Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Mental models and attachment systems: Predictions of health and relational outcomes of the elderly
(USC Thesis Other)
Mental models and attachment systems: Predictions of health and relational outcomes of the elderly
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
MENTAL MODELS AND ATTACHMENT SYSTEMS: PREDICTIONS OF
HEALTH AND RELATIONAL OUTCOMES OF THE ELDERLY
by
Risa E. Dickson
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Communication Arts and Sciences)
December 1991
Copyright December 1991 Risa E. Dickson
UMI Number: DP22467
All rights reserved
IN FO R M A TIO N TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMT
■^^O I ssartM i en Publ i shi ng” " ^
UM I DP22467
Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90089-4015
This dissertation, written by
Risa E. Dickson,
under the direction of h..?.v....... Dissertation
Committee, and approved by all its members,
has been presented to and accepted by The
Graduate School, in partial fulfillm ent of re
quirements fo r the degree of
Ph.b.
CM
’ 9 1
D5S4
D O C T O R O F P H IL O S O P H Y
Dean of Graduate Studies
Date Novem1aer 1991
DISSERTATION COMMITTEE
Chairperson
ACKNOWLEDGMENTS
There are many people who helped me complete this
work and to them I owe great thanks. My greatest
appreciation is for all those who volunteered their
participation in this study. Of course, this includes
the administrators of the Weingart Center, Long Beach
Senior Center, Beverly Hills Parks and Recreation,
Joclyn Adult Center, and the USC Volunteer Network.
The completion of this project would have been
much more difficult without the help and guidance of
more than a few very special people. To these people I
am also appreciative. First, to my mother, whose
support made it possible for me to achieve this goal,
and my stepfather John, whose gentle words of
experience often made my own experiences more
tolerable* My siblings, Elana, Andrew, and Mariah, all
provided emotional support that helped me through the
graduate school experience.
Michael Cody, my dissertation chair, advisor, and
friend provided support, guidance, and patience. I
appreciate the time and guidance of my dissertation
advisors, Carl Broderick and Lynn C. Miller.
Without the help Of my friends, this experience
would have been far less rewarding. Friends such as
Dorothy Gould, whose generosity and love literally
clothed me through graduate school, and who continues
i
to provide me with a family apart from my own. I am
deeply indebted to all my dear friends who tolerated my
inability to be much of a friend at times, but who,
regardless, were there when I needed then.
My partner John provided me with unconditional
love and acceptance. I know there were times he
wondered if I would ever finish, and if we would ever
have a moment together without "my dissertation." The
I
j security he fostered in the relationship helped me
5 complete this project.
j Finally there is Jolane, with whom I lived through
i
i
j some of my most memorable experiences, who stood by me
j through achievements and failures, and whose memory
will be with me always.
i i i
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS................................
LIST OF TABLES.......................................
ABSTRACT..
CHAPTER 1 Attachment, Social Support, and Health..
Introduction...........................
Attachment Theory................*.... .. .5
Individual Attachment Styles...........
Secure............................ ..15
Avoidant..........................
Ambivalent........................ ..21
Attachment Styles and the Elderly......
Hypotheses on Attachment...............
Social Support.........................
Social Support and Attachment..........
Objective vs. Subjective Social Support. . . 3 4
Objective Support.................
Subjective Support................
Social Support and the Elderly.......... . . 3 8
Hypotheses on Social Support...........
Health...............................
Self-Rated Health......................
Attachment, Social Support, and
Self-Rated Health.................
Hypotheses on Self-Rated Health........
CHAPTER 2 Methods.................. ............
General Procedure....................
Respondents...........................
Procedure..............................
Independent Variable...................
Dependent Variables................... . .58
Analyses..........................
Hypotheses........................
Research Questions..............
Page
CHAPTER 3
CHAPTER 4
REFERENCES
APPENDIX A
Results .... ............. 74
Sample Characteristics................... 74
Demographic ...... 74
Social Network........ 74
Adult Children................ .75
Hypothesis One......... .77
Hypothesis Two...... 78
Hypothesis Three ...................... 78
Hypothesis Four.. . .......*............ 79
Hypothesis Five................. ...... . . 80
Research Questions.................. .81
Discussion and Future Directions.......... 85
Overview. ...... .85
Discussion.............. 86
Attachment. ..................... .86
Perceived Social Support............. 90
Physical Well-Being ............. 91
Gender...................... 94
Limitations........ .97
Future Directions............. 100
Conclusion.............................. 103
....................................... .106
Questionnaire........................... 116
v
LIST OF TABLES
TABLE 1
TABLE 2
TABLE 3
TABLE 4
Page
Factor Solution for Attachment
to Child Scales............................ 70
Hazan and Shaver's (1986) Parental
Caregiving Paragraph Descriptions.. ....... 71
Parental Caregiving Scale
Factor Solution.......... 72
Child Caregiving Scale
Factor Solution............. 73
ABSTRACT
Based on Bowlby's (1982) theory of attachment,
this dissertation explores the nature of the attachment
bond in a population of older adults. The attachment
bond formed by an elderly parent to their adult child
was compared with the elders' perceptions of their own
parents caregiving style, perceptions of their child as
a caregiver, and the style with which they had attached
to the adult child. Additionally, attachment to the
child was examined as a predictor of the perceptions
one holds of social support received from the child,
overall physical well-being, and self-ratings of
Respondents included adults (N = 111) over the age
of sixty who were the parent of at least one living
child. Measures were completed for a) attachment to
their adult child; b) perceived caregiving style of
their own parents; c) perceived caregiving style of
their child; d) perceived social support received from
their child; and e) overall physical well-being. It
was hypothesized that attachment to one's child would
be related to each of these measures, and that there
would be similarities in how one's parents and one's
health
v i i
child were viewed as caregivers.
Results indicate that perceptions of one's own
parents caregiving style was not significantly related
to those of the child's caregiving style. Attachment
to one's child, however, was predictive of both
perceived social support, and physical health outcomes.
Attachment style was predictive of ratings of parents
as caregivers only for those who rated high on measures
of anxious attachment to their child. Those who were
securely attached to their child perceived greater
social support from the child and rated the child's
caregiving style more positively than those who were
less securely attached. Respondents with highly secure
attachment styles rated their health more positively,
reported fewer serious illnesses or conditions, and
reported using fewer non-necessary medications.
It was concluded that the attachment system
remains an important organizing factor throughout the
life span, and that attachment to one's adult child is
predictive of both physical health and relational
outcomes.
CHAPTER ONE
Attachment. Social Support: and Health
Introduction
Americans live longer today than at any time in
history - twice as long as they did during the
nineteenth century (Butler, 1981). Increased
longevity, however, entails both positive and negative
consequences. For one, the process of aging is often
viewed as an undesirable life event that impacts
negatively on the communication among family members.
The elderly can be neglected or avoided, and the
quality of communication with family members may suffer
as they age (Dunkel-Schetter & Wortman, 1981).
Additionally, old age is often filled with stressful
life events, such as retirement, failing health,
illness, decreased mobility, and increased loneliness
due to the deaths of friends and family members. Kin
relationships, particularly those with one's children,
become increasingly important, and potentially
problematic.
Most elderly will become dependent upon help from
others (Cicirelli, 1981, 1983a), and will need others
1
for advice on health, legal and financial matters, for
companionship, and for assistance in transportation
(Hanson & Saur, 1983). As an elder becomes
increasingly dependent upon his or her child, the
preestablished parent-child role may be renegotiated,
with the child assuming more of a caregiving role. Not
surprisingly, "role reversal" is a frequently stated
stress associated with aging (Hanson & Saur, 1983).
Implicit in the reformulation of the attachment
pair bond are issues of a person's original attachment
to his or her parent figure, current attachment of the
elder to a child, and perceptions of the child's
ability and willingness to provide various types of
social support to the elderly parent. Further, there
is reason to believe that both issues of attachment and
perceived social support are related to healthy
psychological (e.g., Bowlby, 1980, 1982; Henderson,
Byrne, Duncan-Jones, Adcock, Scott, & Steele, 1978;
Krause, Liang, & Yatomi, 1989) and physical functioning
(e.g., Arling, 1987; Bowlby, 1977; Miller & Ingham,
1976; Ward, Sherman, & LaGory, 1984).
The attachment system developed in infancy between
the parent and child is active throughout a person's
life span (Bowlby, 1982; Marris, 1982; Henderson, 1977;
1982; Kobak & SCeery, 1988). Bowlby (1982) has posited
2
that as one matures, his or her attachment system
shifts focus: The child who at one time formed an
attachment to the parent must now switch roles.
Unfortunately, this speculation has not been confirmed
experimentally. While some attachment research focuses
on young adults, little of this line of research
extends to a population of the elderly.
The limited literature on the relationships
between attachment styles and adult functioning (e.g.,
Cicirelli, 1981, 1983a, 1983b; Collins & Read, 1990;
Cooke & Miller, 1990; Feeney & Noller, 1990; Hazan &
Shaver, 1988) has indicated strong explanatory and
predictive links between various measures of attachment
style and variables such as marital satisfaction and
conversational nonverbal support (Cooke & Miller,
1990), and helping behaviors provided to an aging
parent (Cicirelli, 1981, 1983b). Although little
research has dealt with this area, Cicirelli (1981,
1983b) argues that attachment theory provides a useful
theoretical orientation within which to explore the
changing dynamics of the parent-child relationship in
old age.
A relatively strong link has been forged between
attachment theory and social support. Henderson (1974,
1977; Henderson et al., 1978; Henderson, Duncan-Jones,
3
Byrne, & Scott, 1981) and the Sarasons' (B.R. Sarason,
Pierce, & Sarason, 1990; I.G. Sarason, Levine, Basham,
& Sarason, 1983; I.G. Sarason, Sarason, & Pierce, 1990)
have found a strong relationship between the strength
of the attachment bond and perceptions of available
social support. Thus, attachment theory also provides
a useful organizing framework within which to study
perceptions of available social support that may also
generalize to older populations.
Researchers have often studied the link between
social support and health (e.g., Jung, 1984; Krause et
al., 1989; Henderson et al., 1981), and this area is
presently receiving increased interest in the field of
Gerontology. A review of the literature suggests that
the measure of social support most predictive of
psychological health outcomes is perceptions of the
availability of social support (Jung, 1986; Krause et
al., 1989; Wethington & Kessler, 1986). Considering
the links between attachment theory and social support,
and social support and health, there is a probable link
between one's attachment style and perceptions of one's
physical health. This relationship is particularly
interesting when considered in light of the changing
nature of the parent-child relationship over time.
This dissertation seeks to examine the effect of
4
attachment to an adult child on perceptions of the
caregiving style of one's parents, perceptions of the
caregiving style of the child, perceptions of social
support received from the child and perceptions of
one's overall physical health.
The following section is written with two goals in
mind: To introduce the reader to the literatures of
Attachment Theory, Perceived Social Support, and (among
the elderly) Self-Rated Health, and to integrate these
three research areas.
Attachment Theory
The theory of attachment is founded on the idea of
a strong and enduring attachment bond formed initially
between an infant and the person with whom it has the
greatest amount of social interaction, or the primary
caregiver, during the first year of life. The bond
established in infancy affects the pattern of
attachment behavior throughout the life span of an
individual (Bowlby, 1977, 1982). And, attachment
theory has proven a useful framework within which to
study relationships involving enduring affectional
bonds. This theory has been used both by those
interested in child development (e.g., Kobak & Sceery,
1988? Main & Weston, 1981, 1982; Sroufe & Waters,
5
1977), and, more recently, by those scholars who are
interested in adult relationships (e.g., Collins &
Read, 1990; Feeney & Nailer., 1990; Weiss, 1982).
There is one important difference in the research
of attachment in childhood versus attachment in
adulthood. A majority of the research on childhood
attachment is longitudinal and based on a developmental
paradigm. On the other hand, research to date on the
role of attachment in adult populations tends to focus
on specific characteristics or personality traits which
can be attributed to an individual style of attachment.
The following section will provide an overview of
attachment theory, a description of the three prominent
attachment styles and a summary of research findings in
the attachment literature. This will be followed with
discussions of attachment in both young and late
adulthood, and a delineation of hypotheses pertinent to
this section.
The theory of attachment developed out of the
ground breaking work of Bowlby (1977, 1980, 1982) who
approached the study of human development from an
evolutionary-ethological perspective. Bowlby (1982)
suggested that certain infant attachment behaviors,
such as crying or clinging, function to elicit security
and protection from predators and other forces of
6
nature. Bowlby's basic theory has been extended by the
longitudinal and developmental work of Ainsworth and
her colleagues (Ainsworth, Blehar, Waters, & Wall,
1978) as a means of studying the development of the
infant-caregiver bond.
Ainsworth et al. (1978) found that attachment to a
primary caregiver occurs with the development of the
attachment system during the first year of life, and is
collaboratively generated between the expressed needs
of an infant and the sensitivity and responsiveness of
the caregiver in satisfying those needs. Longitudinal
research has found attachment styles in childhood
remarkably enduring and consistent across both time and
context (Ainsworth et al., 1978; Arend, Gove, & Sroufe,
1979; Main, Tomasini, & Tolan, 1979; Main & Weston,
1981; Waters, 1978). And, it has been suggested that
this attachment style, developed in infancy, will guide
further attachment behavior well into old age (Bowlby,
1982; Troll & Smith, 1976). There are three central
concepts around which attachment theory is organized:
Attachment, attachment behavior, and attachment
behavior systems (Ainsworth et al., 1978; Bowlby, 1982;
Hinde, 1982).
The first concept, attachment, refers to the
enduring affectional bond formed between a person and
7
an attachment figure (Ainsworth et al. 1978). In
childhood this occurs between an infant and a primary
caregiver. In adulthood, attachment bonds are found in
emotionally close relationships, such as family or love
relationships, in which the relationship is considered
to be of central importance (Bowlby, 1977; Weiss,
1982).
The second concept, attachment behavior, refers to
those behaviors which function to form or maintain an
attachment bond. The goal of attachment behavior is
gaining or maintaining proximity with an attachment
figure (Ainsworth et al., 1978; Bowlby, 1977, 1982;
Main & Weston, 1982). In childhood, behaviors such as
crying or clinging fulfill this function, and are
readily observable in stressful situations such as when
a child is tired, frightened or ill (Bowlby, 1982).
While attachment behaviors are most clearly evident in
the child-caregiver pair, these behaviors are also
exhibited in adult relationships (Henderson, 1982;
Henderson et al . . , 1981). In adulthood, attachment
behaviors become evident under stressful conditions
such as anxiety or illness (Bowlby, 1977, 1982). The
display of specific attachment behaviors is highly
dependent upon situational and contextual variables.
As a result, consistency in cross-situational behavior
8
can be expected to be particularly low, however,
identifiable individual attachment patternsf or styles,
have been found to exhibit remarkable stability across
both time and context in adulthood as well as in
infancy (Ainsworth, et al. 1978; Ainsworth, 1982; Arend
et al. , 1979; Troll & Smith, 1976; Waters, 1978).
The third organizing concept of attachment theory
is the attachment behavioral system. Through continued
interaction and experience with a caregiver, an infant
learns that certain behaviors are likely to elicit
specific desired responses (e.g., a hungry infant
cries). During development, attachment systems are
organized around fairly simple needs such as food,
comfort, and safety (Bowlby, 1982). By the end of the
first year discrete attachment behaviors have begun to
cohere into a stable goal-corrected system in reference
to the infant/caregiver interactions (Ainsworth et al.,
1978; Bowlby, 1982). These early principles of
organization form working models of the self in
relation to others and begin the foundational framework
within which new figures are assimilated, life is
experienced and acted on, and a sense of meaning is
created (Bowlby, 1982; Marris, 1982).
With the development of the system, attachment
behaviors are increasingly mediated by both goal-
9
f --------------------------------------------------------------------------
!
j corrected behavioral systems which incorporate
i
i representational models of the self and the
J environment, and anticipated interactions between the
two (Ainsworth et al. 1978? Bowlby, 1977, 1982). The
attachment system is sustained strongly throughout the
\
first three years of life, at which time (in the course
of healthy development), a child comes to tolerate
temporary separations from the attachment figure and
begins to focus attention on competing systems, such as
! autonomy and exploration (Bowlby, 1977, 1982). With
I
j time, the attachment system becomes one of many other
: goal-directed systems in the cognitive framework of an
I
1
! individual. The attachment system is of unique
I
importance, however, because as the first of the
I
systems to develop, it lays the organizational
foundation for those that follow,
j Throughout adolescence, as new life experiences
[ are incorporated within the attachment framework,
' previously developed perceptions and meanings are
| continually reinforced or modified. After adolescence
J the attachment system persists relatively unchanged
I
throughout the life span (Bowlby, 1977). The mental
models of interaction developed up to this point in
life provide enduring impressions and predictions of
normative, appropriate relational behaviors, and
10
expectations of how one may expect to be treated by
others. Persistence of these mental models is high,
and affect the evaluation of all future relationships
(Shaver & Hazan, 1988; Bowlby, 1977).
There are three important distinctions between the
attachment process in childhood and in adulthood
(Weiss, 1982). First, adults have more control over
choosing their attachment figures, and have the option
of breaking an unsatisfactory attachment bond. Second,
the attachment system is not as primary in adult
relationships as it is in infancy. For instance, in
the event of a threat to an attachment bond, an adult
(unlike a child) is often able to function reasonably
well within other behavioral systems, such as caring
for important others. Third, adult attachment is often
directed toward a love or sexual interest (not a
caregiver).
In childhood the attachment bond functions to keep
a child in close proximity to protection. For adults,
attachment bonds function primarily to ensure a social
support network: To ensure that there are several close
potentially helpful adults who will be available in
time of need (Weiss, 1982). B.R. Sarason et al.,
(1987) have defined the function of the adult
attachment bond as fostering feelings that one is
11
worthwhile, capable, and valued as a member of a
particular group.
Individual Attachment Style
As the attachment system emerges into a coherent
pattern, or style, healthy development is dependent
upon the extent to which a child is able to perceive a
secure base in the attachment figure. According to
current theory, an attachment style is a direct
reflection of the level of security felt by an infant
within the caregiver/infant relationship. Felt levels
of security are based primarily on perceptions of the
availability and responsiveness of the caregiver in
relation to the needs of the infant. As a result, it
is the pattern of interactions between an infant and
caregiver that determines the resulting individual
attachment style.
A secure attachment style is important for healthy
development in that it provides the infant with a
secure base for exploration and autonomy. The
development of a secure attachment style depends upon a
caregiver who is consistent in their responsiveness to
the needs and signals of the infant, provides the child
with ample physical and interactional attention, and
who is able to physically comfort the child when it is
12
distressed (Ainsworth et al. 1978; Bowlby, 1982).
By one year of age a majority of infants have
adopted one of three distinct patterns of attachment
behavior, the origins of which can be traced to
behavioral interactions in the infant/caregiver
relationship (Ainsworth et al., 1978). There are clear
indications that the development of a particular
attachment style is highly correlated with the pattern
of caregiving received in infancy (Ainsworth et al.,
1978; Bowlby, 1982). For instance, caregivers who
display avoidant attachment express less affect and
more aversion to physical contact with the infant, and
are significantly more angry or otherwise threatening
in their caregiving behavior compared to those with
other styles (Main, Tomasini, & Tolan, 1979; Main &
Weston, 1982).
On the other hand, caregivers of securely attached
infants are significantly more sensitive and accepting
of the child (Main et al., 1979). The way in which
caregiving is perceived by a child is also an important
determinant in the development of an adult who has a
capacity to create and maintain affectional bonds
(Bowlby, 1977). Consistent with attachment theory,
memories of one's relationship with parents are related
to levels of felt security in adulthood (Collins &
13
Read, 1990; Kazan & Shaver, 1987).
The three styles of attachment can be understood
as various conceptualizations of rules, or "models,"
that guide individual responses in a given situation
(Bowlby, 1980; Kobak & Sceery, 1988). The individual
attachment styles are related to different perceptions
of, and responses to, various situations, indicating a
difference in the way experiences and meanings are
organized. Feeney and Noller (1990) found, in a sample
of adults, that statements in reference to a general
view of the self was the most powerful discriminator
among the three attachment styles. For instance, the
different attachment styles were predictive of views
about benefits and liabilities involved with
interpersonal relationships, and of beliefs in the
trustworthiness of others.
Attachment styles are categorized into one secure
style and two insecure, or anxious, styles of
attachment. These styles have been labeled Secure,
Anxious/Ambivalent, and Anxious/Avoidant (Collins &
Read, 1990; Hazan & Shaver, 1987; Levy & Davis, 1988;
Shaver & Hazan, 1988; Waters, 1978). A review of the
literature indicates that between fifty-five and sixty
percent of those studied can be classified as having a
secure attachment style, followed by the avoidant style
14
which represents between twenty-three and thirty
percent of the samples, and ambivalent which accounts
for fifteen to twenty percent of those studied
(Ainsworth et al., 1978; Arend et al., 1979; Bowlby,
1982; Collins & Read, 1990; Feeney & Noller, 1990;
Hazan & Shaver, 1987; Mikulincer, Florian, & Tolmacz,
1990). In the following paragraphs each of the three
styles will be discussed in detail, beginning first
with the development of the attachment system, followed
by discussions of perceptions of parents, the caregiver
role, love styles, and personality variables associated
with each of the attachment styles.
Secure attachment. Ainsworth et al. (1978)
observed that infants who exhibit a secure attachment
style experience primary caregivers who are
consistently available and responsive to their needs.
As a result, a securely attached child has been able to
develop a reliably predictable mental model of his or
her environment, and a representational model of
themselves as both self-sufficient and as worthy of
assistance in time of need (Bowlby, 1982). This infant
is active in play, uses its' caregiver as a secure base
from which to explore, and will seek out the caregiver
when frightened or distressed (Ainsworth et al., 1978).
The securely attached infant is readily comforted by
15
its' caregiver figure and easily returns to exploration
or play after having been comforted. Research on
preschoolers has linked secure attachment with high
levels of ego-resiliency and curiosity, effectiveness
in interpersonal problem-solving, and social competence
in preschool (Arend et al., 1979).
Adults with secure attachment styles consistently
report perceptions of their parents as warm and
supportive, the relationship between their parents as
positive, and have memories of their parents as readily
available during distressing times (Feeney & Noller,
1990; Hazan & Shaver, 1987; Kobak & Sceery, 1988;
Mikulincer et al., 1990; Shaver & Hazan, 1988). Kobak
and Sceery (1988) found that securely attached
individuals reported childhood experiences in a
coherent pattern, had good recall of attachment
experiences, and were not likely to idealize either
parent figure.
Having received appropriate amounts of care and
comfort in childhood, a securely attached adult is
likely to be comfortable with closeness and sharing,
and with both receiving and giving care and comfort
(Shaver & Hazan, 1988). In describing romantic
relationships, those with a secure style are likely to
recount important love relationships in terms of
16
friendship, trust, and happiness, and to emphasize
acceptance and support of their relational partner
(Feeney & Noller, 1990; Hazan & Shaver, 1987? Shaver &
Hazan, 1988). Several researchers have found that a
securely attached adult is less likely than one who is
anxiously attached to engage in game-playing or
pragmatic styles of loving (Collins & Read, 1990;
Feeney & Noller, 1990; Levy & Davis, 1988; Shaver &
Hazan, 1988). A game-playing style of love is one in
which at least one partner is in the relationship for
fun, not for emotional support Or commitment, while a
pragmatic style of love entails finding a mate who
'fits' necessary qualifications and then establishing a
relatively satisfying relationship. The commonality
between these two styles is that neither entails
"falling in love" in the romantic sense. Those with a
secure style are also more likely than others to
believe that romantic feelings wax and wane over time
(Hazan & Shaver, 1987).
The securely attached also appear to be more
satisfied in their marital relationships. When
compared with anxiously attached individuals, securely
attached individuals report higher levels of marital
satisfaction and more reciprocity of intimate
disclosure (Cooke & Miller, 1990), enjoy longer
17
enduring relationships (Hazan & Shaver, 1987), report
more positive relational characteristics, and engage in
more constructive approaches to conflict (Levy & Davis,
1988). A person with a secure attachment style is also
likely to be married to a partner who also has a secure
attachment style (Cooke & Miller, 1990).
When compared with the anxiously attached, a
secure attachment style is predictive of higher ratings
of self-esteem and ego-resilienee (Kobak & Sceery,
1988), higher levels of trust, a greater willingness to
stand up for one's beliefs, greater levels of felt
control (Collins & Read, 1990; Feeney & Noller, 1990),
and a greater ability to cope constructively with
negative emotions by acknowledging distress and seeking
support (Kobak & Sceery, 1988; Mikulincer et al.,
1990). Additionally, the securely attached score lower
on measures of self-conscious anxiety and unfulfilled
hopes (Feeney & Noller, 1990), and on measures of fear
of personal death (Mikulincer et al., 1990). Secure
individuals are likely to describe themselves as easy
to get to know, as liked by most people, and believe
that others are generally well intentioned and good-
hearted (Hazan & Shaver, 1987).
Avoidant attachment. The avoidant style develops
out of caregiver-infant interactions in which the
18
primary caregiver is consistently unresponsive, or even
rejecting, to the expressed needs of the infant
(Ainsworth et al., 1978). An avoidantly attached
infant has formed an image of the environment as
generally unresponsive, and perhaps even threatening in
time of need. An avoidant child perceives a lack of
support, and may develop a style that is autonomous and
compulsively self-reliant. This self-reliance may
serve in part to dismiss the importance of attachment
(Mikulincer et al., 1990). An avoidant infant is not
easily comforted after a distressing separation from
the caregiver, and, upon reunion, may respond in a
friendlier manner to a stranger than to the caregiver.
Individuals with avoidant attachment styles
remember very few distressing events from childhood;
however, they report considerable rejection and of lack
of love from their parents (Kobak & Sceery, 1988).
They are more likely than others to recall their
parents in a less favorable manner (Mikulincer et al.,
1990), and are more likely to describe their mother as
cold and rejecting (Hazan & Shaver, 1987). Avoidant
individuals may also attempt to avoid negative affect
or minimize distressing memories by idealizing past
relational interactions (Kobak & Sceery, 1988; Shaver &
Hazan, 1988).
19
Because the avoidant individual did not receive
adequate comfort and care as an infant, he or she may
be uncomfortable attending to the needs of others, or
being cared for by others. As a result, the caregiving
style of the avoidant individual is likely to be one of
either unwillingness or inability to provide or receive
comfort or care. An avoidant adult may also encounter
difficulties in sustaining intimate relationships
because of a relatively high need to maintain autonomy.
Those with an avoidant style are far less likely
than either of the other styles to idealize present
love partners, and are more likely to view love as a
form of friendship (Feeney & Noller, 1990). They are
more likely than securely attached adults to engage in
a game-playing love style (see Levy & Davis, 1988),
perhaps in an effort to avoid emotional intimacy
(Shaver & Hazan, 1988). Those who embody an avoidant
style tend to view romantic love as hard to find and as
rarely lasting, and to describe important love
relationships in terms of fear of intimacy, jealousy,
emotional extremes, and a lack of acceptance (Hazan &
Shaver, 1987; Shaver & Hazan, 1988).
Avoidant individuals have a greater tendency to
mistrust others, and to feel distance from others
(Feeney & Noller, 1990). They are also more likely to
20
be rated by their peers as hostile (Kobak & Sceery,
1988), and to have a rather strong subconscious fear of
death (Mikulincer et. al., 1990). Avoidant individuals
score high on loneliness scales; however, Collins and
Read (1990) found that while they distance themselves
from others, avoidant individuals did not report
feeling lonely.
Ambivalent attachment. The ambivalent infant
characteristically has had a primary caregiver who was
inconsistent and sporadic in providing care. The
caregiver may have been sometimes unavailable and at
other times intrusive in caring for the child. An
ambivalently attached infant has been unable to form
predictable expectations with regard to the
responsiveness or consistency of either the caregiver
or the environment in meeting it's needs. An
ambivalently attached infant can be observed
incorporating mixed behaviors when distressed, and is
likely to be somewhat resistant in seeking proximity to
and contact with the caregiver figure. For instance,
this infant is likely to cling to the caregiver in an
attempt to receive comfort; however, at the same it may
resist being comforted.
Ambivalent individuals are likely to remember
their parents as unpredictable, unfair, unsupportive,
21
and intrusive (Shaver & Hazan, 1988). The ambivalent
individual is more likely than the others to describe
their parents as loving but role-reversing, in that the
child may be saddled with providing emotional support
or security for the caregiver figure. Ambivalent
individuals have no difficulty recalling distressing
events from their childhood, but the memories do not
appear consistent and are likely to be recalled in a
confused or incoherent manner (Kobak & Sceery, 1988).
When providing care to others, one with an
ambivalent style may exhibit a caregiving style that is
compulsive and self-sacrificing. However, this same
individual is likely to be dissatisfied with care
offered to or received by them (Shaver & Hazan, 1988).
In terms of romantic relationships, the ambivalent
style is predictive of a passionate, or manic, style of
loving (Levy & Davis, 1988): A love that happens
quickly and easily (Shaver & Hazan, 1988). The
ambivalently attached are not likely to perceive love
as friendship (Feeney & Noller, 1990) and are more
likely to describe past love relationships in terms of
obsession and jealousy, extreme emotional highs and
lows, intense sexual attraction, and an unwillingness
to commit on the part of their partner (Hazan & Shaver,
1987? Shaver & Hazan, 1988). Those with ambivalent
22
styles have been found to express dependence on their
partner, a desire for deeply committed love
relationships (Feeney & Noller, 1990), and a
preoccupation with attachment relationships (Kobak &
Sceery, 1988).
Ambivalently attached individuals tend to be less
satisfied in their marriages (Cooke & Miller, 1990),
and encounter the highest rates of divorce among the
three styles (Hazan & Shaver, 1987; Shaver & Hazan,
1988). Those with an ambivalent style are also more
likely to be married to an avoidant mate (Cooke &
Miller, 1990). This union, of an ambivalent with an
avoidant mate, works to ensure that the ambivalent
partner can continue in a self-sacrificing caregiving
style while at the same time maintaining a level of
frustration with the amount and type of care provided
by the avoidant partner. On the other hand, the
avoidant mate is provided with a partner who can be
perceived as smothering and needy.
Ambivalent individuals have more self-doubts and
feelings of being misunderstood (Hazan & Shaver, 1987),
score higher on measures of anxiety (Collins & Read,
1990), and are likely to be rated by their peers as
anxious (Kobak & Sceery, 1988). Ambivalent individuals
show greater hypervigilence upon separation from an
23
attachment figure and experience greater distress when
an affectional bond is broken (Mikulincer et al.,
1990). Of the three attachment styles, the ambivalent
individual is the most consciously afraid of personal
death (Mikulincer et al., 1990). Anxious attachment is
also predictive of low self-confidence (Shaver & Hazan,
1988), low levels of confidence in others, and a
greater likelihood to conform to social pressures
(Collins & Read, 1990).
In summary, significant differences have been
found among the three attachment styles beginning with
caregiver/infant interactions, and extending to the way
in which one remembers their primary caregivers, the
way they approach and respond to caregiving and love
relationships, and a number of adult personality
traits. Generally speaking, those who are anxiously
attached were exposed to a caregiving style
administered in a manner that magnified levels of felt
anxiety during childhood (Bowlby, 1977). Those with a
secure style remember caregivers who were supportive
and caring, while those with anxious styles remember
caregivers who were either cold and rejecting, or
inconsistent in providing care. With regard to adult
relationships, anxious attachment is predictive of
lower marital satisfaction, negative relationship
24
characteristics and less constructive orientations
towards dealing with relational conflict. Finally,
those with an anxious style of attachment have
personality characteristics which indicate higher
levels of anxiety, and perhaps less social competence
in maintaining positive relationships.
Attachment Styles and the Elderly
Little is known about the function and impact of
attachment in adulthood, and the need is great for
additional research (Shaver & Hazan, 1988). One
essential avenue of research deals with the role of
communication in forming and maintaining the attachment
bond (Kobak & Sceery, 1988). Attachment is a cognitive
phenomenon manifest primarily through communication. A
child communicates attachment by seeking proximity and
contact with the primary caregiver, while adult
attachment behaviors are somewhat more complex. An
adult may communicate closeness not only through
proximity, but through many other forms which can serve
to maintain the bond over long distances (Bowlby, 1980;
Cicirelli, 1983b).
It is widely accepted that the attachment system
remains activated at some level throughout the life
span. And, much of the literature in aging suggests an
25
enduring continuity of basic personality variables,
such as attachment, throughout the life span
(Antonucci, 1976? Troll & Smith, 1976). While there is
no research to date on how the attachment system
functions as a person ages, theorists have speculated
about several probable outcomes.
Bowlby (1982) suggests that "when attachment
behavior can no longer be directed towards members of
an older generation, or even the same generation, it
may come instead to be directed towards members of a
younger one" (p. 207). On the other hand, Weiss (1982)
speculates "it may be that the attachment behavior
system, so powerful in the early years of our lives,
retains much of its power through our young adult
years; then, as we enter the later years...begins
gradually to release its hold on our motivation and
behavior" (p. 181). Only one of these divergent
predictions can be true.
The role of attachment in an aging population is
not yet clear. The literature in Gerontology (e.g.,
Cicirelli, 1981, 1983a, 1983b? Hanson and Saur, 1983?
A.S. Rossi & Rossi, 1990), indicates that many elderly
will at some point become dependent upon one or more of
their children for some type of support. An empirical
link has been established in younger populations,
26
between individual attachment style and the perceptions
of one's parents as caregivers. However, this
relationship has not been tested in older populations.
If attachment is correctly classified as an enduring
cognitive model, it can be expected that one's
attachment style will affect how others who assume the
caregiving role (such as an adult child) are viewed and
experienced.
Hypotheses on Attachment
The following hypotheses are concerned with
testing the relationship of attachment to one's child
with perceptions of one's own parents, and perceptions
of one's child as a caregiver.
Research suggests that those Who have a secure
attachment style remember their parents as responsive
(Feeney and Noller, 1990). At the same time, those
with an ambivalent style of attachment remember their
parents as inconsistent, and those with an avoidant
style are reported to remember their parents as
unresponsive or rejecting. Based on these findings,
this study hypothesizes that memories of the
responsiveness of one's parents will be related to the
type of attachment they have to their child.
27
Hla: Respondents who are more securely
attached to their child will rate the
caregiving style of their parents more
positively than those who are less
securely attached to their child.
Hlb: Respondents who are more ambivalently
attached to their child will rate the
caregiving style of their parents as
more inconsistent than those who are
less ambivalently attached.
Hlc: Respondents who are more avoidantly
attached to their child will rate the
caregiving style of their parents as
more rejecting than those who are less
avoidantly attached to their child.
This dissertation extends the work on the
, , caregiverH role to include an adult child who can
serve as a caregiver to the aging parent. The idea of
attachment as enduring cognitive models (Bowlby, 1977)
suggests expected role-appropriate behaviors in
specific situations. If a "working model" of a
caregiver is developed in childhood based on
28
infant/caregiver interactions, one can expect some
similarity in the way a recent, or prospective,
caregiver (the adult child) and a past caregiver (the
parents) are perceived. Perceptions of one's parents
as caregivers may endure across both time and context
to include the care provided to them by their children.
Additionally, those who have negative perceptions of
their parents may be less secure in their ability to
relate to their own child as a potential caregiver.
H2: There will be a significant positive
correlation between the way the parents
are viewed as caregivers and the way the
child is viewed as a caregiver.
Based on the assumption that an attachment style
initially develops out of perceptions of receptiveness
and responsiveness of a caregiver, there should be a
strong relationship between attachment and perceptions
of that person's caregiving style. Previous research
has found adult attachment style related to perceptions
of one's parents as caregivers. It could also be
expected that attachment to any figure would be
accompanied by perceptions and expectations of a level
of receptiveness and responsiveness, namely a
caregiving style. This study predicts that attachment
to one's child will be accompanied by certain
perceptions of that child's receptiveness and
responsiveness to the needs of the parent.
H3: Those who are more securely attached to
their child will perceive the caregiving
style of their child as more responsive
than those who are less securely
attached to their child.
An additional variable of interest in this study
is that of gender. While previous attachment research
has not found gender differences in attachment style to
a romantic figure (e.g., Collins and Read, 1990), this
may be one area in which parent-child attachment
differs from romantic attachment. Researchers in
Gerontology (e.g., A.S. Rossi & Rossi, 1990) have found
substantive gender differences across parent-child
relationships in late life. It is this finding in the
Gerontology literature that warrants the addition of a
research question.
RQ1: Is there a gender difference in
attachment style to one's adult child?
30
Social Support
As one ages, emotional and instrumental support
from one's children become increasingly important.
Research indicates that for a majority of families, the
parent-child relationship among aging parents and their
children is both highly central and important (Hanson &
Saur, 1985; Troll & Smith, 1976). As health problems
and need for assistance increase, one's child is often
the preferred provider of necessary help (Cicirelli,
1983a; Hanson & Saur, 1985). The purpose of the
following section is to highlight the relationship
between social support and attachment, discuss current
issues in the conceptualization of social support, and
to develop the relationship between social support and
health.
Social Support and Attachment
Social support has been defined by I.G. Sarason et
al. (1983) as "the existence or availability of people
on whom we can rely, people who let us know that they
care about, value, and love us" (p. 127). The
Sarason's and their colleagues (B.R. Sarason et al.,
1990; B.R. Sarason, Shearin, Pierce, & Sarason, 1987;
I.G. Sarason et al., 1990) have argued that social
support in adulthood operates in much the same way as
31
does attachment in childhood. For an adult, social
support, or the knowledge of the availability of
supportive other, provides a secure base from which to
engage in exploration and risk taking behaviors. This
connection between social support and attachment has
not gone unnoticed by others in the literature (e.g.,
Antonucci, 1976, 1985; Barrera, 1986; Cutrona, 1986;
Henderson, 1977, 1982; Henderson et al., 1978, 1980,
1981; Weiss, 1982).
Kahn and Antonucci (1980; Antonucci, 1985a) have
articulated a convoy model of social support. This
model, based largely on the work of Bowlby, argues for
the importance of a developmental approach to the study
of social support. A premise of this model is that
individuals assess the adequacy of their social support
networks, which is then translated into outcomes (i.e.,
mental and physical health).
If social support in adulthood is conceptualized
as equivalent to attachment in childhood, an individual
who perceives high levels of social support is likely
to also hold the belief that there are specific people
who will be available and accepting under most
conditions (I.G. Sarason et al., 1990). This also
suggests that one's perceptions of social support may
be, like attachment, an enduring cognitive model of
32
oneself in relation to others. In this line of
reasoning, researchers have recently begun to explore
the idea that one's sense of support may be rooted in
the childhood development of the attachment system
(Antonucci, 1990; Dunkel-Schetter & Bennett, 1990; B.R.
Sarason et al., 1990; I.G. Sarason et al., 1986).
Whether, or when, an individual decides to seek
support in a difficult situation has been found related
to perceptions of the support they believe is available
(Barrera, 1986). This suggests that the manner in
which experiences are organized within the cognitive
framework of an individual may indicate and explain how
distressing situations are dealt with (Mikulincer et
al., 1990). There are implications for this line of
reasoning in terms of the three attachment styles. For
instance, when distressed, an individual with a secure
style of attachment can be expected to seek out others.
This individual is likely to perceive several important
others as available, accepting, and helpful under most
conditions. At the same time, an ambivalent individual
is likely to turn attention on to the cause of
distress, as well as to an attachment figure in an
hypervigilant manner. This, in turn, may intensify
distress, and serve to inhibit the development of
autonomy and self-confidence in the individual (Kobak &
33
Sceery, 1988). On the other hand, an avoidant
individual may believe that there is no one to whom
they can turn without facing some measure of rejection.
As a result, in an effort to avoid dependence on
others, an avoidantly attached individual is not likely
to request help from others.
Social support, like attachment, develops out of
an interactional process. Individuals differ in their
needs and desires for support and, as a result, differ
in the extent to which they are willing to seek out and
accept support (Dunkel-Schetter & Bennett, 1990). In
this sense, social support consists of two elements:
Perceptions that there are a sufficient number of
available others on whom one can depend, and a level of
satisfaction with the perceived support that is
available (I.G. Sarason et al., 1983). Whether or not
supportive behaviors are recognized as helpful and
supportive is based largely on perceptions of the
receiver of the support, and their view of how loved
and accepted they are by their attachment figures.
Objective vs. Subjective Social Support
There is a present concern in the social support
literature over issues of conceptualization and
measurement (Antonucci, 1976; Barrera, 1986? Krause,
34
1989; Markides & Cooper, 1988; Procidano & Heller,
1983; I.G. Sarason & Sarason, 1985). A diversity of
approaches in the study of social support has often led
to inconsistent, or an apparent lack of, findings.
Because of this, there is a general move to abandon
traditional notions of a globally defined concept of
social support and to focus on more narrowly defined
and precise definitions (Barrera, 1986; Dunkel-Schetter
& Bennett, 1990). A common distinction in the
literature is that of objective and subjective support
(Barrera, 1986; Rook, 1984; B.R. Sarason et al., 1987).
Conceptualizations of objective support are based on
constructs such as the actual number of others in a
social network, or the number of supportive behaviors
provided by others. Subjective conceptions of support
are based on individual perceptions of the amount and
usefulness of available support. These two
conceptualizations have proven to be largely orthogonal
to one another, lending credence to current
reconceptualizations (Barrera, 1986; Dunkel-Schetter &
Bennett, 1990; B.R. Sarason et al., 1987).
Objective social support. Objective support is
that which can be , , objectively, , measured, such as
social embeddedness (i.e. network size and density), or
enacted' support (i.e. provision of information or
35
emotional support) (Krause, 1989; Procidano & Heller,
1983). Social embeddedness, or network size, refers to
the number of supportive connections an individual
maintains with others and is assessed through
information such as marital status, reported numbers of
close friends and relatives, and frequency of
participation in social organizations (Antonucci,
1985a; Barerra, 1986; Krause et al., 1989; Krause,
1989). The size of one's support network has been
found to remain fairly stable throughout the life span
(Schultz & Rau, 1985). While this aspect of social
support has not shown a direct relationship with stress
or health issues (Barerra, 1986), it is likely that
social embeddedness mediates issues of well-being
through perceptions of an adequate number of others who
are available for support.
Enacted support is a measurement of the actual
number of supportive behaviors provided to an
individual during a specified period of time (Barerra,
1986; Krause et al., 1989; Krause, 1989). This
measurement is difficult to obtain and is implicitly
biased by individual perceptions of received support.
Whether a behavior is interpreted as supportive is
mediated by a receivers perceptions of the
supportiveness of the behavior. Like social
36
embeddedness, enacted support is neither directly nor
consistently related to health or well-being outcomes.
Subjective support. Subjective support, also
referred to as perceived support, is defined as the
extent to which one believes that those in his or her
social network are available and able to provide
support. Perceived support is the cognitive
conceptualization of social support. This measure
reflects the level of confidence that adequate support
is available should a need arise. This measure has
been found to remain fairly stable over time (I.G.
Sarason et al. , 1986) and to be relatively independent
of life events (Krause, 1989). This suggests there is
a relationship between the amount of support one
perceives to be available and enduring cognitive models
of the self in relation to others.
While it is not clear what underlies perceptions
of support, there is an apparent link with the memories
one has of their parents. For instance, those who
perceive high levels of social support describe their
parents as interested in them as children, and report
receiving more parental care and protection than those
who perceive lower levels of social support (Dunkel-
Schetter & Bennett, 1990; I.G. Sarason et al., 1990;
I.G. Sarason et al., 1986). As might be expected,
37
measures of perceived social support are also
correlated with measures of attachment (Henderson,
Duncan-Jones, Byrne, and Scott, 1980; Henderson et al.,
1981; B.R. Sarason et al., 1987), individual
personality variables, and the way in which a person is
experienced by others.
High levels of perceived social support have been
linked with heightened interpersonal skills, a sense of
self-efficacy, a positive self-image, low levels of
anxiety, and positive expectations of interactions with
others (B.R. Sarason et al., 1990; I.G. Sarason, 1985).
Those who are high in perceived social support report
more optimism about life (I.G. Sarason et al., 1990),
are rated by others as more attractive and socially
skilled (B.R. Sarason, et al., 1987), and as more
competent leaders and problem solvers (I.G. Sarason et
al., 1986). Those who perceived low levels of social
support tend to have an external locus of control,
encounter difficulty in persisting on demanding tasks,
experience increased cognitive interference and report
relative dissatisfaction with life (I.G. Sarason et
al., 1983).
Social Support and the Elderly
Previous research on social support among the
38
elderly has focused primarily on objective measures of
support provided, rather than on subjective assessments
of available support (Krause et al., 1989). Among the
elderly, perceived support is correlated with affective
states such as anxiety and depression (Dunkel-Schetter
& Bennett, 1990), and older adults who are satisfied
with perceived support are less likely to suffer
psychological distress and depression (Krause et al.,
1989). Marital status has been found to be an
important determinant of perceived support in older
populations (Cutrona, 1986), and a strong predictor of
perceived support among the elderly is frequency of
contact with kin, which has been found to be largely a
function of marital status.
Hypothesis on Social Support
In an outline of how the adult attachment system
functions with other adult behavioral systems, Shaver
and Hazan (1988) postulated that secure individuals are
more comfortable both giving and receiving care and
comfort, and as a result tend to perceive high levels
of available support. Ambivalent individuals are
depicted as rather self-sacrificing in providing
support (relative to the secure and avoidant groups)
but tend to be dissatisfied with care that is provided
39
to them, and as a result can be expected to perceive
low levels of available support. Avoidant individuals
are uncomfortable both giving and receiving care and
are likely to perceive low levels of available social
support. The following hypothesis seeks to test these
assumptions in an elderly population.
H4: Respondents who are more securely
attached to their child will perceive
higher levels of social support
from their child than will those who are
more anxiously attached to their child.
A.S. Rossi and Rossi (1990) found gender to be an
important variable in the exchange of help, or social
support, across the parent-child relationship. As
such, gender may be an important variable in the
relationship between attachment and the perceptions a
parent has of the social support provided to them by
their adult child. The fourth research question
addresses this issue.
RQ2: Is gender a significant variable factor
in the relationship between attachment
and social support?
40
Health
Health is of primary concern for the elderly.
When an elder is confronted with a health problem there
is a tendency to reach out for necessary support,
which, in turn, functions to limit (or buffer) the
total effect of the problem (Krause, 1990). Social
support is increasingly linked with physical and
psychological well-being outcomes in older populations
(Markides & Cooper, 1988). For instance, the absence
of social support is found related to psychological
symptoms such as depression (e.g., Dunkel-Schetter &
Bennett, 1990; Henderson et al., 1978, 1981; Krause et
al., 1989; Miller & Ingham, 1976), and low levels of
perceived support are related to various aspects of
both physical and psychological health (e.g., Bowlby,
1977, 1982; Henderson, 1977; Henderson et al., 1978;
I.G. Sarason et al., 1990).
Self-Rated Health
Perceptions of one's own health are known to play
an important role in the life of the elderly (LaRue,
Bank, Jarvik, & Hetland, 1979), and are recognized as a
distinct aspect of health important to various aspects
of aging (Tissue, 1972). One's rating of their health
can be thought of as a summary of various subjective
41
and objective aspects of health as they are represented
in the perceptual framework of an individual (Tissue,
1972). A review of the literature indicates that the
perceptions one holds of the status of their health are
important indicators of other health related aspects of
aging. For instance, self-ratings of health have
emerged as an important of predictor mortality among
the elderly (LaRue et al., 1979; Mossey & Shapiro,
1982; Seigler & Costa, 1985). Longitudinal research
has found that those elderly who perceive their health
as poor are almost three times more likely to die
sooner than those who believe their health is excellent
(Mossey & Shapiro, 1982; Seigler & Costa, 1985).
Self-ratings of health are a widely used measure
of health in Gerontology (Markides & Martin, 1979).
Among the most commonly used measures of self-rated
health is a single question in which a respondent rates
their health on a Likert-type scale from poor to
excellent. Respondents may also be asked to identify
symptoms or illnesses they presently believe they have
(see Linn & Linn, 1984).
There has been some debate over the accuracy and
validity of self-ratings of health as they compare to
more objectively defined ratings of physical health.
However, Rosencranz and Pihlblad (1970) have argued
42
that many symptoms reported by respondents are based on
previous physician diagnoses. For instance, it is
unlikely that someone would report having an illness
such as high blood pressure without having been
previously diagnosed as such. Additionally, there is
substantial evidence to indicate a significant
relationship between self-ratings and more objective
ratings of health (Ferraro, 1980; LaRue et al., 1979;
Linn and Linn, 1984; Maddox & Douglass, 1973; Mossey &
Shapiro, 1982; Salthouse, Kausler, & Saults, 1990).
Objective health measurements used in previous studies
include: physicians ratings (LaRue, et al., 1979),
symptom or illness checklists (e.g., Liang, 1986; Linn
& Linn, 1984; Pilpel, Carmel, & Galinsky, 1988), and
scales of physical functioning (e.g., Krause, 1990;
Maddox & Douglass, 1973; Ward et al., 1984).
Self-ratings of health appear to remain stable
over time (Maddox & Douglass, 1973; Mossey & Shapiro,
1982; Tissue, 1972) and as a result, provide both a
valid index of health (Tissue, 1972) and an economical
means of gaining health information in an elderly
population (Ferraro, 1980). The most common
discrepancy between self and objective ratings of
health is a tendency for the elderly to assess their
health more positively than may objectively be the case
43
(Cockerham, Sharp, & Wilcox, 1983; LaRue et al., 1979;
Maddox & Douglass, 1973; Seigler & Costa, 1985; Tessler
& Mechanic, 1978). And, old-old respondents (80+) are
repeatedly found to perceive their health more
positively than younger respondents (Cockerham et al.,
1983; Pilpel et al., 1988), despite the fact that those
who are considered old-old report a greater number of
chronic physical and mental health related problems,
and more disability than those in younger age groups
(Ferraro, 1980; Perlmutter S Nyquist, 1990; Seigler &
Costa, 1985).
Self-Rated Health. Social Support, and Attachment
Consistent with attachment theory, there is a
strong negative relationship between perceptions of
available social support and both psychological and
physical health related outcomes (e.g., Barrera, 1986;
Bowlby, 1977; Cutrona, 1986; Dunkel-Schetter and
Bennett, 1990; Henderson, 1977; Henderson et al., 1978,
1980; Rosencranz, & Pihlblad, 1970; B.R. Sarason et
al., 1990; I.G. Sarason et al., 1983; Miller & Ingham,
1976; Ward et al., 1984). In terms of attachment
theory both Bowlby (1977) and Henderson (1974, 1977:
Henderson et al., 1978; Henderson et al., 1981) have
noted a relationship between anxious adult attachment
44
and physical and psychological symptomology. These
researchers have argued that those with an anxious
attachment often have unexpressed feelings of anger and
anxiety which surface in a way that sacrifices their
physical and/or emotional health. This self-
sacrificing behavior, in forms such as hypochondria or
anorexia nervosa, is an attempt to coax others into
fulfilling deficiencies in perceived levels of social
support. Hence, it is likely that those with an
anxious attachment style are likely to perceive
themselves as poorer in health and to report more
symptomology than are those who are more securely
attached.
Social support affects health outcomes through
both main and buffer effects (Dunkel-Schetter &
Bennett, 1990). The main effects of social support on
health have been most widely studied in terms of
psychological health and well-being. Henderson et al.
(1978, 1981) found a strong link between perceived
deficiencies in social bonds and both psychological
health and physical morbidity. Buffering effects of
social support are most consistently found linked with
perceptions of available social support (Wethington &
Kessler, 1986). Buffering effects of social support
suggest that perceptions of high levels of support are
45
accompanied by lower levels of symptomatology. These
findings have lead some to argue that adequate levels
of perceived social support are a prerequisite for
psychological well-being (Blazer, 1980; Henderson et
al., 1978; Sarason et al., 1983; Dunkel-Schetter,
Folkman, and Lazarus, 1987).
Tessler and Mechanic (1978) have argued that most
people respond to their overall well-being to the
extent that a condition interferes with their ability
to engage in life activities. As a result, individual
perceptions of health are likely to significantly
impact social interaction and participation in social
roles. This effect is probably reciprocal in nature in
that those who are socially isolated and those who feel
neglected rate their health lower than others (Arling,
1987; Tessler & Mechanic, 1978). Cockerham et al.
(1983) suggest that being alive and having the ability
to function adequately promotes a subjective impression
of good health among the elderly.
While there is a positive relationship between
social ties and health (Pilpel et al., 1988), large
social networks and frequent contact with family and
friends are not strong predictors of self-rated health
in elderly populations (Liang et al., 1980; Ward et
al., 1984). Self-rated health is influenced by the
46
frequency of contact with children and with friends in
that those who rate themselves as either healthy or as
constantly ill, report seeing family members more often
than those who rate their health as closer to average
(Pilpel et al., 1988).
It is difficult to predict the direction of the
relationship between well-being and perceived health
status (Krause, 1990). While many elders have some
degree of health impairment, level of impairment does
not seem to affect negatively how they perceived their
health (LaRue et al., 1979). For the elderly,
perceptions of the status of their health are strongly
correlated with levels of life satisfaction (Cockerham
et al., 1983; Mossey & Shapiro, 1982; Pilpel et al,
1988), perceptions of overall quality of life (Arling,
1987), and psychological well-being (Seigler & Costa,
1985). The reporting of physical symptoms may reflect
more than the physical state of one's health, and is
likely dependent on the psychological state of the
respondent (LaRue et al., 1979). The belief that one
is healthy or ill may be an important predictor of
psychological state and behavior (Maddox and Douglass,
1972). However, some researchers have speculated that
those who report the most health problems have
personality traits or psychological states, such as
47
depression or anxiety, that make such behavior
predictable (i.e., Seigler & Costa, 1985; Tessler &
Mechanic, 1978). This line of reasoning is consistent
with the arguments made of the impact of attachment
style on overall well-being (e.g., Bowlby, 1977;
Henderson, 1974).
Hypotheses on Self-Rated Health
These following hypotheses test the relationship
between attachment style and health. The first
hypothesis is an attempt to draw out a relationship
between self-ratings of health and one's attachment
style. The links have been made in previous research
between attachment and perceived social support, and
between perceived social support and self-rated health.
It is with these relationships in consideration that
the first hypothesis is made.
H5a: Respondents with a more secure style of
attachment will perceive themselves as
healthier than those with more anxious
attachment styles
Both Bowlby (1977) and Henderson (1974) have
suggested that attachment style is related to health
48
outcomes. The arguments suggest that those with a more
anxious attachment style will not only perceive
themselves to be poorer in health than their more
securely attached counterparts, but will also report
more physical symptoms. The hypothesis is followed by
a research question concerned with determining
differences in the nature of the symptoms reported by
those who are securely versus those who are anxiously
attached to their child.
H5b: Respondents with more anxious styles of
attachment will report more symptoms
than will those who are more securely
attached to their child.
RQ3: Is there a difference in the nature of
the symptoms reported by those who are
securely attached versus those who are
avoidantly attached to their child?
If it is expected that those who are more securely
attached will report fewer symptoms, it might also be
expected that those who are more anxiously attached
will report the use of more medications. Hypothesis 5c
tests this assumption. A second research question was
49
added to determine differences in the nature of the
reported medications used by those of different
attachment styles.
H5c: Respondents who are more anxiously
attached to their child will report
the use of more medications that will
those who are more secure in their
attachment to their child.
RQ4: Is there a difference in the nature of
the medication used by those who are
securely attached versus those who are
anxiously attached to their child?
Gender has proven to be a significant variable in
the study of self-rated health. It has been found that
women report more physical and mental illnesses and
utilize hospital services at higher rates than do men
(Ferraro, 1980; Seigler & Costa, 1985; Tessler &
Mechanic, 1978). In keeping with these findings, men
report fewer illnesses (Ferraro, 1980). However, women
who rate their health as excellent generally report a
greater number of illnesses than men who rate their
health as excellent. While this dissertation is
50
interested primarily in the effects of attachment style
on self-rating or health, there may also be a gender
effect mediating the findings. In light of previous
findings on the effects of gender on self-ratings of
health, a fifth research questions is posed.
RQ5: What is the effect of gender on self-
ratings of health?
51
CHAPTER TWO
Methods
General Procedure
Volunteers for this study were recruited from
senior activity centers in the Los Angeles, California
area. Senior centers are community sponsored centers
which provide various social programs for area senior
residents. The services offered by a center varies
considerably, dependent upon the center, but tend to
focus on providing social services and opportunities
for social interaction. Community centers were chosen
as a way to recruit volunteers because of the large
number of seniors who utilize the services.
Volunteers were recruited in each of the centers
either through an announcement in the dining hall
during the lunch hour or over a building loudspeaker.
Respondents were compensated for their participation
according to the agreement made between the researcher
and the senior center administration. For instance,
some respondents were individually paid for their
participation, while for others a donation was made in
their name to an organizational activity fund.
52
Respondents
One hundred and eleven people volunteered to
participate in this study. Of this sample 40 were male
(36%) and 71 were female (64%). The age distribution
spanned 37 years and ranged between 52 and 89, the
median age was 74.
Original criteria for inclusion in the study was
that the respondent be a minimum of sixty years of age
and the parent of at least one living child. Previous
research on the elderly has tended to focus on those
aged 60 years or older (e.g., LaRue, et al., 1979; Linn
& Linn, 1984; Maddox & Douglass, 1973; Markides &
Martin, 1979; Arling, 1987; Ward, et al., 1984). For
most of the sample this was true. However, in five
instances the volunteers were in their fifties. These
younger respondents were people who regularly visited a
senior center and for whom the center was an integral
part of their social lives. Additionally, it has
become common practice in Gerontology to split age
groups according to young-old (50-65), middle-old (65-
80) and old-old (80+)(Kahn & Antonucci, 1981). With
this age breakdown in consideration, the five responses
„ were left in the data set for inclusion in the
analysis.
53
Procedure
Each volunteer was told that the purpose of the
study was to obtain information about changes in the
parent-child relationship as both the parent and child
grew older. Respondents were asked to complete a
questionnaire packet containing six measurement
instruments. Confidentiality was assured and each
volunteer completed a Consent to Participate form. In
the event a respondent wished to participate but could
not complete the questionnaire by themselves (i.e.,
poor eyesight or difficultly reading), the
questionnaire was read to them in an interview format
by the researcher. Those interviewed were informed
that they were free to decline an answer to any
question, or to end the interview at any time. The few
interviews that were conducted strictly followed the
format of the questionnaire.
Independent Variable
The independent variable in this study is
attachment to one's child. In the childhood
development literature attachment to a caregiver is
measured by observing a child's experiences during a
distressing separation from his or her primary
caregiver coupled with the way in which the caregiver
54
is greeted by the child at reunion (Ainsworth et al.,
1978; Arend, et al., 1979; Main et al., 1979; Main &
Weston, 1981). Longitudinal research has proven this
method reliable and consistent across both context and
time in the determination of attachment style in
children. In adult populations attachment style is
often determined through the use of the Hazan and
Shaver (1987) self-report measure (e.g., Collins &
Read, 1990; Feeney & Noller, 1990; Hazan & Shaver,
1987; Mikulincer, et al., 1990). Other methods used to
determine attachment style in adult populations include
questions about geographic proximity and frequency of
interaction with an elderly parent (Cicirelli, 1983a),
questions addressing the nature of attachment
relationships (Kobak & Sceery, 1988) and questions
about close affectional ties (Henderson et al., 1981).
The most commonly used attachment instrument to
date is a trichotomous measure designed by Hazan and
Shaver (1987). This instrument includes three separate
paragraph descriptions representing each of the three
attachment styles. In an extension of the Hazan and
Shaver attachment measure, Collins and Read (1990)
deconstructed the various paragraphs and developed an
18-item attachment scale rated by respondents along a
5-point Likert-type scale. The Collins and Read
55
attachment measurement has proven to have adequate
reliability and internal consistency for determining
attachment style in adult populations.
Both the Hazan and Shaver (1987) and the Collins
and Read (1990) attachment scales measure adult
romantic attachment. Because this research is
interested in parent's attachment to a child, not to a
romantic partner, the Collins and Read (1990) scale was
revised to apply to attachment to a child (see Table
1). For instance, an item originally written ”1 often
worry that my partner does not really love me" was
reworded to read "I often worry that my child does not
really love me."
The revised 18-item scale was Factor analyzed
using SPSS. An initial SPSS Principal Components
Analysis (PCA) solution resulted in five factors with
eigenvalues equal to or greater than l. Only two of
the factors were meaningful. A Skree test indicated
that the amount of variance accounted for in the
overall solution diminished greatly after the first two
factors. Additionally, theoretical and conceptual
clarity for the factor solution was weak, and a number
of items loaded on more than one factor. After
consideration of the eigenvalues, results of the skree
test, and the theoretical underpinnings of the study,
56
two factors were extracted for the final solution.
A forced two-factor analysis yielded a solution in
which 16 of the 18 items loaded .40 or greater on only
one factor. An additional criteria for inclusion in
the factor solution was that an item which loaded on
one factor could not load at more than one-half of the
loading value on the other factor. Factor loadings,
eigenvalues and variance accounted for are shown in
Table 1.
The first factor, comprised of ten items,
consisted of items reflective of an anxious attachment.
This factor indicated an attachment with a low levels
of confidence in the availability and responsiveness of
their child and a fear of abandonment, reflecting both
an avoidant and ambivalent attachment style. Because
both ambivalent and avoidant attachment are classified
in the literature under the broader concept of anxious
attachment, this factor was labeled 'Anxious.' The
second factor, labeled 'Secure,' was comprised of four
items and reflected an attachment that indicated a
level of comfort depending on a child and having the
child depend on the parent.
It was originally expected that the factor
analysis would result in a three factor solution; One
factor for each of the attachment styles. However,
57
this was not the case. As a result, the hypotheses for
ambivalent and avoidant attachment were combined and
tested in terms of anxious attachment.
The two factors were only moderately correlated
(.266) suggesting the existence of two distinct
factors. Cronbach's alpha for the Anxious and Secure
factors was .886 and .685 respectively. For data
analysis the items defining each factor were summed to
form two composite scores, one representative of secure
attachment and one representative of anxious
attachment.
Dependent Variables
The first dependent variable in this study was
parental caregiving. Hazan and Shaver (1986) developed
a trichotomous parental caregiving style measure
similar in structure to their attachment paragraphs
(see Table 2). Three paragraphs, based on childhood
memories of parental responsiveness, are written to
represent a Warm style, an Inconsistent style, and a
Cold style of parental caregiving. A goal of this
present study is to undertake a task similar to that by
Collins and Read (1990), in that the three paragraphs
were deconstructed and restructured into a 16-item
scale. Each statement from the paragraphs was made to
58
represent a scale item. Two separate scales were
devised, one to measure maternal caregiving and one to
measure paternal caregiving.
Results from the two scales were combined to form
one overall Parental Caregiving Scale. Each item of
the maternal caregiving scale was added to the
corresponding item on the paternal caregiving scale
resulting in a composite scale. The composite scale
was then factor analyzed using the SPSS factor
solution. An initial orthogonal solution produced a
solution in which three factors had eigenvalues greater
than 1. Several items loaded on more than one factor
in this solution suggesting a correlation among the
factors. The results of a Skree test indicated that a
third factor did not contribute substantively to the
overall solution. An oblique rotation (delta = 0)
produced a cleaner and more theoretically sound
solution, and was therefore used in the final analysis.
Factor loadings, eigenvalues and variance
accounted for are presented in Table 3. As with the
attachment factor solution, only items loading .40 or
higher were selected for inclusion in the final
solution. Thirteen of the original 16 items loaded on
a two factor solution. The first factor (Cold) was
comprised of 8 items reflecting a cold and unresponsive
59
caregiving style. The second factor (Warm) was
comprised of five items reflective of a warm and loving
caregiving style in which the respondents felt loved
and accepted in the relationship with their parents.
Cronbach's alpha was .935 for Cold and .907 for
Warm. The two factors proved to be highly correlated
(.73), suggesting the possibility of a unifactor
solution. Based on the high correlation and the
results of a skree test the two factors were combined
to form one composite factor. Cronbach's alpha for the
combined Parental Caregiving Factor (PCG) was .945.
Composite scores on the combined one factor solution
were used for analysis.
The second dependent variable in this study was
the perceptions one has of their child as a caregiver.
A child caregiving scale was constructed based on the
Hazan and Shaver (1986) parental caregiving paragraphs.
A 15-item scale was constructed from the three
paragraph descriptions. (Two items were inadvertently
combined into one - hence, the difference from the
Parental measure, however in the Parental Caregiving
scales these items both loaded on the same factor.)
An initial SPSS factor run yielded three factors
with eigenvalues over 1. Based on the results of a
skree test and a review of the factor loadings, the
60
number of factors was reduced to two. All fifteen
items loaded on only one factor with a minimum loading
of .40.
Factor loadings, eigenvalues and variance
accounted for are listed in Table 4. The first factor,
(Chcold), was comprised of 10-items measuring a cold
and nonresponsive caregiving style. This factor
differed in content from the Parental caregiving factor
by two items. Both of these items were reworded for
use in the Child Caregiving Scale, and it is possible
that the underlying meanings were changed as a result.
For instance, one of the items which read "It is
possible she/he would just as soon not have had me" in
the two Parental scales was reworded as "It is possible
she/he would just as soon not have to deal with me" for
the Child scale.
The second factor, (ChWarm), was comprised of 5
items reflective of a loving and warm caregiving style.
This scale differed from the Parental Caregiving scale
in only one of the items. The item "I have no major
reservations or complaints about my relationship with
my child" loaded on this factor. However the item "I
have no major reservations or complaints about my
relationship with my mother/father" did not load on the
Warm factor for the Parental Caregiving scale.
61
The factor solutions proved to be highly reliable.
Cronbach's alpha was .932 for Chcold, and .906 for
ChWarm. The two factors were moderately correlated
(.593)* Based on the moderate correlation between the
factors, results of a skree test, and a desire to
maintain consistency with the Parental Caregiving
scale, the two factors were combined to form a
composite unifactor solution. The Chronbach's alpha
for the composite scale was .944. The composite score
for each respondent was used in the analyses of this
study.
The third dependent variable in this study was
perceived social support. Previous research on social
support and the elderly has primarily used measures of
social embeddedness or a summary index of assistance
received - such as advice or transportation (Arling,
1987; Krause, et al., 1989). However, the aspect of
social support most important to this study is
perceived support.
In this study perceived support was
operationalized as a score on a modified version of
Procidano and Heller's perceived social support from
family scale (PSS-FA). The items on the scale were
modified so they would be more focused on social
support from one's child rather than from the family as
62
a whole. For instance, an item originally written MMy
family gives me the moral support I need" was modified
to read "My child gives me the moral support I need."
As designed, the PSS-FA measurement responses are
Yes/No/Don't Know. In order to broaden the range of
responses and increase the sensitivity of the measure,
this study asked volunteers to rate their responses on
a 5-point scale ranging from strongly disagree to
strongly agree. This modified scale proved to be
internally consistent and reliable in this study with a
Chronbach's alpha of .953. The responses were summed
across the variables and the total was used as a
measure of perceived social support.
The fourth dependent variable in this study is
self-rated health. Because this study was interested
the relationship of attachment to the perceptions one
holds of his/her health, it is the subjective
evaluation of one's health that is of interest. The
most frequently used measure of perceived health in
epidemiological and gerontological research is the use
of a single item measure rated on a 4 to 5-point scale
(LaRue et al., 1979; Liang, 1986). Generally this
single measure item asks a respondent for a general
rating of their health, such as "In general, how would
you rate your health?" (Ferraro, 1980; Linn and Linn,
63
1984; LaRue, et al., 1979; Maddox and Douglass, 1973;
Mossey and Shapiro, 1982; Pilpel, et al., 1988;
Rosencranz and Pihlblad, 1970; Salthouse, et al., 1990;
Tessler and Mechanic, 1978; Tissue, 1972). Other
measurements have asked respondents to rate their
health as compared to others who are in the same age
group (Cockerham, et al., 1983; Linn and Linn, 1983;
Mossey and Shapiro, 1982; Rosencranz and Pihlblad,
1970; Tissue, 1972), to rate their health as compared
to the previous year (Linn and Linn, 1983; Mossey and
Shapiro, 1982), or to rate their health as compared
with five years ago (Krause, 1990).
In this study perceived health was operationalized
as a composite of four questions. These questions
asked the respondent to rate their overall health, to
rate their health as compared to others in their age
group, their health compared to one year ago, and their
health compared to five years ago. Responses to these
questions were summed to create a measure of perceived
health status. The resultant score was used in the
final analyses. This scale proved to be internally
consistent and reliable with a Cronbach's alpha of
.907.
The fifth and final dependent variable in this
study was objective health. Objective health has most
64
often been operationalized in the self-rated health
literature as number and severity of reported health
conditions (Mossey and Shapiro, 1982). Generally,
respondents are questioned about the nature and
seriousness of any health problems or illnesses.
Previous research has presented respondents with
checklists of symptoms or diagnoses (Cockerham, et al.,
1983; Ferraro, 1980; Linn and Linn, 1984; Pilpel, et
al., 1988; Tissue, 1972), and prescription medications
(Linn & Linn, 1984; Salthouse, et al., 1990).
Respondents are then asked to indicate the
symptoms/conditions from which they currently suffer,
and any medications they presently use. This study
used the symptom and prescription check-lists extracted
from the Linn and Linn (1984) Self Evaluation of Life
Function Scale (SELF).
The symptom list presented twenty diagnoses and
two blank spaces marked "other.1 1 Respondents simply
indicated any symptoms or conditions they believed they
presently had. In this study responses from the
symptom list were split into two conditions. The
scores for each respondent were summed into a score
reflective of either a serious and potentially life
threatening illness (such as heart disease) or an
illness affecting the quality of one's life but which,
65
with rare exception, is not life threatening (such as
arthritis). The responses included by the respondent
were coded into one of the two categories at the time
of data entry. The two individual scores were used in
the analysis along with the total number of symptoms
reported.
The prescription list presented twenty common
medications such as heart medication and aspirin, and
two blank spaces marked "other." The responses from
this scale were also summed into one of two scores.
The first score included medications which are
necessary for well-being and must be taken on a regular
basis (such as pills for high blood pressure). The
second score represented medication taken either
occasionally or only as needed (such as asthma
medication or vitamins). As with the symptom
checklist, the medications added by a respondent were
coded into one of the two categories by the researcher
at the time of data entry. The two scores were used in
the analyses of this study along with a total score of
the number of prescriptions taken.
Analyses
Hypotheses. Hypotheses la-c were tested with a 2x2
Analysis of variance (ANOVA). Hypotheses lb and lc
66
were originally written with the expectation that the
attachment scale would result in a three factor
solution. Because the two anxious styles loaded
together to form one factor, these hypotheses were
combined and one ANOVA was used to test the
relationship between anxious attachment and scores on
the parental caregiving scales.
For hypothesis 1 the two factors on the attachment
scale, secure and anxious, were split into high and low
scores. This created a 2x2 matrix comprised of the
conditions of high secure/high avoidant, high
secure/low avoidant, low secure/low avoidant, and low
secure/high avoidant. This matrix was used to test for
differences between the groups in obtained scores on
the parental caregiving measure.
For hypothesis 2, a Pearson product moment
correlation was used to test for a relationship between
scores on the total parental caregiving scale and
scores on the attachment to child scale, both of which
were constructed for use in this study.
Hypothesis 3 and 4 were also tested with a 2x2
ANOVA. The same matrix was used as in Hypothesis 1.
For hypothesis 3 the independent variables were high
and low levels of secure and anxious attachment and the
dependent variable was the caregiving style of their
67
child. The independent variable in Hypothesis 4 was
high and low levels of secure and anxious attachment
and the dependent variable was perceived social
support.
Hypotheses 5a-c were tested with separate 2x2
ANOVA's. Hypotheses 5a tested the four attachment
conditions against the self-rated health measure.
Hypothesis 5b tested for the effect of anxious
attachment on the number of symptoms reported, and
hypothesis 4c tested for the effects of attachment
style on total number of medications used.
Research Questions. For the first research
question a t-test was used to determine any significant
differences between the attachment styles of men and
women.
The second research question, interested in the
effects of gender on attachment and perceptions of
social support, was tested by a 2x2x2 ANOVA. The ANOVA
cells were Attachment, Perceived Social Support, and
Gender.
The third and fourth questions were tested through
separate 2x2 ANOVA's. The first question looked at two
types of symptoms across the four attachment
conditions. The second ANOVA examined two levels of
reported medication usage across the four attachment
68
conditions.
The fifth research question was tested with a
2x2x2 ANOVA. This question sought to determine if
gender was a significant variable in self-ratings of
health. In this analysis the cells in the analysis
were Attachment, Self-ratings of Health, and Gender.
69
Table 1
Attachment to Child Scale Items and Factor Matrix
Oblique Solution
Factor Factor
Scale Items 1 2
1. I often worry that my child does
not really love me .8269 -.0079
2. I'm not sure I can always depend
on my child to be there when I
need him/her .8045 .2349
3. I often worry my child will not
want to take care of me . 7751 . 2482
4. I find it difficult to trust
my child completely .7329 .1645
5. I find my child is reluctant to
get as close as I would like .7081 .3168
6. I am somewhat uncomfortable
being close to my child .6398 .1087
7. This child is never there when
need him/her .6217 .3234
8. My desire to bond sometimes
scares people away .4934 -.1763
9. I am nervous when anyone gets too
close .4492 .0131
10. I often worry about being abandoned .4983 -.1148
11. I am comfortable depending
on this child .0996 .6139
12. I know my child will be
there when I need him/her . 3178 .5907
13. I find it relatively easy to
get close to my child .0761 .5674
14. I am comfortable having my
child depend on me -.0482 .5509
15. Often, my child wants to provide
more help than I feel comfortable
receiving .1672 -.3253
16. I find it difficult to allow
myself to depend on this child .1843 . 3102
17. I want to bond completely with my
with my child .2806 -.2822
18. I often worry about someone
getting too close to me .2863 -.1416
Eigenvalue 5.596 2.422
Percent of variance 31.1 13.5
70
Table 2
Descriptions of Parental Carecrivina Style
(From Hazan and Shaver's (1986) unpublished
questionnaire)
1. Warm/Responsive - She/he was generally warm and
responsive; she/he was good at knowing when to be
supportive and when to let me operate on my own; our
relationship was almost always comfortable, and I
have no major complaints about it*
2. Cold/Rejecting - She/he was fairly cold and distant,
or rejecting, not very responsive; I wasn't her/his
highest priority, her/his concerns were often
elsewhere, it's possible that she/he would just as
soon not have had me.
3. Ambivalent/Inconsistent - She/he was noticeably
inconsistent in her/his reactions to me, sometimes
warm and sometimes not; she/he had her/his own
agendas which sometimes got in the way of her/his
receptiveness and responsiveness to my needs; she/he
definitely loved me but didn't always show it in the
best way.
Table 3
Total Parental Caregiving Scale Items and Factor
Pattern
(Combined maternal and paternal scales)
Oblimin rotation
Factor
Scale Items 1
Factor
2
1. He/She was not very responsive
to my needs .9868
2. He/She was rejecting .9421
3. I was not his/her highest priority .8764
4. His/Her concerns were often
somewhere else .8612
5. He/She was fairly cold and distant .8361
6. He/She had his/her own needs and
agendas which sometimes got in the
way of his/her receptiveness and
responsiveness to my needs .6624
7. He/She didn't always show love
in the best way .6601
8. It is possible he/she would just
as soon not have had me .5306
9. He/She was noticeably inconsistent
in his/her reactions to me .4623
10. He/She was sometimes warm and
sometimes not .4396
11. He/She was good at knowing when to
let me operate on my own -.1846
12. He/She was warm and responsive .1742
13. Our relationship was almost always
comfortable .1788
14. He/She definitely loved me .1521
15. He/She was good at knowing when
to be responsive .3211
16. I have no major reservations or
complaints about my relationship
with my mother/father .3969
2485
0902
0354
1006
0375
2251
0396
3655
3759
2703
8830
8672
7555
6482
5928
2920
Eigenvalues before rotation
Variance accounted for before
rotation1
9.541
59.6
1.455
9.1
Variance accounted for is only approximate because
factors were rotated (Tabachnik & Fide11, 1983).
72
Table 4
Child Caregiving Scale Items and Factor Loadings
Oblimin rotation
Scale Items
Factor
1
Factor
2
1
2
3
4
5
8
He/She is fairly cold and distant
He/She is rejecting
He/She is not very responsive
to my needs
I am not his/her highest priority
His/Her concerns are often
somewhere else
It is possible he/she would just
as soon not have to deal with me
He/She is noticeably inconsistent
in his/her reactions to me
He/She is sometimes warm and
sometimes not
He/She has his/her own needs and
agendas which sometimes get in the
way of his/her receptiveness and
responsiveness to my needs
He/She definitely loves me
He/She doesn't always show love
in the best way
He/She is warm and responsive
13. He/She is good at knowing when to
to be responsive and when to let me
operate on my own
Our relationship is almost always
comfortable
I have no major reservations or
complaints about my relationship
with this child
10.
11.
12.
14.
15.
Eigenvalues before rotation
Variance accounted for before
rotation2
.6497
.8810
. 2618
.1884
.7367
.8365
.0689
-.1849
. 9017 -.1989
.5901 .3159
.6721 . 2851
.5088 . 3058
.7213
.0981
.1054
.7467
.4189
.0599
.2624
.9423
.1051 .6594
.2187 .7350
.0603 . 8059
.45
.0
1.40
10. 3
2 Variance accounted for is only approximate because
factors were rotated (Tabachnik & Fidell,1983).
73
CHAPTER 3
Results
Sample Characteristics
Demographics. Forty men and 71 women volunteered
to participate in this research. Median age of the
respondents was 74, and ranged from 52-89.
Thirty-three percent (36) of the respondents were
presently married, 23% (25) were divorced, 40% (44)
were widowed, and 4% (5) were living with a partner.
Thirty two percent (35) had been married more than
once.
Fifty-seven percent of the respondents (63)
reported that they lived alone, 33% (36) lived with a
spouse, and only 3% (4) lived with an adult child.
The average level of education in the sample was
approximately equal to a high school education (12
years), and ranged from no formal education to the
completion of a masters degree.
Social network. Ninety-seven percent (108) of
those studied reported the existence of friends or
relatives with whom they had contact on a regular
basis, while 3% (3) reported that they did not engage
74
in regular interaction with friends or relatives. The
median number of reported close friends and relatives
was 2.8 and ranged betweem 0 and 25.
Ninety-six percent (106) of the sample reported
that they had someone who they considered a confidant.
Thirty-seven percent (41) indicated that this person
was a friend, 25% (29) indicated this person was one or
more of their children, 14% (15) named a professional
(M.D. or counselor), while only 6% (7) named their
spouse.
Fifty-six percent (62) of those studied reported
attending some type of social organization on a daily
basis, 35% (39) attended weekly, and 5% (5) had regular
social activities on a monthly basis.
Adult Children. Seventy-three percent of the
sample had at least two children, 31% had at least
three children, and 20% had more than three children.
None of those studied had more than six children. The
average age of first born children was 49 years and
ranged from 14 to 66. Second born children ranged in
age from 22 to 63, with a median age of 47. Those who
were third in the birth order were on average 43 years
of age, with a range of 27 to 62. Children who were
fourth through sixth in the birth order were a median
age of 36 and ranged in age from 24 to 53. There were
75
slightly more female children reported in this study
(130 or 53%) than male children (114 or 47%).
Fifty-six percent of the sample (62) chose their
first born as the "study" child, while 32% (36) chose
their second born child. None of the respondents chose
a child who was not one of their first two children.
Fifty-three percent (58) chose a female child, while
forty-one percent (45) chose a male child as the study
child.
The number of miles between the parents and their
offspring varied considerably. The median number of
miles that a respondent lived from their study child
was 40, and ranged between 0 and 8000.
The largest percentage of the sample, 25%,
reported that they are in contact with their child two
times per month, followed by those who are in weekly
(17%), daily (16%), and monthly contact (8%). Twenty-
five percent of the respondents reported contact with
their child less than monthly. For 79% of the sample,
the telephone was the most frequent mode of
communicating with their child, followed by visiting
(11%) and writing (5%). Fifty-seven percent (63) of
the sample reported that efforts to communicate were
made equally both themselves and their child, 20%
reported that it was usually the child who contacted
them, and 17% claimed that it was they who made the
effort to contact their child.
Hypothesis One
The first set of hypotheses predicted significant
differences between attachment to one's child and
ratings of their own parents as caregivers. Hypothesis
la predicted that those who were more securely attached
to their child would rate their parents more positively
than those who were less securely attached. Hypotheses
lb and lc were combined in this analysis. Because the
factor solution for attachment produced only two
factors, compared with an expected three factors
solution, both avoidant and ambivalent attachment were
tested under the more general heading of anxious
attachment. Additionally, because the factor solution
for the Parental Caregiving Scale produced a two factor
solution, as opposed to the expected three factor
solution, it was not possible to differentiate between
inconsistent and rejecting parenting styles.
A 2x2 ANOVA indicated no significant main effects
for type of attachment to the child (Anxious or Secure)
and ratings of one's parents as caregivers for either
secure (F (l, 110) =.700, p =.405) or anxious
attachment (F (1, 110) = .659, p =.419). However,
77
there was a significant interaction effect (F (1, 110)
= 2.919/ e = .028).
Those who scored high on anxious and low on secure
attachment to their child rated the caregiving style of
their parents the least positively (M = 3.32), while
those with high scores on both anxious and secure
attachment to their child rated the caregiving style of
their parents most positively (M = 3.81). Those who
scored low on both anxious and secure attachment (M =
3.78) and those who scored high on measures Of secure
attachment and low on measures of anxious attachment (M
= 3.57) rated the caregiving style of their parents
more moderately.
Hypothesis Two
The second hypothesis in this study stated that
there would be a significant positive correlation
between ratings of one's parents as caregivers and
one's perceptions of their child as a caregiver. This
hypothesis was not supported (r=.0795, p <.203).
Hypothesis Three
Hypothesis three predicted that those who were
more securely attached to their child would perceive
the caregiving style of their child more positively
than those who were either less securely attached or
anxiously attached to their child. This hypothesis was
confirmed. Results of a 2x2 ANOVA indicated main
effects for both secure (F (1, 110) = 30.628, p = .000)
and anxious attachment (F (1, 110) = 30.628, p = .000).
Interaction effects for this hypothesis were not
significant (£ (1, 110) = .088, p = .767).
Those who were more securely attached perceived
the caregiving style of their child more positively (M
= 4.18) than those who were less securely attached to
their child (M = 3.42). Along the same lines, those
who were less anxiously attached to their child rated
the caregiving style of their child more positively (M
— 4.23) than those who were more anxiously attached to
their child (M = 3.35).
Hypothesis Four
Hypothesis four predicted that those who are more
securely attached to their child would perceive higher
levels of social support from their child. This
hypothesis was supported. Results from a 2x2 ANOVA
produced significant main effects for both secure (F
(1, 110) = 17.002, p = .000) and avoidant attachment (F
(1, 110) = 19.977, p = .000). Those who scored low on
measures of secure attachment to their child perceived
79
the lowest levels of support from that child (M =
3.13), while those who scored high on secure attachment
to their child perceived the most social support from
their child (M = 3.96). Those who were less anxiously
attached to their child perceived significantly more
social support from their child (M - 3.93) than did
those who were more anxiously attached to their child
(M = 3.93). Interaction effect for this analysis were
not significant (F (1, 110) = 3.445, p = .066).
Hypothesis Five
The fifth set of hypotheses concerned the
relationship of attachment style and physical well
being. Hypothesis 5a predicted that those who were
more securely attached to their child would perceive
themselves as healthier than those who were less
securely attached. A 2x2 ANOVA indicated significant
main effects only for anxious attachment (F (1, 110) =
5.798, p = .018). Those who are more anxiously
attached to their child perceived their health as
poorer (M = 3.35) than those who are less anxiously
attached (M = 3.89). Neither main effects for secure
attachment (F (l, 110) = 2.475, p = .119) or
interaction effects (£ (1, 110) = .403, p = .527) were
significant.
80
Hypothesis 5b predicted that those with more
anxious attachment to their child would report more
overall symptoms than those who were more securely
attached to their child. This hypothesis was not
supported. Neither main effects for anxious (F (1,
110) = .475, p =.492) nor secure (F (1, 110) = .324, e
=.570) attachment were significant. The interaction
effect (E (1, 110) = 454, e - .502) was not
significant.
Hypothesis 5c predicted that those who were more
anxiously attached to their child would report using
more overall medications. Results of a 2x2 ANOVA did
not support this hypothesis. Neither main effects for
anxious (F (1, 110) = .594, e - -443) nor for secure (F
(1, 110) = .563, e = .455) were significant for
attachment and number of symptoms. Interaction effects
(F (1, 110) = .270, e = -604) were not significant.
Research Questions
Gender and attachment. Question one sought to
determine if attachment to one's adult child varied by
gender of the respondent. Results of a T-test indicate
no significant differences between ratings of either
anxious (T(2,109) = -.99, e = .325) or secure
attachment (T(2, 109) = (T(2,109) = .33, e = .741).
81
Gender, attachment and social support. The second
question, interested in gender differences in
attachment and perceived social support produced no
significant main effect (F(l,110) = .056, p = .814).
However, there was a significant two-way interaction
effect between secure attachment and gender (F(1,110) =
6.21, p = .014). Males who scored low on secure
attachment perceived much less support (M = 2.81) than
women who scored low on the same measures (M = 3.30).
And, males who scored high on measures of secure
attachment perceived greater social support (M = 4.15)
than women who scored high on secure attachment to
their child (M = 3.85).
SymptomoloCTY. The third research question was
interested in any difference in the nature of symptoms
reported between the four attachment conditions.
Results from a 2x2 ANOVA indicate significant main
anxious attachment style (F (1, 110) = 5.433, p =.022).
Those who are more anxiously attached reported having
more serious illnesses (M = 1.03) than those who were
less anxiously attached (M = .56). The main effects
for secure attachment and reporting of serious illness
were not significant (F (1, 110) = .106, p = .745) nor
were interaction effects (F (1, 110) = 008. p = .928).
For quality of life illnesses, neither main
82
effects for either anxious (F (1, 110) = .022, p =
.884) or secure attachment (F = (1, 110) = .676, p =
.413) were significant. The interaction effect for
this analysis was not significant (F (1, 110) = .908, £
= .343).
Medications. A fourth research question asked if
there was a difference among those in the attachment
groups in the nature of medications they used. A 2x2
Anova indicated significant main effects for secure
attachment and the use of occasional medications (F (1,
110) = 4.168, p = .043). Those who were more securely
attached to their child reported the use of fewer
occasional medications (M = .77) than those who were
less securely attached to their child (M = 1.15). The
main effects for anxious attachment were not
significant (F (1, 110) = .450, p. — .504) and the
interaction effects were also not significant (F (1,
110) = . 3.3 3 , p = . 565) .
An ANOVA for the use of necessary daily medication
did not provide significant main effects for either
anxious (F (l, 110) = 3.490, p = .064) or secure
attachment (F (1, 110) = .012, p = .914). Interaction
effects for this analysis were also not significant (F
(1, 110) = 3.428, p = .065).
83
Gender. attachment and self-ratinas of health.
The ANOVA for the fifth question indicated no
significant main (F(1,110) = .598, e — -441) effects
for the effect of gender on self-ratings of health. The
interaction effects for secure (£(1,110) = .499, e -
.481), and anxious attachment (F(1,110) = .385, e —
.536) were also not significant.
CHAPTER FOUR
Discussion and Future Directions
Overview
This dissertation was interested in the effects of
attachment style on health and relational outcomes in
an elderly population. Specifically, this study sought
to determine if attachment to an adult child was
predictive of ratings of one's own parents as
caregivers, perceived social support from the adult
child, and physical health outcomes. There have been
conflicting assertions as to the role of the attachment
system during the later part of the life span. Bowlby
(1982) suggested that as an individual ages, the
attachment process shifts from an older to a younger
generation and one's offspring become important
attachment figures. On the other hand, Weiss (1982)
has suggested that as one ages the need for attachment
is replaced with a desire for solitude. Results from
this study provide support for Bowlby's theory and
indicate that individual attachment style remains an
important predictor of both health and relational
outcomes well into old age.
85
Discussion
Attachment:. Much recent attention has come to the
study of attachment in adult populations. With few
exception (e.g., Cicirelli, 1981, 1983a, 1983b; Hazan &
Shaver, 1987; Henderson, et al., 1981), this research
has focused exclusively on young college age adults.
It was argued in this dissertation that the attachment
system remains activated throughout the human life
span, well past young adulthood and into old age.
Specifically, this study investigated how one's
attachment to an adult child affects both relational
and physical health outcomes in a population of older
!
adults. |
Previous research has demonstrated that, in post j
adolescent populations, the style with which one
attaches to a romantic partner is highly correlated
with the perceptions one holds of their parents
caregiving style (see Feeney & Noller, 1990; Hazan &
Shaver, 1987; Kobak & Sceery, 1988; Mikulincer et al.,
1990). With an understanding of adult attachment as an
extension of the cognitive models formulated in the |
I
beginning years of life, this study predicted that this !
i
relationship between attachment and perceptions of
parental caregiving would hold throughout the life
span. The first hypothesis in this study predicted
86
that those who were securely attached to their adult
"Study" child would have more positive perceptions of
their own parents as caregivers than would those who
were more anxiously attached to their child.
Attachment was measured in this study with a modified
Collins and Read (1990) adult attachment scale. This
measure was modified to represent attachment to an
adult child, rather than to a romantic partner.
Perceptions of parental caregiving were measured with a
scale instrument constructed for use in this study and
based on the Hazan and Shaver (1987) paragraph
descriptions of parental responsiveness.
Results indicate that those who scored high on
measures of anxious attachment and low on measures of
secure attachment to their child perceived their
parents least positively, while those with high scores
on both measures of anxious and secure attachment
perceived their parents caregiving style most
positively. The first result was as expected: It
appears that, at least for the anxiously attached,
perceptions of one's parents as caregivers remain an
important organizing factor late into life.
The interaction of a high score on secure coupled
with a high score on anxious attachment is interesting.
This finding is consistent with an Anxious/Ambivalent
87
style: An attachment that expresses both high levels of
anxiety (in terms of fear of rejection and/or
abandonment) along with high levels of dependence and
love. This result is in line with previous findings
that those who are anxiously attached are most likely
to idealize one or both of their parent figures (e.g.,
Kobak & Sceery, 1988). It is possible that those who
scored high on both measures can be classified as
Anxious/Ambivalent in their attachment to their child.
However, this could not confirmed in this study because
of an inability of the factor solution to discriminate
between the two anxious styles of attachment.
An alternative explanation for this finding
concerns the nature of parent-child attachment.
Previous research using adult populations has most
often studied romantic attachment. Attachment among
family members, particularly a parent's attachment to
their child, is probably substantively different from
attachment to a romantic partner. For instance,
romantic attachment, especially among young adults, is
often imbued with idealized expectations for future
interactions. On the other hand, attachment among
family members is based on past history and
interactions, and on more realistic expectations for
future interactions.
88
1
It was reasoned that people hold role specific
expectations, based on past experiences, for role
specific behaviors, such as that of a caregiver. Based
on the argument of attachment style as an extension of
mental representations of the responsiveness and
receptiveness of a primary caregiver, it was argued
that those in similar roles would be perceived
similarly. A second hypothesis predicted that
perceptions of an adult child's caregiving style would j
I
be positively correlated with perceptions one held of J
their parents as caregivers. A child caregiving scale
was constructed for use in this research based on a j
modified version of the parental caregiving scale
described above. This hypothesis was not confirmed.
Ratings of one's parents caregiving style are
related to high scores on measures of anxious
attachment, however, these perceptions do not appear to
influence evaluations of the caregiving style of the
child. Consistent with Bowlby's theory, attachment to
an older population (one's parents) may play a less j
I
significant role in how life is experienced later in j
j
the life course. This finding is also consistent with j
i
i
previous research findings that the benefits of the j
!
parent-child relationship to a child diminish as the j
i
l
adult child ages beyond young adulthood, while at the
same time the benefits to the parent remain fairly
stable over the life course of their adult child
(Roberts, 1990).
While the parent-child relationship remains
important over the life-course of the parent, having
lived through parenthood themselves, older parents may
come to understand and to accept the shortcomings of
their own parents. For a majority of older adults, the
shift in attachment to a younger population may be
accompanied by a letting go of, and coming to terms
with, issues in the relationships with their own
i
parents. j
Percieved social support. Individual differences j
in attachment style suggest differences in the belief j
of the availability and helpfulness of important others I
I
in times of need. Shaver and Hazan (1987) postulated a
relationship between attachment style and comfort with
giving and receiving comfort and care. And, a
relationship between attachment style and perceived
social support from family and friends has been
1
established in the literature (e.g., Henderson, et al., |
1981; I.G. Sarason & Sarason, 1985). The third
hypothesis in this study sought to determine if this
relationship held true in a population of older adults, j
This study predicted that those who were more securely
90
attached to their child would perceive higher levels of
social support from this child than would those who
were less securely attached. This hypothesis was
supported.
This finding supports the idea of individual
attachment style as enduring cognitive models of the
anticipated receptiveness and responsiveness of others.
Those individuals who are anxiously attached perceived
significantly less available support from their child
i
than did those who were securely attached. Whether {
these perceptions translate into actual behaviors must j
j
be determined in future research. Results from this
study lend credence to the theory of a continuity of
attachment related phenomenon across the life-span.
Results also suggest some measure of continuity across
both age groups and context (parent-child versus
romantic attachment) in how the attachment system
functions in adulthood. This finding provides a useful
predictor into the likelihood that an elder will ask
for assistance in time of need, and perhaps their
receptiveness in accepting assistance.
Physical well-being. A third area of interest was
the relationship of adult attachment on overall
physical well-being. While little attachment research
has focused on health outcomes, previous research has
91
found a relatively strong relationship between adult
measures of attachment and psychological well-being
(e.g., Henderson, 1977; Henderson et al., 1978, 1981).
And, both Bowlby (1977) and Henderson (1974) have
postulated a relationship between attachment and
physical well-being outcomes. In light of the
substantial relationships between attachment and
perceived social support, and perceived social support
and physical health outcomes, this study predicted a
relationship between attachment style and overall
physical well-being.
Results indicate that those who were more
anxiously attached to their child perceived themselves
as in poorer health than did those who were less
anxiously attached. This result supports the theory of
a causal relationship between anxious attachment and
physical well-being. However, further research
determine the strength and nature of this link. This
result is striking in light of previous findings of the
relationship between self-rated health and mortality in
elderly populations, and suggests a consideration of a
patients relationship with their offspring as a part of
medical treatment.
Henderson (1974) has argued that those with
anxious attachment styles are likely to use their
92
health to elicit care from important others. Results
from this study did not find a difference in the number j
of reported symptoms between those who were anxiously j
versus those who were securely attached to their child.
However, there was a significant difference in the
nature of the symptoms reported by the respondents.
In this study health conditions and symptoms were j
separated into two types: Those that are by nature
serious or life threatening (such as heart disease),
and those that primarily affect the quality of one's !
i
life but are rarely life threatening (such as j
arthritis). Respondents who were more anxiously I
attached to their child reported significantly more 1
serious symptoms and conditions than those who were
less anxiously attached. This finding lends support to
Henderson's (1974) theory that those with less secure
attachment styles may indeed use their physical health
and well-being as a tool to gain desired support from
important others.
It was also hypothesized that those who were
anxiously attached to their child would report using a
greater amount of medication. This was not found to be
the case. There were no significant differences in the
reported amount of medication used by those who were
anxiously attached versus those who were securely
93
attached to their child. However, there was a
substantive difference in the type of medications used
by those with different attachment styles. Respondents
who were less securely attached to their child reported
the use of more occasional medications such as
vitamins, diuretics, and asthma medications. The
reported usage of necessary medications, such as heart
or high blood pressure medication, was not significant
between the groups.
Gender. As noted in Chapter One, the role of
gender has proven significant in both the study of the
i
social exchange within the parent-child relationship
and in self-ratings of health. Three research
questions were posed concerning the importance of
gender as a predictor variable in parent-child
attachment.
The first question sought to determine gender
differences in reported attachment styles. A
comparison of male and female attachment measures found
no significant differences in one's attachment style as j
»
discriminated by gender of the respondent. This
finding is consistent with the results of Collins and
Read (1990).
A second question addressed the role of gender in
i
the perceptions one has of the social support provided
to them by their adult child. Previous research (e.g.
A.S. Rossi & Rossi, 1990) has found gender to be a
strong predictor of the type of support provided to an
aging parent. A comparison of measures in this study
indicates that men may perceive available support from
their children in more extreme terms than do women.
Men who scored high on measures of secure attachment
perceived greater support from their child than did
women with similar scores. On the other hand, men who
rated low on secure attachment perceived far less
support from their child than did women who were less
securely attached to their child.
Rossi & Rossi found that mothers overwhelmingly
receive more support from their offspring, both in type
of support and amount provided. They attribute this to
the investment of time, energy, and affect into the
lives of the child which is implicit in the maternal
caregiving role. Women typically are in contact more
frequently with their adult children, resulting perhaps
in a relationship with a higher level of familiarity
and comfort, in fostering more moderate and stable
perceptions of available support. In accounting for
the above finding, a father with a highly secure
attachment to his child may believe he has a
relationship that is stronger than the traditional
95
father-child relationship. As a result, these men may
perceive higher levels of help available should it
become necessary. On the other hand, men who score low
on secure attachment to their child may feel
particularly alienated from the child, thus accounting
for the low ratings of perceived support.
The third question sought to determine if gender
was a significant predictor of self-rated health. In
this sample, gender did not prove significant in self-
ratings of overall well-being. This finding is in
contrast with previous research which has found
significant gender effects (e.g., Ferraro, 1980).
There are two readily available reasons this finding
may differ from previous research. First, the sample
size in this study is considerably smaller than that of
previous research, and this, coupled with the uneven
distribution of males and females may be responsible
for a lack of findings. An alternative explanation
stems from the sampling method. The respondents for
this research were recruited from senior activity
centers. Those who attend activity centers tend to be
in relatively good health, and as such may perceive
themselves as healthier than others in their cohort.
Limitations
This study provides evidence that attachment, as
conceptualized in the adult attachment literature, is
an important factor in how one's child is perceived in
terms of their caregiving behaviors, social support
provided to the parent, and perhaps also an important
determinant in overall physical functioning. However,
several notes of caution must be raised as to the !
generalizability of the results of this study. j
First, the nature of the attachment examined in !
9 1
this research is substantively different from that in !
other studies which have used similar measures. As
noted previously, attachment to one's child can be
expected to differ from romantic attachment. As a
result, the health outcome indications in this study
may not generalize to other types or measures of
attachment. Results from the parental caregiving and
social support measures do indicate, however, that
there is some degree of generalizability between
attachment to a romantic partner and attachment to i
one's child. However, the similarities and differences I
' i
are not yet clear enough to confidently generalize
across either populations or attachment figure types.
Second, a limitation in the sampling method is
notable. Perhaps the most limiting aspect of this
study was the lack of information about the respondents
relationships with their other children. In asking a
respondent to simply choose one child on whom to
report, a measure of control was lost. This became
painfully obvious during the data collection when one
of the respondents asked "Do you want me to tell you
about the kid I love, or the kid I hate?"
In this study the underlying context of the
specific parent-child relationship (compared with the
respondents relationships with their other children) J
was not taken into consideration. As a result, it is j
impossible to draw strong conclusions about the !
relationship of the attachment the respondent had to
the study child with perceptions of their parents j
caregiving style, and general physical health outcomes. I
Because of how the data were gathered, it is not
clear if respondents with more than one child would
have reported more than one attachment style. It is
conceivable that a parent who is securely attached to
one child may be avoidantly attached to another. An
attempt should be taken in future research to
determine, in the case of more than one offspring, how
a respondents relationship with their chosen study
child compares to their relationships with their other
children. This information would vastly increase the
98
strength of any conclusions and provide increased
insight into how attachment functions differently
across similar relationships.
An additional limitation of this study is that,
unlike previous research in attachment, it was not
possible to differentiate between the two anxious
styles of attachment. Because of this, the conclusions
of this study were limited to the two basic types of
attachment, Secure and Anxious, and finer distinctions
could not be made. One explanation for this is the
size of the sample. It may be that the number of
respondents in the study was too small to obtain a
strong factor solution on the Anxious attachment scale.
This will be determined in future research.
It may be also, that anxious attachment to an
adult child differs from other forms of anxious
attachment, with the result being that there are not
two clear styles (at least in terms of how it has been
traditionally measured). The important elements of a)
attachment to one's adult child, and b) attachment in
late life, are still relatively unclear. It is as yet
unclear if there are important differences among
attachments to different figures, and/or attachment at
various points in the life-span.
A third explanation focuses on the nature of
99
Anxious/Avoidant attachment. A person with an Avoidant
attachment style is characterized as having difficulty
establishing and maintaining relationships, having a
high need for autonomy and indendence, and perhaps a
latent mistrust of others. Older adults attend senior
centers primarily for the social interaction, and as
such one might not expect to find a large number of
people with an anxious style in attendance. And, those
who are present are not likely to volunteer their
participation in a study on the nature of human
relationships.
Future Directions
This study has provided empirical support for a
relationship between the attachment system and
important health and relational outcomes among the
elderly. Several suggestions are made for future
research.
First, this research has illustrated that there
are strong similarities between attachment in early
adulthood to a romantic figure and attachment in later
life to an adult child. This conclusion can be drawn
on the basis of the good "fit" of the modified Collins
and Read (1990) attachment measure with the current
population. However, it is naive to think that there
are not also important differences. Two differences
which should be explored in future research are: That
between romantic attachment and attachment to one's
offspring; and differences between attachment in young
adulthood and in later adulthood.
For an older parent the relationship with their
child is a product of many years of interaction and
communicated expectations. Unlike romantic attachment,
a parent often takes much credit or blame for the
"final product" in their adult child. As such, there
is a much greater investment in, and perhaps even
greater expectations for, desired relational outcomes.
The attachment system probably undergoes several
significant changes during various stages of life, such
!
i
as changes in attachment figures due to both expected '
i
and unexpected losses and additions. It is yet unclear j
what changes occur in the attachment system as a result j
of life experiences. For instance, do those with
particular attachment styles experience life in a way
that facilitates this style, or do certain experiences
(such as marriage or retirement) change the attachment
system in predictable ways?
A second area worth further attention is the
relationship between attachment and health outcomes.
This finding is striking in light of the previous
101
research into self-rated health and actual health and
mortality outcomes. The broader implications of this
finding must be examined. While results of this study j
indicate a relationship between attachment to an adult j
child and self-ratings of health, it is not clear if
|
this result comes out of an individual attachment |
i
style, or one's reported attachment to the child. j
Future research should determine the extent to which 1
I
1
attachment to one child differs from attachment to
one's other children, or from another figure, such as a
spouse or a close friend in terms of health related
i
outcomes.
Finally, the study of attachment lends itself more j
I
readily to interview, rather than questionnaire, j
methods of data collection. Much is to be gained not
only through interviews with those in a sample
population, but also with those who are acquainted with
them. This too became clear during data collection
when at various times it was obvious that important
information was not made available for analysis. The
most extreme example of this came from one woman who
insisted on holding the forearm of the researcher with
her free hand while she completed the questionnaire.
When she encountered negative statements on the measure |
about either her parents or her child (a daughter) she !
I
102 ;
would pause and exclaim "Oh MY! NO!" After she
completed the questionnaire (still clutching the
researchers arm) she began discussing the relationship
with her daughter. Within minutes she was complaining
about the lack of time spent with her daughter and
commenting on her dissatisfaction with their
interactions. Soon she was emulating the daughter, "Oh
MO-ther. Do you HAVE to tell me about that A-GAIN?"
Interestingly enough, once the researcher had
freed herself from the grip of this woman, one of the
woman's long-time acquaintances snickered and remarked
that it was amazing to have gotten away so soon. While
the group seemed to tolerate her, she had evidently
alienated many of them throughout the years as a result
of her demanding and clinging "style." The people with
whom this woman socialized may have been able to
provide an accurate portrayal of her long term
attachment behavior. Despite the fact that interviews
are much more labor intensive than questionnaires,
interviewing a respondent will provide more accurate
information within which to understand the attachment
system in later life.
!
Conclusion J
This dissertation was designed as an initial probe }
103 !
into how attachment operates in a population of older
adults. The larger goal of this research was to
confirm that the attachment system remains functional
into old age. Results of this study have empirically
established that the attachment system (as measured in
younger adult populations) is active late into the
human life span. And, also, that the attachment system
remains an important organizing factor, as hypothesized
by Bowlby.
A more narrowly defined goal of this dissertation
was to explore health and relational outcomes as
predicted by attachment to an adult offspring.
Previous research has illustrated a link between
attachment and psychological health outcomes. The
current research has demonstrated that a relationship
also exists between attachment style and physical
health outcomes. Consistent with previous research,
this study also confirmed that the parent-child
relationship remains an important part of the life of
the parent (e.g., Roberts, 1990). And, evidence has
been provided that attachment plays a significant role
for the parent in this relationship.
This dissertation sought to determine if
attachment was an important element in the health and
relationships of the elderly. This research has
provided an initial insight into the nature and
function of the attachment system in the later years.
It is important, however, that future research devote
more energy into exploring the important elements of
attachment during the later part of the life span, and
on attachment other than to a romantic partner. While
answering several questions, this study confirmed the
most important: The study of attachment in older adult !
i
populations has much to offer those interested in the ;
i
!
study of interpersonal relationships across the life !
span. j
10 5
References
Ainsworth, M. D. S. (1982). Attachment: retrospect and
prospect. In C.M. Parkes and J. Stevenson-Hinde
(Eds.), The place of attachment in human
behavior (pp. 3-29). New York: Basic Books.
Ainsworth, M. D. S., Blehar, M. €., Waters, E. & Wall,
S. (1978). Patterns of attachment: A psychological
study of the strange situation. Hillsdale, New
Jersey: Lawrence Erlbaum Associates, Publishers.
Antonucci, T. C. (1990). Social supports and social
relationships. In R.H. Binstock & L.K. George
(Eds.), Handbook of aging and the social sciences
(3rd ed. , pp. 205 - 226). San Diego: Academic
Press, Inc.
Antonucci, T. C. (1985a). Social support: Theoretical
advances, recent findings and pressing issues. In
I.G. Sarason & B.R. Sarason (Eds.), Social
support:Theory. research and applications (pp.
21-37). Dordrecht, The Netherlands: Martinus
Nijhoff Publishers.
Antonucci, T. C. (1985b). Personal characteristics,
social support, and social behavior. In R.H.
Binstock & E. Shanas (Eds.), Handbook of aaina
and the social sciences (2nd ed., pp. 94-128). New
York: Van Nostrand-Reinhold.
Antonucci, T. (1976). Attachment: A life-span
concept. Human Development. 19, 135-142.
Arend, R., Gove, F. L., and Sroufe, L. A. (1979).
Continuity of individual adaptation from infancy
to kindergarten: A predictive study of ego-
resiliency and curiosity in preschoolers. Child
Development. 50, 950-959.
Arling, G. (1987). Strain, social support, and distress
in old age. Journal of Gerontology. 42(1), 107-
113.
Barrera, M. Jr. (1986). Distinctions between social
support concepts, measures, and models. American
Journal of Community Psychology, 14(4), 413-445.
106
U
Barrera, M. Jr., Sandler, I. N., and Ramsay, T. R.
(1981). Preliminary development of a scale of
social support: Studies on college students.
American Journal of Community Psychology. 9(4),
435-447.
Blazer, D. (1980). Life events, mental health
functioning and the use of health care services by
the elderly. American Journal of Public Health,
70, 1174-1178.
Bowlby, J. (1977). The making and breaking of
affectional bonds: Aetiology and psychopathology
in the light of attachment theory. British Journal
of Psychiatry. 130, 201-210.
Bowlby, J. (1980). Attachment and loss. Volume 3: Loss,
sadness and depression. New York: Basic Book, Inc.
Bowlby, J. (1982). Attachment and loss. Volume 1:
Attachment (2nd ed.). New York: Basic Books, Inc.
Butler, R. N. (1981). Overview on aging: Some
biomedical, social, and behavioral perspectives.
In J.G. March, S.B. Kiesler, J.N. Morgan, and V.
K. Oppenheimer (Eds.), Aging: Social change (pp.
1-8). New York: Academic Press.
Cicirelli, V. G. (1981). Helping elderly parents: The
role of adult children. Boston: Auburn House
Publishing.
Cicirelli, V. G. (1983a). Adult children's attachment
and helping behavior to elderly parents: A path
model. Journal of Marriage and the Family, 815-
826.
Cicirelli, V. G. (1983b). Adult children and their
elderly parents. In T.H. Brubaker (Ed.), Family
relationships in later life (pp. 31-46). Beverly
Hills: Sage.
Cockerham, W. C., Sharp, K., & Wilcox, J. S. (1983).
Aging and perceived health status. Journal of
Gerontology f 38(3), 349-355.
107
Collins, N. L. , & Read, S . J. (1990). Adult attachment,
working models, and relationship quality in dating
couples. Journal of Personality and Social
Psycholoay. 58(4), 644-663.
Cooke, L., & Miller, L. C. (1990). Mental models of
parental caregiving styles: Do they affect marital
satisfaction and communication? Unpublished
manuscript, University of Southern California.
Costa, P. T. Jr., McCrae, R. R., Locke, B. Z. (1990).
Personality factors. In J.C. Cornoni-Huntley, R.R.
Huntley, & J.J. Feldman (Eds.), Health status and
well-being of the elderly: National health survey-
I epidemiologic follow-up study (pp. 210-220)
New York: Oxford University Press.
Cutrona, C. E. (1986). Objective determinants of
perceived social support. Journal of Personality
and Social Psychology. 50(2), 349-355.
Dunkel-Schetter, C., and Bennett, T. L. (1990).
Differentiating the cognitive and behavioral \
aspects of social support. In Sarason, B. R., j
Sarason, I. G., and Pierce, G. R. (Eds.), Social t
Support: An Interactional View, (pp. 267-295). New j
York: John Wiley and Sons ]
i
Dunkel-Schetter, C., Folkman, S., & Lazarus, R. S. I
(1987). Correlates of social support receipt. j
Journal of Personality and Social Psychology.
53(1), 71-80.
Dunkel-Schetter, C., and Wortman, C. B. (1981).
Dilemmas of social support: Parallels between
victimization and aging. In J. G. March, S. B.
Kiesler, J. N. Morgan, V. K. Oppenheimer (Eds.),
Aging: Social Change. (pp. 349-389). New York:
Academic Press
Ferraro, K. R. (1980). Self ratings of health among
the old and the old-old. Journal of Health and
Social Behaviorf 21, 377-383.
Feeney, J. A., & Noller, P. (1990). Attachment style as
a predictor of adult romantic relationships.
Journal of Personality and Social Psychology.
58(2), 281-291.
108
George, L. K. (1990). Social structure, social
processes, and social-psychological states. In R.
M. Binstock & L. K. George (Eds.), Handbook of
aaina and the social sciences (pp. 186-204). San
Diego, CA: Academic Press.
Hanson, S. M., & Saur, W. J. (1985). Children and their
elderly parents. In W. J. Saur & R. T. Coward
(Eds)., Social Support Networks and the Care of
the Elderly, (pp. 41-66). New York: Springer.
Hazan. C. & Shaver, P. (1986). [Parental caregiving
style questionniare]. Unpublished questionnaire.
Hazan, C. & Shaver, P. (1987). Romantic love
conceptualized as an attachment process. Journal
of Personality and Social Psychology. EL2(3), 511-
524.
Henderson, S. (1974). Care-eliciting behavior in man.
The Journal of Nervous and Mental Disease. 159(3),
172-181.
Henderson, S. (1977). The social network, support, and
neurosis: The function of attachment in adult
life* British Journal of Psychiatry. 131, 185-191.
Henderson, S. (1982). The significance of social
relationships in the etiology of neurosis. In C.
M. Parkes & J. Stevenson-Hinde (Eds.). The place
of attachment in human behavior (pp. 205-231).
New York: Basic Books
Henderson, S., Byrne, D. G., Duncan-Jones, P., Adcock,
S., Scott, R., and Steele, G. P. (1978). Social
bonds in the epidemiology of neurosis: A
preliminary communication. British Journal of
Psychiatry. 132, 463-466.
Henderson, S., Byrne, D. G., Duncan-Jones, P. (1981).
Neurosis and the Social Environment. Sydney
Australia: Academic Press.
Henderson, S., Duncan-Jones, P., Byrne, D. G., and
Scott, R. (1980). Measuring social relationships:
The interview schedule for social interaction.
Psychological Medicine. 10, 723-734.
109
Hinde, R. A. (1982)* Attachment: Some conceptual and
biological issues. In C.M. Parkes, and J.
Stevenson-Hinde (Eds.), The place of attachment
in human behavior (pp. 60-76). New York: Basic
Books.
Jung, J. (1984). Social support and its relation to
health: A critical evaluation. Basic and Applied
Social Psychology, 5(2). 143-169.
Kahn, R. L. , and Antonucci, T. C. (1981). Convoys of
social support: A life course approach. In J. G.
March, S. B. Kiesler, J. N. Morgan, and V. K.
Oppenheimer (Eds.), Aaincr: Social change, (pp.
349-389). New York: Academic Press
Kobak, R. R., & Sceery, A. (1988). Attachment in late
adolescence: Working models, affect regulation,
and representations of self and others. Child
Development. 59, 135-146.
Krause, N. (1990). Perceived health problems,
formal/informal support, and life satisfaction
among older adults. Journal of Gerontology: Social
Sciences. 45(5), S193-205.
Krause, N. (1989). Issues of measurement and analysis
in studies of social support, aging and health. In
K.S. Markides & C.L. Cooper (Eds.), Aaina. stress
and health, (pp. 43-66). New York: John Wiley &
Sons.
Krause, N., Liang, J., and Yatomi, N. (1989).
Satisfaction with social support and depressive
symptoms: A panel analysis. Psychology and Aaina.
4(1), 88-97.
LaRue, A., Bank, L., Jarvik, L., & Hetland, M. (1979).
Health in old age: How do physicians' ratings and
self-ratings compare? Journal of Gerontology. 41,
248-260.
Levy, M. B., & Davis, K. E. (1988). Lovestyles and
attachment styles compared: Their relations to
each other and to various relationship
characteristics. Journal of Social and Personal
Relationships. 5, 439-471.
110
Liang, J. (1986). Self-reported physical health among
aged adults. Journal of Gerontology, 41(2), 248-
260.
Linn, M. W., & Linn, B. S. (1984). Self-evaluation of
life function (SELF) scale: A short, comprehensive
self-report of health for elderly adults. Journal
of Gerontology. 39(5), 603-612.
Maddox, G. L., & Douglass, E. G. (1973). Self
assessment of health. Journal of Health and Social
Behavior. 14, 87-93.
Main, M., Tomasini, L., & Tolan, W. (1979). Differences
among mothers of infants judged to differ in
security. Developmental Psychology, 15(4), 472-
473.
Main, M., & Weston, D. R. (1982). Avoidance of the
attachment figure in infancy: Descriptions and
interpretations. In. C.M. Parkes & J.Stevenson-
Hinde (Eds.). The place of attachment in human
behavior (pp. 31-59). New York: Basic Books.
Main, M. and Weston, D. R. (1981). The quality of the
toddler's relationship to mother and to father:
Related to conflict behavior and the readiness to
establish new relationships. Child Developmentf
52, 932-940.
Markides, K. S., & Cooper, C. L. (1989). Aging, stress,
social support and health: An overview. In K.S.
Markides & C.L. Cooper (Eds.^ Aaina. stressf and
health (pp. 1-10). New York: Wiley and Sons.
Markides, K. S., & Martin, H. W. (1979). Predicting
self-rated health among the aged. Research On
Aging, 1, 97-112.
Marris, P. (1982). Attachment and society. In C.M.
Parkes & J. Stevenson-Hinde (Eds.). The place of
attachment in human behavior (pp. 185-201). New
York: Basic Books.
Mikulincer, M., Florian, V. , & Tolmacz, R. (1990).
Attachment styles and fear of personal death: A
case study of affect regulation, Journal of
Personality and Social Psychology, 58(2), 273-280.
Ill
Miller, P. McC., and Ingham, J. G. (1976). Friends,
confidants and symptoms. Social Psychiatry. 11,
51-58.
Mossey, J. M., & Shapiro, E. (1982). Self-rated health:
A predictor of mortality among the elderly.
American Journal of Public Health. 72(8). 800-808.
Parkes, C. M. (1982). Attachment and the prevention of
mental disorders. In C.M. Parkes and J. Stevenson-
Hinde (Eds.). The place of attachment in human
behavior (pp. 295-310). New York: Basic Books.
Perlmutter, M., & Nyquist, L. (1990). Relationships
between self-reported physical and mental health
and intelligence performance across adulthood.
Journal of Gerontology: Psychological Sciences.
45(4), P145-155.
Pilpel, D., Carmel, S. & Galinsky, D. (1988). Self-
rated health among the elderly. Comprehensive
Gerontology. Sect B, 2, 110-116.
Procidano, M. E., and Heller, K. (1983). Measures of
perceived social support from friends and from
family: Three validation studies. American Journal
of Community Psychology. 11(1), 1-24.
Reis, H. T. (1990). The role of intimacy in
interpersonal relations. Journal of Social and
Clinical Psychology. 9(1), 15-30.
Roberts, R. E. L. (1990). Intergenerational affection
and psychological well-being: Implications of the
changing salience of work and family over the
adult life course. (Unpublished Doctoral
Dissertation, University of Southern California,
1990).
Rook, K. S. (1984). Research on social support,
loneliness, and social isolation: Toward an
integration. In P. Shaver (Ed.), Review of
personality and social psychology: Emotions,
relationships, and health (pp. 239-264). Beverly
Hills: Sage.
Rosencranz, H. A. & Pihlblad, C. T. (1970). Measuring
the health of the elderly. Journal of Gerontology.
25(2), 129-133.
112
Rossi, A. S., & Rossi, P. H. (1990). Of human bonding:
Parent-child relations across the life course. New
York: Aldine de Gruyter.
Salthouse, T. A., Kausler, D. H., and Saults, J. S.
(1990). Age, self-assessed health status, and
cognition. Journal of Gerontology: Psychological
Sciences. 4J5(4), P156-160.
Sarason, B. R., Pierce, G. R., & Sarason, I, G. (1990).
Social Support: The sense of acceptance and the
role of relationships. In B. R. Sarason, I. G.
Sarason and G. R. Pierce (Eds.)., Social Support:
An Interactional View. Somerset, N.J.: Wiley and
Sons.
Sarason, B. R., Shearin, E. N., Pierce, G. R., &
Sarason, I. G. (1987). Interrelations of social
support measures: Theoretical and practical
implications. Journal of Personality and Social
Psychology. H2(4), 813-832.
Sarason, I. G., Levine, H. M., Basham, R. B., and
Sarason, B. R. (1983). Assessing social support:
The social support questionnaire. Journal of
Personality and Social Psychology. 44(1), 127-139.
Sarason, I. G., & Sarason, B. R. (1985). Social
support: Insights from assessment and
experimentation. In I.G. Sarason & B.R. Sarason
(Eds.), Social support: Theory, research and
application (pp. 39-50). Dordrecht, The
Netherlands: Martinus Nijhoff Publishers.
Sarason, I. G., Sarason, B. R., & Pierce, G. R. (1990).
Social support: The search for theory. Journal of
Social and Clinical Psychology. 9(1), 133-147.
Sarason, I. G., Sarason, B. R., & Shearin, E. N.
(1986). Social support as an individual difference
variable: Its stability, origins, and relational
aspects. Journal of Personality and Social
Psychology. 50(4), 845-855.
Saur, W. J., & Coward, R. T. (1985). The role of social
support networks in the care of the elderly. In W.
J. Saur & R. T. Coward (Eds.), Social support
networks and the care of the elderly (pp. 3-20).
New York: Springer.
113
Schultz, R., & Tompkins, C. A. (1990). Life events and
changes in social relationships: Examples,
mechanisms, and measurement. Journal of Social and
Clinical Psychology. 9(10) 69-77.
Seigler, I. C., and Costa, P. T., Jr., (1985). Health
behavior relationships. In J. E. Birren & K. W.
Schaie (Eds.), Handbook of the psychology of aaina
(pp. 144-166, 2nd ed.). New York: Van Nostrand
Reinhold.
Shaver, P. R., & Hazan, C. (1988). A biased overview of
the study of love. Journal of Social and
Personal Relationships. 5, 473-501.
Sroufe, L. A., and Waters, E., (1977). Attachment as an
organizational construct. Child Development. 48,
1184-1199.
Tabachnik, B. G., & Fidell, L. S. (1983). Using
Multivariate Statistics (pp. 409-411). New York:
Harper & Row.
Tessler, R., & Mechanic, D. (1978). Psychological
distress and perceived health status. Journal of
Health and Social Behavior. 19, 254-262.
Tissue, T. (1972). Another look at self-rated health
among the elderly. Journal of Gerontology. 27, 91-
94.
Troll, L. & Smith, J. (1976). Attachment through the
life span: Some questions about dyadic bonds among
adults. Human Development 19(2), 156-170.
Ward, R. A., Sherman, S. R., & LaGory, M. (1984).
Subjective network assessments and subjective
well-being. Journal of Gerontology. 39(1), 93-101.
Waters, E. (1978). The reliability and stability of
individual differences in infant-mother
attachment. Child Development. 49, 483-494.
Weiss, R. S. (1982). Attachment in adult life. In C. M.
Parkes and J. Stevenson-Hinde (Eds.), The place of
attachment in human behavior (pp. 171-184). New
York: Basic Books.
114
Wethington, E., Kessler, R. C. (1986). Perceived
support, received support, and adjustment to
stressful life events, Journal of Health and
Social Behavior. 27, 78-89.
Appendix A
Questionnaire
Parent-Child Research
Thank you in advance for your participation in this
study.
DEMOGRAPHIC INFORMATION
Check: Ma 1 e_
1. When were you born? ■ /____/___ Female.
month day year
2. EDUCATION: Highest year of school completed
Highest degree earned
SOCIAL NETWORK INFORMATION - Please check the correct
response
3. MARITAL STATUS
year event occurred
Married...................................
Separated....................... . _____
. Divorced........................ _________
Widowed................... ._____
Living with Partner............ ...........
4. Have you ever been married before, not counting
your present marriage? Yes________No
How many t i m e s ? _________
How long did the longest marriage last?________ _
What was the reason it ended?_______ ____
5. Who lives with you? _____live alone
_____spouse/partner
_____child
_____roommate
______other__________ _
6. Do you have close friends or relatives that you see
on a regular basis?
Yes: How m a n y ? _______
___ No
7. How often do you attend events at various
organizations such as a senior center, or business
meetings?
Daily Weekly Monthly Other._______________
117
8. Do you have someone with whom you can talk about
private matters and who you can call on for help?
Yes: Who is this person
No
The following are general questions about your overall
health.
1. Are you currently taking
medications?
1. Arthritis medication
2. Pain Killers
3. Sleeping Pills
4. Allergy Pills
5. Blood Pressure Pills
6. Pills for Diabetes
7. Heart Pills
8. Insulin
9. Stomach Medication
10. Tranqui1i zers
11. Cortisone
any of the following
.12. Antibiotics
.13. Thyroid Pills
.14. Seizure Pills
15. Chest Pain Pills
(NITRO)
,16. Water Pills
.17. Laxatives
18. Blood Thinner
Medication
.19. Pills for
Breathing
20. Circulation Pills
21. OTHER
22. OTHER_____ ; ______
2. Which, if any, of the following conditions do you
currently have at this time?
1. Heart Condition
2. Circulation Problems
3. High Blood Pressure
4. Anemia
5. Diabetes
6. Emphysema/Bronchitis
7. Cataracts
. 8. Stomach Ulcers
. 9. Broken Bones
.10. Gall Bladder Problems
11. Hernia
12. Liver Disease
13. Kidney Disease
14. Urinary Problems
15. Parkinsons
Disease
16. Stroke
17. Arthritis
18. Emotional
Problems
19. Skin Problems
20. Cancer
21. OTHER , ________
118
The following questions are concerned with how you feel
about your physical health.
VERY POOR POOR AVERAGE VERY GOOD EXCELLENT
1 2 3 4 5
1 2 3 4 5 1. How do you rate your current health?
1 2 3 4 5 2. How would you rate your health as
compared to those who are your age?
1 2 3 4 5 3. How is your health compared to what
it was last year?
1 2 3 4 5 4. How is your health compared to what
is was five years ago?
5. How often do you visit the doctor?
Every Day Other_________
Once a week
Twice a month
Once a month
6. In your opinion, have you or anyone in your family
ever had a drug or alcohol problem?
____ Yes: Who___________________
_No
7. Do you smoke cigarettes? Yes: How many years?___
No
119
The next set of questions concern the amount of control
you feel over certain aspects of your life. Circle the
number that is closest to how you feel about the
question.
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5
4 5 1. Becoming a success is a matter of
hard work? Luck has little or
nothing to do with it.
4 5 2. What happens to me is my own
doing.
4 5 3. Most people don't realize how much
their lives are controlled by
accidental happenings.
4 5 4. I often feel that I have little
influence over the things that
happen to me.
120
Take a minute and think about the relationship between
you and your MOTHER while you were growing up. On a
scale of one to five, circle the answer that best fits
how well the statement reflects your relationship with
your MOTHER.
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5
1 2 3 4 5 1.
1 2 3 4 5 2.
1 2 3 4 5 3.
1 2 3 4 5 4.
1 2 3 4 5 5.
1 2 3 4 5 6.
1 2 3 4 5 7.
1 2 3 4 5 8.
1 2 3 4 5 9.
1 2 3 4 5 10.
1 2 3 4 5 11.
1 2 3 4 5 12.
1 2 3 4 5 13.
1 2 3 4 5 14.
1 2 3 4 5 15.
1 2 3 4 5 16.
She was fairly cold and distant.
She was rejecting.
She was not very responsive to my
needs
I wasn't her highest priority.
Her concerns were often somewhere
else.
It is possible she would just as
soon not have had me.
She was noticeably inconsistent in
her reactions to me.
She fluctuated in how warm she was
to me.
She had her own needs and agendas
which sometimes got in the way of
her receptiveness and
responsiveness to my needs.
She definitely loved me.
She didn't always show love in the
best way.
She was warm and responsive.
She was good as knowing when to
be responsive.
She was good at knowing when to let
me operate on my own.
Our relationship was almost always
comfortable.
I have no major reservations or
complaints about my relationship
with my mother.
121
Take a minute and think about the relationship between
you and your FATHER while you were growing up. On a
scale of one to five, circle the answer that best fits
how well the statement reflects your relationship with
your FATHER.
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5
1 2 3 4 5 1.
1 2 3 4 5 2.
1 2 3 4 5 3.
1 2 3 4 5 4.
1 2 3 4 5 5.
1 2 3 4 5 6.
1 2 3 4 5 7.
1 2 3 4 5 8.
1 2 3 4 5 9.
1 2 3 4 5 10.
1 2 3 4 5 11.
1 2 3 4 5 12.
1 2 3 4 5 13.
1 2 3 4 5 14.
1 2 3 4 5 15.
1 2 3 4 5 16.
He was fairly cold and distant.
He was rejecting.
He was not very responsive to my
needs
I wasn't his highest priority.
His concerns were often somewhere
else.
It is possible he would just as
soon not have had me.
He was noticeably inconsistent in
his reactions to me.
He fluctuated in how warm he was to
me.
He had her own needs and agendas
which sometimes got in the way of
his receptiveness and
responsiveness to my needs.
He definitely loved me.
He didn't always show love in the
best way.
He was warm and responsive.
He was good as knowing when to
be responsive.
He was good at knowing when to let
me operate on my own.
Our relationship was almost always
comfortable.
I have no major reservations or
complaints about my relationship
with my father.
122
This research is interested in finding out about the
relationship you have with one of your children.
Because of the complexities of the parent-child
relationship, questions are limited to only one child.
If you have more than one child, choose one to be the
"study child" for this questionnaire.
First, we would like to know how many children you
have. Please list your children from oldest to
youngest.
First Name Year of
Birth
Oldest ____________: __________ _________
2nd_____ . . ______
3rd _____ ■
4th .
5th__________________ . ____________
6th_______________________ . _____ .
Who is your designated "Study"
C h i l d ? ___________ ___
1. How many miles does this child live from you? __
2. How often are you in contact with this child?___
3. How do you communicate most frequently with this
child?
Telephone Writing visiting other___
4. Which one of you is usually the first to make
contact?
Them Me ___ About Equal
123
For each of the following statements circle the
response that best describes how you feel about your
communication with this child
STRONGLY
DISAGREE
1
1 2 3
1 2 3
1 2 3
1 2 3
DISAGREE NEUTRAL AGREE STRONGLY
AGREE
5 1. The communication between this
child and myself is very good
5 2. I am very satisfied with the
communication between this child
and myself
5 3. There is very little
misunderstanding between this
child and myself
5 4. There is very little conflict,
tension, or disagreement between
this child and myself
Take a minute and think about your relationship with
your child. On a scale of one to five, how much do you
agree with the following statements?
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5 1.
1 2 3 4 5 2.
1 2 3 4 5 3.
1 2 3 4 5 4.
1 2 3 4 5 5.
1 2 3 4 5 6.
1 2 3 4 5 7.
1 2 3 4 5 8.
1 2 3 4 5 9.
1 2 3 4 5 10.
1 2 3 4 5 11.
1 2 3 4 5 12.
1 2 3 4 5 13 .
1 2 3 4 5 14.
1 2 3 4 5 15.
1 2 3 4 5 16.
1 2 3 4 5 17.
1 2 3 4 5 18.
I find it difficult to allow myself
to depend on this child.
This child is never there when I
need them.
I am comfortable depending on this
child.
I know my child will be there when
I need him or her.
I find it difficult to trust my
child completely.
I'm not sure I can always depend on
my child to be there when I need
him or her
I often worry about being
abandoned.
I often worry that my child does
not really love me.
I find my child is reluctant to get
as close as I would like.
I often worry my child will not
want to take care of me.
I want to bond completely with my
child.
My desire to bond sometimes scares
people away.
I find it relatively easy to get
close to my child.
I often worry about someone getting
too close to me.
I am somewhat uncomfortable being
close to my child.
I am nervous when anyone gets too
close, relatively easy to
I am comfortable having my child
depend on me
Often, my child wants to provide
more help than I feel comfortable
receiving.
125
The following questions are interested in how
responsive you feel this child is to your needs.
Circle the number next to the statement that best
describe how you feel.
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5
1 2 3 4 5 1.
1 2 3 4 5 2.
1 2 3 4 5 3.
1 2 3 4 5 4.
1 2 3 4 5 5.
1 2 3 4 5 6.
1 2 3 4 5 7.
1 2 3 4 5 8.
1 2 3 4 5 9.
1 2 3 4 5 10.
1 2 3 4 5 11.
1 2 3 4 5 12.
1 2 3 4 5 13.
1 2 3 4 5 15.
1 2 3 4 5 16.
needs
I am not his/her highest priority.
His/Her concerns were often
somewhere else.
It is possible he/she would just as
soon not have to deal with me.
He/She is noticeably inconsistent
in his/her reactions to me.
He/She fluctuates in how warm
he/she is to me.
He/She has his/her own needs and
agendas which sometimes get in the
way of his/her receptiveness and
responsiveness to my needs.
He/She definitely loves me.
He/She doesn't always show love in
the best way.
He/She was warm and responsive.
He/She was good as knowing when to
be responsive and when to let me
operate on my own.
Our relationship was almost always
comfortable.
I have no major reservations or
complaints about my relationship
with this child.
126
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5
1 2 3 4 5 1. My child gives me the moral
support I need
1 2 3 4 5 2.1 get good ideas about how to do
things or make things from my
child
When I confide in this child I get
the idea that it makes him/her
uncomf ortable
My child enjoys hearing what I
think
My child shares many of my
interests
My child comes to me when he/she
has problems or needs advice
I rely on this child for emotional
support
I can go to this child if I were
just feeling down, without
feeling funny about it later
My child and I are very open about
what we think about things
My child is sensitive to my
personal needs
This child comes to me for
emotional support
This child is good at helping and
solving problems
I have a deep sharing relationship
with this child
This child gets good ideas about
how to do things or make things
from me
When I confide in this child it
makes me uncomfortable
This child seeks me our for
companionship
I think that this child feels
that I'm good at helping him/her
solve problems
I don't have a relationship with
my child that is as close as other
people's relationships with their
children
1 2 3 4 5 3.
1 2 3 4 5 4.
1 2 3 4 5 5.
1 2 3 4 5 6.
1 2 3 4 5 7.
1 2 3 4 5 8.
1 2 3 4 5 9.
1 2 3 4 5 10.
1 2 3 4 5 11.
1 2 3 4 5 12.
1 2 3 4 5 13.
1 2 3 4 5 14.
1 2 3 4 5 15.
1 2 3 4 5 16.
1 2 3 4 5 17.
1 2 3 4 5 18.
127
STRONGLY DISAGREE NEUTRAL AGREE STRONGLY
DISAGREE AGREE
1 2 3 4 5 19.
1 2 3 4 5 20.
1 2 3 4 5 21.
1 2 3 4 5 22.
1 2 3 4 5 23.
1 2 3 4 5 24.
I wish this relationship was
different
Taking everything into
consideration, I am very satisfied
with the amount of contact I have
with this child
I often feel like I give this
child more support than they can
or will give back
I often feel that this child makes
too many demands on me
The only time this child ever
calls is when he or she needs
something from me
I often feel that my family or
friends want too much from me
25. As people age, relationships between parents and
children often change. In the past few years
would you say that your relationship with this
child has
.Declined considerably
.Declined
.Stayed the same
.Improved
.Improved considerably
THANK YOU SO MUCH FOR PARTICIPATION IN THIS STUDY
If you wish to receive a copy of the results of this
study (sometime in the Fall of 1991), have the
researcher add you to the mailing list, or you may
write her at the numbers on the consent form.
THANKS A BUNCH!!!
128
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
AIDS preventive sexual behavior in college students: An empirical test of the health belief model
PDF
Factors influencing acceptance of telecommunications innovations: A study of vocational rehabilitation practitioners
PDF
The alcoholic/coalcoholic marriage: A communication rules approach defining relational coherence and control
PDF
Evidence of age-based rationing of health care to the elderly: The effect of age and death on the use of health care services by elderly enrollees in an HMO
PDF
The role of the clear and present danger doctrine in prejudicial publicity cases before the United States Supreme Court: 1941-1976
PDF
Statistical analysis of health care expenditure and health status in the well-elderly study
PDF
Intercultural communication: A testbench of the theory of activity
PDF
Vowel comparison of normal, deaf intelligible and deaf unintelligible speakers
PDF
Service leadership: Feedback and action in the organization
PDF
James Henley Thornwell and the theological justification of slavery: A study in the development of a proslavery ideology
PDF
Gender, blame, and dependence as determinants of relational dissolution and disengagement communication tactics
PDF
Physicians' use of compliance-gaining message strategies: Their effect upon patient medical communication satisfaction and patient adherence to treatment regimen instructions
PDF
Recovering the discourse: Genres of sexual harassment in organizations
PDF
Why didn't you use a condom? On the coherence of gay men's explanations for lapse behavior
PDF
Cinematic concepts and techniques in the adaptation and staging of chamber theatre
PDF
The theory and application of presumption in public debate
PDF
Effects of echoic, short-term, and long-term memory on validity and reliability of self-identification of authentic and simulated stuttering
PDF
Red waves of grain: An analysis of radical farm movement rhetoric in Montana, 1918-1937
PDF
The relationship between perceived communication competence and choice of interpersonal conflict message strategies
PDF
Bringing social worlds together: Information systems as catalysts for new interactions in health care organizations
Asset Metadata
Creator
Dickson, Risa E
(author)
Core Title
Mental models and attachment systems: Predictions of health and relational outcomes of the elderly
Degree
Doctor of Philosophy
Degree Program
Communication Arts and Sciences
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Gerontology,Health Sciences, Aging,health sciences, public health,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c17-732721
Unique identifier
UC11344167
Identifier
DP22467.pdf (filename),usctheses-c17-732721 (legacy record id)
Legacy Identifier
DP22467.pdf
Dmrecord
732721
Document Type
Dissertation
Rights
Dickson, Risa E.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
Health Sciences, Aging
health sciences, public health