of middle-age. In B. L. Neug- arten (Ed.), Middle age and aging (pp. 93-98). Chicago: University of Chicago Press.

Paikoff, R. L., & Brooks-Gunn, J. ( 1 9 9 1 ) . Do parent-child relation-

ships change during puberty? Psychological Bulletin, 110, 47-66.

606 FINGERMAN

Rook, K. S. (1984). The negative side of social interaction: Impact on

psychological well-being. Journal of Personality and Social Psychol-

ogy, 5, 1097-1108.

Rook, K. S. (1990). Parallels in the study of social support and social

strain. Journal of Social and Clinical Psychology, 9, 118-332.

Rossi, A. S. ( 1 9 9 3 ) . Intergenerational relations: Gender, norms, and be-

havior. In V. L. Bengtson & W. A. Achenbaum (Eds.). The changing

contract across generations (pp. 191-212). New York: Aldine de

Gruyer.

Rossi, A. S., & Rossi, P. H. (1990). Of human bonding: Parent-child

relations across the life course. New York: Aldine de Gruyter.

Ryff, C. D., Lee, Y. H., Essex, M. }., &Schmutte, P. S. ( 1 9 9 4 ) . My chil-

dren and me: Midlife evaluations of grown children and of self. Psy-

chology and Aging. 9, 195-205.

Seltzer, M. M. (1990). Role reversal: You don't go home again. Journal

of Geronlological Social Work, 15, 5-14.

Semple. S. J. (1992). Conflict in Alzheimer's caregiving families: Di-

mensions and consequences. The Geromologist, 32. 648-655.

Steinman, L. A. ( 1 9 7 9 ) . Reactivated conflicts with aging parents. In

P. K. Ragan (Ed.), Aging parents (pp. 126-143). Los Angeles: Uni-

versity of Southern California Press.

Streib, G. E, & Beck, R. W. (1980). Older families: A decade review.

Journal of Marriage and the Family, 42, 937-956.

Stull, D. E., Kosloski, K., & Kercher, K. (1994). Caregiver burden and

generic well-being: Opposite sides of the same coin? Gerontologist,

34, 88-94.

Talbott, M. M. (1990). The negative side of the relationship between

older widows and their adult children: The mothers1 perspective. Ger-

onlologixl. 30, 595-603.

Thompson, L., & Walker, A . J . ( 1 9 8 2 ) . The dyad as u n i t of analysis:

Conceptual and methodological issues. Journal of Marriage and the

Family, 44, 899-900.

Thompson, L., & Walker, A. J. ( 1 9 8 4 ) . Mothers and daughters: Aid

patterns and attachment. Journal of Marriage and the Family, 46,

313-322.

Troll, L. E. (1987). Mother-daughter relationships through the life-

span. In S. Oskamp (Ed.), Family processes and problems: Social

psychological aspects (pp. 284-306). Newbury Park, CA: Sage.

Troll, L. E. (1988). New thoughts on old families. Geronlologist, 28,

586-591.

Troll, L. E., & Smith, J. (1976). Attachment through the lifespan. Hu-

man Development, 4, 67-74.

Troll, L. E., & Stapley, J. (1985). Elders and the extended family sys-

tem: Health, family salience, and affect. In J. A. Munnichs (Ed.),

Lifespan and change in a gerontologicalperspective, (pp. 211-238).

Orlando: Academic Press.

Walker, A. J., & Thompson, L. (1983). Intimacy and aid and contact

among mothers and daughters. Journal of Marriage and the Family.

-#5,841-848.

Weishaus, S. S. ( 1 9 7 8 ) . Determinants of affect of middle-aged women

tOH-ards their aging mothers. Unpublished doctoral dissertation. Uni-

versity of Southern California.

Whitbeck, L., Hoyt, D. R., & Huck, S. M. (1994). Early family rela-

tionships, intergenerational solidarity, and support provided to par-

ents by their adult children. Journal of Gerontology: Social Science,

49, S85-S94.

Wodak, R., & Schulz, M. (1986). The language of love and guilt:

Mother-daughter relationships from a cross-cultural perspective.

Amsterdam: John Benjamins.

Zarit, S., Rccvcr, K.., & Bach-Peterson, J. (1980). Relatives of the im-

paired elderly: Correlates of feelings of burden. Gerontologist, 20,

649-655.

Received February 27, 1995

Revision received February 12, 1996

Accepted February 12, 1996 •