University of Southern California Dissertations and Theses

Family relationships and their influence on health outcomes over time among older adults in rural China

(USC Thesis Other)

Family relationships and their influence on health outcomes over time among older adults in rural China

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1

Family Relationships and Their Influence on Health Outcomes over Time among Older Adults in
Rural China

by

Weiyu Mao

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(SOCIAL WORK)

Committee Members
Iris Chi, DSW
William Vega, PhD
Chih-Ping Chou, PhD
Merril Silverstein, PhD

August 2015
2
Table of Contents
Dedication ........................................................................................................................................ 5

Acknowledgments ........................................................................................................................... 6

List of Tables ................................................................................................................................... 8

List of Figures ................................................................................................................................ 10

Abstract .......................................................................................................................................... 11

Chapter 1: Introduction and Theoretical Framework .................................................................... 13

Introduction ............................................................................................................................. 13

Family Relationships and Health ............................................................................................. 16

Theoretical Framework ........................................................................................................... 19

Social Support Theory ....................................................................................................... 19

Social Exchange Theory .................................................................................................... 21

Intergenerational Solidarity Paradigm ............................................................................... 22

Significance of the Study ......................................................................................................... 23

Specific Aims .......................................................................................................................... 24

Chapter 2: A Gendered Perspective of Intergenerational Relationships and Functional
Limitations: Four-Wave Autoregressive and Cross-Lagged Panel Analysis ................................ 26

Introduction ............................................................................................................................. 26

Methods ................................................................................................................................... 29

Sample ............................................................................................................................... 29

Measures ............................................................................................................................ 30

Analyses ............................................................................................................................ 31

Results ..................................................................................................................................... 32

3
Discussion ................................................................................................................................ 34

Chapter 3: Multidimensional Intergenerational Instrumental Support and Self-Rated Health:
Trajectories and Correlated Change over 11 Years ....................................................................... 37

Introduction ............................................................................................................................. 37

Instrumental Support and Health in the Chinese Context ................................................. 39

The Present Study .............................................................................................................. 41

Methods ................................................................................................................................... 42

Sample ............................................................................................................................... 42

Measures ............................................................................................................................ 43

Analyses ............................................................................................................................ 45

Results ..................................................................................................................................... 47

Discussion ................................................................................................................................ 51

Chapter 4: Dynamic Interplay between Intergenerational Instrumental Support and Self-Rated
Health among Older Adults: An Application of Bivariate Latent Change Score Modeling ......... 55

Introduction ............................................................................................................................. 55

Methods ................................................................................................................................... 57

Sample ............................................................................................................................... 57

Measures ............................................................................................................................ 57

Analyses ............................................................................................................................ 58

Results ..................................................................................................................................... 59

Discussion ................................................................................................................................ 62

Chapter 5: Conclusions and Implications ...................................................................................... 65

Summary of Major Research Findings .................................................................................... 66

4
Implications for Future Research ............................................................................................ 70

Implications for Policies and Social Work Practice ................................................................ 72

References ..................................................................................................................................... 76

Tables ............................................................................................................................................ 94

Figures ......................................................................................................................................... 106

5
Dedication
I dedicate this dissertation to my family, my mother, Xiuzhi Xie, my father, Xuecai Mao,
and my brother, Weinan Xie. I love you. You have provided me with encouragement, support,
and love. You have helped me stay focused and believe in myself. You have always said that I
can do it. You have encouraged me to pursue my dreams and hopes. You have made scarifies to
support me. Words are not enough to express my deep appreciation for your unshakable belief in
me and your tremendous support throughout the years.
I would also like to dedicate this dissertation to my maternal grandmother, Yun Zhang,
my maternal grandfather, Qi Xie, and my paternal grandfather, Fuyun Mao. They inspired me to
write my dissertation on this topic. Grandmother, you have been so strong, resilient, and caring.
Grandfathers, you were so kind and giving. It was difficult to lose you to health issues. I remind
myself frequently that I can never lose the time I enjoyed with you. I sincerely hope my research
will benefit other older adults and their families.

6
Acknowledgments
This has been a journey for me in which I learned that there are good moments and there
are bad moments in life and I am very fortunate to have so many people by my side who believe
in me and cheer me up. With support from many people, writing this dissertation has been a
memorable process during which I not only grew as a scholar but also as a person.
I would like to start by giving thanks and appreciation from the bottom of my heart to my
mentor and dissertation committee chair, Dr. Iris Chi. It would not be possible for me to finish
my dissertation without her guidance, inspiration, and patience. Her dedication, vision, and
enthusiasm in gerontology research has encouraged and inspired me to pursue a career in this
field. It has been a privilege and honor to work with her during my years as a doctoral student.
She has helped me in many ways that I could never have imagined throughout the years. I am so
grateful and fortunate that she has always been there for me throughout this process and I know I
can always count on her for many years to come. She helped me adapt and adjust to an
unfamiliar environment, pushed me to achieve and maintain high academic standards, and
provided me with candid and invaluable feedback and enormous support during my search for a
scholarly identity. She is more than a mentor to me; she has nurtured me like a mother and
provided a safe harbor for me to grow as a person. For that, I am forever grateful and thankful.
I want to thank my dissertation committee members, Dr. William Vega, Dr. Chih-Ping
Chou, and Dr. Merril Silverstein for their understanding, patience, and support. Throughout this
process, they asked intriguing questions, helped me capitalize on their expertise, and provided
critical feedback and insights. I also want to acknowledge that Dr. Silverstein sponsored my visit
to Xi’an Jiaotong University to present a portion of my dissertation and partake in the planning
for the upcoming wave of the Anhui study (my dissertation used longitudinal data generated
7
from this study). I would like to thank Dr. Shuzhuo Li from Xi’an Jiaotong University for the
help and support I received from him and his team. I also want to thank Dr. Lawrence Palinkas,
Dr. Julie Cederbaum, Dr. Eric Rice, and Dr. Lei Duan for their support throughout the course of
my doctoral studies.
I would like to thank my parents and my brother. They have been my biggest supporters
and fans. They encourage me to pursue my own dreams and hopes. They encourage me to be
strong when I feel weak. They believe in me when I struggle. They nurture me to become the
person I am today. I am indebted to them for everything they have done for me. I also would like
to thank my grandparents. My interactions, observations, and memories with them drove me to
ponder how family could potentially improve health in the later years of life and inspired me to
pursue my current research agenda.
I would like to give thanks to my friends Ling Xu, Man Guo, Jinyu Liu, Dongmei Zuo,
Yawen Li, Hsin-Yi Hsiao, Hsun-Ta Hsu, Hyunsung Oh, Yura Lee, and Mee-Young Um for their
help and support. I would like to give special thanks to my cohort friends, Anamika Barman-
Adhikari, Armando Barragan, Diana Ray-Letourneau, Mercedes Hernandez, and Caroline Lim,
for their help and support. We shared happy moments and sad moments. With all their friendship
and support, the past few years have been unforgettable.
I want to thank all the people who have helped and supported me during my years as a
doctoral student. My special thanks to the doctoral program of School of Social Work,
University of Southern California for the generous support and help throughout the years. I also
want to give special thanks to Malinda Sampson, who has been instrumental at every step of the
way, and Eric Lindberg, who provided me with generous support in editing my manuscripts and
dissertation.
8
List of Tables
Table 2.1. Characteristics of Older Adults with at least One Son and One Daughter at Baseline
(N = 1,322)
Table 2.2. Means and Standard Deviations of Gendered Intergenerational Support Receipt and
Functional Limitations among Older Adults from 2001 to 2009
Table 2.3. Autoregressive and Cross-Lagged Panel Analysis for Gendered Intergenerational
Support and Functional Limitations among Older Adults: Unstandardized Parameter
Estimates with Adjustments of Control Variables
Table 3.1. Characteristics of Older Adults with at least One Child at Baseline (N = 1,636)
Table 3.2. Means and Standard Deviations of Multidimensional Intergenerational Instrumental
Support and Self-Rated Health from 2001 to 2012
Table 3.3. Bivariate Latent Growth Curve Model for Receipt of Household Chore Help and Self-
Rated Health: Unstandardized Parameter Estimates with and without Adjustments for
Control Variables
Table 3.4. Bivariate Latent Growth Curve Model for Receipt of Personal Care Help and Self-
Rated Health: Unstandardized Parameter Estimates with and without Adjustments for
Control Variables
Table 3.5. Bivariate Latent Growth Curve Model for Provision of Household Chore Help and
Self-Rated Health: Unstandardized Parameter Estimates with and without Adjustments
for Control Variables
Table 3.6. Bivariate Latent Growth Curve Model for Provision of Personal Care Help and Self-
Rated Health: Unstandardized Parameter Estimates with and without Adjustments for
Control Variables
9
Table 4.1. Bivariate Latent Change Score Model Fit Indices for Receipt of Instrumental Support
and Self-Rated Health from 2001 to 2012
Table 4.2. Bivariate Latent Change Score Model for Receipt of Household Chore Help and Self-
Rated Health: Unstandardized Parameter Estimates
Table 4.3. Bivariate Latent Change Score Model for Receipt of Personal Care Help and Self-
Rated Health: Unstandardized Parameter Estimates
10
List of Figures
Figure 2.1. Path Diagram of an Autoregressive and Cross-Lagged Panel Analysis for Gendered
Intergenerational Support and Functional Limitations among Older Adults at Four Time
Points from 2001 to 2009
Figure 3.1. Diagram of a Parallel Process Latent Growth Curve Model for Intergenerational
Instrumental Support and Self-Rated Health at Five Occasions over 11 Years
Figure 3.2. Quadratic Latent Growth Curve Model Estimates for Individual Growth Curves of
Provision of Household Chore Help over Time (N = 50)
Figure 3.3. Quadratic Latent Growth Curve Model Estimates for Individual Growth Curves of
Provision of Personal Care Help over Time (N = 50)
Figure 4.1. Diagram of a Bivariate Latent Change Score Model for Intergenerational
Instrumental Support and Self-Rated Health at Five Occasions over 11 Years
11
Abstract
Most individuals will face developing morbidity, declining health, and increasing
mortality risk as they grow old. Family relationships, as a central and primary experience of
human relationships, are multidimensional and dynamic in nature and each domain tends to have
unique characteristics and differential consequences on health outcomes. Theoretical
perspectives such as social support theory, social exchange theory, and intergenerational
solidarity paradigm shed light on the linkages between structural and functional aspects of family
relationships and health. Family relationships and health are not static but changing over time.
The dynamic relationships between family relationships and health warrant scholarly attention.
Cultural context helps shape how individuals interact and form relationships. Culturally sensitive
patterns of structure, interaction, and exchange in family relationships need further investigation.
Rural China provides a unique context for studying family relationships and health among older
adults over time, because it is shaped by a distinct cultural, economic, and political environment.
This dissertation focused on intergenerational support among older parents in rural China,
investigated culturally distinctive patterns of support provided by children, and examined the
dynamic relationship between multidimensional intergenerational support and multiple health
outcomes including functional limitations and self-rated health over time. This dissertation
featured three studies that addressed different aspects of the relationship between
intergenerational support and health among older adults in rural China using different advanced
analytic approaches: (a) culturally informed gendered intergenerational support and functional
limitations using four-wave autoregressive and cross-lagged panel analysis; (b) trajectories of
multidimensional intergenerational instrumental support and self-rated health and their correlated
change over 11 years using parallel process latent growth curve modeling; and (c) changes in
12
intergenerational instrumental support and self-rated health among older adults and the direction
of changes over a period of 11 years using bivariate latent change score modeling.
The findings of this dissertation show the pattern of effect that gendered intergenerational
support exerts on functional limitations among older adults over time, reveal the developmental
nature of multidimensional intergenerational support, demonstrate how the processes of different
dimensions of intergenerational support are correlated with the process of perceptions of health
in older adults as they unfold over time, reveal the direction of two processes of intergenerational
support and health, and indicate how change in intergenerational support predicts change in
health over time.
This dissertation contributes to existing knowledge on support and health in the following
ways. This dissertation used a multidimensional and nuanced approach for the conceptualization
and measurement of family relationships and emphasized the different dimensions of
intergenerational support. It examined the dynamics between different aspects of family
relationships and late-life health over time and focused on questions that addressed
developmental nature of intergenerational support and health, patterns of change between those
two constructs, and the direction in which one process influenced the other. This dissertation
used quality longitudinal regional data collected using multistratified random sampling and
advanced analytic approaches to specify different longitudinal structural equation models. This
dissertation investigated culturally distinctive features of family relationships in a specific
context and focused on the adult children and older parent relationship in a cultural context
featuring filial piety; the gendered nature of support provision among adult children in particular
was examined in one study and the importance of support and care exchange between adult
children and older parents was examined in the other two studies.
13
Chapter 1: Introduction and Theoretical Framework
Introduction
Most individuals will face developing morbidity, declining health, and increasing
mortality risk as they grow old. Health is a salient topic, especially among older adults, who tend
to shoulder a heavier disease and disability burden compared to other populations (McLaughlin,
Connell, Heeringa, Li, & Roberts, 2010). Avoidance of disease and disability and maintenance of
high physical functioning are considered integral components of successful aging (Rowe &
Kahn, 1997). Self-rated health and functional limitations have been key themes in gerontological
research (e.g., Antonucci, Ajrouch, & Birditt, 2014) because they represent various aspects of
health and have differential influences on individuals in later life.
Social relationships help shape health (Braveman, Egerter, & Williams, 2011). Among
social determinants of health, social ties have a profound impact throughout the human lifespan
(Marmot, 2005; Umberson & Montez, 2010). Social relationships have been associated with
important health outcomes such as morbidity, mortality, recovery from illness, and the ability to
cope with stress (Sarason & Sarason, 2009). There is also accumulative evidence of the health-
promoting effects of social relationships among older adults (Fratiglioni, Paillard-Borg, &
Winblad, 2004). Quantity and quality aspects of social relationships have been shown to be
associated with all-cause mortality in the general population and the older adult population in
particular (Barger, 2013). The importance of staying socially connected has been emphasized in
studies exploring the protective effects of social ties on well-being among older adults (Carr &
Springer, 2010).
There is increasing evidence of how different social factors, including the structure of
networks, support from other individuals, and the quality and quantity of interactions, influence
14
health via unique mechanisms (Cohen, 2004), but how those relationships sustain or improve
health remains equivocal (Thoits, 2011). Family relationships, a central and primary experience
of human relationships, have been acknowledged in the study of social relations (Antonucci et
al., 2014; Carr & Springer, 2010). However, family relationships as a subset of social
relationships have been far less investigated with respect to health, especially among older adults
(e.g., Ryan & Willits, 2007). Family and health research has focused on families, child health,
marriage, and adult health (Carr & Springer, 2010) while largely ignoring other family
relationships and older parent health. Another relevant area of research is family caregiving
among frail older adults. Families have been emphasized in conjunction with caregiving for older
adults with physical or mental caregiving needs or both, yet existing research has generally
focused on caregiver stress and the needs of care recipients (Pinquart & Sörensen, 2005), rather
than health outcomes of older care recipients.
The effect of family on health among community-dwelling older adults has been
relatively understudied compared to its influence in a broader context such as social
relationships. How different dimensions of family relationships influence health remains unclear,
and further research on family relationships using a multidimensional approach is warranted
(e.g., Guo, Chi, & Silverstein, 2012). For instance, how the provision of support influences
health among older adults has been less examined compared to the effects of receiving support,
despite research indicating that providing support could be promising for health and well-being
in later life (Brown, Nesse, Vinokur, & Smith, 2003; Krause, Herzog, & Baker, 1992). In
addition, little is known about the pathway through which family relationships influence health.
Research with a focus on how family relationships influence health over time has also been
lacking compared to cross-sectional studies (e.g., Thoits, 2011; Umberson & Montez, 2010).
15
Family relationships have also been shown to have important and varying influences
across diverse contexts. The contextual background in which relationships form and interactions
take place has been shown to affect how relationships influence health in general (e.g.,
Zunzunegui et al., 2004). Most studies conducted in North America (mostly among Caucasians)
have shown that family relationships have little or negative influence on well-being among older
adults. On the other hand, family relationships have been shown to be beneficial among minority
and disadvantaged populations in North America and other countries (e.g., Antonucci et al.,
2014; S.-T. Cheng, Lee, Chan, Leung, & Lee, 2009; Chi & Chou, 2001; Fiori, Antonucci, &
Akiyama, 2008; Stephens, Alpass, Towers, and Stevenson, 2011; Zunzunegui et al., 2004).
Family elder care is equally important, particularly for ethnic minority groups given that the vast
majority of elder care in North America is provided by the family regardless of ethnicity (e.g.,
Kosberg, Kaufman, Burgio, Leeper, & Sun, 2007).
Culturally sensitive patterns of structure, interaction, and exchange in family
relationships have been less empirically examined in China compared to other countries (S.-T.
Cheng et al., 2009). In addition, scant research has been devoted to family life and the health of
older adults in rural areas compared to research focused on urban populations (e.g., Silverstein,
Cong, & Li, 2006). China, which is characterized by an emphasis on filial piety, provides a
unique context in which to explore this topic. Family relationships are central to the well-being
and health of older Chinese adults (Chi & Chou, 2001; Chou, 2010; Y. Li & Chi, 2011). In
essence, family is often the sole support system for older adults in rural China given the general
lack of public assistance programs and pension systems (e.g., Dong & Simon, 2010).
This dissertation investigates the role of families in reducing health disparities in later life
and specifically examines the influence of family relationships on multiple health outcomes over
16
time among older adults in rural China. This dissertation features three overarching research
questions: How do family relationships influence health outcomes over time among older adults
in rural China? How do different dimensions of family relationships exert influence on health
outcomes over time among older adults in rural China? What are the pathways through which
family relationships influence health outcomes over time among older adults in rural China?
Family Relationships and Health
Families contribute the most central and primary experience that individuals have in
terms of social relationships (Franks, Campbell, & Shields, 1992). Research on family
relationships and adult health has long been of scholarly interest and has been traced to French
sociologist Émile Durkheim’s 1897 book, Suicide (Carr & Springer, 2010). Studies that
improved understanding of the context of families in the middle and later years of life, primarily
in Western societies, outlined multidimensional family support both in a variety of forms and
from different sources; the gendered nature of family relations with the salient relationship
between mothers and daughters; complicated parent–child relationships following transitions of
children such as marriage (normative) and cohabitation (nonnormative); and significant marital
transitions, typically widowhood (Allen, Blieszner, & Roberto, 2000). There are different types
of specific familial relationships, including intergenerational relationships, sibling relationships,
and in-law relationships (C. L. Johnson, 2000). In this dissertation, intergenerational
relationships between parents and adult children are of particular interest.
Pathways through which family relationships influence health can be extrapolated from
general research on social relationships and health, due to limited research with a sole focus on
the effects of family relationships on health. Structural aspects of social relationships, i.e., social
integration, are believed to directly influence health outcomes. Individuals involved in social
17
networks tend to experience social controls and peer pressure that could contribute to normative
health behaviors (Cohen, 2004). The most striking evidence in this arena is emerging research
linking relationships and mortality risk during the past two decades. Previous studies have shown
that community-dwelling adults who are married and have close family ties are more likely to
survive compared to their more isolated counterparts during an extended period of time
(Berkman & Glass, 2000). Current empirical research has demonstrated that relationships exert
an independent influence on mortality risk, which is comparable with other well-documented
risk factors for mortality (Holt-Lunstad, Smith, & Layton, 2010).
In addition, social support, both received and provided, has been shown to have direct
effects on health, including self-rated health and functional limitations (e.g., Everard, Lach,
Fisher, & Baum, 2000; Liu, Liang, & Gu, 1995). Receiving social support, such as tangible and
informational support, typically increases as people age, whereas providing support declines over
time (Shaw, Krause, Liang, & Bennett, 2007). Embedded in the literature on social support and
health, family support has been correlated with health-related outcomes such as better
physiological processes, including reliable effects on blood pressure regulation specifically and
cardiovascular regulation in general, across correlational, prospective, and laboratory studies
(Uchino, Cacioppo, & Kiecolt-Glaser, 1996). However, few studies have examined changes in
health over time and the effects of psychosocial determinants of health during the course of
change (Bailis, Segall, & Chipperfield, 2003; Wolinsky et al., 2008). The few studies that
investigated the relationship between social support and health have shown some protective
effects over time (e.g., Bailis et al., 2003). Nevertheless, those studies used social support as a
general term and did not examine the differential effects of various dimensions of social support
on health outcomes.
18
In terms of the effects of family support on health, existing research has demonstrated
cultural differences. Studies conducted in the Western context among ethnic majorities such as
Caucasians have tended to emphasize the importance of friends instead of families, whereas
among ethnic minorities and in other countries, especially East Asian countries with a collectivist
culture, family plays a central role in promoting health and well-being in later life (e.g.,
Antonucci et al., 2014; S.-T. Cheng et al., 2009; Chi & Chou, 2001; Fiori et al., 2008; Stephens
et al., 2011; Zunzunegui et al., 2004). Dominant generational bonds between parents and adult
children in the family system have long been found in societies featuring Confucianism (C. L.
Johnson, 2000). A gendered perspective of support transfers across generations has shown that
sons are expected to shoulder the primary responsibility for parental well-being in the Chinese
context (Lin et al., 2003). This pattern is unique when compared to the commonly found mother–
daughter bond in the Western context; that is, mothers have more frequent contact with
daughters and tend to live near each other (e.g., C. L. Johnson, 2000).
The term relationships has been used loosely and typically without strict definition.
Relationships could refer to connections among individuals and could encompass different
aspects such as the relative absence of relationships, overall involvement level in relationships,
functional aspects of relationships, and structural features of relationships (e.g., Umberson &
Montez, 2010). The term family relationships in this dissertation is used to describe functional
(support) and structural (network) aspects of connections (based on Berkman, Glass, Brissette, &
Seeman, 2000) in the context of families in later life.
Family networks include both structural and interactional components of the relational
ties that surround individuals; for instance, number of children. Family support may include
instrumental and emotional support. Because support exchange between adult children and older
19
parents could occur in long-term and short-term settings, the concept of reciprocity is used
(Leopold & Raab, 2011). Instrumental support involves help with daily tasks such as personal
care and household chores, both received and provided; emotional support includes the cohesive
aspect of relationships such as feeling close, having a confidant, and getting along (based on
Berkman et al., 2000).
Theoretical Framework
Three theoretical perspectives, namely social support theory, social exchange theory, and
intergenerational solidarity paradigm, served as the general guide for the studies in this
dissertation. Those theoretical perspectives provide a general understanding of why and how
familial relationships across generations influence one another and health during the life course.
Social Support Theory
Since the classic works by John Cassel (1976), Gerald Caplan (1974), and Sidney Cobb
(1976), social support has been viewed as having main and indirect effects on health and well-
being; more importantly, support provided by primary groups is considered to be most important
to an individual and serves an important protective function regarding health (Vaux, 1988). In
response to the early atheoretical approaches of studying support, the convoy model of support
has been suggested as a global theoretical framework of social relationships over time, especially
among older adults (Antonucci & Akiyama, 1987). The convoy model incorporates both a life
span and a multigenerational perspective and considers people as having a dynamic network of
close ties with family and friends; concentric circles are used to describe relationships around an
older adult with the strongest relationships closest to the older individual and weaker
relationships outside (Antonucci, Akiyama, & Takahasi, 2004). Those interpersonal relationships
20
form a convoy that shapes and protects individuals, travels with individuals throughout the life
course, and exchanges social support and assistance (Novak, 2006).
The convoy model also identifies personal and situational characteristics that help define
and shape the convoy of relations including social networks, social support, and satisfaction with
support (Antonucci, Birditt, & Ajrouch, 2011). With the increasing interest in and need to better
understand social relations within and across cultural contexts, the convoy model is also useful in
delineating culturally specific characteristics of social relations. Specifically, the convoy of
social relations may form based on different inclusion criteria in different cultural contexts, such
as family-focused versus family- and friend-focused convoys, and this model emphasizes the
role and importance of culture in specifying situational characteristics in social interactions and
exchanges (Antonucci et al., 2011).
The convoy model has helped move the field of social relationships forward and provided
invaluable insights in gerontological research. This model conceptualizes social relations as
multidimensional, uses specific measures of structural or compositional characteristics, regards
social relations as crucial at any given time point and with additional significance over time,
recognizes the need to understand social relations from multiple perspectives, and contextualizes
the nature of social relations (Antonucci et al., 2014). The convoy model also recognizes the
importance and need to examine changes in the structure and function of family relationships
over time. By contextualizing close family relationships in a specific cultural context, a greater
understanding of how relationships between adult children and aging parents influence multiple
health outcomes over time can be achieved.
21
Social Exchange Theory
Social exchange theory involves the benefits and costs of social interaction. The
characteristics of social interactions result from what individuals value and need in their social
life. During the process of social exchange, people provide valued resources to others and
depend on one another for such valuable resources (Molm & Cook, 1995). George Homans,
Peter Blau and Richard Emerson are three of the most influential scholars in the development of
social exchange theory (Ritzer & Goodman, 2004). Several key principles are of particular
interest. The norm of reciprocity mandates meeting at least two requirements in the value
system: People should help other people who have helped them and people should not harm
others who have helped them. The norm of reciprocity has been internalized into the value
system of many individuals; the norm urges people who have received a benefit to repay it
someday and assures individuals who have provided the benefit that they will be repaid at some
time. This normative principle of reciprocity facilitates the formation of social relations and
exchange (Ritzer & Goodman, 2004). During the exchange process, resources can refer to
anything exchanged in an interpersonal situation. They can be either concrete or symbolic and
can be grouped into six classes: love, status, information, money, goods, and services. Emotions
have also been incorporated in the process of social exchange. Actors engaged in social
exchange have emotions, try to understand the source of those emotions, and respond with
emotions during the exchange process. Emotions are involuntarily internal responses and affect
group solidarity (Ritzer & Goodman, 2004).
From the social exchange theory perspective, actors engage in social behaviors that are
rewarding and relationships between actors develop and persist over time. Individuals who are
involved in the exchange process strive to provide resources of value to others. Some individuals
22
may have advantages, with many resources and favorable exchange conditions, whereas others
may have little to offer and have little choice but to engage in unfavorable exchanges. Social
relationships require reciprocity over time, and they develop and persist in an ecological context.
The benefits derived from social relationships are diverse and may be of personal and situational
value (Vaux, 1988).
Social exchange theory has provided a useful theoretical framework for understanding
direct or delayed transfers of resources across generations over time. In response to changing
family structures and evolving intergenerational patterns of resource allocation, social forces
such as reciprocity and altruism influence the exchange process between intergenerational
linkages, thus influencing the viability of family support systems for older adults (Silverstein &
Giarrusso, 2010).
Intergenerational Solidarity Paradigm
The intergenerational solidarity paradigm specifies the behaviors and sentiments that
keep generations connected, typically the relationships between children and parents, and
considers families unique in their commitment to intergenerational transfers and
intergenerational solidarity (Bengtson, Olander, & Haddad, 1976). The intergenerational
solidarity paradigm has guided research in family studies for decades and describes various
forms of family support across generations. The classification of family relationships derived
from the intergenerational solidarity model is in accordance with social support theory,
especially in terms of structural (extent of interaction), functional (support exchange), and
affectual (closeness) dimensions of solidarity (Bengtson, Giarrusso, Mabry, & Silverstein, 2002).
This paradigm provides a useful lens to understand various aspects of intergenerational
23
relationships and the importance of providing support from younger generations to older
generations to maintain solidarity within families.
Significance of the Study
This dissertation contributes to existing knowledge in the following ways. It helps unfold
the complex associations between family relationships and multiple health outcomes by
examining functional aspect of family relationships controlling for structural aspect of family
relationships. Prior research has demonstrated that greater social integration and support is
correlated with multiple health outcomes such as better self-rated health, improved life
expectancy, and lower morbidity and mortality (Antonucci, Fuhrer, & Dartigues, 1997; Holt-
Lunstad et al., 2010; Uchino, 2004). However, less is known regarding how or why this
association exists due to lack of specificity in measurement and the related contextual
environment (Antonucci, Birditt, & Webster, 2010). For instance, the association between
relationships and mortality was strongest when complex measurements were used (Holt-Lunstad
et al., 2010).
This dissertation contributes to the process of knowledge building by examining how
family relationships influence health. Based on existing research, one key question regarding the
mechanism through which family relationships influence health remains unanswered. The
differential effects of each dimension of family relationships have drawn attention in the field.
However, how each dimension of family relationships influences health over time, especially
among older adults, remains unclear.
This dissertation used quality longitudinal data and advanced statistical analyses to
expand understanding of the dynamic interplay between family relationships and health
outcomes over time. Longitudinal data and a rigorous and appropriate analytic approach in the
24
field of family relationships and health has been long desired but largely unfulfilled due to data
constraints and limited application of advanced analyses (e.g., Thoits, 2011; Umberson &
Montez, 2010).
This dissertation addresses family relationships and health in later life in a specific
cultural context. Cultural norms, roles, and relationships shape how people interact with one
another; in other words, how family relationships influence health in later life needs to be studied
in context (e.g., Zunzunegui et al., 2004). Older adults in rural China in particular deserve more
scholarly interest to generate a better understanding of family relationships and health-related
outcomes in their specific context (e.g., Silverstein et al., 2006). Culturally specific expectations
and normative experiences and their effect on health are examined in detail.
Specific Aims
This dissertation investigated the role of families in reducing health disparities in later
life and focuses on the influence of family relationships on health outcomes over time among
older adults in rural China. Health outcomes include self-rated health and functional limitations.
Three aims guided this dissertation:
1. To examine in patrilineal family systems how gendered support exchange across
generations influences functional limitations among aging parents over time.
Instrumental support provided by sons, daughters, and daughters-in-law to aging
parents over time is examined. In the Chinese context of filial piety, sons and their
family are expected to provide support to their aging parents. The crucial role of
daughters-in-law in the support system has drawn more and more scholarly attention.
Instrumental support from sons-in-law is not a normative experience and rarely
occurs for aging parents in this context; therefore it was not included in the analyses.
25
Emotional cohesion between sons and parents and between daughters and parents was
also examined.
2. To investigate the dynamic relationship between multidimensional intergenerational
instrumental support and self-rated health among older adults in rural China over 11
years. Instrumental support includes provision and receipt of household chore help
and personal care help. Trajectories of multidimensional instrumental support and
perceptions of health were examined. Correlated changes in multidimensional
instrumental support and perceptions of health were tested.
3. To investigate the dynamic interplay between instrumental support received and self-
rated health among older adults in rural China over 11 years. Changes in household
chore help received, personal care help received, and perceptions of health over time
and the direction of those associations were examined. To what extent household
chore help received and personal care help received predict health over time was
assessed. How those two types of instrumental support improve health among older
adults over time were further examined.
26
Chapter 2: A Gendered Perspective of Intergenerational Relationships and Functional
Limitations: Four-Wave Autoregressive and Cross-Lagged Panel Analysis
Introduction
Functional capability is crucial for older adults to maintain their independence and stay
connected with family, friends, and community (Avlund, Lund, Holstein, & Due, 2004). As
people get older, there is a greater chance of developing limitations in physical functioning and
consequently an increased risk of being dependent on others, disrupting lifelong patterns of
activity and decreasing quality of life (Atchley & Scala, 1998). Functional limitations typically
involve restrictions in activities that are fundamental to the performance of a wide range of roles
and their importance in the disability process has been recognized (Litwin, Shrira, & Shmotkin,
2012; Long & Pavalko, 2004). Social relations are among the factors associated with functional
decline and the development of disability and have been shown to have a beneficial influence on
maintenance or improvement of functional ability (Berkman et al., 2000).
However, existing studies have generated mixed findings on the association between
social relations, including structural and functional aspects, and physical functioning. For
instance, social network size has been considered a protective factor of functional ability among
community-dwelling older adults in some studies (e.g., Camacho, Strawbridge, Cohen, &
Kaplan, 1993; Mendes de Leon et al., 1999; Unger, McAvay, Bruce, Berkman, & Seeman,
1999), yet this was not the case in other studies (e.g., Seeman, Bruce, & McAvay, 1996;
Strawbridge, Camacho, Cohen, & Kaplan, 1993). Social support has been regarded as protective
against functional limitations (e.g., Boult, Kane, Louis, Boult, & McCaffrey, 1994), whereas a
similar effect was not found in other studies (e.g., Mendes de Leon et al., 1999; Seeman et al.,
1996). It has been suggested that differential effects on functioning exist across different
27
dimensions of social support in a manner that the impact of instrumental support received is not
explicit (Seeman et al., 1995) or had an association with increased disability (Seeman et al.,
1996; Weinberger, Tierney, Booher, & Hiner, 1990), whereas receiving emotional support was
shown to be beneficial (Seeman et al., 1995).
Social relations are subject to influences from personal and situational characteristics, and
there has been an increasing interest in better understanding social relations within and across
cultural contexts and an emphasis on the role and importance of culture in specifying situational
characteristics in social interactions and exchanges (Antonucci et al., 2011). Based on existing
studies carried out among ethnic majorities such as Caucasians in the Western context, the
importance of friends instead of families on health in later life tend to be emphasized, whereas
among ethnic minorities and in other countries such as East Asian countries, families are crucial
for maintaining and improving health and well-being among older adults (e.g., Antonucci et al.,
2014; S.-T. Cheng et al., 2009; Chi & Chou, 2001; Fiori et al., 2008; Stephens et al., 2011;
Zunzunegui et al., 2004). In societies featuring Confucianism, dominant generational bonds
between parents and adult children in the family system are often long lasting (C. L. Johnson,
2000).
Rural China provides a unique context in which to study the influence of family
relationships on functional limitations among older adults over time. In Chinese societies, a
prominent family-centered cultural value—filial piety, or xiao—regulates the attitudes and
behaviors of children toward their parents and fosters an inherent sense of obligation for children
to support and care for their parents to ensure parental well-being (Mao & Chi, 2011). A
gendered perspective of support transfers across generations has been observed in this cultural
context and sons are expected to shoulder the primary responsibility for parental well-being (Lin
28
et al., 2003; Zhang, Li, & Feldman, 2005). In concordance with traditional gender norms in
which women typically take responsibility for domestic work, the important role of daughters-in-
law in the support system has drawn more and more scholarly attention (Cong & Silverstein,
2008). Daughters tend to participate in support provision to aging parents out of affection rather
than culturally prescribed responsibility in the patrilineal family system, and support from sons-
in-law is not normative nor expected (Wong, 2005).
In rural areas of China, where there is scant formal support, the expectations that children
will practice filial piety are especially strong and most older adults rely primarily on their adult
children for financial, instrumental, and emotional support (Cong & Silverstein, 2008).
Receiving instrumental support from children has been correlated with increased functional
limitations, whereas receiving emotional support has been correlated with decreased functional
limitations among older adults in rural China (P. Wang & Li, 2011). Investigations using a
gendered perspective indicated that receiving instrumental support from sons and daughters has
differential effects on well-being among older adults in rural China. For example, assistance
received from sons was correlated with increased depressive symptoms, help from daughters was
not explicitly correlated with depressive symptoms, and assistance from daughters-in-law was
correlated with decreased depressive symptoms (Cong & Silverstein, 2008).
It should be noted that there has been a lack of research investigating the relationship
between gendered intergenerational support and functional limitations over time in later life. It is
of theoretical and practical importance to fill this gap. This nuanced approach to studying the
differential effects of different sources of support on health outcomes among older adults over
time will contribute to further understanding of how social support influences health in later life.
Recognizing the differential effects of gendered intergenerational support on health could
29
provide useful insights and inform the development of programs and policies to promote the
health protective function of intergenerational support.
The present study aims to investigate the relationship between gendered intergenerational
support and functional limitations over time among older adults in rural China. Given the
differential effects of different dimensions of support on functional limitations noted in previous
studies, measures of gendered instrumental support included instrumental support received from
sons, daughters, and daughters-in-law and emotional cohesion between sons and parents and
between daughters and parents.
Methods
Sample
Data used in this study came from four waves of a longitudinal study titled The Well-
Being of Older People in Anhui Province conducted in 2001, 2003, 2006, and 2009. Data were
collected jointly by the School of Gerontology and School of Social Work at the University of
Southern California and the Population Research Institute of Xi’an Jiaotong University. Adults
aged 60 or older residing in the rural region of Chaohu in Anhui province were randomly
selected based on their administrative records using stratified multistage sampling. Standard
back-translation was used to ensure the accuracy of the questionnaire in Mandarin. The survey
was conducted in each respondent’s home and covered topics such as family relations and
physical health status. In the initial survey, 1,715 of 1,800 older adults completed the survey at
baseline in 2001, yielding a satisfactory response rate of 95.3% (Zuo, Li, Mao, & Chi, 2014).
There were 1,322 respondents who reported having at least one son and at least one daughter at
baseline (2001), 1,080 respondents in 2003, 834 respondents in 2006, and 631 respondents in
30
2009. Attrition rate ranged from 18.31% to 24.34% across the four waves. The primary reason
for attrition was mortality.
Measures
Gendered intergenerational instrumental support. Older adults were asked to report
how often they received help with household chores (e.g., cleaning house, washing clothes, and
washing dishes) and any personal care (e.g., taking a bath, putting on clothes) during the
previous 12 months from sons, daughters, and daughters-in-law based on a 5-point scale (0 =
none, 1 = seldom, 2 = several times per month, 3 = at least once per week, 4 = every day).
Gendered intergenerational instrumental support was measured by summated scores of
frequencies of help received from each source separately. The theoretical range for each source
was 0 to 8. Higher scores indicated higher levels of instrumental support received.
Gendered intergenerational emotional support. Emotional cohesion between sons and
parents and between daughters and parents was measured by a 3-item scale assessing how close
respondents felt toward each child, how much they felt that each child would be willing to listen
when respondents needed to talk about their worries and problems, and how well respondents
and each child get along together based on a 3-point response set (0 = not at all, 1 = somewhat, 2
= very). A satisfactory reliability (Cronbach’s α = .950) was reported using the initial survey
data (Guo et al., 2012). An emotional cohesion score was calculated for each child with a range
of 0 to 6. Higher scores indicated closer parent–child relationships. Maximum emotional
cohesion scores represent the closest parent–child relationships (Guo, Chi, & Silverstein, 2011).
Separate maximum emotional cohesion scores for sons and daughters were used in the analysis.
Functional limitations. Functional limitations were measured by asking respondents to
rate their level of difficulty in performing 11 activities of daily living (ADLs) and instrumental
31
activities of daily living (IADLs) with a 3-point response set: 0 = none, 1 = some, and 2 = cannot
do it without help. The theoretical range of this scale was 0 to 22. The Cronbach’s alpha was
.945 for functional limitations in this sample.
Control variables. Gender was a dichotomous variable consisting of male (referent)
versus female. Age was a continuous variable. Education was dichotomized into no formal
education versus some education (referent). Marital status was regrouped into married versus not
married (referent) at baseline. Satisfaction with economic situation was dichotomized into
satisfied versus dissatisfied (referent) at baseline. Chronic conditions were measured as the
number of diagnosed chronic conditions, given their association with functional limitations
among older adults (e.g., Béland & Zunzunegui, 1999). Number of sons was measured as the
number of living sons at baseline. Number of daughters was measured as the number of living
daughters at baseline. Instrumental support from spouses at each data collection point was also
used as a control variable.
Analyses
Descriptive statistics were used to assess the characteristics of the sample at baseline.
Means and standard deviations of key measurements across four waves were also reported.
Autoregressive and cross-lagged panel analysis was the focal analytic approach in this study.
Autoregressive effects in this model represent the effects of a construct on itself at a later time
and describe the stability of individual differences in constructs from one occasion to the next,
whereas cross-lagged effects represent the effects of one construct on another at a later time and
are typically estimated after controlling for the prior level of the construct being predicted (Selig
& Little, 2012). Autoregressive and cross-lagged panel analysis can simultaneously address
reciprocal influences on gendered intergenerational support and functional limitations in this
32
study and the pattern of effects would be conceptually replicated at each time point as suggested
in other studies (e.g., Cacioppo, Hawkley, & Thisted, 2010).
[Insert Figure 2.1 about here]
Figure 2.1 provides a simplified diagram of the autoregressive and cross-lagged panel
analysis in this study. Paths a and b represents the autoregressive portion of the model, meaning
the influence of one variable at one occasion on the same variable at the next occasion. Paths c
and d represents the cross-lagged portion of the model, meaning the influence of one variable at
one occasion on another variable at the next occasion. Analyses were conducted using Mplus
6.12. Maximum likelihood estimation was used to address attrition. Model fit was assessed
according to multiple goodness-of-fit indices such as chi-square statistics, comparative fit index
(CFI), root mean square error of approximation (RMSEA), and standardized root mean square
residual (SRMR). Generally speaking, when the value of CFI is above .90, a good model fit is
suggested (Bentler, 1992). When the values of SRMR and RMSEA are less than .06, a good
model fit is indicated (Buhi, Goodson, & Neilands, 2007; Kline, 2005).
Results
[Insert Table 2.1 about here]
Table 2.1 shows the sample characteristics at baseline (2001). The average age of older
adults was 71, falling in the young-old adult category. Women comprised 51.7% of the sample.
Of the older adults in this sample, 78.6% had no formal education. Married older adults
comprised 54.90% of the sample. Most older adults (68.5%) were satisfied with their economic
situation. Respondents reported having at least one chronic condition on average. Older adults
had an average of two sons and two daughters in this sample.
[Insert Table 2.2 about here]
33
Table 2.2 shows the means and standard deviations of gendered intergenerational support
and functional limitations at four time points. At baseline, older adults reported average
instrumental support scores of 1.44 for sons, 1.57 for daughters, 1.88 for daughters-in-law, and
2.33 for spouses. Older adults reported average maximum emotional cohesion scores of 4.56 for
sons and 4.82 for daughters. Older adults had some functional limitations (M = 3.52, SD = 5.09,
range = 0–22) at baseline. Based on sample means across four waves, instrumental support from
daughters and daughters-in-law decreased over time. Instrumental support from sons and spouses
fluctuated over time. Maximum emotional cohesion with sons stayed relatively stable over time,
whereas it decreased over time with daughters. Functional limitations also fluctuated over time.
[Insert Table 2.3 about here]
Table 2.3 shows findings from the autoregressive and cross-lagged panel analysis with
adjustments for control variables. The model fit indices suggested a good fit with the data
(χ
2
(457) = 1390.088, p < .001; CFI = .900; RMSEA = .039; SRMR = .063). Based on
statistically significant and positive autoregressive effects, gendered intergenerational support
and functional limitations had consistent positive effects over time (instrumental support from
sons: β = .174; instrumental support from daughters: β = .148; instrumental support from
daughters-in-law: β = .177; instrumental support from spouses: β = .279; maximum emotional
cohesion with sons: β = .235; maximum emotional cohesion with daughters: β = .219; functional
limitations: β = .455).
Regarding cross-lagged effects from gendered intergenerational support to functional
limitations at a later time, receiving more instrumental support from sons was significantly
correlated with more functional limitations at a later time (β = .297), whereas receiving more
instrumental support from daughters-in-law was significantly correlated with fewer functional
34
limitations at a later time (β = -.247). Receiving more instrumental support from daughters was
not significantly correlated with functional limitations at a later time. Having emotional cohesion
with sons or daughters was not significantly correlated with functional limitations at a later time.
Regression coefficients for cross-lagged effects demonstrated significant positive effects
of functional limitations on instrumental support from sons, daughters, daughters-in-law, and
spouses at a later time (instrumental support from sons: β = .179; instrumental support from
daughters: β = .110; instrumental support from daughters-in-law: β = .184; instrumental support
from spouses: β = .115). Having more functional limitations was also significantly correlated
with less maximum emotional cohesion with daughters (β = -.028).
Discussion
In this study, findings suggest that gendered intergenerational instrumental support at a
later time tended to be positively correlated with the same dimension of support at a prior time.
This indicates that older adults who received those support at one point tended to receive more
support from the same source at a later time. Older adults with functional limitations at one point
tended to have increased functional limitations at a later time. Stability across four waves was
seen in gendered intergenerational instrumental support and functional limitations among older
adults over a period of 8 years.
More importantly, receiving more instrumental support from sons was significantly
correlated with increased functional limitations over time, whereas receiving more instrumental
support from daughters-in-law was significantly correlated with decreased functional limitations
over time. The effect of receiving instrumental support from daughters and having close
emotional relationships with one son or one daughter at a prior time were not significantly
correlated with functional limitations at a later time. These findings highlight the gendered nature
35
of intergenerational support and health, and are consistent with previous findings on depressive
symptoms (Cong & Silverstein, 2008) and partially consistent with the reverse association
between receipt of instrumental support from children and functional limitations (Seeman et al.,
1996; P. Wang & Li, 2011; Weinberger et al., 1990). In the patrilineal family system, sons and
their families are the normative support providers for aging parents, yet the meaningful
contributions of daughters-in-law in the support system have been recognized and preferred by
older adults (Cong & Silverstein, 2008). This could help explain the protective function of
instrumental support received from daughters-in-law and the detrimental function of instrumental
support received from sons. Men are typically socialized to be providers, whereas women are
typically socialized to be care providers; in this case, a gender-specific division of filial tasks
existed between married sons and their spouses (Lin et al., 2003) and this might help explain the
differential effects on physical functioning.
In addition, findings differ from research in the Western context, which found a strong
beneficial effect of emotional social support on physical functioning (e.g., Seeman et al., 1995).
This might be conditioned by the normative expectation of support provision from children, i.e.,
children need to provide support and care to older parents irrespective of emotional closeness
between generations. In the Chinese culture, people tend to be emotionally restrictive and value
self-control of strong emotions (S. W.-H. Chen & Davenport, 2005). This might be another
possible explanation for the nonsignificant effect of emotional support on functional limitations
in this sample.
Having more functional limitations at a prior time was significantly correlated with more
instrumental support from sons, daughters, daughters-in-law, and spouses at a later time and less
emotional closeness with daughters. This suggests that receipt of instrumental support tended to
36
be need based and influenced by prior level of functional limitations among older adults. The
relationship between gendered intergenerational support and functional limitations is likely to be
reciprocal and a unidirectional approach might be limited. This helps to identify older adults at
risk of poor outcomes, and providers should assess sources and amount of support to ensure
adequate support and care.
There are several limitations of the present study. No causal relationship between
gendered intergenerational support and functional limitations was established. Functional
limitations were self-reported and might be limited compared to performance-based assessments.
Data came from a regional study in a rural area of China, thus generalizability might be limited
regarding older adults in other areas or other older populations.
Despite these limitations, this study’s contributions include its recognition of the
protective function of instrumental support from daughters-in-law and the nonprotective function
of support from sons on functional limitations among older adults. Functional limitations are
believed to be an intermediate pathway to disablement (R. J. Johnson & Wolinsky, 1993).
Research on functional limitations is relevant to individuals in terms of increased risk of personal
dependence on others and their direct influence on the quality of life of older adults; it is also
relevant to public policies regarding negative public perceptions of the aging process, the public
cost of long-term care, the ability to meet the needs of an increasing number of oldest old adults,
and increasing prevalence of late-life impairment (Atchley & Scala, 1998; Litwin et al., 2012).
37
Chapter 3: Multidimensional Intergenerational Instrumental Support and Self-Rated
Health: Trajectories and Correlated Change over 11 Years
Introduction
Understanding the developmental nature of the relationship between social support and
health over time in later life has important theoretical and practical implications. Self-rated
health has been a key theme in gerontological research because of its association with objective
health and functional status, its relationship with functional decline and mortality, and its
correlation with health services use (Jylhä, 2009; Menec, Shooshtari, & Lambert, 2007; Pinquart,
2001). Among the correlates of self-rated health among older adults, the importance of social
support has been recognized in diverse contexts (e.g., García, Banegas, Pérez-Regadera, Cabrera,
& Rodríguez-Artalejo, 2005; C.-W. Wang, Chan, Ho, & Xiong, 2008). Social support and self-
rated health are not static but change over time (Craigs, Twiddy, Parker, & West, 2014). Despite
increasing evidence that social support is associated with health in later life, little is known
regarding the social support process and how this process is associated with changes in late-life
health over time (Shaw et al., 2007).
Studies of changes in receipt and provision of social support in later life have indicated
different and sometimes inconsistent trajectories for different types of social support (e.g.,
Bergeman, Neiderhiser, Pedersen, & Plomin, 2001; Bossé, Aldwin, Levenson, Spiro, &
Mroczek, 1993; Cornman, Lynch, Goldman, Weinstein, & Lin, 2004; Miller & McFall, 1991;
Shaw et al., 2007; van Tilburg, 1998). Based on the few longitudinal studies that specifically
focused on trajectories of social support, receiving social support such as instrumental support
from families and friends tends to increase as people age (Shaw et al., 2007; van Tilburg, 1998),
whereas provision of such support tends to decline over time (Shaw et al., 2007). Instrumental
38
support typically includes practical or tangible forms of support, such as household help (e.g.,
Shaw et al., 2007; van Tilburg, 1998) or personal care help (e.g., Silverstein et al., 2006). It is
important to note that some older adults may receive increased instrumental support as their
health declines; in other words, receipt of instrumental support tends to be reversely correlated
with self-rated health (Deeg & Kriegsman, 2003).
The association between receipt or provision of support and perceptions of health over
time in later life has been inconclusive. The few studies that examined the relationship between
support received from family or friends and perceptions of health among community-dwelling
older adults over time reported some evidence of an inverse relationship. An increase in help
with daily chores over time was associated with worsening self-rated health over time, whereas
having worse self-rated health at baseline was found to be associated with receiving more help
with daily chores across three observations (van Tilburg, 1998). A similar pattern of change in
support and health was not found by Minkler and Langhauser (1988) in that receipt of support
was not significantly correlated with subsequent self-rated health. It should be noted that these
two studies did not provide evidence of the association between changes in support and health
due to the lack of assessment of changes over time. On the other hand, in the few studies that
have investigated the relationship between support provided to family or friends and perceptions
of health in community-dwelling older adults over time, changes over time were rarely
examined. A significant correlation between better self-rated health and support provided to
family and friends at each wave was found, but changes over time were not investigated
(Hinterlong, Morrow-Howell, & Rozario, 2007). Boerner and Reinhardt (2003) found no
significant correlation between self-rated health at baseline and change in support provision.
39
Different sources of support serve specific functions in later life. The task-specific model
developed by Litwak (1985) suggests different sources of support typically provide older adults
with different types of support. Family members more often provide instrumental support to
older adults (Corhan & Antonucci, 1989). The functional specificity model (Weiss, 1974) also
indicates that requirements for specific forms of support can only be met in certain relationships.
Similar patterns are also indicated by the social convoy model (Kahn & Antonucci, 1980) and a
recent emphasis on cultural influences on the formation of convoys of social interactions and
exchanges (Antonucci et al., 2011). In accordance with specific cultural norms and expectations
in collectivist cultures, the importance of family-focused convoys has been emphasized.
Intergenerational transfers of resources are influenced by a country’s level of economic
development, political structure, and cultural norms (Silverstein & Giarrusso, 2010). In rural
China, where public resources are scarce, families survive through exchanges within families,
and older adults primarily rely on adult children and children-in-law for support and care. The
exchange of support among older adults in rural China is more likely to involve instrumental
support compared to other types of support (Shi, 1993).
Instrumental Support and Health in the Chinese Context
There is a growing body of research that has investigated the relationship between social
support and health in rural Chinese contexts. However, these predominantly regional cross-
sectional studies have been unable to detect changes and reveal the dynamic relationships
between instrumental support and perceptions of health over time (e.g., Liu, Liang, & Gu, 1995;
Song, Li, Zhang, Feldman, 2008). In addition to these limitations, there have been mixed
findings with respect to the association between instrumental support and health among older
adults in rural China. Receipt of instrumental support has been positively correlated with poor
40
health (Liu, Liang, & Gu, 1995) or not significantly associated with perceptions of health (Song
et al., 2008) and subjective well-being (Silverstein et al., 2006). On the other hand, provision of
instrumental support has been negatively correlated with poor health (Liu, Liang, & Gu, 1995)
and subjective well-being (X. Chen & Silverstein, 2000) or not significantly associated with
perceptions of health (Song et al., 2008) and subjective well-being (Silverstein et al., 2006).
Few longitudinal studies have examined the relationship between social support and
health among older adults in the Chinese context. Cornman, Goldman, Glei, Weinstein, and
Chang (2003) found that the protective effects of social support such as perceived support and
satisfaction with received support on health outcomes disappeared once prior health status was
controlled. It should be noted that change in social support was not included in the analysis by
Cornman et al. (2003), making it difficult to speculate how change in support influences health
over time. S. Li, Song, and Feldman (2009) provided some evidence of the association between
changes from one time point to the next regarding instrumental support across three time points
and its effect on self-rated health among older adults in rural areas of Chaohu. Based on random-
effect logistic regression models with pooled data from two survey intervals, increased
instrumental support from children was associated with deterioration in health among older men,
whereas increased instrumental support provided to children was associated with improvements
in health among older women (S. Li et al., 2009). Using hierarchical linear models, F. Chen and
Liu (2012) found that the effect of providing instrumental support to children in the form of
caregiving for grandchildren on self-rated health over time among older adults differed by level
of support provided and other individual characteristics. However, the authors did not assess
correlated change between provision of instrumental support and health. How changes in
41
instrumental support influence changes in self-rated health over time in later life warrants further
exploration.
There are other variables that may influence self-rated health in the Chinese context,
including age, gender, marital status, education, satisfaction with economic situation, number of
children, functional limitations, and depressive symptoms. Adults who are older tend to report
worse self-rated health (e.g., Mjelde-Mossey, Chi, Lubben, & Lou, 2009; C.-W. Wang et al.,
2008). Gender also has been shown to influence self-rated health among older adults in China
(e.g., Mjelde-Mossey et al., 2009). Marital status such as widowhood was found to have direct
negative effects on self-rated health among Chinese older adults (Krochalk, Li, & Chi, 2008).
Having fewer years of education has been positively associated with poorer self-assessed health
among older adults in rural China (Zimmer & Kwong, 2004). Having satisfaction with one’s
economic situation was shown to be significantly correlated with better self-rated health among
older individuals (Y. H. Cheng, Chi, Boey, Ko, & Chou, 2002; Mjelde-Mossey et al., 2009).
Functional limitations had strong negative influences on self-rated health among Chinese older
adults (Liu, Liang, & Gu, 1995). Higher levels of depressive symptoms have been correlated
with worse self-rated health (Song et al., 2008). The association of between number of children
and health among older adults was significant in one study (Beckett, Goldman, Weinstein, Lin,
& Chuang, 2002) yet not significant in another (Cornman et al., 2003). Given the interest of this
study in instrumental support from all adult children and the structural aspect of family
relationships, the number of children was included as a covariate.
The Present Study
Despite the importance of this issue, there has been limited research on the trajectories of
and association between support and health over time among older adults. The existing
42
knowledge on this subject has been limited in terms of examining the association between
changes in support and changes in health over time among older adults. This study aimed to
understand the dynamic relationship between multidimensional intergenerational instrumental
support, both received and provided, and self-rated health over time. This approach is
theoretically relevant to debates about the protective or maintenance function of social support in
health among older adults (e.g., Cohen & Wills, 1985; Krause, 2004; Liu, Liang, & Gu, 1995;
Zunzunegui et al., 2004) and will provide useful insight into the trajectories of different
dimensions of intergenerational instrumental support (household chore help and personal care
help) and their correlated change with self-rated health in later life. It is important to recognize
that different dimensions of instrumental support serve different functions and fulfill different
tasks in later life. With this background in mind, the following research questions were
addressed:
1. What are the trajectories of multidimensional intergenerational instrumental support
among older adults over time?
2. What is the trajectory of health among older adults over time?
3. How are changes in each dimension of intergenerational instrumental support
correlated with changes in health among older adults over time?
Methods
Sample
Data used in this study came from five waves of a longitudinal study titled The Well-
Being of Older People in Anhui Province conducted in 2001, 2003, 2006, 2009, and 2012 and
collected jointly by the School of Gerontology and School of Social Work at the University of
Southern California and the Population Research Institute of Xi’an Jiaotong University. Adults
43
aged 60 or older residing in the rural region of Chaohu in Anhui province were randomly
selected based on administrative records using stratified multistage sampling. Standard back-
translation was used to ensure the accuracy of the questionnaire in Mandarin. The survey was
conducted in each respondent’s home and covered topics such as family relations and physical
health status. In the initial survey, 1,715 of 1,800 older adults completed the survey, yielding a
satisfactory response rate of 95.3% (Zuo et al., 2014). At Time 1, 1,636 respondents reported
having at least one child, compared to 1,324 respondents at Time 2, 1,016 respondents at Time 3,
765 respondents at Time 4, and 567 respondents at Time 5. Attrition rate ranged from 19.07% to
25.88% across the five waves. The primary reason for attrition was mortality.
Measures
Instrumental support. Instrumental support referred to receipt and provision of help
with household chores and personal care. Child care was included in the provision of
instrumental support because older adults tend to provide care to their grandchildren to help their
adult children in rural regions, particularly in the context of mass outmigration of young adults
seeking employment opportunities in urban settings.
Receipt of household chore help. Respondents were asked to report how often they
received household help (e.g., cleaning house, washing clothes, and washing dishes) during the
previous 12 months from sons, daughters, daughters-in-law, and sons-in-law on a 5-point scale
(0 = none, 1 = seldom, 2 = several times per month, 3 = at least once per week, 4 = every day). A
summated score was calculated by adding frequencies of help received from each source, with a
theoretical range of 0 to 16. Higher scores indicated higher levels of household help received.
Receipt of personal care help. Respondents were asked to report how often they
received any help with personal care (e.g., taking a bath, putting on clothes) during the previous
44
12 months from sons, daughters, daughters-in-law, and sons-in-law on a 5-point scale (0 = none,
1 = seldom, 2 = several times per month, 3 = at least once per week, 4 = every day). A summated
score was calculated by adding frequencies of help received from each source, with a theoretical
range of 0 to 16. Higher scores indicated higher levels of personal care received.
Provision of household chore help. Respondents were asked to report how often they
provided household help (e.g., cleaning house, washing clothes, washing dishes) during the
previous 12 months to sons, daughters, daughters-in-law, sons-in-law, and grandchildren on a 5-
point scale (0 = none, 1 = seldom, 2 = several times per month, 3 = at least once per week, 4 =
every day). A summated score was calculated by adding frequencies of help provided to each
group, with a theoretical range of 0 to 20. Higher scores indicated higher levels of household
help provided.
Provision of personal care help. Respondents were asked to report how often they
provided assistance with personal care (e.g., taking a bath, putting on clothes) during the
previous 12 months to sons, daughters, daughters-in-law, sons-in-law, and grandchildren on a 5-
point scale (0 = none, 1 = seldom, 2 = several times per month, 3 = at least once per week, 4 =
every day). A summated score was calculated by adding frequencies of help provided to each
group, with a theoretical range of 0 to 20. Higher scores indicated higher levels of personal care
provided.
Self-rated health. Self-rated health was measured by one global item: “How do you
assess your current health status?” This item featured a 4-point response set (1 = poor, 2 = fair, 3
= good, 4 = very good). Higher scores indicated better self-rated health. This single-item global
rating has been found to have robust validity in older Chinese populations (Y. Li, Aranda, & Chi,
2007; Zeng, Vaupel, Xiao, Zhang, & Liu, 2002).
45
Covariates. Gender was a dichotomous variable consisting of male (referent) versus
female. Age was a continuous variable. Education was dichotomized into no formal education
versus some education (referent). Marital status was regrouped into married versus not married
(referent) at baseline. Number of children was measured by the number of living children at
baseline. Satisfaction with economic situation was dichotomized into satisfied versus dissatisfied
(referent) at baseline. Baseline functional limitations were measured by asking respondents to
rate their level of difficulty performing 15 ADLs, IADLs, and activities requiring physical
strength, mobility, and flexibility, with a 3-point response set of 0 = none, 1 = some, and 2 =
cannot do it without help (Guo, Aranda, & Silverstein, 2009). Cronbach’s alpha was .945 for
functional limitations in this sample. A summated score was created with a theoretical range of 0
to 30, and higher scores indicated more functional limitations.
Baseline depressive symptoms were measured by the 9-item translated version of the
Center for Epidemiologic Studies Depression Scale, including positive affect, negative affect,
marginalization, and somatic symptoms assessed on a 3-point scale (0 = rarely or none of the
time, 1 = some of the time, 2 = most of the time; Silverstein et al., 2006). Cronbach’s alpha was
.765 for depressive symptoms in this sample. A summated score was created with a theoretical
range of 0 to 18, and higher scores indicated higher levels of depressive symptoms.
Analyses
Descriptive statistics were used to describe the characteristics of the sample at baseline.
Means and standard deviations of key measurements across five waves were also calculated.
Latent growth curve models were used as the analytic approach to explore changes over time.
These models can be used to test the degree to which the data follow a specific trajectory over
time; to what extent individuals deviate from the average trajectory; whether there are
46
correlations in changes between variables; and whether there is interindividual variability in
intraindividual changes (McArdle, 1998; Voelkle, 2007). The average change in each dimension
of intergenerational instrumental support and self-rated health in this sample over 11 years was
first evaluated. Latent intercepts and slopes capture the initial level and linear rate of change in
key constructs, respectively. In addition, a quadratic slope can estimate nonlinear rates of change
in a given construct (Muthén & Muthén, 2010). Given the nuanced approach of this study and
limited evidence on the trajectory and rate of change in intergenerational instrumental support
and health over time, a standard linear latent growth curve model was fitted for each of the key
constructs of this study. Nonlinear models such as quadratic growth curve models were tested
when appropriate.
Parallel process latent growth curve models can be used to estimate the growth in
intergenerational instrumental support and health simultaneously and assess the association (i.e.,
covariances) between the two growth processes (i.e., rate of change, or latent slopes). In the
context of longitudinal structural equation modeling, a series of individual latent growth curve
models and parallel process latent growth curve models, both unadjusted and adjusted for other
covariates, were conducted using Mplus 6.12. Covariates such as age, gender, marital status,
education, satisfaction with economic status, number of children, functional limitations, and
depressive symptoms at baseline were included in the analyses. Covariates were grand mean
centered. Maximum likelihood estimation was used to address attrition. Figure 1 provides a
diagram of the parallel process latent growth curve models without covariates used in this study.
Model fit was assessed according to multiple goodness-of-fit indices, such as chi-square
statistics, CFI, RMSEA, and SRMR. Generally speaking, CFI values above .90 suggested a good
model fit (Bentler, 1992). SRMR and RMSEA values less than .06 also indicate a good model fit
47
(Buhi et al., 2007; Kline, 2005).
[Insert Figure 3.1 about here]
Results
[Insert Table 3.1 about here]
Table 3.1 shows the sample characteristics at baseline (2001). The average age of older
adults was 71, falling in the young-old adult category. Women accounted for 52.4% of the
sample. Most older adults (79.2%) in this sample had no formal education. Older adults who
were married comprised 52.80% of the sample, and 65.4% were satisfied with their economic
situation. Respondents had an average of four children. Older adults had some functional
limitations (M = 5.89, SD = 7.35, range = 0–30) and depressive symptoms (M = 7.49, SD = 2.62,
range = 0–16).
[Insert Table 3.2 about here]
Table 3.2 shows the means and standard deviations of key measurements over time.
According to changes in sample means, receipt of household chore help decreased over time.
Receipt of personal care help fluctuated, with an upward trend over time. There was no clear
linear pattern for either provision of household chore help or personal care; they tended to
increase to some extent and then decreased over time. Self-rated health decreased over time.
In the current study, model estimation proceeded as follows. Univariate latent growth
curve models without covariates were fitted to the observed scores of each dimension of
intergenerational instrumental support and self-rated health. As expected, estimated individual
linear latent growth curve models for the two dimensions of instrumental support received and
self-rated health demonstrated a good fit to the data (household chore help received: χ
2
(10) =
10.535, p = .395; CFI = .997; RMSEA = .006; SRMR = .028; personal care help received: χ
2
(10)
48
= 24.201, p = .007; CFI = .911; RMSEA = .030; SRMR = .039; self-rated health: χ
2
(10) =
53.466, p < .001; CFI = .928; RMSEA = .052; SRMR = .048). On the contrary, estimated
individual linear latent growth curve models for the two dimensions of instrumental support
provided demonstrated a poor fit to the data (household chore help provided: χ
2
(10) = 219.619, p
< .001; CFI = .470; RMSEA = .115; SRMR = .108; personal care help provided: χ
2
(10) =
182.396, p < .001; CFI = .280; RMSEA = .104; SRMR = .097). In an effort to capture the
potentially nonlinear trend of the two dimensions of instrumental support provided, individual
quadratic latent growth curve models were tested. The estimated quadratic latent growth curve
model for provision of household chore help demonstrated a good fit to the data (χ
2
(6) = 28.769,
p < .001; CFI = .942; RMSEA = .049; SRMR = .037). After adjusting the model specification for
the estimated quadratic latent growth curve model for provision personal care help provided, a
relatively good model fit according to RMSEA and SRMR values was reached compared to the
previous linear latent growth curve model (χ
2
(9) = 56.055, p < .001; CFI = .803; RMSEA = .057;
SRMR = .059).
[Insert Table 3.3 about here]
To estimate the concurrent development and associations of changes between each
dimension of intergenerational instrumental support and self-rated health among older adults
over time, parallel process latent growth curve models, both unadjusted and adjusted for
covariates, were fitted. As shown in Table 3.3, an estimated parallel process latent growth curve
model was found to be a good fit for the data regarding receipt of household chore help (χ
2
(41) =
102.940, p < .000; CFI = .936; RMSEA = .030; SRMR = .043). Older adults on average received
some household chore help from children and children-in-law at baseline (β = 3.264, p < .001).
Receipt of household chore help among older adults decreased over time (β = -.191, p < .001).
49
Older adults who started with a lower initial level of household chore help received at baseline
tended to receive increasingly more help with household chores over time than others (β = -.655,
p = .013). Older adults reported having fair health status on average at baseline (β = 2.093, p <
.001). Self-rated health of older adults decreased over time (β = -.042, p < .001). Older adults
who reported better health status tended to decline at a slower rate over time (β = -.021, p =
.027). The change between receipt of household chore help and self-rated health over time was
significantly reversely correlated (β = -.048, p < .001), suggesting that older adults who reported
worsening health status received household chore help at a faster rate over time. When
controlling for covariates, a similar pattern held.
[Insert Table 3.4 about here]
As shown in Table 3.4, an estimated parallel process latent growth curve model was
found to be a good fit for the data regarding receipt of personal care help (χ
2
(41) = 110.974, p <
.001; CFI = .922; RMSEA = .032; SRMR = .044). Older adults on average received some
personal care help from children and children-in-law at baseline (β = 1.454, p < .001). Receipt of
personal care help among older adults increased over time (β = .125, p = .001). The initial level
of personal care help received was not significantly correlated with changes in receipt of
personal care help over time (β = -.043, p = .805). The change between personal care help
received and self-rated health over time was significantly reversely correlated (β = -.035, p =
.001), suggesting that older adults who reported worsening health status gained help with
personal care at a faster rate over time. When controlling for covariates, a similar pattern held.
[Insert Table 3.5 about here]
[Insert Figure 3.2 about here]
50
As shown in Table 3.5, an estimated parallel process latent growth curve model with a
quadratic term was found to be a good fit for the data regarding provision of household chore
help (χ
2
(35) = 117.415, p < .001; CFI = .923; RMSEA = .038; SRMR = .041). Older adults on
average provided some household chore help to children and children-in-law at baseline (β =
2.339, p < .001). Provision of household chore help among older adults increased and then
decreased over time (linear latent slope: β = .577, p < .001; quadratic latent slope: β = -.232, p <
.001). Figure 3.2 is an illustration of the change in provision of household chore help over time.
The change between household chore help provided and self-rated health over time was not
significantly correlated (linear latent slope: β = .017, p = .589; quadratic latent slope: β = -.002, p
= .816). When controlling for covariates, a similar pattern held.
[Insert Table 3.6 about here]
[Insert Figure 3.3 about here]
As shown in Table 3.6, an estimated parallel process latent growth curve model with a
quadratic term was found to be an acceptable fit for the data regarding provision of personal care
help (χ
2
(40) = 139.786, p < .001; CFI = .887; RMSEA = .039; SRMR = .048). Older adults on
average provided some personal care help to children and children-in-law at baseline (β = .980, p
< .001). Personal care help provided by older adults increased and then decreased over time
(linear latent slope: β = .479, p < .001; quadratic latent slope: β = -.148, p < .001). Refer to
Figure 3.3 for an illustration of this change. The change between personal care help provided and
self-rated health over time was not significantly correlated (linear latent slope: β = .000, p = .993;
quadratic latent slope: β = .001, p = .895). When controlling for covariates, a similar pattern
held.
51
Discussion
This study investigated the trajectories of and correlated change between
multidimensional intergenerational instrumental support and self-rated health. Receipt of
household chore help decreased over time, whereas receipt of personal care help increased over
time. Provision of household chore help and personal care help increased and then decreased
over time. The self-rated health of older adults decreased over time. The findings also indicated
that an increase in receipt of both household chore and personal care help from children and
children-in-law was associated with a slower increase in health among older adults over time,
and an increase in health was associated with a slower increase in receipt of both household
chore and personal care help from children and children-in-law over time. A similar pattern did
not hold for provision of household chore help or personal care help.
This study employed a nuanced approach to examine multiple dimensions of
intergenerational instrumental support and advanced analytic methods to assess changes over
time in later life. The trajectory of instrumental support in terms of receipt of household chore
help was inconsistent with other studies (Shaw et al., 2007; van Tilburg, 1998). Considering that
children are the primary source of support and care for aging parents in rural China, limited
resources in families might be commonly divided into help with household chores and personal
care and the amount of each kind of help might shift based on the health condition of older
adults. Given that personal care help received by older adults increased over time, it might be
expected that receipt of household chore help would decrease over time.
The trajectory of instrumental support provided by older adults in terms of household
chore help was somewhat consistent with the general downward trajectory in a previous study
(Shaw et al., 2007) and may be influenced by the unique situation of older adults in rural China,
52
which is characterized by mass outmigration of young adults and the traditional roles and
expectations of grandparenting (Lou, Lu, Xu, & Chi, 2013). In this context, older adults may
provide more intensive support and care to their children and children-in-law, especially in the
form of support and care for grandchildren when they are young, and provide less care as
grandchildren grow up. This might help explain why the trajectories of provision of household
chore help and personal care help increased and then decreased over time.
In terms of investigating the dynamic relationship between each dimension of
instrumental support and health in later life, this study found a reverse correlation between
received instrumental support and health over time, similar to other studies (e.g., Deeg &
Kriegsman, 2003; S. Li et al., 2009). The potential health benefits of support provision over time
in later life constitute an important aspect of health research and warrant more scholarly attention
(Gruenewald & Seeman, 2010). In this study, change in the provision of instrumental support
was not significantly correlated with change in self-rated health over time. This is consistent with
findings on the association between provision of instrumental support and health in some cross-
sectional studies (e.g., Song et al., 2008), yet inconsistent with some longitudinal studies (e.g., F.
Chen & Liu, 2012; S. Li et al., 2009). This may be related to a sense of obligation among older
adults in rural China to provide instrumental support, especially care and help to grandchildren,
when their children migrate to urban areas to seek jobs, rather than a health-based assessment of
the provision of instrumental support.
This study contributes to the knowledge building process in the following ways. Based on
task-specific and functional specificity models and informed by the social convoy model, this
study examined the exchange of culturally informative multidimensional instrumental support
between adult children and older parents, demonstrating different trajectories of intergenerational
53
instrumental support and correlations with changes in health over time. Both intergenerational
instrumental support and health were dynamic over time. This nuanced approach demonstrated
the differential changes in each dimension of intergenerational instrumental support in later life.
More importantly, developmental change and the correlated change between each dimension of
intergenerational instrumental support and health over time in later life were examined,
expanding on previous longitudinal studies on this topic. Interventions should focus on
encouraging and maintaining instrumental support, especially receipt of intergenerational
instrumental support, among older adults to ensure well-being in later life.
There are several limitations that need to be noted. No causal relationships between
multidimensional intergenerational instrumental support and self-rated health could be
established. Self-rated health was a subjective indicator of health and evaluative in nature. Self-
rated health was treated as a continuous variable using a single item with a 4-point scale.
Distribution of each dimension of instrumental support was skewed to some extent and the
measurements only captured one aspect of support exchange, i.e., frequency. Selection effect or
attrition may have generated biased results despite maximum likelihood estimation because of
the assumption that data were missing at random.
Future studies should continue to explore the relationship between different dimensions
of social support and physical health outcomes and examine potential influences of social
support on self-rated health over time. Future studies should also extend beyond a correlational
relationship between support and health over time to examine the sequence of change and how
these dimensions influence one another during an extended period of time. Using established
measurements and testing factorial invariance in the context of structural equation modeling
would be a potential next step. Testing the potential influences of a third variable, such as other
54
dimensions of social support, in the relationship between instrumental support and self-rated
health is also recommended, as is further analyzing change over time and taking advantage of
longitudinal data.

55
Chapter 4: Dynamic Interplay between Intergenerational Instrumental Support and Self-
Rated Health among Older Adults: An Application of Bivariate Latent Change Score
Modeling
Introduction
A critical yet often understudied theoretical question in gerontological research concerns
the study of change, namely how certain constructs develop or change over time or how two
processes are interrelated with each other as they unfold over time (e.g., Ferrer & McArdle,
2010). A growing body of scientific evidence has shown the health benefits of social
relationships, including consistent findings of prospective studies of mortality in industrialized
nations (Berkman et al., 2000; Umberson & Montez, 2010). Social support has been shown to
have powerful effects on health among older adults in different contexts (Auslander & Litwin,
1991; Liu, Liang, & Gu, 1995; Stephens et al., 2011; White, Philogene, Fine, & Sinha, 2009). In
these examples, however, the goal of assessing changes in social support and consequences on
health over time has been left unfulfilled.
The dynamic nature of the relationship between an individual and that individual’s social
relationships is depicted in the social convoy model, which posits that individuals tend to receive
more instrumental support as they age and the source of this support is often the closest members
in one’s convoy of social relationships (Antonucci, 1985). Receipt of instrumental support has
been shown to increase over time as people age (Shaw et al., 2007). Prominent sources of health-
related support change over the life course, with adult children taking an increasingly important
role in later life (Umberson, Crosnoe, & Reczek, 2010).
Research on changes in personal relationships and perceptions of health over time and the
direction of these associations needs to be based on longitudinal data (Craigs et al., 2014),
56
namely the examination of the sequence of change in both intergenerational instrumental support
and self-rated health among older adults in this study. Limited research has investigated the
association between receiving support from family or friends and perceptions of health among
community-dwelling older adults using longitudinal data collected from the same participants.
Cornman et al. (2003) did not examine changes in social support over time, including satisfaction
with received support. Minkler and Langhauser (1988) did not find a significant association
between received support and self-rated health over time. Others found some evidence of an
association between change in support received and change in self-rated health over time (S. Li
et al., 2009; van Tilburg, 1998). An increase in help with daily chores over time was associated
with worsening self-rated health over time, irrespective of gender (van Tilburg, 1998). Similarly,
an increase in instrumental support, including help with chores or personal care, between two
time points was shown to be significantly associated with worse self-rated heath over time
among men (S. Li et al., 2009). On the other hand, having worse self-rated health at baseline was
found to be associated with receiving more help with daily chores across three observations (van
Tilburg, 1998). However, these two studies did not examine the sequence of those changes, thus
preventing inferences regarding a unidirectional relationship between received support and
health among older adults over time.
This study examined the dynamic interplay between intergenerational instrumental
support received and self-rated health among older adults over time. Specifically, changes in
both instrumental support and health over time and, more importantly, the sequence of those
changes over time were investigated. This research is theoretically relevant and will provide
invaluable insights into the debate about the protective or maintenance function of social support
57
in health among older adults (e.g., Cohen & Wills, 1985; Krause, 2004; Liu, Liang, & Gu, 1995;
Zunzunegui et al., 2004). The following research questions were addressed in this study:
1. Does intergenerational instrumental support received predict health over time or is the
relationship reciprocal?
2. If so, how does instrumental support improve health among older adults over time?
Methods
Sample
Data used in this study came from five waves of a longitudinal study titled The Well-
Being of Older People in Anhui Province conducted in 2001, 2003, 2006, 2009, and 2012 and
collected jointly by the School of Gerontology and School of Social Work at the University of
Southern California and the Population Research Institute of Xi’an Jiaotong University. Adults
aged 60 or older residing in the rural region of Chaohu in Anhui province were randomly
selected based on their administrative records using stratified multistage sampling. Standard
back-translation was used to ensure the accuracy of the questionnaire in Mandarin. The survey
was conducted in each respondent’s home and covered topics such as family relations and
physical health status. Of the initial 1,636 respondents who reported having at least one child,
567 of them were successfully followed over a period of 11 years.
Measures
Receipt of household chore help. Respondents were asked to report how often they
received household help (e.g., cleaning house, washing clothes, and washing dishes) during the
previous 12 months from sons, daughters, daughters-in-law, and sons-in-law on a 5-point scale
(0 = none, 1 = seldom, 2 = several times per month, 3 = at least once per week, 4 = every day). A
58
summated score was calculated by adding frequencies of help received from each source, with a
theoretical range of 0 to 16. Higher scores indicated higher levels of household help received.
Receipt of personal care help. Respondents were asked to report how often they
received any help with personal care (e.g., taking a bath, putting on clothes) during the previous
12 months from sons, daughters, daughters-in-law, and sons-in-law on a 5-point scale (0 = none,
1 = seldom, 2 = several times per month, 3 = at least once per week, 4 = every day). A summated
score was calculated by adding frequencies of help received from each source, with a theoretical
range of 0 to 16. Higher scores indicated higher levels of personal care received.
Self-rated health. Self-rated health was measured by one global item: “How do you
assess your current health status?” This item featured a 4-point response set (1 = poor, 2 = fair, 3
= good, 4 = very good). Higher scores indicated better self-rated health. The single-item global
rating has been found to have robust validity in Chinese older adult populations (Y. Li et al.,
2007; Zeng et al., 2002).
Analyses
Latent change score modeling (Ferrer & McArdle, 2010; McArdle, 2009) is an advanced
analytic approach for examining dynamics in one or several variables over time and combines
assessment of change such as growth or decline and dynamics among multiple processes. It is
especially helpful when the interrelations among various constructs and changes in those
constructs over time are the foci. Compared to other techniques for multivariate longitudinal data
(i.e., cross-lagged regression models, hierarchical linear models) that might not capture dynamic
features over time, latent change score modeling works extremely well when the data involve
both growth and interrelations between processes (Ferrer & McArdle, 2010). This model directly
defines the model of change and latent changes are the essential features of latent change score
59
models (McArdle, 2009).
In bivariate latent change score models, latent changes are composed of three elements:
an additive constant change, scores for the same variable at a previous occasion, and scores for
another variable at the previous occasion. The last component, the coupling parameter,
represents the influence of one variable at one occasion (e.g., receipt of instrumental support
from children) on changes in another variable at the next occasion (e.g., perceptions of health in
older adults) as the two interrelated dynamic processes unfold over time. In other words, it
represents the effect of one variable on subsequent intraindividual change in another variable
(e.g., Orth, Berking, Walker, Meier, & Znoj, 2008). In this study, those dynamic coefficients
jointly reflect the dynamic interplay between instrumental support and health over time.
Figure 4.1 is an illustration of the path diagram of a bivariate latent change score model.
Path coefficients a and b represent the coupling effect component. Path coefficient a represents
the effect of intergenerational instrumental support at one occasion on self-rated health at the
next occasion; path coefficient b represents the effect of self-rated health at one occasion on
intergenerational instrumental support at the next occasion. Path coefficients c and d represent
the self-feedback component. Path coefficient c represents the effect of intergenerational
instrumental support at one occasion on itself at the next occasion. Path coefficient d represents
the effect of self-rated health at one occasion on itself at the next occasion.
[Insert Figure 4.1 about here]
Results
[Insert Table 4.1 about here]
Table 4.1 demonstrates the model fit indices for bivariate latent change score models of
intergenerational instrumental support (i.e., receipt of household chore help and personal care
60
help) and self-rated health among older adults. Regarding receipt of household chore help and
self-rated health, bivariate latent change score models demonstrated a good fit to the data (no
coupling effect: χ
2
(46) = 117.669, p < .001; CFI = .925; TLI = .927; RMSEA = .031; coupling
effect from support to health: χ
2
(45) = 82.333, p < .001; CFI = .961; TLI = .961; RMSEA = .023;
coupling effect from health to support: χ
2
(45) = 117.034, p < .001; CFI = .925; TLI = .925;
RMSEA = .031). Based on the model fit statistics, chi-square test statistics for model comparison
were calculated. The bivariate latent change score model with a coupling effect from support to
health fit the data significantly better (∆χ
2
(1) = 35.336, p < .001).
Regarding receipt of personal care help and self-rated health, bivariate latent change
score models demonstrated an acceptable fit to the data (no coupling effect: χ
2
(46) = 136.376, p
< .001; CFI = .899; TLI = .902; RMSEA = .035; coupling effect from support to health: χ
2
(45) =
134.142, p < .001; CFI = .901; TLI = .901; RMSEA = .035; coupling effect from health to
support: χ
2
(45) = 135.387, p < .001; CFI = .899; TLI = .899; RMSEA = .035). Based on the
model fit statistics, chi-square test statistics for model comparison were calculated. The bivariate
latent change score model with a coupling effect from support to health did not fit the data
significantly better (∆χ
2
(1) = 2.234, p > .10).
[Insert Table 4.2 about here]
Table 4.2 features results from bivariate latent change score models for receipt of
household chore help and self-rated health among older adults. These results show initial means
of 3.36 for receipt of household chore help and 2.05 for self-rated health. They also show
significant substantial variation in initial levels of receipt of household chore help (β = 3.302, p <
.001) and self-rated health (β = .228, p < .001). The growth estimates indicate that changes in
both variables were a function of: (a) a positive linear slope (household chore help: β = 1.486, p
61
< .01; self-rated health: β = .999, p < .05), with some individual variability (household chore
help: β = 1.026, p < .10; self-rated health: β = 1.211, p < .01); (b) a negative autoproportion
(household chore help: β = -.563, p < .001; self-rated health: β = -1.102, p < .001), representing
the influence of the variable on itself over time; and (c) a positive coupling parameter (β = .410,
p < .001), representing the influence of support at the previous occasion on self-rated health.
These findings indicate that the level of receipt of household chore help is a positive predictor of
subsequent change in self-rated health at all four time intervals over 11 years. Older adults who
received lower levels of household chore help at one time point were expected to report
decreased self-rated health at the subsequent time point. This model specified the model of
change in which the coefficients reflected the growth of the latent change, which differed from
the model in previous chapter that specified the growth of the data.
[Insert Table 4.3 about here]
Table 4.3 features results from bivariate latent change score models of receipt of personal
care help and self-rated health among older adults. These results show initial means of 1.49 for
receipt of personal care help and 2.10 for self-rated health. They also show significant substantial
individual variation in initial levels of receipt of personal care help (β = 1.579, p < .001) and self-
rated health (β = .273, p < .001). The growth estimates did not indicate any significant changes in
either variable, reflected by a nonsignificant linear slope for both variables, a nonsignificant
autoproportion, and a nonsignificant coupling effect from receipt of personal care help to
changes in self-rated health. There was no apparent unique association between prior levels of
receipt of personal care help and subsequent changes in self-rated health in this sample.
62
Discussion
This study investigated changes in receipt of instrumental support and perceptions of
health over time, the direction of these associations, and the sequence of changes in both
variables among older adults over 11 years, using high-quality longitudinal data and advanced
analytic tools. This longitudinal study demonstrated that receiving household chore help from
adult children was a leading predictor of subsequent self-rated health over time. In other words,
older adults who received lower levels of household chore help at one time point were expected
to experience a decrease in self-rated health at the subsequent time point. A similar pattern did
not hold for receipt of personal care help.
The findings in this study differed from the limited and inconclusive research on the
association between changes in receipt of support and self-rated health over time. Some studies
reported nonsignificant association between receipt of support and self-rated health over time
(Minkler & Langhauser, 1988), some reported that an increase in help with daily chores over
time was associated with worsening self-rated health over time (S. Li et al., 2009; van Tilburg,
1998), and others reported that worse self-rated health at baseline was associated with receiving
more help with daily chores across three observations (van Tilburg, 1998). Because the sequence
of these changes was not examined in previous studies, this might be a case in which latent
change scores assessed concurrent growth and dynamic relationship between multiple processes
and provided different results.
On the other hand, the findings in this study align well with the protective or maintenance
function of social support in health among older adults (e.g., Cohen & Wills, 1985; Krause,
2004; Liu, Liang, & Gu, 1995; Zunzunegui et al., 2004). Receiving household chore help from
children played a protective role in the subsequent perceptions of health among older adults over
63
time. In rural China, there are few public programs and interventions available to promote health
among older adults. Families and adult children in particular are expected to support and care for
older adults as they age, according to the cultural value and norm of filial piety and as outlined in
contemporary laws. It is of great importance to recognize the health maintenance function of
intergenerational support and develop programs, interventions, and policies that could help
maintain and strengthen such support.
There are several limitations that need to be noted. To establish a causal relationship
between support and perception of health, the following criteria need to be met: an association
exists between the variable, one factor precedes the other factor, and no other factor can account
for the association. This study used longitudinal data and an advanced analytic approach to
explore the association between changes in support and health over time, thus meeting the first
two criteria. However, causal relationships between intergenerational instrumental support and
self-rated health could not be established because the association between the two variables
might be related to other factors. Self-rated health was a subjective indicator of health and is
evaluative in nature. It was also treated as a continuous variable using a single item with a 4-
point scale. Distribution of each dimension of instrumental support was skewed to some extent
and the measurements only captured one aspect of support exchange, i.e., frequency. Selection
effect or attrition may have led to biased results despite maximum likelihood estimation because
of the assumption that data were missing at random. There is limited generalizability given the
use of a regional sample of older adults residing in a rural area of China.
Future studies should continue to explore the relationship between different dimensions
of social support and physical health outcomes and examine potential pathways between social
support and self-rated health over time using high-quality longitudinal data and advanced
64
analytic approaches. Using a rigorous experimental design to further establish the causal
relationship between support and health in later life is encouraged. Mixed-method approaches
might also offer more insight into needs-based support provision and the protective function of
support from adult children.
65
Chapter 5: Conclusions and Implications
Family relationships are important in later life given the fact that the majority of support
and care comes from family in countries such as the United States and China. Aging in China
provides a unique context for studying how family relationships influence health outcomes,
because it is shaped by a distinct historical, cultural, economic, and political contextual
environment (F. Chen & Liu, 2009). Due to the one-child policy and low mortality, adults aged
60 or older (approximately 177 million) account for 13.3% of the total population in China
according to the latest national census in 2010 (National Bureau of Statistics of China, 2011).
That population is expected to reach 350 million by 2030 and account for roughly one third of
the total population, or 459 million people, by 2050 (He, Sengupta, Zhang, & Guo, 2007). In
Chinese societies characterized by filial piety, family support is central to the well-being of older
adults (Chi & Chou, 2001; Chou, 2010; Li & Chi, 2011) and has been reinforced by government
policies, such as the Law of the People’s Republic of China on Protection of the Rights and
Interests of the Elderly, in a manner that providing emotional, financial, and instrumental support
to older adults has become a lawful responsibility of families (National People’s Congress of the
People’s Republic of China, 1996). In the patrilineal family system, sons’ families are expected
to shoulder the primary responsibility for support and care provision. The aging trend in China
has led to increased pressure on health care infrastructure, elevated health care costs, and an
increasing burden on family support systems (F. Chen & Liu, 2009; Li & Chi, 2011; Zimmer &
Kwong, 2004), which in turn has demonstrated negative effects on health among older adults
(e.g., Sun, 2004).
This dissertation focused on intergenerational support among older parents in rural China,
investigated culturally distinctive patterns of support provided by children, and examined the
66
dynamic relationship between multidimensional intergenerational support and multiple health
outcomes including functional limitations and self-rated health over time. This dissertation
featured three studies that addressed different aspects of the relationship between
intergenerational support and health among older adults in rural China using different advanced
analytic approaches: (a) culturally informed gendered intergenerational support and functional
limitations using four-wave autoregressive and cross-lagged panel analysis; (b) trajectories of
multidimensional intergenerational instrumental support and self-rated health and their correlated
change over a decade using parallel process latent growth curve modeling; and (c) changes in
intergenerational instrumental support and self-rated health among older adults and the direction
of those changes over a period of 11 years using bivariate latent change score modeling.
Summary of Major Research Findings
The first study of this dissertation focused on the gendered perspective of
intergenerational support and functional limitations among older adults using four-wave data
from 2001 to 2009 of the longitudinal study The Well-Being of Older People in Anhui Province.
Specifically, it examined instrumental support from sons, instrumental support from daughters,
instrumental support from daughters-in-law, maximum emotional cohesion with sons, and
maximum emotional cohesion with daughters. Sociodemographic variables, chronic conditions,
number of sons, number of daughters, and instrumental support from spouses across four waves
were controlled for in the autoregressive and cross-lagged panel analysis. Major findings suggest
that receiving more instrumental support from sons leads to increased functional limitations at a
later time, whereas receiving more instrumental support from daughters-in-law leads to
decreased functional limitations at a later time. These findings highlight the gendered nature of
intergenerational support and health, and are consistent with previous findings on gendered
67
intergenerational support and depressive symptoms (Cong & Silverstein, 2008) and partially
consistent with the reverse association between receipt of instrumental support from family and
friends and functional limitations (Seeman et al., 1996; P. Wang & Li, 2011; Weinberger et al.,
1990). In the patrilineal family system, sons and their families are the normative support
providers for aging parents and the meaningful contributions of daughters-in-law in the support
system have been recognized and preferred by older adults (Cong & Silverstein, 2008). This
could help explain the protective function of instrumental support received from daughters-in-
law and the detrimental function of instrumental support received from sons. Having more
functional limitations at a prior time lead to more instrumental support from sons, daughters,
daughters-in-law, and spouses at a later time and less emotional closeness with daughters. This
suggests that receipt of instrumental support might also be need based and subject to the
influence of prior level of functional limitations among older adults. The relationship between
gendered intergenerational support and functional limitations is likely to be reciprocal and a
unidirectional approach is likely limited to some extent.
The second study of this dissertation focused on the trajectories of multidimensional
intergenerational instrumental support and self-rated health among older adults and the
correlated change between multidimensional intergenerational instrumental support and self-
rated health over a course of 11 years using parallel process latent growth curve modeling.
Receipt of household chore help decreased over time. Receipt of personal care help increased
over time. Provision of household chore help and personal care help increased and then
decreased over time. Self-rated health of older adults decreased over time. These findings
indicate that an increase in receipt of both household chore and personal care help from children
was associated with a slower increase in health among older adults over time, and an increase in
68
health was associated with a slower increase in receipt of both household chore and personal care
help from children and children-in-law over time. A similar pattern did not hold for provision of
household chore help or personal care help.
The trajectory of instrumental support in terms of receipt of household chore help was
inconsistent with existing studies (Shaw et al., 2007; van Tilburg, 1998). Considering that
children are the primary source of support and care for aging parents in rural China, limited
resources within families might be divided into help with household chores and personal care,
and the amount in each kind of help might shift based on the health condition of older adults.
The trajectory of provision of instrumental support in terms of household chore help was
somewhat consistent with the general downward trajectory in a previous study (Shaw et al.,
2007) and may have been influenced by the unique situation of older adults in rural China, which
is characterized by mass out-migration of young adults and the traditional role and expectations
of grandparenting (Lou et al., 2013). Older adults might provide more intensive support and care
to their children and children-in-law, especially in the form of support and care for grandchildren
when they are young and less support as grandchildren grow up.
These findings were similar to other studies noting a reverse correlation between received
instrumental support and health over time (e.g., Deeg & Kriegsman, 2003; S. Li et al., 2009).
The potential health benefits of support provision over time are an important aspect of health in
later life and warrant more scholarly attention (Gruenewald & Seeman, 2010). In this study,
change in the provision of instrumental support was not found to be significantly correlated with
change in self-rated health over time. This is consistent with findings on the association between
instrumental support provided and health in some cross-sectional studies (e.g., Song et al., 2008),
yet inconsistent with some longitudinal studies (e.g., F. Chen & Liu, 2012; S. Li et al., 2009).
69
This could be related to a sense of obligation among older adults in rural China to provide
instrumental support, especially care and help for grandchildren when their children migrate to
urban areas to seek jobs, and less a health-based assessment of the provision of instrumental
support.
The third study of this dissertation focused on changes in receipt of instrumental support
and perceptions of health over time, the direction of these associations, and the sequence of the
change among older adults over a period of 11 years using bivariate latent change score
modeling. Receiving household chore help from children was a leading predictor of subsequent
self-rated health over time. In other words, receiving lower levels of household chore help at an
earlier time point predicted a decrease in self-rated health at the subsequent time interval. A
similar pattern did not hold for receipt of personal care help.
The findings in this study differed from the limited inconclusive existing research on the
association between change in support received and change in self-rated health over time. Some
studies reported a nonsignificant association between received support and self-rated health at
next occasion (Minkler & Langhauser, 1988), some reported an increase in help with daily
chores over time was associated with worsening self-rated health over time (S. Li et al., 2009;
van Tilburg, 1998), and others reported that having worse self-rated health at baseline was
associated with receiving more help with daily chores across three observations (van Tilburg,
1998). Because the sequence of these changes was not examined in the previous studies, this
might be a case in which latent change scores assessed concurrent growth and dynamic
relationship between multiple processes and provided a different perspective. On the other hand,
the findings align well with health protective or maintenance function of social support among
older adults (e.g., Cohen & Wills, 1985; Krause, 2004; Liu, Liang, & Gu, 1995; Zunzunegui et
70
al., 2004). Receiving household chore help from children played a protective role in the
subsequent perceptions of health among older adults over time. In rural China, there are few
public programs and interventions available to promote health among older adults. Their families
and adult children in particular are expected to support and care for them as they get old, as
mandated by the cultural value and norm of filial piety and outlined in contemporary laws. It is
of great importance to recognize the health maintenance function of intergenerational support
and develop programs, interventions, and policies that help maintain and strengthen such
support.
Implications for Future Research
There has been a lack of studies on family relationships, particularly between adult
children and older parents, and multiple health outcomes over time in later life. This dissertation
is one of few studies that focused on intergenerational relationship and health among older adults
and examined the dynamic relationship between multidimensional support and health outcomes
over time. This dissertation sought to understand how intergenerational support and health
develop or change over time, contributing to the understanding of the developmental nature of
the relationship between intergenerational support and health in later life and how those two
processes are interrelated with each other as they unfold over time. Culturally sensitive patterns
of support exchange and change in support and health in later life warrant more scholarly
attention. Based on the findings of this dissertation, future research should focus on the following
areas with respect to intergenerational support and health over time.
The theoretical approach to studying dynamic relationship between intergenerational
support and health in older adults over time generates valuable insights into the developmental
nature and interrelationships between change in late-life family support and health, from the
71
literature on developmental psychology to the study of late-life family relationships. It extends
the theoretical configurations of the relationship between family relationships and health by
advancing beyond the stationary view of family relationships and health and encouraging
conceptualization of the dynamic relationship between those two constructs over time. This is
also consistent with tenets of the life course perspective, such as interactions remaining of great
importance throughout human life span and skills, routines, knowledge, and values acquired
throughout the years tending to be sustained and reproduced in daily life (Dannefer & Kelley-
Moore, 2009). Future research should continue to address how changes in family relationships
influence health in later life over time and what characteristics and under what circumstances
these dynamic relationships unfold over time.
The use of high-quality longitudinal data provides an opportunity to examine
developmental changes in family relationships and health over times. This dissertation used
valuable data collected in a traditional yet highly mobile sociocultural context, providing a
unique examination of how societal factors help shape interactions and support exchange within
families. The findings of this dissertation point to the importance of incorporating the influences
of unique historical, cultural, economic, and political contextual environments into the
conceptualization and operationalization of different dimensions of family relationships and how
those dimensions exert influence on late-life health. This created an opportunity to compare the
culturally sensitive patterns of interactions and support exchange among aging Chinese
populations with those from research in different contexts, generating a better understanding of
similarities and differences in the relationship between support and health across diverse cultural
contexts. Future research should focus on cross-cultural comparative studies and investigate the
72
dynamic relationship between family support and health while accounting for sociocultural
characteristics.
Application of advanced analytic approaches can helps address theoretical questions, test
research hypotheses, and take advantage of unique and valuable longitudinal data in the field of
gerontology. More advanced analytic approaches have focuses on the examination of change
over time, but they have been used much less frequently than more traditional analyses. This
dissertation used several different analytic approaches to address different questions regarding
the dynamic relationship between intergenerational support and health over time, such as
autoregressive and cross-lagged panel analysis, parallel process latent growth curve modeling,
and bivariate latent change score modeling. Future studies should continue to use advanced
analytic approaches given their ability to address research questions that would not have been
possible otherwise.
Implications for Policies and Social Work Practice
Family relationships are central to the health and well-being among older adults in the
Chinese context. In the coming years, family support systems will face greater challenges
because of the fast-growing aging population, decreasing family size, increasing dispersion of
family members, and increasing need for long-term care and support. Because there are scarce
public resources for older adults in rural areas, families are the primary source of support for
older adults. Studies on the association between family support and its influence on late-life
health and well-being in a changing context are especially valuable to maintaining and
strengthening the family support system for older adults. The findings of this dissertation show
the pattern of effect that gendered intergenerational support exerts on functional limitations
among older adults over time. This dissertation revealed the developmental nature of
73
multidimensional intergenerational support and its importance in relation to maintaining and
improving health among aging parents over time. It demonstrated how the processes of different
dimensions of intergenerational support are correlated with the process of perceptions of health
in older adults as they unfold over time. This dissertation also revealed the direction of two
processes of intergenerational support and health and indicated how change in support predicted
change in health over time.
One important lesson from this dissertation is the gendered nature of source of
intergenerational support and its influence on physical functioning among older parents over
time. The patrilineal feature in the family support system still exists, and the role of sons in
providing support and care to aging parents remains significant. The inclusion of support and
care provided by daughters-in-law further informs the gender-specific support division in the
family support system. Social work practitioners need to be aware of different sources of family
support and the differential effect of corresponding support. Findings regarding the protective
function of support from daughters-in-law on health are quite different compared to research in
the Western context, where support and care from daughters play a major role in the health of
older adults.
Another important lesson from this dissertation is that intergenerational instrumental
support is multidimensional and each dimension presents a unique course of growth over time.
For example, receipt of household chores decreased over time, whereas receipt of personal care
increased over time. Provision of household chores and personal care both increased and
decreased. Providers must assess the stability of support from children; having support from
children at one time does not necessarily mean continuing and stable support at a later time.
Differences in the trajectories of support also suggest the need for targeted formal help to
74
complement existing family support. Given the limited resources of family systems, sole reliance
on family members for support in later life might place too much of a burden on family
members. Social work practitioners need to advocate for more support to help adult children
sustain support and care for their older parents, such as advocating for more family-friendly
work policies to allow adult children to fulfill their filial responsibilities with more flexibility,
other formal sources of support and care for aging parents such as senior centers and adult day
care centers, and supplementary support services to prevent role overload and burnout among
adult children, including respite services and peer support groups.
One other important lesson from this dissertation is that change in intergenerational
support is correlated with change in perceptions of late-life health; intergenerational support
predicts perceptions of health at a later time. This emphasizes the importance of the role of
intergenerational support in promoting health among older adults. Social work practitioners
should help raise awareness and educating the general public about the health benefits of
intergenerational support in later life, equip themselves with such knowledge and perspective to
conduct comprehensive and culturally sensitive assessments, and design programs and
interventions that reinforce family connections and help maintain and strengthen
intergenerational support. Interventions and programs should not be limited to the family setting
and should emphasize the importance of intergenerational features, such as culturally sensitive
health promotion programs and senior centers that involve young volunteers and
intergenerational programs that help younger generations transfer useful health information to
older generation.
One other important lesson from this dissertation is its emphasis on the important role
that culture plays in shaping interactions and exchanges across generations. As societies continue
75
to evolve, the traditional cultural value of filial piety is subject to challenges and changes in the
dynamic context of East Asian countries. Social work practitioners need to have a good
understanding of traditional values and beliefs and prepare themselves to work with families that
possess different and even contradictory values. Education programs that help facilitate the
dialogue between younger and older generations and intergenerational programs that promote
open communication and positive interaction between the two generations will help adult
children and older parents cope and adapt in new situations.
It is also important to advocate for changes at a policy level. In developing countries such
as China, formal care systems remain underdeveloped and the need for long-term care will
increase drastically in the near future. Currently, family support for older adults has been
reinforced by legislation. It is wise to recognize the strengths and potential weaknesses of family
support system in terms of ensuring late-life health and well-being. This dissertation provides a
picture of changing family support and health in later life. Social work practitioners should also
advocate at the macro level and work with policy makers with the mindset of preserving and
strengthening the valuable family system and initiating and promoting a dialogue regarding the
need for a more established formal care system.
76
References
Allen, K. R., Blieszner, R., & Roberto, K. A. (2000). Families in the middle and later years: A
review and critique of research in the 1990s. Journal of Marriage and Family, 62, 911–
926. doi:10.1111/j.1741-3737.2000.00911.x
Antonucci, T. C. (1985). Personality characteristics, social support, and social behavior. In R.
Binstock & E. Shanas (Eds.), Handbook of aging and the social sciences (pp. 94–128).
New York, NY: Nostrand Reinhold.
Antonucci, T. C., Ajrouch, K. J., & Birditt, K. S. (2014). The convoy model: Explaining social
relations from a multidisciplinary perspective. Gerontologist 54, 82–91.
doi:10.1093/geront/gnt118
Antonucci, T. C., & Akiyama, H. (1987). Social networks in adult life and a preliminary
examination of the convoy model. Journal of Gerontology, 42, 519–527.
doi:10.1093/geronj/42.5.519
Antonucci, T. C., Akiyama, H., & Takahashi, K. (2004). Attachment and close relationships
across the life span. Attachment & Human Development, 6, 353–370.
doi:10.1080/1461673042000303136
Antonucci, T. C., Birditt, K. S., Ajrouch, K. (2011). Convoys of social relations: Past, present,
and future. In K. L. Fingerman, C. A. Berg, J. Smith, & T. C. Antonucci (Eds.),
Handbook of lifespan development (pp. 161–182). New York, NY: Springer.
Antonucci, T. C., Birditt, K. S., & Webster, N. J. (2010). Social relations and mortality: A more
nuanced approach. Journal of Health Psychology, 15, 649–659.
doi:10.1177/1359105310368189
77
Antonucci, T. C., Fuhrer, R., & Dartigues, J.-F. (1997). Social relations and depressive
symptomatology in a sample of community-dwelling French older adults. Psychology
and Aging, 12, 189–195. doi:10.1037/0882-7974.12.1.189
Atchley, R. C., & Scala, M. A. (1998). Long-range antecedents of functional capability in later
life. Journal of Aging and Health, 10, 3–19. doi:10.1177/089826439801000101
Auslander, G. K., & Litwin, H. (1991). Social networks, social support, and self-ratings of health
among the elderly. Journal of Aging and Health, 3, 493–510.
doi:10.1177/089826439100300404
Avlund, K., Lund, R., Holstein, B. E., & Due, P. (2004). Social relations as determinants of onset
of disability in aging. Archives of Gerontology and Geriatrics, 38, 85–99.
doi:10.1016/j.archger.2003.08.003
Bailis, D. S., Segall, A., & Chipperfield, J. G. (2003). Two views of self-rated general health
status. Social Science & Medicine, 56, 203–217. doi:10.1016/S0277-9536(02)00020-5
Barger, S. D. (2013). Social integration, social support and mortality in the US National Health
Interview Survey. Psychosomatic Medicine, 75, 510–517.
doi:10.1097/PSY.0b013e318292ad99
Beckett, M., Goldman, N., Weinstein, M., Lin, I.-F., & Chuang, Y.-L. (2002). Social
environment, life challenge, and health among the elderly in Taiwan. Social Science &
Medicine, 55, 191–209. doi:10.1016/S0277-9536(01)00161-7
Béland, F., & Zunzunegui, M. V. (1999). Predictors of functional status in older people living at
home. Age and Ageing, 28, 153–159. doi:10.1093/ageing/28.2.153
Bengtson, V. L., Giarrusso, R., Mabry, J. B., & Silverstein, M. (2002). Solidarity, conflict, and
ambivalence: Complementary or competing perspectives on intergenerational
78
relationships? Journal of Marriage and Family, 64, 568–576. doi:10.1111/j.1741-
3737.2002.00568.x
Bengtson, V. L., Olander, E. B., & Haddad, A. A. (1976). The generation gap and aging family
members: Toward a conceptual model. In J. F. Gubrium (Ed.), Time, roles, and self in old
age (pp. 237–263). New York, NY: Human Sciences Press.
Bentler, P. M. (1992). On the fit of models to covariances and methodology to the Bulletin.
Psychological Bulletin, 112, 400–404. doi:10.1037/0033-2909.112.3.400
Bergeman, C. S., Neiderhiser, J. M., Pedersen, N. L., & Plomin, R. (2001). Genetic and
environmental influences on social support in later life: A longitudinal analysis.
International Journal of Aging & Human Development, 53, 107–135.
doi:10.2190/CTKR-F6FF-Q8AW-E68V
Berkman, L. F., & Glass, T. (2000). Social integration, social networks, social support, and
health. In L. F. Berkman & I. Kawachi (Eds.), Social epidemiology (pp. 137–173). New
York, NY: Oxford University Press.
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to
health: Durkheim in the new millennium. Social Science & Medicine, 51, 843–857.
doi:10.1016/S0277-9536(00)00065-4
Boerner, K., & Reinhardt, J. P. (2003). Giving while in need: Support provided by disabled older
adults. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences,
58, S297–S304. doi:10.1093/geronb/58.5.S297
Bossé, R., Aldwin, C. M., Levenson, M. R., Spiro, A., III, & Mroczek, D. K. (1993). Change in
social support after retirement: Longitudinal findings from the Normative Aging Study.
Journal of Gerontology, 48, P210–P217.
79
Boult, C., Kane, R. L., Louis, T. A., Boult, L., & McCaffrey, D. (1994). Chronic conditions that
lead to functional limitation in the elderly. Journal of Gerontology, 49, M28–M36.
doi:10.1093/geronj/49.1.M28
Braveman, P., Egerter, S., & Williams, D. R. (2011). The social determinants of health: Coming
of age. Annual Review of Public Health, 32, 381–398. doi:10.1146/annurev-publhealth-
031210-101218
Brown, S. L., Nesse, R. M., Vinokur, A. D., & Smith, D. M. (2003). Providing social support
may be more beneficial than receiving it: Results from a prospective study of mortality.
Psychological Science, 14, 320–327. doi:10.1111/1467-9280.14461
Buhi, E. R., Goodson, P., & Neilands, T. B. (2007). Structural equation modeling: A primer for
health behavior researchers. American Journal of Health Behavior, 31, 74–85.
doi:10.5993/AJHB.31.1.8
Cacioppo, J. T., Hawkley, L. C., & Thisted, R. A. (2010). Perceived social isolation makes me
sad: 5-year cross-lagged analyses of loneliness and depressive symptomatology in the
Chicago Health, Aging, and Social Relations Study. Psychology and Aging, 25, 453–463.
doi:10.1037/a0017216
Camacho, T. C., Strawbridge, W. J., Cohen, R. D., & Kaplan, G. A. (1993). Functional ability in
the oldest old: Cumulative impact of risk factors from the preceding two decades. Journal
of Aging and Health, 5, 439–454. doi:10.1177/089826439300500402
Caplan, G. (1974). Support systems and community mental health. New York, NY: Behavioral
Publications.
Carr, D., & Springer, K. W. (2010). Advances in families and health research in the 21st century.
Journal of Marriage and Family, 72, 743–761. doi:10.1111/j.1741-3737.2010.00728.x
80
Cassel, J. (1976). The contribution of the social environment to host resistance: The Fourth
Wade Hampton Frost Lecture. American Journal of Epidemiology, 104, 107–123.
Chen, F., & Liu, G. (2009). Population aging in China. In P. Uhlenberg (Ed.), International
handbook of population aging (pp. 157–172). New York, NY: Springer.
Chen, F., & Liu, G. (2012). The health implications of grandparents caring for grandchildren in
China. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences,
67, 99–112. doi:10.1093/geronb/gbr132
Chen, S. W.-H., & Davenport, D. S. (2005). Cognitive-behavioral therapy with Chinese
American clients: Cautions and modifications. Psychotherapy: Theory, Research,
Practice, Training, 42, 101–110. doi:10.1037/0033-3204.42.1.101
Chen, X., & Silverstein, M. (2000). Intergenerational social support and the psychological well-
being of older parents in China. Research on Aging, 22, 43–65.
doi:10.1177/0164027500221003
Cheng, S.-T., Lee, C. K. L., Chan, A. C. M., Leung, E. M. F., & Lee, J.-J. (2009). Social network
types and subjective well-being in Chinese older adults. Journals of Gerontology, Series
B: Psychological Sciences and Social Sciences, 64, 713–722. doi:10.1093/geronb/gbp075
Cheng, Y. H., Chi, I., Boey, K. W., Ko, L. S. F., & Chou, K. L. (2002). Self-rated economic
condition and the health of elderly persons in Hong Kong. Social Science & Medicine,
55, 1415–1424. doi:10.1016/S0277-9536(01)00271-4
Chi, I., & Chou, K.-L. (2001). Social support and depression among elderly Chinese people in
Hong Kong. International Journal of Aging & Human Development, 52, 231–252.
doi:10.2190/V5K8-CNMG-G2UP-37QV
81
Chou, R. J.-A. (2010). Willingness to live in eldercare institutions among older adults in urban
and rural China: A nationwide study. Ageing & Society, 30, 583–608.
doi:10.1017/S0144686X09990596
Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic Medicine, 38, 300–
314. doi:10.1097/00006842-197609000-00003
Cohen, S. (2004). Social relationships and health. American Psychologist, 59, 676–684.
doi:10.1037/0003-066X.59.8.676
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis.
Psychological Bulletin, 98, 310–357. doi:10.1037/0033-2909.98.2.310
Cong, Z., & Silverstein, M. (2008). Intergenerational support and depression among elders in
rural China: Do daughters-in-law matter? Journal of Marriage and Family, 70, 599–612.
doi:10.1111/j.1741-3737.2008.00508.x
Corhan, S., & Antonucci, T. C. (1989). Friends as a source of social support in old age. In R. G.
Adams & R. Blieszner (Eds.), Older adult friendships: Structure and process (pp. 129–
146). Newbury Park, CA: Sage.
Cornman, J. C., Goldman, N., Glei, D. A., Weinstein, M., & Chang, M.-C. (2003). Social ties
and perceived support: Two dimensions of social relationships and health among the
elderly in Taiwan. Journal of Aging and Health, 15, 616–644.
doi:10.1177/0898264303256215
Cornman, J. C., Lynch, S. M., Goldman, N., Weinstein, M., & Lin, H.-S. (2004). Stability and
change in the perceived social support of older Taiwanese adults. Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 59, S350–S357.
doi:10.1093/geronb/59.6.S350
82
Craigs, C. L., Twiddy, M., Parker, S. G., & West, R. M. (2014). Understanding causal
associations between self-rated health and personal relationships in older adults: A
review of evidence from longitudinal studies. Archives of Gerontology and Geriatrics,
59, 211–226. doi:10.1016/j.archger.2014.06.009
Dannefer, D., & Kelley-Moore, J. A. (2009). Theorizing the life course: New twists in the paths.
In V. Bengtson, M. Silverstein, N. Putney, & D. Gans (Eds.), Handbook of theories of
aging (2nd ed., pp. 389–411). New York, NY: Springer.
Deeg, D. J. H., & Kriegsman, D. M. W. (2003). Concepts of self-rated health: Specifying the
gender difference in mortality risk. Gerontologist, 43, 376–386.
doi:10.1093/geront/43.3.376
Dong, X., & Simon, M. A. (2010). Health and aging in a Chinese population: Urban and rural
disparities. Geriatrics & Gerontology International, 10, 85–93. doi:10.1111/j.1447-
0594.2009.00563.x
Everard, K. M., Lach, H. W., Fisher, E. B., & Baum, M. C. (2000). Relationship of activity and
social support to the functional health of older adults. Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 55, S208–S212.
doi:10.1093/geronb/55.4.S208
Ferrer, E., & McArdle, J. J. (2010). Longitudinal modeling of developmental changes in
psychological research. Current Directions in Psychological Science, 19, 149–154.
doi:10.1177/0963721410370300
Fiori, K. L., Antonucci, T. C., & Akiyama, H. (2008). Profiles of social relations among older
adults: A cross-cultural approach. Ageing & Society, 28, 203–231.
doi:10.1017/S0144686X07006472
83
Franks, P., Campbell, T. L., & Shields, C. G. (1992). Social relationships and health: The relative
roles of family functioning and social support. Social Science & Medicine, 34, 779–788.
doi:10.1016/0277-9536(92)90365-W
Fratiglioni, L., Paillard-Borg, S., & Winblad, B. (2004). An active and socially integrated
lifestyle in late life might protect against dementia. Lancet Neurology, 3, 343–353.
doi:10.1016/S1474-4422(04)00767-7
García, E. L., Banegas, J. R., Pérez-Regadera, A. G., Cabrera, R. H., & Rodríguez-Artalejo, F.
(2005). Social network and health-related quality of life in older adults: A population-
based study in Spain. Quality of Life Research, 14, 511–520. doi:10.1007/s11136-004-
5329-z
Gruenewald, T. L., & Seeman, T. E. (2010). Social support and physical health: Links and
mechanisms. In W. D. Gentry (Ed.), Handbook of behavioral medicine: Methods and
applications (pp. 225–236). New York, NY: Springer.
Guo, M., Aranda, M. P., & Silverstein, M. (2009). The impact of out-migration on the inter-
generational support and psychological wellbeing of older adults in rural China. Ageing
& Society, 29, 1085–1104. doi:10.1017/S0144686X0900871X
Guo, M., Chi, I., & Silverstein, M. (2011). Family as a context: The influence of family
composition and family geographic dispersion on intergenerational relationships among
Chinese elderly. International Journal of Social Welfare, 20, S18–S29.
doi:10.1111/j.1468-2397.2011.00793.x
Guo, M., Chi, I., & Silverstein, M. (2012). The structure of intergenerational relations in rural
China: A latent class analysis. Journal of Marriage and Family, 74, 1114–1128.
doi:10.1111/j.1741-3737.2012.01014.x
84
He, W., Sengupta, M., Zhang, K., & Guo, P. (2007). Health and health care of the older
population in urban and rural China: 2000. Washington, DC: U.S. Government Printing
Office.
Hinterlong, J. E., Morrow-Howell, N., & Rozario, P. A. (2007). Productive engagement and late
life physical and mental health: Findings from a nationally representative panel study.
Research on Aging, 29, 348–370. doi:10.1177/0164027507300806
Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A
meta-analytic review. PLOS Medicine, 7, e1000316. doi:10.1371/journal.pmed.1000316
Johnson, C. L. (2000). Perspectives on American kinship in the later 1990s. Journal of Marriage
and Family, 62, 623–639. doi:10.1111/j.1741-3737.2000.00623.x
Johnson, R. J., & Wolinsky, F. D. (1993). The structure of health status among older adults:
Disease, disability, functional limitation, and perceived health. Journal of Health and
Social Behavior, 34, 105–121. doi:10.2307/2137238
Jylhä, M. (2009). What is self-rated health and why does it predict mortality? Towards a unified
conceptual model. Social Science & Medicine, 69, 307–316.
doi:10.1016/j.socscimed.2009.05.013
Kahn, R. L., & Antonucci, T. C. (1980). Convoys over the life course: Attachment, roles and
social support. In P. B. Baltes & O. Brim (Eds.), Life span development and behavior (pp.
253–286). New York, NY: Academic Press.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New
York, NY: Guilford Press.
85
Kosberg, J. I., Kaufman, A. V., Burgio, L. D., Leeper, J. D., & Sun, F. (2007). Family caregiving
to those with dementia in rural Alabama: Racial similarities and differences. Journal of
Aging and Health, 19, 3–21. doi:10.1177/0898264306293604
Krause, N. (2004). Stressors arising in highly valued roles, meaning in life, and the physical
health status of older adults. Journals of Gerontology, Series B: Psychological Sciences
and Social Sciences, 59, S287–S297. doi:10.1093/geronb/59.5.S287
Krause, N., Herzog, A. R., & Baker, E. (1992). Providing support to others and well-being in
later life. Journal of Gerontology, 47, P300–P311. doi:10.1093/geronj/47.5.P300
Krochalk, P. C., Li, Y., & Chi, I. (2008). Widowhood and self-rated health among Chinese
elders: The effect of economic condition. Australasian Journal on Ageing, 27, 26–32.
doi:10.1111/j.1741-6612.2007.00269.x
Leopold, T., & Raab, M. (2011). Short-term reciprocity in late parent-child relationships.
Journal of Marriage and Family, 73, 105–119. doi:10.1111/j.1741-3737.2010.00792.x
Li, S., Song, L., & Feldman, M. W. (2009). Intergenerational support and subjective health of
older people in rural China: A gender-based longitudinal study. Australasian Journal on
Ageing, 28, 81–86. doi:10.1111/j.1741-6612.2009.00364.x
Li, Y., Aranda, M. P., & Chi, I. (2007). Health and life satisfaction of ethnic minority older
adults in mainland China: Effects of financial strain. International Journal of Aging &
Human Development, 64, 361–379. doi:10.2190/9053-2G67-4X22-1211
Li, Y., & Chi, I. (2011). Correlates of physician visits among older adults in China: The effects
of family support. Journal of Aging and Health, 23, 933–953.
doi:10.1177/0898264311401390
86
Lin, I.-F., Goldman, N., Weinstein, M., Lin, Y.-H., Gorrindo, T., & Seeman, T. (2003). Gender
differences in adult children’s support of their parents in Taiwan. Journal of Marriage
and Family, 65, 184–200. doi:10.1111/j.1741-3737.2003.00184.x
Litwak, E. (1985). Helping the elderly: The complementary roles of informal networks and
formal systems. New York, NY: Guilford Press.
Litwin, H., Shrira, A., & Shmotkin, D. (2012). Self-reported functional status among the old-old:
A comparison of two Israeli cohorts. Journal of Aging and Health, 24, 846–862.
doi:10.1177/0898264312436714
Liu, X., Liang, J., & Gu, S. (1995). Flows of social support and health status among older
persons in China. Social Science & Medicine, 41, 1175–1184. doi:10.1016/0277-
9536(94)00427-U
Liu, X., Liang, J., Muramatsu, N., & Sugisawa, H. (1995). Transitions in functional status and
active life expectancy among older people in Japan. Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 50, S383–S394.
doi:10.1093/geronb/50B.6.S383
Long, J. S., & Pavalko, E. K. (2004). The life course of activity limitations: Exploring indicators
of functional limitations over time. Journal of Aging and Health, 16, 490–516.
doi:10.1177/0898264304265776
Lou, V. W. Q., Lu, N., Xu, L., & Chi, I. (2013). Grandparent–grandchild family capital and self-
rated health of older rural Chinese adults: The role of the grandparent–parent
relationship. Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 68, 599–608. doi:10.1093/geronb/gbt040
Mao, W., & Chi, I. (2011). Filial piety of children as perceived by aging parents in China.
87
International Journal of Social Welfare, 20, S99–S108. doi:10.1111/j.1468-
2397.2011.00826.x
Marmot, M. (2005). Social determinants of health inequalities. Lancet, 365, 1099–1104.
doi:10.1016/S0140-6736(05)71146-6
McArdle, J. J. (1998). Modeling longitudinal data by latent growth curve methods. In G. A.
Marcoulides (Ed.), Modern methods for business research (pp. 359–406). Mahwah, NJ:
Lawrence Erlbaum.
McArdle, J. J. (2009). Latent variable modeling of differences and changes with longitudinal
data. Annual Review of Psychology, 60, 577–605.
doi:10.1146/annurev.psych.60.110707.163612
McLaughlin, S. J., Connell, C. M., Heeringa, S. G., Li, L. W., & Roberts, J. S. (2010).
Successful aging in the United States: Prevalence estimates from a national sample of
older adults. Journals of Gerontology, Series B: Psychological Sciences and Social
Sciences, 65, 216–226. doi:10.1093/geronb/gbp101
Mendes de Leon, C. F., Glass, T. A., Beckett, L. A., Seeman, T. E., Evans, D. A., & Berkman, L.
F. (1999). Social networks and disability transitions across eight intervals of yearly data
in the New Haven EPESE. Journals of Gerontology, Series B: Psychological Sciences
and Social Sciences, 54, S162–S172.
Menec, V. H., Shooshtari, S., & Lambert, P. (2007). Ethnic differences in self-rated health
among older adults: A cross-sectional and longitudinal analysis. Journal of Aging and
Health, 19, 62–86. doi:10.1177/0898264306296397
Miller, B., & McFall, S. (1991). Stability and change in the informal task support network of
frail older persons. Gerontologist, 31, 735–745. doi:10.1093/geront/31.6.735
88
Minkler, M., & Langhauser, C. (1988). Assessing health differences in an elderly population: A
five-year follow-up. Journal of the American Geriatrics Society, 36, 113–118.
doi:10.1111/j.1532-5415.1988.tb01779.x
Mjelde-Mossey, L. A., Chi, I., Lubben, J., & Lou, V. W. Q. (2009). Relationship between
productive activities, family relations, and aging well for elders in China. Journal of
Ethic & Cultural Diversity in Social Work, 18, 276–292.
doi:10.1080/15313200903310742
Molm, L. D., & Cook, K. S. (1995). Social exchange and exchange networks. In K. S. Cook, G.
A. Fine, & J. S. House (Eds.), Sociological perspectives on social psychology (pp. 209–
235). Needham Heights, MA: Allyn and Bacon.
Muthén, L. K., & Muthén, B. O. (2010). Mplus user’s guide (6th ed.). Los Angeles, CA: Muthén
& Muthén.
National Bureau of Statistics of China. (2011, April 28). Communiqué of the National Bureau of
Statistics of People’s Republic of China on major figures of the 2010 population census.
Retrieved from
http://www.stats.gov.cn/english/NewsEvents/201104/t20110428_26449.html
National People’s Congress of the People’s Republic of China. (1996, August 29). Law of the
People’s Republic of China on Protection of the Rights and Interests of the Elderly.
Retrieved from http://www.china.org.cn/english/government/207403.htm
Novak, M. (2006). Issues in aging. Boston, MA: Pearson.
Orth, U., Berking, M., Walker, N., Meier, L. L., & Znoj, H. (2008). Forgiveness and
psychological adjustment following interpersonal transgressions: A longitudinal analysis.
Journal of Research in Personality, 42, 365–385. doi:10.1016/j.jrp.2007.07.003
89
Pinquart, M. (2001). Correlates of subjective health in older adults: A meta-analysis. Psychology
and Aging, 16, 414–426. doi:10.1037/0882-7974.16.3.414
Pinquart, M., & Sörensen, S. (2005). Ethnic differences in stressors, resources, and psychological
outcomes of family caregiving: A meta-analysis. Gerontologist, 45, 90–106.
doi:10.1093/geront/45.1.90
Ritzer, G., & Goodman, D. J. (2004). Modern sociological theory (6th ed.). New York, NY:
McGraw-Hill.
Rowe, J. W., & Kahn, R. L. (1997). Successful aging. Gerontologist, 37, 433–440.
doi:10.1093/geront/37.4.433
Ryan, A. K., & Willits, F. K. (2007). Family ties, physical health, and psychological well-being.
Journal of Aging and Health, 19, 907–920. doi:10.1177/0898264307308340
Sarason, I. G., & Sarason, B. R. (2009). Social support: Mapping the construct. Journal of Social
and Personal Relationships, 26, 113–120. doi:10.1177/0265407509105526
Seeman, T. E., Berkman, L. F., Charpentier, P. A., Blazer, D. G., Albert, M. S., & Tinetti, M. E.
(1995). Behavioral and psychosocial predictors of physical performance: MacArthur
studies of successful aging. Journals of Gerontology, Series A: Biological Sciences and
Medical Sciences, 50, M177–M183. doi:10.1093/gerona/50A.4.M177
Seeman, T. E., Bruce, M. L., & McAvay, G. J. (1996). Social network characteristics and onset
of ADL disability: MacArthur studies of successful aging. Journals of Gerontology,
Series B: Psychological Sciences and Social Sciences, 51, S191–S200.
doi:10.1093/geronb/51B.4.S191
Selig, J. P., & Little, T. D. (2012). Autoregressive and cross-lagged panel analysis for
longitudinal data. In B. Laursen, T. D. Little, & N. A. Card (Eds.), Handbook of
90
developmental research methods (pp. 265–278). New York, NY: Guilford Press.
Shaw, B. A., Krause, N., Liang, J., & Bennett, J. (2007). Tracking changes in social relations
throughout late life. Journals of Gerontology, Series B: Psychological Sciences and
Social Sciences, 62, S90–S99. doi:10.1093/geronb/62.2.S90
Shi, L. (1993). Family financial and household support exchange between generations: A survey
of Chinese rural elderly. Gerontologist, 33, 468–480. doi:10.1093/geront/33.4.468
Silverstein, M., Cong, Z., & Li, S. (2006). Intergenerational transfers and living arrangements of
older people in rural China: Consequences for psychological well-being. Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 61, S256–S266.
doi:10.1093/geronb/61.5.S256
Silverstein, M., & Giarrusso, R. (2010). Aging and family life: A decade review. Journal of
Marriage and Family, 72, 1039–1058. doi:10.1111/j.1741-3737.2010.00749.x
Song, L., Li, S., Zhang, W., & Feldman, M. W. (2008). Intergenerational support and self-rated
health of the elderly in rural China: An investigation in Chaohu, Anhui province. In
Zeng, Y., Poston, D. L., Jr., Vlosky, D. A., & Gu, D. (Eds), Healthy longevity in China:
Demographic, socioeconomic, and psychological dimensions (pp. 235–249). New York,
NY: Springer.
Stephens, C., Alpass, F., Towers, A., & Stevenson, B. (2011). The effects of types of social
networks, perceived social support, and loneliness on the health of older people:
Accounting for the social context. Journal of Aging and Health, 23, 887–911.
doi:10.1177/0898264311400189
Strawbridge, W. J., Camacho, T. C., Cohen, R. D., & Kaplan, G. A. (1993). Gender differences
in factors associated with change in physical functioning in old age: A 6-year
91
longitudinal study. Gerontologist, 33, 605–609. doi:10.1093/geront/33.5.603
Sun, R. (2004). Worry about medical care, family support, and depression of the elders in urban
China. Research on Aging, 26, 559–585. doi:10.1177/0164027504266467
Thoits, P. A. (2011). Mechanisms linking social ties and support to physical and mental health.
Journal of Health and Social Behavior, 52, 145–161. doi:10.1177/0022146510395592
Uchino, B. N. (2004). Social support and physical health: Understanding the health
consequences of relationships. New Haven, CT: Yale University Press.
Uchino, B. N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). The relationship between social
support and physiological processes: A review with emphasis on underlying mechanisms
and implications for health. Psychological Bulletin, 119, 488–531. doi:10.1037/0033-
2909.119.3.488
Umberson, D., Crosnoe, R., & Reczek, C. (2010). Social relationships and health behaviors
across the life course. Annual Review of Sociology, 36, 139–157. doi:10.1146/annurev-
soc-070308-120011
Umberson, D., & Montez, J. K. (2010). Social relationships and health: A flashpoint for health
policy. Journal of Health and Social Behavior, 51, S54–S66.
doi:10.1177/0022146510383501
Unger, J. B., McAvay, G., Bruce, M. L., Berkman, L., & Seeman, T. (1999). Variation in the
impact of social network characteristics on physical functioning in elderly persons:
MacArthur studies of successful aging. Journals of Gerontology, Series B: Psychological
Sciences and Social Sciences, 54, S245–S251. doi:10.1093/geronb/54B.5.S245
92
van Tilburg, T. (1998). Losing and gaining in old age: Changes in personal network size and
social support in a four-year longitudinal study. Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 53, S313–S323.
Vaux, A. (1988). Social support: Theory, research, and intervention. New York, NY: Praeger.
Voelkle, M. C. (2007). Latent growth curve modeling as an integrative approach to the analysis
of change. Psychology Science, 49, 375–414.
Wang, C.-W., Chan, C. L. W., Ho, A. H. Y., & Xiong, Z. (2008). Social networks and health-
related quality of life among Chinese older adults with vision impairment. Journal of
Aging and Health, 20, 804–823. doi:10.1177/0898264308321083
Wang, P., & Li, S. (2011). Dai ji zhi chi dui nong cun lao ren sheng huo zi li neng li de zong
xiang ying xiang [A longitudinal study of the effect of intergenerational support on
activity of daily living of rural elderly]. Population & Economics, 2, 13–22.
Weinberger, M., Tierney, W. M., Booher, P., & Hiner, S. L. (1990). Social support, stress and
functional status in patients with osteoarthritis. Social Science & Medicine, 30, 503–508.
doi:10.1016/0277-9536(90)90353-T
Weiss, R. S. (1974). The provisions of social relationships. In Z. Rubin (Ed.), Doing unto others
(pp. 17–26). Englewood Cliffs, NJ: Prentice Hall.
White, A. M., Philogene, G. S., Fine, L., & Sinha, S. (2009). Social support and self-reported
health status of older adults in the United States. American Journal of Public Health, 99,
1872–1878. doi:10.2105/AJPH.2008.146894
Wolinsky, F. D., Miller, T. R., Malmstrom, T. K., Miller, J. P., Schootman, M., Andresen, E. M.,
& Miller, D. K. (2008). Self-rated health: Changes, trajectories, and their antecedents
93
among African Americans. Journal of Aging and Health, 20, 143–158.
doi:10.1177/0898264307310449
Wong, O. M. (2005). Gender and intimate caregiving for the elderly in Hong Kong. Journal of
Aging Studies, 19, 375–391. doi:10.1016/j.jaging.2004.07.007
Zeng, Y., Vaupel, J. W., Xiao, Z., Zhang, C., & Liu, Y. (2002). Sociodemographic and health
profiles of the oldest old in China. Population and Development Review, 28, 251–273.
doi:10.1111/j.1728-4457.2002.00251.x
Zhang, W., Li, S., & Feldman, M. W. (2005). Gender difference in activity of daily living of the
elderly in rural China: Evidence from Chaohu. Journal of Women & Aging, 17, 73–89.
doi:10.1300/J074v17n03_06
Zimmer, Z., & Kwong, J. (2004). Socioeconomic status and health among older adults in rural
and urban China. Journal of Aging and Health, 16, 44–70. doi:10.1177/0898264303260440
Zunzunegui, M. V., Koné, A., Johri, M., Béland, F., Wolfson, C., & Bergman, H. (2004). Social
networks and self-rated health in two French-speaking Canadian community dwelling
populations over 65. Social Science & Medicine, 58, 2069–2081.
doi:10.1016/j.socscimed.2003.08.005
Zuo, D., Li, S., Mao, W., & Chi, I. (2014). Family caregiving for older parents at the end of life
in China’s rural Anhui province. Canadian Journal on Aging, 33, 448–461.
doi:10.1017/S0714980814000373
94
Tables

Table 2.1
Characteristics of Older Adults with at Least One Son and One
Daughter at Baseline (N = 1,322)
n % or M (SD) Range
Age 1,322 70.56 (7.11) 60–92
Female 1,322 51.70 0–1
No formal education 1,322 78.60 0–1
Married 1,322 54.90 0–1
Satisfaction with economic situation 1,287 68.50 0–1
Chronic conditions 1,320 1.73 (1.48) 0–8
Number of sons 1,322 2.24 (1.08) 1–7
Number of daughters 1,322 2.16 (1.09) 1–7
95

Table 2.2
Means and Standard Deviations of Gendered Intergenerational Support Receipt and
Functional Limitations among Older Adults from 2001 to 2009
Time 1 Time 2 Time 3 Time 4
n = 1,322 n = 1,080 n = 834 n = 631
M (SD) M (SD) M (SD) M (SD)
Instrumental support from
Sons 1.44 (2.92) 1.50 (3.11) 1.37 (2.89) 1.53 (2.97)
Daughters 1.57 (3.13) 1.51 (3.48) 1.15 (2.58) 1.08 (2.24)
Daughters-in-law 1.88 (3.11) 1.79 (3.31) 1.69 (3.00) 1.59 (3.15)
Spouse (Control) 2.33 (3.04) 2.58 (3.23) 2.33 (3.18) 2.21 (3.32)
Maximum emotional cohesion
Sons 4.56 (1.50) 4.58 (1.48) 4.54 (1.49) 4.54 (1.44)
Daughters 4.82 (1.35) 4.92 (1.31) 4.79 (1.30) 4.78 (1.38)
Functional limitations 3.52 (5.09) 3.55 (5.27) 4.44 (5.55) 3.13 (5.44)
96

Table 2.3
Autoregressive and Cross-Lagged Panel Analysis for Gendered
Intergenerational Support and Functional Limitations among Older Adults:
Unstandardized Parameter Estimates with Adjustments of Control Variables
Estimate SE
Autoregressive effects
Instrumental support from sons .174*** .016
Instrumental support from daughters .148*** .015
Instrumental support from daughters-in-law .177*** .017
Instrumental support from spouse .279*** .017
Maximum emotional cohesion with sons .235*** .017
Maximum emotional cohesion with daughters .219*** .018
Functional limitations .455*** .027
Cross-lagged effects: Support to functional limitations
Instrumental support from sons .297** .105
Instrumental support from daughters -.032 .051
Instrumental support from daughters-in-law -.247* .102
Instrumental support from spouse -.005 .054
Maximum emotional cohesion with sons .002 .064
Maximum emotional cohesion with daughters .015 .074
Cross-lagged effects: Functional limitations to support
Instrumental support from sons .179*** .015
Instrumental support from daughters .110*** .014
Instrumental support from daughters-in-law .184*** .016
Instrumental support from spouse .115*** .018
Maximum emotional cohesion with sons -.007 .006
Maximum emotional cohesion with daughters -.028*** .006
Note. Model fit statistics for the adjusted model: χ
2
(457) = 1390.088, p < .001; CFI = .900;
RMSEA = .039; SRMR = .063.
*p < .05. **p < .01. ***p < .001.

97

Table 3.1
Characteristics of Older Adults with at least One Child at Baseline (N =
1,636)
n % or M (SD) Range
Age 1,636 70.93 (7.31) 60–92
Female 1,636 52.40 0–1
No formal education 1,635 79.20 0–1
Married 1,636 52.80 0–1
Satisfaction with economic situation 1,583 65.40 0–1
Number of children 1,634 4.01 (1.60) 1–10
Functional limitations 1,624 5.89 (7.35) 0–30
Depressive symptoms 1,578 7.49 (2.62) 0–16
98

Table 3.2
Means and Standard Deviations of Multidimensional Intergenerational Instrumental
Support and Self-Rated Health from 2001 to 2012
Time 1 Time 2 Time 3 Time 4 Time 5
n = 1,636 n = 1,324 n = 1,016 n = 765 n = 567
M (SD) M (SD) M (SD) M (SD) M (SD)
Received HH 3.32 (3.80) 3.05 (3.93) 2.66 (3.46) 2.42 (3.58) 2.16 (3.78)
Received PC 1.44 (3.09) 1.60 (3.23) 1.48 (2.99) 1.57 (3.05) 1.71 (3.55)
Provided HH 2.36 (3.41) 2.80 (3.78) 2.59 (3.50) 2.79 (3.76) 1.02 (2.03)
Provided PC 1.00 (2.30) 1.24 (2.66) 1.53 (2.81) 1.16 (2.47) 0.54 (1.44)
Self-rated health 2.05 (0.85) 2.19 (0.93) 2.03 (0.89) 2.00 (0.83) 2.05 (0.87)
Note. HH = household help; PC = personal care.
99

Table 3.3
Bivariate Latent Growth Curve Model for Receipt of Household Chore Help
and Self-Rated Health: Unstandardized Parameter Estimates with and
without Adjustments for Control Variables
Unadjusted Adjusted
Estimate SE Estimate SE
Means
HH intercept 3.264*** 0.094 3.152*** 0.091
SRH intercept 2.093*** 0.020 2.100*** 0.017
HH slope -0.191*** 0.045 -0.115* 0.047
SRH slope -0.042** 0.008 -0.056*** 0.009
Covariances
HH intercept ↔ HH slope -0.655* 0.263 -0.526* 0.250
SRH intercept ↔ SRH slope -0.021* 0.010 -0.003 0.008
HH intercept ↔ SRH intercept -0.570*** 0.077 -0.153* 0.061
HH slope ↔ SRH slope -0.048*** 0.013 -0.029* 0.012
SRH intercept ↔ HH slope 0.051 0.035 -0.012 0.029
HH intercept ↔ SRH slope 0.092** 0.032 0.031 0.028
Variances
HH intercept 4.006*** 0.669 2.461*** 0.623
SRH intercept 0.294*** 0.028 0.148*** 0.022
HH slope 0.468*** 0.122 0.414*** 0.119
SRH slope 0.009* 0.004 0.006 0.004
Note. Model fit statistics for unadjusted model: χ
2
(41) = 102.940, p < .001; CFI = .936;
RMSEA = .030; SRMR = .043. Model fit statistics for adjusted model: χ
2
(89) = 156.096, p <
.001; CFI = .961; RMSEA = .022; SRMR = .031. Adjusted model controlled for age, gender,
marital status, education, number of children, satisfaction with economic situation,
functional limitations, and depressive symptoms at baseline. HH = receipt of household
chore help from children; SRH = self-rated health.
*p < .05. **p < .01. ***p < .001.
100

Table 3.4
Bivariate Latent Growth Curve Model for Receipt of Personal Care Help
and Self-Rated Health: Unstandardized Parameter Estimates with and
without Adjustments for Control Variables
Unadjusted Adjusted
Estimate SE Estimate SE
Means
PC intercept 1.454*** 0.072 1.393*** 0.069
SRH intercept 2.093*** 0.020 2.100*** 0.017
PC slope 0.125*** 0.038 0.193*** 0.039
SRH slope -0.043*** 0.008 -0.056*** 0.009
Covariances
PC intercept ↔ PC slope -0.043 0.175 0.016 0.162
SRH intercept ↔ SRH slope -0.021* 0.009 -0.003 0.008
PC intercept ↔ SRH intercept -0.415*** 0.059 -0.068 0.046
PC slope ↔ SRH slope -0.035*** 0.010 -0.025* 0.010
SRH intercept ↔ PC slope 0.026 0.029 -0.009 0.024
PC intercept ↔ SRH slope 0.059* 0.024 0.017 0.021
Variances
PC intercept 1.357** 0.434 0.335 0.391
SRH intercept 0.296*** 0.028 0.149*** 0.022
PC slope 0.217* 0.087 0.192* 0.083
SRH slope 0.009* 0.004 0.006 0.004
Note. Model fit statistics for unadjusted model: χ
2
(41) = 110.974, p < .001; CFI = .922;
RMSEA = .032; SRMR = .044. Model fit statistics for adjusted model: χ
2
(89) = 156.802, p
< .001; CFI = .959; RMSEA = .022; SRMR = .030. Adjusted model controlled for age,
gender, marital status, education, number of children, satisfaction with economic situation,
functional limitations, and depressive symptoms at baseline. PC = receipt of personal care
help from children; SRH = self-rated health.
*p < .05. **p < .01. ***p < .001.
101

Table 3.5
Bivariate Latent Growth Curve Model for Provision of Household Chore Help and
Self-Rated Health: Unstandardized Parameter Estimates with and without
Adjustments for Control Variables
Unadjusted Adjusted
Estimate SE Estimate SE
Means
HH intercept 2.339*** 0.088 2.408*** 0.088
SRH intercept 2.094*** 0.020 2.100*** 0.017
HH slope 0.577*** 0.110 0.495*** 0.125
HH quadratic slope -0.232*** 0.026 -0.227*** 0.030
SRH slope -0.043** 0.008 -0.055*** 0.009
Covariances
HH intercept ↔ HH slope 1.053 0.842 -- --
HH intercept ↔ HH quadratic slope -0.387* 0.163 -- --
HH slope ↔ HH quadratic slope -0.463* 0.192 -0.992*** 0.124
SRH intercept ↔ SRH slope -0.019* 0.009 -0.002 0.008
HH intercept ↔ SRH intercept 0.223** 0.071 -- --
HH slope ↔ SRH slope 0.017 0.031 0.018 0.025
HH quadratic slope ↔ SRH slope -0.002 0.007 -0.003 0.006
SRH intercept ↔ HH slope 0.176* 0.085 0.141* 0.061
SRH intercept ↔ HH quadratic slope -0.052* 0.020 -0.034* 0.016
HH intercept ↔ SRH slope -0.018 0.028 -- --
Variances
HH intercept 2.625** 0.934 0.000 0.000
SRH intercept 0.290*** 0.027 0.147*** 0.022
HH slope 1.782* 0.910 4.178*** 0.470
HH quadratic slope 0.132** 0.044 0.238*** 0.034
SRH slope 0.008† 0.004 0.005 0.004
Note. Model fit statistics for unadjusted model: χ
2
(35) = 117.415, p < .001; CFI = .923; RMSEA =
.038; SRMR = .041. Model fit statistics for adjusted model: χ
2
(80) = 256.178, p < .001; CFI = .901;
RMSEA = .038; SRMR = .038. Adjusted model controlled for age, gender, marital status, education,
number of children, satisfaction with economic situation, functional limitations, and depressive
symptoms at baseline. HH = provision of household chore help to children; SRH = self-rated health.
†p < .10. *p < .05. **p < .01. ***p < .001.
102

Table 3.6
Bivariate Latent Growth Curve Model for Provision of Personal Care Help and Self-
Rated Health: Unstandardized Parameter Estimates with and without Adjustments
for Control Variables
Unadjusted Adjusted
Estimate SE Estimate SE
Means
PC intercept 0.980*** 0.057 1.033*** 0.058
SRH intercept 2.094*** 0.020 2.100*** 0.017
PC slope 0.479*** 0.088 0.385*** 0.088
PC quadratic slope -0.148*** 0.021 -0.135*** 0.022
SRH slope -0.041** 0.008 -0.054*** 0.009
Covariances
PC intercept ↔ PC slope -- -- -- --
PC intercept ↔ PC quadratic slope -- -- -- --
PC slope ↔ PC quadratic slope -0.679*** 0.071 -0.538*** 0.067
SRH intercept ↔ SRH slope -0.020* 0.010 -0.002 0.008
PC intercept ↔ SRH intercept -- -- -- --
PC slope ↔ SRH slope 0.000 0.020 0.003 0.018
PC quadratic slope ↔ SRH slope 0.001 0.005 0.000 0.005
SRH intercept ↔ PC slope 0.258*** 0.055 0.125** 0.045
SRH intercept ↔ PC quadratic slope -0.066*** 0.015 -0.034** 0.012
PC intercept ↔ SRH slope -- -- -- --
Variances
PC intercept 0.000 0.000 0.000 0.000
SRH intercept 0.291*** 0.028 0.147*** 0.022
PC slope 2.714*** 0.270 2.126*** 0.253
PC quadratic slope 0.175** 0.020 0.141*** 0.019
SRH slope 0.008† 0.004 0.005 0.004
Note. Model fit statistics for unadjusted model: χ
2
(40) = 139.786, p < .001; CFI = .887; RMSEA =
.039; SRMR = .048. Model fit statistics for adjusted model: χ
2
(80) = 178.481,p < .001; CFI = .938;
RMSEA = .028; SRMR = .030. Adjusted model controlled for age, gender, marital status, education,
number of children, satisfaction with economic situation, functional limitations, and depressive
symptoms at baseline. PC = provision of personal care help to children; SRH = self-rated health.
†p < .10. *p < .05. **p < .01. ***p < .001.

103

Table 4.1
Bivariate Latent Change Score Model Fit Indices for Receipt of
Instrumental Support and Self-Rated Health from 2001 to 2012
χ
2
df CFI TLI RMSEA
HH and SRH
No coupling 117.669* 46 .925 .927 .031
HH → SRH change 82.333* 45 .961 .961 .023
SRH → HH change 117.034* 45 .925 .925 .031
PC and SRH
No coupling 136.376* 46 .899 .902 .035
PC → SRH change 134.142* 45 .901 .901 .035
SRH → PC change 135.387* 45 .899 .899 .035
Note. Based on model fit statistics, chi-square test statistics for model comparison
were calculated. For receipt of household help and self-rated health, the model
with a coupling effect from support to change in self-rated health fit the data
significantly better (∆χ
2
(1) = 35.336, p < .001). For receipt of personal care and
self-rated health, the model with a coupling effect from support to change in self-
rated health did not fit the data significantly better (∆χ
2
(1) = 2.234, p > .10. HH =
receipt of household chore help; PC = receipt of personal care help; SRH = self-
rated health.
*p < .001.
104

Table 4.2
Bivariate Latent Change Score Model for
Receipt of Household Chore Help and Self-
Rated Health: Unstandardized Parameter
Estimates
Estimate SE
Regression coefficients
HH self-feedback -0.563*** 0.172
SRH self-feedback -1.102*** 0.183
HH → SRH change 0.410*** 0.093
SRH → HH change -- --
Means
HH intercept 3.362*** 0.099
SRH intercept 2.050*** 0.021
HH slope 1.486** 0.526
SRH slope 0.999* 0.437
Variances
HH intercept 3.302*** 0.379
SRH intercept 0.228*** 0.031
HH slope 1.026† 0.621
SRH slop 1.211** 0.398
HH error 10.784*** 0.304
SRH error 0.505*** 0.017
Note. HH = receipt of household chore help; SRH = self-
rated health.
†p < .10. *p < .05. **p < .01.***p < .001.
105

Table 4.3
Bivariate Latent Change Score Model for
Receipt of Personal Care Help and Self-
Rated Health: Unstandardized Parameter
Estimates
Estimate SE
Regression coefficients
PC self-feedback 0.597 0.569
SRH self-feedback 0.779 0.504
PC → SRH change 0.191 0.198
SRH → PC change -- --
Means
PC intercept 1.487* 0.069
SRH intercept 2.096* 0.020
PC slope -0.826 0.901
SRH slope -1.948 1.298
Variances
PC intercept 1.579* 0.307
SRH intercept 0.273* 0.021
PC slope 0.633 0.893
SRH slope 0.132 0.165
PC error 7.911* 0.258
SRH error 0.516* 0.014
Note. PC = receipt of personal care help; SRH = self-
rated health.
*p < .001.
106
Figures

Figure 2.1. Path diagram of an autoregressive and cross-lagged panel analysis for gendered
intergenerational support and functional limitations among older adults at four time points from
2001 to 2009.

Note. Paths a and b represent the autoregressive portion of the model, meaning the influence of one variable at one
occasion on the same variable at the next occasion. Paths c and d represent the cross-lagged portion of the model,
meaning the influence of one variable at one occasion on the other variable at the next occasion.
107

Figure 3.1. Diagram of a parallel process latent growth curve model for intergenerational
instrumental support and self-rated health at five occasions over 11 years.

Note. IS = instrumental support; SRH = self-rated health.
108

Figure 3.2. Quadratic latent growth curve model estimates for individual growth curves of
provision of household chore help over time (N = 50).
Baseline Wave 2 Wave 3 Wave 4 Wave 5
Time
109

Figure 3.3. Quadratic latent growth curve model estimates for individual growth curves of
provision of personal care help over time (N = 50).
Baseline Wave 2 Wave 3 Wave 4 Wave 5
Time
110

Figure 4.1. Diagram of a bivariate latent change score model for intergenerational instrumental
support and self-rated health at five occasions over 11 years.

Note. Support = instrumental support; SRH = self-rated health; Lsupport = latent variable of instrumental support;
Lsrh = latent variable of self-rated health; ∆ = change in corresponding variable between two time points.
Path coefficients a and b represent the coupling effect component. Path coefficient a represents the effect of
intergenerational instrumental support at one occasion on self-rated health at the next occasion; path coefficient b
represents the effect of self-rated health at one occasion on intergenerational instrumental support at the next
occasion. Path coefficients c and d represent the self-feedback component. Path coefficient c represents the effect of
intergenerational instrumental support at one occasion on itself at the next occasion. Path coefficient d represents the
effect of self-rated health at one occasion on itself at the next occasion.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Intergenerational support between grandparents and grandchildren in rural China and its effect on the psychological well-being of older adults

PDF

The effects of time and space on health status of older adults in China

PDF

Children 's migration and the financial, social, and psychological well-being of older adults in rural China

PDF

Life transitions, leisure activity engagement, and cognition among older adults

PDF

Intergenerational transfers in rural China: Do children's gender and birth order matter?

PDF

The relationship of expectation and satisfaction of filial piety with stroke-related outcomes of Chinese older adults in stroke rehabilitation

PDF

Intergenerational transmission of values and behaviors over the family life course

PDF

Examining the longitudinal influence of the physical and social environments on social isolation and cognitive health: contextualizing the role of technology

PDF

Mechanisms of the effect of involuntary retirement on older adults' health and mental health

PDF

The role of protective factors on outcomes for Latinos with schizophrenia

PDF

Longitudinal relationships of cognitive deficits, symptoms, and social functioning outcomes in community-based psychosocial rehabilitation programs: mechanisms of longitudinal change

PDF

The impact of post-traumatic stress and post-traumatic growth on young adult cancer survivors

PDF

A study of depression disclosure among Latino older adults in Los Angeles

PDF

The measurement, life course patterns, and outcomes of intergenerational ambivalence among parent-adult child dyads

PDF

My Life, My Wishes, a death education program for community-dwelling Chinese older adults: development and pilot testing

PDF

Investigating the effectiveness of a social work intervention on reducing hospital readmissions among older adults

PDF

Incarceration trajectories of mothers in state and federal prisons and their relation to the mother’s mental health problems and child’s risk of incarceration

PDF

Impacts of caregiving on wellbeing among older adults and their spousal caregivers in the United States

PDF

Cultural resources and health among Asian Americans: results from the National Latino and Asian American Study

PDF

COACH: Connecting Older Adults to Community for Health - an evidence-based program to address older adult social isolation and loneliness

Asset Metadata

Creator Mao, Weiyu (author)

Core Title Family relationships and their influence on health outcomes over time among older adults in rural China

School School of Social Work

Degree Doctor of Philosophy

Degree Program Social Work

Publication Date 07/23/2015

Defense Date 05/12/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag family relationships,health,longitudinal,OAI-PMH Harvest,older adults,rural China

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Chi, Iris (committee chair), Chou, Chih-Ping (committee member), Silverstein, Merril (committee member), Vega, William (committee member)

Creator Email butterflynankeyimeng@hotmail.com,weiyumao@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-604229

Unique identifier UC11298788

Identifier etd-MaoWeiyu-3682.pdf (filename),usctheses-c3-604229 (legacy record id)

Legacy Identifier etd-MaoWeiyu-3682.pdf

Dmrecord 604229

Document Type Dissertation

Format application/pdf (imt)

Rights Mao, Weiyu

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 a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA