35......
3:5 7? x.

L...
. . 3
09.»)

1o.

5.
i ‘3
~.

L.

4....
1.:

 

"r6598

llBRARY
Michigan State
University

This is to certify that the

dissertation entitled

SOCIAL CAPITAL, PARENTAL BEHAVIORS AND CHILDREN'S
SELF-ESTEEM IN THE FAMILY CONTEXT

presented by

Shu-Yao Hsu

has been accepted towards fulfillment
of the requirements for

 

Ph‘D- degree in Sociology—

c/ 1“, £427 ' (alga/L
// Major professor
DateflflQZTQJZLL—

MS U is an Affirmative Action/ Equal Opportunity Institution 0—12771

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

DATE DUE DATE DUE DATE DUE
2 7 P531 43 2&06 5
M395 5 9

 

601 c ICIRCIDateDue.965-p,1 5

SOCIAL CAPITAL, PARENTAL BEHAVIORS AND CHILDREN’S
SELF-ESTEEM IN THE FAMILY CONTEXT
By

Shu-Yao Hsu

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Sociology and
Urban Affairs Programs

2001

ABSTRACT

SOCIAL CAPITAL, PARENTAL BEHAVIORS AND CHILDREN’S
SELF-ESTEEM IN THE FAMILY CONTEXT

By

Shu-Yao Hsu

The increasing of the one-parent families has been argued to decrease the
availability of social capital for children due to the lack of social ties from the other
parent. The growing number of mothers participating in the labor force can negatively
effect the parent-child relationship due the limited time spending with children. Parents
have long been seen as the primary social agents to bring and secure resources for their
children in the process of socialization. Do parents still play the primary role in the

process of child development in the current American families?

Research has not systematically examined the relationships among social capital,
parental behaviors, and a child’s socio-emotional development. Some research has
focused on the parental intellectual and other resources contributing to family capital that
effects child outcomes, but neglected parental behaviors, while some emphasized the
impact of parental behavior on the child’s psychological development without
consideration of other broader family factors. This current study attempts to apply
Alejandro Portes’ theory of social capital and examine the impact of social capital and
other forms of family capital on parental behaviors that contribute to children’s self-

esteem.

The major research questions are (1) what types of capital have the more effects on

parental behaviors, social capital, human capital of parents, or financial capital of a

family? (2) whether the lack of any of the three types of capital can be compensated by
either one of the other two types of capital in terms of shaping parental behaviors; (3)

what types of parental behaviors lead to higher self-esteem for children.

The cLu'rent study will use the data set collected by the Child Development
Supplement (CDS) to the Panel Study Income Dynamics (PSID). The CD8 data were
collected by the University of Michigan Survey Research Center in 1997 with funding
from the National Institute of Child Health and Human Development (NICHD) as a

separate project of the Panel Study Income Dynamics (PSID).

Support reciprocity between parents and people outside the household, family
income, and parental education predict supportive parental behaviors including parental
warmth, parental monitoring, and “telling others positive things about the child”. The
family financial strain predicts parental monitoring. The less the family financial strain is,
the higher the parental monitoring is. “The higher degree of supportive parental
behaviors, the higher children’s self-esteem” is supported by three supportive parental
behaviors that are parental warmth, parental monitoring, and “telling others positive

things about the child”.

This study reveals that social support, family income, and parental human capital
are the important factors to lead to supportive parental behaviors. The supportive parental
behaviors contribute to the higher children’s self-esteem. It has been argued that the best
parenting style should spare the rod or spare the hug. No matter we spare the rod or not,

the hug should never be spared.

ACKN OLEDGMENT S

The years as a graduate student in Sociology Department and Urban Affairs
Programs have been very stimulating and instructive. First of all I would like to thank
Clifford L. Broman, my adviser, not only for his support, encouragement and wise
guidance, but also for providing me the opportunities to join his research. Consequently, I
learned what I needed for finishing my dissertation smoothly.

I would like to thank Steve Gold, Stan Kaplowitz and Dozier Thornton, members
of my committee, who provided me important information and suggestions.

In addition, I would like to give special thanks to Ruth S. Hamilton, who has been
encouraging during my years as a graduate student.

Finally, I would like to thank my husband, my daughter, my parents, the
secretaries in related departments and offices including Fran Anna Fowler, Joan Reid,

Tamara Dennany, and friends who have been supportive, instrumentally or emotionally.

iv

TABLE OF CONTENTS

 

 

 

 

I?“ IN I.—

If" I.“ I?” 11" I.“

L
L

L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF TABLES- vii
LIST OF FIGURES- ix
CHAPTER I INTRODUCTION - 1
Introduction 1
Purpose of the Disscnation 7
Literature Review 8
3.1 Support Reciprocity 8
3.2 Social Capital 11
3.3 Different Forms ofCapitaI in the Family Context 17
3.4 Parental Behaviors 19
3.5 Self-Esteem in Child Development 23
Research 11v pothescs 26
4 l Reciprocitv of Social Support and Parental Behaviors 27
_4._.. Financial Capital and Parental Behaviors 29
1.3 Human Capital and Parental Behaviors 3O
_4_.4 Parental Behaviors and Children’s Self-Esteem 31
CHAPTER II METHOD - 13
[Late 33
Data Collection and the Instruments 36
Sample 39
Variable Reconstruction 43
Measures 44
_S__l Social Capital and Support Reciprocity (as the Independent Variable) 44
5.2 Parental Behaviors (as the Mediator) 47

5.3 Human Capital and Other Parental Characteristics (as Independent
Variables) 49
Ll Financial Capital (as the Independent Variable) 49
_5_._3: C hildrcn's Self-Esteem (as the dependent variable) 49
:16 Socio-Demographic Variables (as Control Variables) 50
Mean Differences between the Target Subsample and Missing Data 50
Profile of Respondents in the Target Subsample 58
CHAPTER III RESULTS -- 60
Social Capital and Parental Behaviors 61
I_.I_ Overall Social Support Reciprocity and Parental Behaviors 62
1.1.1 Overall instrumental support reciprocity and parental behaviors 63

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.1.2 Overall emotional support reciprocity and parental behaviors 67

_.2_ Suppon Reciprocity and Parental Behaviors by Social Tie 70

1.2.1 Instrumental support reciprocity and parental behaviors by social ties 71

1.2.2 Emotional support reciprocity and parental behaviors by social tie 77

A Financial Capital and Parental Behaviors 83
2_.I FamiILFinaneiaI Strain and Parental Behaviors 83

3 Family Income and Parental Behaviors 86

; Human Capital and Parental Behaviors 88

4.- Parental Behaviors and C hildren’s Self-Esteem 91
CHAPTER IV CONCLUSION - -- 94
L Summary 94

; Implications 97

i Limitations and Future Research 99
APPENDICES 103
REFERENCES 114

 

 

vi

LIST OF TABLES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table l. Gains and Losses Mediated by Social Capital (modified from Portes, 1998) ..13
Table 2. Questionnaires and Booklets for Data Collection ............................................. 38
Table 3. Frequency Distribution of Age of Child by Number of Children Answering

Questionnaires from the Same Household ........................................................ 42
Table 4. Means of Socio-Demographic Variables and Children’s Self-Esteem for

the Targgt Subsample and the Missing Group .................................................. 52
Table 5. ANOVA of Socio-Demographic Variables and Children’s Self-Esteem

between the Target Subsample and the Missing Grow; .................................... 53
Table 6. Means of Parental Warmth, Monitoring, and Positive Parental Behaviors

for the Target Subsample and the Missing Group ............................................ 54
Table 7. ANOVA of Parental Warmth, Parental Monitoring, and Positive Parental

Behaviors between the Target Subsample and the Missing Group ................... 55
Table 8. Means ofNegative Parental Behaviors for the Target Subsample and the

Missing Group ................................................................................................... 56
Table 9. ANOVA of Negative Parental Behaviors between the Target Subsmle

and the Missing Group ...................................................................................... 57
Table 10. Distribution of Socio-Demographic Variables of the Target Subsample ......... 58
Table l 1. Regression of Parental Warmth and Parental Monitoring on Overall Social

Support Reciprocity .......................................................................................... 64
Table 12. Logistic Regression of Specific Parental Behaviors on Overall Instrumental

 

Table 13.

Table 14.

Table I 5..

Support Reciprocity .......................................................................................... 66

Logistic Regression of Specific Parental Behaviors on Overall Emotional
Suppon Recitlrocm .......................................................................................... 69

Regression of Parental Warmth and Parental Monitoring on Instrumental
Support Reciprocity by Social Tie .................................................................... 72

Logistic Regression ofSpecific Parental Behaviors on Instrumental
Support Reciprocity between the Primary Carggiver and His/Her Own
Parent ................................................................................................................. 74

vii

Table 16

Table 17.

Table 18.

Table 19.

Table 20.

Table 21.

Table 22.

Table 23.

Logistic Regression of Specific Parental Behaviors on Instrumental
Support Reciprocity between the Primary Caregiver and His/Her Sibling ....... 75

Logistic Regression of Specific Parental Behaviors on Instrumental
Support Reciprocity between the Primary Caregiver and His/Her Friend ........ 76

Regression of Parental Warmth and Parental Monitoring on Emotional
Suppmt Reciprocity by Social Tie .................................................................... 78

Logistic Regression ofSpecific Parental Behaviors on Emotional Support
Reciprocity between the Primary Caregiver and His/Her Own Parent ............. 80

Logistic Regression of Specific Parental Behaviors on Emotional Support
Reciprocity between the Primary Caregiver and His/Her Sibling .................... 81

Logistic Regression of Specific Parental Behaviors on Emotional Support
Reciprocity between the Primary Caregiver and His/Her Friend ..................... 82

Regression of Parental Warmth and Parental Monitoring on Family
Financial Strain ................................................................................................. 84

 

Logistic Regression of Specific Parental Behaviors on Family Financial

 

 

 

 

Sm ................................................................................................................. 85
Table 24. Regression of Parental Warmth and Parental Monitoring on Family Income..86
Table 25. Logistic Regression of Specific Parental Behaviors on Family Income ........... 87
Table 26. Regression of Parental Warmth and Parental Monitoring on Primary

Caregivers Education ....................................................................................... 89
Table 27. Logistic Regression ofSpccifIc Parental Behaviors on the PrimaIy

 

Table 28.

 

Caregiver‘s Education ....................................................................................... 90
Regression ofChiIdren‘s Self-Esteem on Parental Behaviors .......................... 92

 

viii

LIST OF FIGURES

Figpre 1. Theoretical Model of Social Capital. Parental Behaviors, and C hildren’s
Self-Esteem ...................................................................................................... 26

 

ix

CHAPTER I
INTRODUCTION

1. Introduction

“A Call to Civil Society: Why Democracy Needs Moral Truths”, a report issued by
the Council on Civil Society in 1998, claims that the deterioration of child and adolescent
well-being is one critical dimension indicating social crisis of moral depletion in the
United States of America (Elshtain. 1999). As noted at several places in “A Call to Civil
Society”, children don’t get enough attention from their parents, and parental investment

in their children and parenting skills are in decline.

F urstenberg (1999) argues that the most dramatic changes in the American family
are the disconnection between sexuality and marriage, cohabitation, and less enduring
marriage. These changes have affected parental practices. How to balance the demands of
family life and work has become a challenge for parents. According to the report of the
international adolescent health survey, the Health Behaviors in School-Aged Children
(HBSC) Study, that investigated 11, 13, and 15 year old children's experiences
concerning health behaviors and lifestyle issues in 26 European countries and regions,
Canada, and the United States coordinated by the World Health Organization (WHO),
only 64 percent of US. students reported living with both parents. This proportion is

lower than for all other countries except Greenland (NIH and NICHD, 2000).

Child development can result from both parenting processes and the parental
structure (Furstenberg, 1999). The former refers to parenting styles and parental
behaviors such as warmth, control, and autonomy, while the later refers to who resides
with the child and who are the primary caregivers. One question to be asked is: how to
make a family function to serve the best interests of children. A single parent can be
aided by his/her own parents, relatives, or friends in raising the child. It needs further
investigation if the presence of support from relatives promotes parenting skills, or

compensates for the absence of the biological father or mother.

The increase in one-parent families and the quantitative decline in relationships with
relatives outside the households lead to fewer social ties that can decrease the availability
of social capital for children in the contemporary American society (DeFrain and Olson,
1999; McLanahan and Sandefur, 1994). Another striking fact is the increasing number of
families with mothers in the labor force (DeF rain and Olson, 1999; Furstenberg, 1999).
Mother’s labor force participation increases family income, but decreases the time
spending with children that can lead to negative effects on parent-child relationship and
child cognitive development (Carlin, 1999; Haveman and Wolfe, 1995). However,
maternal working has been linked to the reduction of social capital for children and its

negative effects on child development (Coleman, 1988a, 1988b, 1990).

Parents have long been seen as the primary social agents to bring and secure
resources for their children in the process of socialization. Recently, research has
supported that maternal intellectual ability and parental time allocation contribute to
family capital beneficial in facilitating children's cognitive and socio-emotional

development (Parcel and Menaghan, 1993, 1994). Greater social capital has also been

associated with lower high school dropout (F urstenberg and Hughes, 1995; Smith et al.,
1992; Teachman et al., 1996), and with greater college enrollment (Furstenberg and
Hughes, 1995). Parental support can compensate for the loss of community bonds for
positive child outcomes (Gold, 1995; Hagan et al., 1996). Obtaining greater social capital
in facilitating positive child outcomes appears to be the major task for parents. Such a

task makes the role of parenting challenging.

The concept of social capital has been used across different disciplines such as
economics, education, health and well-being, political science, and sociology in order to
understanding various aspects of human lives from the institutional level to the individual
level (Portes & Landolt, 1996; Woolcock, 1998). However, recently, this concept has
been extended to apply to the organizational level rather than just to the individual level
(Portes & Landolt, 1996). Besides, the concept of social capital is treated in various ways.
The decomposition of social capital into the three elements is usually ignored (Portes,
1998). According to Portes (1998), the three elements of social capital are the sources of
social capital, the possessors of social capital (those making claims), and the resources of
social capital. There are four different sources of social capital including reciprocity
exchange, enforceable trust, value introjection, and bounded solidarity (Portes and
Sensenbrenner, 1993; Portes, 1998). Portes’ decomposition of social capital is applied to

establish the theoretical framework and to guide data analysis of this current study.

Due to the limitation on data sources, this current study is not able to cover all
sources of social capital; therefore, reciprocity exchange of social support is the major
focus rather than other sources. Related literature on social support and reciprocity is

integrated for further understanding the mechanism of reciprocity exchange under the

broad social capital framework. Parents here are treated as the persons who make claims
of social capital for their children. This study will examine the pattern of support
reciprocity between parents (or primary caregivers) and people living outside the
household including instrumental and emotional support, as well as how support

reciprocity relates to parental behaviors.

Research has primarily focused on the positive side of social support to parents who
are constrained by childrearing responsibilities. The possible downsides of social capital
need to be considered (Light and Gold, 2000; Portes and Landolt, 1996; Waldinger,
1995). The downsides refer to the limitations and costs of social capital. Social support
system can both facilitate and constrain the actors. Social relations may lead to
unexpected stress for either support recipients or support providers in the process of
support exchange. In this study, the effects of support reciprocity are examined in both
the general and the specific perspectives. The general perspective investigates the global
effects of support reciprocity on parental behaviors, while the specific perspective looks
into the effects of support reciprocity based on the type of social ties (e.g. own parents,

siblings, and friends).

The unit of analysis is based on the individual child. Corresponding to the individual
child, the primary caregiver of the child at home (most of them are mothers) is the main
focus. A child’s primary caregiver at home (mostly the child’s mother) is treated as a
social agent who claims or secures resources of social capital for the child. However,
other related issues not mentioned above (e. g. solidity within the family, trust between a

parent and people outside the household) are beyond the scope of the dissertation.

Research has not systematically examined the relationships among social capital,
financial capital, and human capital, parental behaviors, and children’s socio-emotional
development. Some research focused on the effects of parental work characteristics or
parental investment on children’s outcomes, but neglected parental behaviors (e.g. Parcel
and Menaghan, 1993, 1994), while some emphasized the impact of parental behaviors on
the children’s psychological development without consideration of other broader family
factors (Coopersmith, 1967). This current study attempts to examine the impact of social
capital and other forms of family capital on parental behaviors that contribute to

children’s self-esteem.

The focus on children’s self-esteem is based on three theoretical and empirical
considerations. First, global self-esteem is viewed as a critical indicator of psychological

well-being (Rosenberg, Schooler, Schoenbach, and Rosenberg, 1995).

Second, self-esteem is found to correlate positively with social support from parents
and peers through different developmental spans (N ewcomb and Keefe, 1997), and, more
important, can increase later social support (Newcomb, 1990). According to Newcomb
and Keefe’s 12-year longitudinal study on self-esteem and social support with a sample
of 7‘“, 8m, and 9th grade students (Newcomb and Keefe, 1997), social support consistently
predicts later self-esteem over a 12-year period at 4-year intervals, but not as good as
predicts earlier self-esteem. Conversely, self-esteem has positive effects on later social
support and the quality of later relationships (Newcomb, 1990). Critically, neither the
impact from social support to self-esteem nor that from self-esteem to social support is
empirically significant at the same interval. The relationship between social support and

self-esteem need further investigation. According to the “looking glass” theory (Cooley,

1902; Mead 1934), individuals define themselves by internalizing evaluative judgements
of others through social interactions. During childhood, the degree to which a child feels
accepted and supported by the parents leads to positive feelings about the self and certain
personal characteristics. Then, the certain personal characteristics (e. g. self-esteem) help

shape the nature of one’s social support system in their future life.

Finally, self-esteem appears to have the characteristic of lower stability than other
personality traits such as social conformity and gregariousness, even though self-esteem
has substantial stability over time in general (Newcomb and Keefe, 1997). Social
conformity refers to individual’s attitudes toward society and traditional values, while
gregariousness refers to individual’s interaction style toward other people and the world.
The lower stability of self-esteem can be a reflection of self-feelings toward perceived
judgement of others that invokes in the self (Hertzog and Nesselroade, 1987; Newcomb
and Keefe, 1997). During the period of development in childhood, it is believed to be of
great importance for the development of a positive sense toward self. Since self-esteem
has more room for change compared to other personality traits; therefore, it is necessary
to understand what factors, in what context contribute more to self-esteem among

children.

Several variables are found to be critically related to child development, for
example, the family financial condition (Mayhew and Lempers, 1998), the relationship
between parents and children (Putallaz and Heflin, 1990), the impact of family support on
a child’s friendship quality (Franco and Levitt, 1998), and the structure of the family
(Parcel and Menaghan, 1993, 1994). Since the family plays the crucial role in the process

of child development, psychologically and intellectually; thus, the focus of this study lies

primarily on the family context. The complex issues surrounding the development and

maintenance of a child's self-esteem demands more research in this area (Shelton, 1990).

2. Purpose of the Dissertation

Existing research evidence is still ambiguous to explain which variables (e.g.
parental behaviors, family support, family financial condition, or other familial factors)
are more powerful in predicting children’s self-esteem. This study aims to refine the
neglected details between the theoretical explanations and the past empirical findings
regarding different forms of capital surrounding a family, and how the different forms of
capital are transformed to shape children’s self-esteem. Specifically, this study
investigates parental social capital, financial capital, human capital, parental behaviors,
and children’s self-esteem. Parental behaviors are treated as a mediator between these
different forms of capital and children’s self-esteem, because parents are viewed as the
social agents who transform or make claims of resources for their children. This research

attempts to solve three questions including:

1. Which form(s) of capital has/have the more effects on parental behaviors? Social
capital? Human capital? Or, financial capital?

2. Can any form of capital be substituted by other forms of capital in terms of shaping
parental behaviors?

3. What parental behaviors lead to higher self-esteem for children?

3. Literature Review
3.1 Support Reciprocity

Current literature on social capital emphasizes that “reciprocity exchange” is a key
variable of facilitating individual actions for generating social capital (Coleman, 1988a,
1990; Femandez-Kelly, 1995; Portes & Sensenbrenner, 1993). Reciprocal relations can
increase solidarity and trust within relationships (Pearlin et al., 1981). The notion of
reciprocity is one important focus on social support (Antonucci, 1985; Lu, 1997).
Individuals whose network is non-reciprocal are likely to perceive their network as not
adequate and to report less positive outcomes such as psychological well-being, self-
esteem, life satisfaction, and happiness. In general, reciprocity between providers and
recipients is more likely to lead to positive and longer relationships, while asymmetrical
relationships are more likely to elicit anger and resentment (Antonucci, 1985). In addition
to being an important focus on social support, reciprocity is a key element that effects the
production of resources during the process of social capital transaction for individuals or

groups (Coleman, 1988a, 1990; Portes, 1998; Portes & Sensenbrenner, 1993).

Social support most commonly refers to helpful functions accessible to an
individual through social ties or from significant others such as family members, friends,
relatives, and co-workers (Lin et al., 1979; Thoits, 1985). The functions of support
typically include emotional aid and instrumental aid. Emotional aid usually refers to
assertions or demonstrations of love, caring, esteem, value, and group belonging (Cobb,
1976; House, 1981; Kaplan et al., 1977). Instrumental aid refers to actions or materials
provided by others that enable the fulfillment of ordinary responsibilities, such as

household, childcare, financial, and job-related obligations (House, 1981).

According to Pearlin and his colleagues (1981), the qualities of support are
especially critical when they “involve the exchange of intimate communications and the
presence of solidarity and trust” (Pearlin et al., 1981, p.340). Particularly, emotional
support from significant others or primary others appears to be the most powerful
predictor of reduced psychological distress regardless of the presence of stressful
circumstances (Cohen & McKay, 1984; Heller, 1979; House, 1981; Thoits, 1985; Turner,
1983). The emotional function of social support appears to be the key aspect of social
support related studies. Social support from our social networks can buffer stress, and
improve well-being in adulthood as well as in childhood and adolescence, maybe even
more than in later life (Heller et al., 1986). Such effects will foster self-esteem, feelings
of security and control over oneself and the environment, and in this way will help to

protect and promote well-being (N estmann and Hurrelmann, 1994).

Social support can also function as a stress buffer for mothers with irritable infants
(Crodkenberg and McCluskey, 1986). Those mothers who have more social support and
who have less frequent negative interactions with significant others provide positively
warm and sensitive care to their children (Cotterell, 1986; Crnic et al., 1983; Pascoe et

al., 1981; Weinraub & Wolf, 1983; Zarling, Hirsch & Landry, 1988).

Social support is important in parenting for several reasons. First, the less social
support can lead to the more frequently negative parental behavior. Research finding
indicates that adolescent mothers who reported less support are found to use restriction
and punishment more often (Garcia-Coll et al., 1986). Second, social support reduces the
negative effects of stress on the parenting behavior of mothers (Kurtz and Derevensky,

1994; Rhodes and Woods, 1995), and cases the stress of parenting by enabling a mother

to share her frustrations, which may, in turn, enable her to be more sensitive with her

children (Cochran and Brassard, 1979).

It has been suggested that there is a down side to giving support as well as to
receiving support (Bierhoff, 1994; Lu, 1997; Lu & Argyle, 1992; Rook, 1992). In other
words, social support might have not only positive but also negative consequences for the
support recipients and providers. Both giving and seeking support require special skills,
and sometimes the process of exchange can be risky (Goldsmith & Parks, 1990).
According to social exchange theory, social interaction can result in both rewards and
costs (Rook, 1984). Within the social support related studies, research evidence indicates
social support may not only fail to buffer the stress of illness, but may add to distress
(Wortman, 1984; Barbee, Derlega, Sherburne, and Grimshaw, 1998). Many supportive
behaviors that are meant to be helpful may turn out to be harmful. For example, child
care assistance offered by others often increases the feelings of inadequacy because the

parent perceives his/her incompetence to perform everyday tasks (Peters-Golden, 1982).

Research findings increasingly suggest that receiving support conflicts with the idea
of self-management, leads to negative reactions from others (Coates, Wortman, & Abbey,
1979; Miller & Steinberg, 1975), damages self-esteem (Williams & Williams, 1983),
evokes feelings of helplessness (McLeroy, DeVellis, DeVellis, Kaplan, & Toole, 1984),
and relates to symptoms of anxiety, feelings of dependence and guilt (Lu and Argyle,
1992), and dissatisfaction of a recipient’s needs (Tracey, Revenson, Schiafino,
Majerovitz, & Gibofsky, 1991). Problems of giving support may involve feelings of
being burdened and frustrated (Lu and Argyle, 1992), fear of possible interpersonal

conflicts (Chesler & Barbarin, 1984; Wortman & Lehman, 1985), personal distress

10

(Wortman & Dunkel-Schetter, 1979), and feelings of inadequacy and incompetence

(Wortman & Lehman, 1985).

Although the mainstream social support research has been largely toward studying
the beneficial effects on well-being, the recognition of the potential harmful effects of

social support need to be considered (La Gaipa, 1990; Lu, 1997).

In addition, the reciprocal relationship between a provider and a recipient is critical
in the effects of social support on an individual. Norms of reciprocity can change
depending upon the level of intimacy between the two parties. Reciprocity with spouse
relates to higher levels of happiness, while reciprocity with friends relates to higher level
of life satisfaction (see Antonucci, 1985). Growing evidence shows that more intimate
support networks increase the vulnerability to the detrimental effects of such stress rather
than buffering the negative effects of life stress (Kessler & McLeod, 1984). In sum,
positive and negative effects of social support, and the reciprocal relationships between

support providers and recipients among parents will be examined.

3.2 Social Capital

The concept of social capital has been broadly used across disciplines in the last
decade. However, the wide use of this concept has led to great variations on its definition
from one scholar to another. This current study applies Alejandro Portes’
conceptualization of social capital (1998). His concept is gradually receiving consensus

in the literature on social capital.

11

Portes’ conceptualization on social capital is embedded in the proposition that
collective expectations affect individual economic behavior (Portes and Sensenbrenner,
1993). For Portes (1998), social capital per se is neither a source nor a resource, but
rather a mediator between the sources and the functions (or resources). The concept of
social capital is systematically treated by distinguishing it into three elements: (1) the
sources of social capital (those agreeing to these demands); (2) the possessors of social
capital (those making claims); and (3) the resources themselves. The distinction between
these elements and the ability to obtain them by virtue of membership in different social
structures are explicitly made. Portes argues that social capital is defined as the
“expectations for action within a collectivity that affect the economic goals and goal-
seeking behavior of its members, even if these expectations are not oriented toward the
economic sphere” (Portes and Sensenbrenner, 1993), and “the ability of actors to secure
benefits by virtue of membership in social networks or other social structures” (Portes,
1998). In sum, expectations for action and the ability of actors are the critical factors to
facilitate actions for securing or claiming benefits. This is, however, consistent with
Coleman’s emphasis that social capital as a variety of entities facilitates actions within
social structure (1988a). One may not be beneficial from any resource, if the resource has

not been claimed or transformed into the form that is needed or appropriate to be used.

There are four different sources of social capital including value introjection,
bounded solidarity, reciprocity exchange, and enforceable trust (Portes and
Sensenbrenner, 1993; Portes, 1998). The former two sources are instrumental, while the

later two are consummatory (see Table 1).

12

Table 1. Gains and Losses Mediated by Social Capital (modified from Portes, 1998)

 

 

Sources Mediator Resources (or Functions)
Instrumental Ability to secure benefits Positive effects:
. Reciprocity through membership in a Family support
exchange networks and other social . Network-mediated benefits
- Enforceable trust structures, and . Upward leveling norms
expectations for action
Consummatory within a collectivity that Negative effects:
. Value lntrojection affect the economic 90318 . Restn'cted access to opportunities
0 Bounded solidarity 3'19 goal-seeking behavror - Restrictions on individual freedom
0f 'ts members . Excessive claims on group members
. Downward leveling norms

 

 

 

Reciprocity exchange refers to norms of giving and receiving favors, information,
approval, and other valued items. The giver’s returns are not necessarily directly from the
individual recipient, but can be from the collectivity as whole in the form of honor or
approval. Enforceable trust is built through effective collective sanctions and rewards to
monitor the behaviors of group members. Value introjection stresses socialization into
consensually established beliefs and urges individuals to behave based on the group's
guiding values rather than pursuing self-interest. Bounded solidarity refers to the
mechanism that appropriates the altruistic dispositions of actors. The sense of solidarity is
generated by a common fate or the common adversity of confrontation with the
community. Literature on social capital has strongly emphasized the positive
consequences, but neglected the negative side. Dense social ties can provide more access
to resources, but bring greater social control and restrict individual freedom. The equal

attention for both positive and negative effects should be noticed in theory and research.

13

Due to the limitations on data sources, this study will not cover all sources of social
capital; therefore, reciprocity exchange of social support is the main focus rather than
other sources. The reciprocal relationships between a parent and people outside the
household will be examined, but not the reciprocal relationship between parents and
children. Parents here are treated as the persons who make claims of social capital for

their children.

In addition to Portes, Pierre Bourdieu (1979, 1980, 1985) and James S. Coleman
(1988a, 1988b, 1990) are important contributors to the conceptualization of social capital.
Pierre Bourdieu is the one who produces the first systematic contemporary analysis of
social capital (Portes, 1998). Bourdieu regards social capital as “the aggregate of the
actual or potential resources which are linked to possession of a durable network of more
or less institutionalized relationships of mutual acquaintance or recognition” (Bourdieu
1985, p. 248; 1980). Bourdieu’s statement clearly decomposes social capital into two
elements: (1) the social relationship itself that individuals use to claim access to
resources, and (2) the quantity and quality of the resources. The treatment of this concept
tends to be instrumental, and emphasizes the benefits accruing to individuals by
participation in-groups and on the construction of sociability for the purpose of creating
resources. Instead of being a natural given resource, social networks are constructed
through investment strategies and become usable as a reliable source of other benefits.
Through social capital, actors can gain direct access to economic resources such as
subsidized loans, investment tips, and protected markets. Throughout, Bourdieu's
emphasis is on the ultimate reduction of all forms of capital to economic capital

(Bourdieu, 1979, 1980, 1985; also see Portes, 1998).

14

Coleman conceptualizes social capital as “a variety of entities with two elements in
common: They all consist of some aspect of social structures, and they facilitate certain
action of actors--whether persons or corporate actors--within the structure” (Coleman
1988a, p. S98; also see Coleman, 1990, p. 302). This definition is based on the function
of social capital (Coleman, 1988a, 1990). For Coleman, reciprocity expectations and
group enforcement of norms are the mechanisms used to generate social capital. Social
organization provides the context for both sources and effects to materialize. Privileged
access to information is as the consequence of its possession. Coleman stresses that social
capital plays the key role in the creation of human capital. He not only sheds the light on
the importance of social capital for the acquisition of human capital and the creation of
human capital for the next generations, but also identifies some of the mechanisms
through which it is generated. Coleman’s concept has been applied by various empirical
research, and a great number of these studies focus particularly on family and child
development (e.g. Hofferth, Boisjoly, & Duncan, 1999; Parcel and Menaghan, 1993,

1994; Zhou and Bankston, 1998).

Portes criticizes that Coleman’s definition is too ambiguous and opens the way for
relabeling some contradictory processes as social capital (Portes, 1998). Although
Bourdieu like Coleman has emphasized social capital as a resource that is generated
through social ties and can facilitate certain actions for actors (Bourdieu, 1980, 1985;
Coleman, 1988a, 1990), Coleman makes the fallacy of equating social capital with the
resources acquired through it. The distinction of social capital from its resources is made

explicit in Bourdieu, but ambiguous in Coleman.

15

After Bourdieu and Coleman, Baker (1990) defined social capital as “a resource that
actors derive from specific social structures and then use to pursue their interests; it is
created by changes in the relationship among actors” Q). 619). Baker’s concept is very
close to Bourdieu’s, but he treats social capital as a resource like Coleman. Schiff’ 5
definition (1992) is “the set of elements of the social structure that affects relations
among people and are inputs or arguments of the production and/or utility function”
(p.161). This definition is broader than others are. Burt (1992) sees social capital as
"friends, colleagues, and more general contacts through whom you receive opportunities
to use your financial and human capital" (p.9). Burt emphasizes the relative absence of
ties, labeled "structural holes," that facilitates social mobility of individuals, because
weaker ties are seen as sources of seeking new knowledge and new resources. This is
completely opposite to Coleman’s emphasis on dense networks as a necessary condition
for the generation of social capital. Hofferth and her colleagues (1999) follow Coleman’s
essay and conceptualize social capital as an important potential resource emerging from
social relationships. “Social capital posits that social relationships form a resource that
individuals can draw upon in their personal and professional lives”(Hofferth et al., 1999,

p.79).

In short, Portes’ conceptualization on social capital is the only one that makes a
clear and systematic differentiation to the key elements of social capital into three
categories. Although his concept is gradually receiving consensus in the literature on
social capital, the empirical application on social capital following his conceptualization
has not been made. This study applies Portes’ concept for providing advanced and

systematic analysis on the construct of social capital.

16

3.3 Different Forms of Capital in the Family Context

According to Coleman (1988a, 1990), there are three major resources of capital in
the family context: (1) social capital, (2) human capital, and (3) financial capital. Social
capital is different from the other two forms of capital including human capital and
financial capital. Forms of capital can be transformed into one another through various
institutional mechanisms (Bourdieu, 1985; Fernandez-Kelly, 1995; Portes, 1998). Social
capital here refers to the “expectations for action within a collectivity that affect the
economic goals and goal-seeking behavior of its members, even if these expectations are
not oriented toward the economic sphere” (Portes and Sensenbrenner, 1993), and “the
ability of actors to secure benefits by virtue of membership in social networks or other
social structures” (Portes, 1998). Human capital refers to parent’s education and related
cognitive skills that assist learning for children (Coleman, 1988a). Financial capital refers
to family income or other financial resources used for the household and the child
(Coleman, 1988a). All three forms of capital can be productive to make possible certain

actions or achievements.

The effect of social capital on the creation of human capital in the next generation
has been emphasized (Coleman, 1988a). Family factors are considered to account for
more variance in the socialization process for most children than school or community
factors (Hofferth et al., 1997). Families provide the context in which most children grow
up, and learn social values, knowledge as well as skills to develop connections to other
people from different aspects of social structure in later life. A child’s psychological and

intellectual development can be regarded as the by-produet of social capital and other

17

forms of family capital in a child’s socialization process (Coleman, 1988a). The role of
social capital in resource transactions is critical for understanding the role of parents in

the creation of human capital for their children.

Social ties not only consist of the interpersonal connection within the family, but
also include those that the family establishes with people outside the immediate family
members. A family rich in connectedness can be beneficial to parents as well as to
children. However, the wide range of social connections and strong bonds between
parents and children need time and effort that demand the physical presence of parents,
their attention, and involvement. For the purpose of building a stock of resources that the
family can call upon when needed, such investment in these connections is necessary
(Hofferth et al., 1997; Hofferth et al., 1998; Hofferth et al., 1999). Coleman (1988a)
argues that the parental human capital can contribute to the child’s development, only
when parents are an important part of child’s life. In other words, the absence of parents
is seen as the lack of investment in the parent-child connectedness. Even when the
parents are rich in their own human capital, children can gain little support from their
parents in the process of cognitive development due to the frequent absence of their
parents. Thus, in this condition the child’s growth can be irrelevant to the parent’s human

capital (Coleman, 1988a, 1990).

Social support has been identified as a crucial determinant of parenting practices
(Belsky, 1990; Cochran & Brassard, 1979; Cotterell, 1986; Pascoe et al., 1981; Simons,
Beaman, Conger & Chao, 1993; Weinraub & Wolf, 1983; Zarling, Hirsch & Landry,
1988). In the past, social capital in families are often the focus, but it is also important to

consider social relationships that are embedded in the larger community through which

18

parents can mobilize their resources that facilitate children’s development (Lin, 1990;
Wong, 1998). Other factors including family income, family financial strain, parental
education, socioeconomic status, and maternal unemployment are considered to be
critically associated with parenting (Brenner & Fox, 1999; Shumow, Vandell & Posner,
1998). Research findings indicate that maternal education and financial resources
contribute to the higher degree of parental warmth and firm control (Brody etal., 1995;
Elder, Eccles, Ardelt, & Lord, 1995). In contrast, greater parental harshness can be the
result of lower family income (Conger et al., 1990; Mirowsky & Ross, 1986), family
financial stress (Takeuchi, Williams & Adair, 1991), lower parental education
(Dombusch et al., 1987; Kelley, Sanchez-Hucles, & Walker, 1993; Raikkonen &
Keltikangas-Jarvinen, 1992), and maternal unemployment (McLoyd, .Iayaratne, Ceballo,
& Borquez, 1994). In other words, the level of financial capital and human capital is

associated with parenting practices.

Research findings have shown that financial problems are less psychologically
central to children than to adults, because family financial hardship is an ascribed status
rather than achievement status for children (Demo & Savin-Williams, 1983; Whitbeck et
al., 1991). Consequently, financial capital is treated as a predictor of parental behaviors

rather that of children’s self-esteem.

3 .4 Parental Behaviors

This section specifically focuses on parental behavior related literature. In an early

study (Coopersmith, 1967: Antecedents ofSelf-Esteem), Coopersmith (1967) has pointed

l9

out the significance of parental behaviors in shaping children’s self-esteem. Parental
reinforcement and the rewards of imitating parental behaviors presumably play important

parts in shaping children’s actual characteristics and behaviors.

Two general dimensions of parental behaviors, support and control, have been
identified by numerous studies and reviews to predict child psychological and intellectual
development since the 19405 (Peterson and Hann, 1999). The dimension of parental
support has been labeled as warmth, verbal affection, physical affection, nurturance,
acceptance and general support (Becker, 1964; Rohner, 1986; Rollins & Thomas, 1979;
Schaefer, 1959; Siegelman, 1965; Stafford & Bayer, 1993). According to Harter (1998),
support in the form of acceptance and approval from parents is a significant source of
self-evaluation for young children. Empirical evidence suggests that parents provide
emotional security, and serve as models of interactive behaviors for their children (Parke
& Buriel, 1998; Putallaz & Heflin, 1990). In addition, support from close family
members has been related to higher self-esteem for children and adolescents (Franco and
Levitt, 1998; Harter, 1998; Levitt et al., 1993; van Aken & Asendorpf, 1997; Wolchik,
Beals & Sandler, 1989). The protection of self-esteem is regarded as a primary function
of social support across the life span (Antonucci, 1990; Franco and Levitt, 1998; Sandler,

Miller, Short, and Wolchik, 1989).

The dimension of parental control has been conceptualized as actions used by
parents to modify the behaviors and internal states of children (Peterson & Rollins,
1987). F irm control and excessive control are the two most frequently identified types of

parental control. Induction and monitoring (or behavior control) are the two frequently

20

identified forms of firm control; while psychological overcontrol and punitiveness are the

two forms of excessive control (Peterson and Hann, 1999).

Induction is a behavior which emphasizes the psychological form used by parents to
explain positive and negative consequences of a child's actions for both the child and
others (Hoffman, 1994; Maccoby & Martin, 1983; Rollins & Thomas, 1979). In this form
of firm control, parents do not impose arbitrary authority on children. Consequently,
children have opportunities to engage in two-way communication and express different
viewpoints with their parents (Peterson & Rollins, 1987). Induction is seen as a means of
fostering internalization and voluntary commitment to social expectations from parents

and the society (Peterson and Hann, 1999).

Monitoring (or behavioral control), the other form of firm control, refers to parental
attempts or supervising behaviors to manage their children's schedules, activities,
physical whereabouts, and peer associations (Barber et al., 1994; Patterson and Capaldi,
1991). Much of the research on parental monitoring has focused on the association of
insufficient supervision with higher frequencies of antisocial, delinquent, and other

problem behaviors (Barber et al., 1994).

Psychological overcontrol, one form of excessive control, refers to behaviors with
intrusive or arbitrary qualities used by parents to manipulate children's emotional
experiences. Love withdrawal is the frequently used techniques of psychological
overcontrol (Maccoby & Martin, 1983; Rollins & Thomas, 1979; Stafford & Bayer,
1993). Love withdrawal techniques such as turning one's back and refusing to speak to

the young are often used as punishment to manipulate children's fears, keep children

21

responsive to parental perspectives, and discourage children to deviate from parental

expectations (Maccoby & Martin, 1983).

Punitiveness. the other form of excessive control, refers to punitive or coercive
behaviors used by parents including either verbal or physical attempts to control children
without the rational explanations (Peterson et al., 1985; Peterson and Hann, 1999). More
severe forms of punitiveness can be characterized as violence between parents children.
Parental punitiveness constrains the development of self-esteem, and the application of
arbitrary force often cause hostility and resistance from children to their parents (Rollins

& Thomas. 1979; Turner & Finkelhor, 1996).

Parental behaviors of firm control such as induction are more likely to enhance
children’s self-esteem through the enforcement of clearly defined limits (Coopersmith,
1967; Openshaw & Thomas, 1986). In other words, induction increases self-esteem by
the parent-child communication in which parental confidence matches children’s abilities
to understand and cope successfully with the social environment (Peterson & Rollins,
1987; Stafford & Bayer, 1993). The result is that clear, consistent, and legitimate parental
control facilitates self-esteem by contributing to the internalization of values and
expectations that, in turn, lead to greater confidence in a child's definition of the
standards for expected outcome (Baumrind, 1978; Coopersmith, 1967; Felson & Reed,
1986; Higgins, 1991; Openshaw & Thomas, 1986; Peterson, Rollins & Thomas, 1985;

Putallaz & Heflin, 1990).

The positive effects of parental firm control on children’s self-esteem have been

challenged (Grusec & Goodnow, 1994; Lepper, 1981; Lewis, 1981). F inn control may

22

apply excessive pressure on children and induce external compliance to parental
authority, rather than encourage the children to internalize norms and become self
responsible. The most noticeable advocate of firm control addressed by Baumrind
(1971,1978) is that the "firm enforcement of rules, effective resistance to the child's
coercive demands, and willingness to guide the child by regime and structured
interventions" (1971, p. 87). Lewis (1981) reinterprets the measures of child compliance
used in Baumrind's research (1971) as the result from the internal responsiveness of the
children rather than from the firm control of parents. Consequently, Lewis (1981) argues
that firm control may not be a necessary antecedent of self-esteem or social competence.
She further suggests that two-way communication between parents and children, rather
than the firm control used by authoritative parents, is the primary mechanism that fosters
self-esteem and social competence in children. In this study, both firm control and

excessive control of parental behaviors are included.

Research findings have consistently showed that low levels of parental warth or
high levels of harsh or punitive parental behaviors are related to the occurrence of
problem behavior in children across demographically diverse groups of families (Berlin

et al., 1995; Florsheim et al., 1996; Ge et al., 1996; Greenberger and Chen, 1996).

3.5 Self-Esteem in Child Development

The period of middle childhood is a significant time for exploring the connection of
different types of social relations. Furthermore, this period corresponds to the time when

various components of the child’s self-concept come together to yield the general

23

affective sense of self-worth referred to as self-esteem (Franco and Levitt, 1998; Harter,
1983). According to Cooley’s looking-glass self formulation (1902), self-esteem is
socially constructed through the interaction between children and their parents or other
significant others who are important in the children’s lives. Thus, approval and
disapproval from significant others becomes incorporated into one’s own esteem for the
self. Developmentally, the internalization of parental approval or disapproval is
particularly crucial to the self-esteem formation of children (Harter, 1990, and 1993). In
other words, the appraisal from significant others is a critical determinant of children’s
self-esteem. This is consistent with the concept of “reflected appraisals”, an important

principle of self-esteem formation, addressed by Rosenberg (1986).

Within this current research framework, self-esteem is conceptualized as “the level
of global regard that one has for the self as a person” (Harter, 1993, p.88). This
definition has much in common with Rosenberg’s conception (1965, 1979, and 1986).
Both deal with self-esteem as self-evaluation in which an individual makes and maintains
with regard to self. Because of the limitations of the questions included in the survey,

this definition emphasizes perception based self-esteem rather than efficacy based.

According to Rosenberg (1979), self-esteem is different from self-acceptance. An
individual with high self-esteem has self-respect, feels worthy, recognizes personal
limitations, and expects to experience personal growth and improvement over time.
Conversely, an individual with low self-esteem exhibits self-dissatisfaction, lacks self-
respect, and has a self-picture that is disagreeable. Block and Robins (1993) and Harter
(1990) suggested that low self-esteem reflects a substantial deficit between the real self

and the ideal self. In Cooley's conception of the looking-glass self. the discrepancy

24

between one's imagination of other’s perception and one’s actual evaluations of self is a
key factor that is usually ignored. However, imagined perceptions and actual perceptions
do not always correspond to each other. Based on the examination of the empirical
evidence for the looking-glass self in over 50 studies, Shrauger and Schoeneman (1979)
find that one's evaluation to self is not always consistent with others’ evaluations to the
particular person, while one's evaluation to self can be substantially based on the
imagination on how others evaluate him/her. The discrepancy between imagined
perceptions and actual perceptions suggests that self-evaluation in the process of self-
concept formation is active rather than passive (Gecas and Schwalbe, 1983). The active

form of the "looking-glass self” is usually neglected.

A similar idea is also argued by Rosenberg (1979). That is the selective activity of
the self within the process of reflected appraisals, as well as within other processes of
self-esteem formation. Under some of the conditions, individuals will even select certain
people as significant others. Most of this selectivity occurs in the service of self-esteem
maintenance or enhancement. For Rosenberg (1979) and Cooley (1902), the self is an
active agent in the process of reflected appraisals. Consequently, the concern here is
whether or not the active form of self-evaluation in the self-esteem formation starts in
childhood; if the early start of active self-evaluation gives more challenges to parents as
social agents in the child’s socialization process. This study includes several crucial
parental behaviors and characteristics, and examines the impacts of these variables on

children’s self-esteem.

25

4. Research Hypotheses

According to the discussion in literature review, social capital, family financial
capital, and parental human capital have been identified as critical determinants of
parenting practices. Hence, social capital, family financial capital, and parental human
capital are the three predictors of parental behaviors. Then, parental behaviors are the
predictors of children’s self-esteem. Parental behaviors consist of parental warmth,
parental monitoring, and other specific behaviors such as praising, showing physical
affection, telling others positive things about the child, sending children to their room,
taking away TV or other privileges, taking away an allowance, spanking, and grounding.
The research hypotheses are formulated with the attempts to answer the research
questions mentioned previously what types of capital (e.g. social capital, human capital,
and financial capital) have the more impact on parental behaviors and what types of
parental behaviors lead to higher self-esteem for children. Based on the theoretical and

empirical literature review above, the theoretical model is illustrated as the follows.

Figure 1. Theoretical Model of Social Capital, Parental Behaviors, and
Children’s Self-Esteem

26

 

 

 

Social Capital

 

 

 

 

. Parental Children’s
Human Capital > Behaviors ’ Self-Esteem

 

 

 

 

 

 

 

 

 

 

 

 

Financial Capital

 

 

 

4.1 Reciprocity of Social Support and Parental Behaviors

This study applies Portes’ decomposition of social capital into the three elements
that are: (1) the sources of social capital (those agreeing to these demands); (2) the
possessors of social capital (those making claims); and (3) the resources themselves.
Regarding the sources of social capital, reciprocity exchange of social support is the
major focus rather than other sources due to the limitations on data sources. Reciprocity
exchange of social support in this study refers to the reciprocal relationship of giving and
receiving favors (e.g. childcare), information, emotional support, and other valued items

between parents (or primary caregivers) and people living outside the household.

For the possessors of social capital (those making claims), parents here are treated
as the persons who make claims of social capital for their children. According to Portes’

definition of social capital (Portes and Sensenbrenner, 1993; Portes, 1998), expectations

27

for action and the ability of actors are the critical factors to facilitate actions for securing
or claiming benefits. Children can be beneficial from resources that are claimed or

transformed by their parents.

The resources in this study refer to parental behaviors that can be supportive or
obstructive to children. In short, the research framework includes four various levels of
social contexts. The first is resources from outside of the family; that is social support
from people outside of the household. The second is resources inside the family including
parental human capital and family income. The third is parents, precisely parental
behaviors. The fourth level focuses on children, in particular children’s self-esteem.
Parents in this case play the key role to transform and claim the resources to increase the

well-being for their children.

According to the perspectives of social support and social capital, reciprocity is
regarded as an important factor that influences the exchange of social support or other
resources for individuals (Antonucci, 1985; Coleman, 1988a; Portes, 1998; Portes &
Sensenbrenner, 1993). Non-reciprocal network is perceived as inadequate and less
positive in terms of life satisfaction and happiness. Reciprocity between providers and
recipients is more likely to lead to positive and longer relationships (Antonucci, 1985),

that may buffer each other’s stress for a long term.

The receiving of emotional or instrumental social support is supposed to relieve the
stress of parenting, while the providing of social support can increase one’s sense of self-
management and prolong the support relationship that also decrease the hardship of being

a parent. Those mothers who have less frequent negative interactions with significant

28

others provide positively affectionate and sensitive care to their children (Cotterell, 1986;
Crnic et al., 1983; Pascoe et al., 1981; Weinraub & Wolf, 1983; Zarling, Hirsch &

Landry, 1988).

Therefore, the reciprocal support exchange can be a critical process in which parents
learn various ways to interact with others supportively; then, the new experience can lead
to more supportive parental behaviors. Support reciprocity between parents and parents’
social ties (e. g. grandmother/grandfather, friends, co-workers) results in warm and

supportive parental behaviors. The first hypothesis is formulated as:

Support reciprocity between parents and people outside the household (6. g. own
mother/father, grandmother/grandfather, friends, siblings, and co-workers) predicts
supportive parental behaviors such as the higher degrees of parental warmth, parental
monitoring, telling others positive things about the child, and praising, and showing

physical affections.

4.2 Financial Capital and Parental Behaviors

Family income and family financial strain, as two indicators of financial capital,
have been identified as determinants of parental behaviors to be supportive or harsh
(Shumow, Vandell & Posner, 1998; Takeuchi, Williams, & Adair, 1991). Research
findings indicate that greater parental harshness relates to lower family income and

higher financial strain (Conger et al., 1990). In other words, the less family financial

29

strain and the higher family income parents have, the more likely the parents provide

supportive parental behaviors to their children. Specifically, the hypotheses are:

The less family financial strain, the more likely parents have supportive parental
behaviors such as the higher degrees of parental warmth, parental monitoring, telling

others positive things about the child, praising, and showing physical affections.

The higher family income, the more likely parents have supportive parental
behaviors such as the higher degrees of parental warmth, parental monitoring, telling

others positive things about the child, praising. and showing physical affections.

4.3 Human Capital and Parental Behaviors

Parental education is considered to be a determinant of supportive parenting
(Kelley, Power, & Wimbush, 1992; Shumow, Vandell & Posner, 1998). Research
findings indicate that supportive, reasoning, and firm-control parenting is related to
higher maternal education (Brody et al.,l995; Elder, Eccles, Ardelt, & Lord, 1995), while
greater parental harshness relates to lower parental education (Dombusch et al., 1987;
Kelley, Sanchez-Hucles, & Walker, 1993; Raikkonen & Keltikangas-Jarvinen, 1992). In
other words, the higher parental education leads to more supportive and reasoning
parenting. Theoretically, the higher levels of parental education and parental self-esteem

predict supportive parental behaviors. The hypothesis is formulated as:

30

The higher the parental education is, the more likely parents have supportive
parental behaviors such as the higher degrees of parental warmth, parental monitoring,

telling others positive things about the child. praising, and showing physical affections.

4.4 Parental Behaviors and Children’s Self-Esteem

Parents have been seen as the primary social agents to bring and secure resources
for their children and to create human capital in the next generation. Parental human
capital has been considered highly related to children's cognitive and socio-emotional
development (Parcel and Menaghan, 1993, 1994). Anyhow, parental human capital must
be transformed by some forms of interaction to their children. Therefore, parental
behaviors are treated as the primary contact between parents and their children in shaping
children’s self-esteem. A great deal of research findings support that the warm, clear,
consistent, and legitimate parental control facilitates children’s self-esteem by
contributing to the internalization of values and expectations that, in turn, lead to greater
confidence in children’s definition of the standards for expected outcome (Baumrind,
1978; Coopersmith, 1967; F elson & Reed, 1986; Higgins, 1991; Openshaw & Thomas,
1986; Peterson, Rollins & Thomas, 1985; Putallaz & Heflin, 1990). Specifically, the

hypothesis is formulated as:

31

The higher degree of supportive parental behaviors (including parental warmth,
parental monitoring, telling others positive things about the child, praising, and showing

physical affections). the higher children’s self-esteem.

32

CHAPTER II
METHOD

Following the introduction of research framework, literature review, and research
hypotheses, this chapter focuses on data used for this study, the data collection procedure,
the instruments used for collecting data, the sample, the measures used to observe
research constructs, and some technical issues including data management (e.g.
extracting the target subsample, and dealing with missing data, and variable

reconstruction).

1. Data

The current study uses the data set collected by the Child Development Supplement
(CDS) to the Panel Study Income Dynamics (PSID). The CDS data were collected by the
University of Michigan Survey Research Center in 1997 with funding from the National
Institute of Child Health and Human Development (NICHD). The CDS data were
released to the public at the CDS project web site in May, 1999
(http://www.isr.umich.edu/src/child-development/home.html). The CDS is a separate
project to the PSID rather than just a supplement study. It is for the first time since 1981
that the comprehensive national time-use information was collected from children of all

ages in several settings. This data set provides information on social support, financial,

33

time, and social-psychological resources with measurement taken at the level of the

family, neighborhood, and school for 0-12 year old children and related childcare givers.

The Panel Study of Income Dynamics (PSID) is a longitudinal survey and has
collected data annually from a representative sample of US. men, women, children, and
the families in which they reside since 1968. The data collected by the PSID include
employment, income, wealth, housing, food expenditures, income transfers, marriage,
and children from 5,000 families in 1968. The study has grown to include over 10,000
families currently. Core funding for the PSID is provided by the National Science
Foundation. The information on children’s development and experience as children has
been limited to demographic characteristics (e.g. age and sex), and schooling. There are
no direct assessments or reports of children's development and experiences as children.
Consequently, it is vague on the childhood process whereby early experiences in various
contexts like family, school, and neighborhood determine children’s successes as young

adults.

As Hofferth and her associates (1997) point out, it has been a major research focus
on the consequences of family circumstances and events such as family structure and
income for children's educational and economic successes as young adults by using the
PSID data in recent years (Duncan, Brooks-Gunn & Klebanov, 1994; Duncan, Yeung,

Brooks-Gunn, & Smith, 1998; Haveman & Wolfe, 1994; McLanahan & Sandefur, 1994).

The Child Development Supplement (CDS) rectifies this situation and collects
additional information that is not contained in the main PSID. The CDS data include: (1)

measurements of economic and demographic conditions at both the individual and the

34

family levels. as well as neighborhood and school characteristics at the time of the
interview; (2) information from the primary caregiver instead of directly targeting on a
child’s mother or father in the household; (3) information on family processes such as
parenting, and social relationships between parents, between parents and children, and
between parents and non-family members; (4) age graded assessments of child cognitive,
behavioral, and health from the children, the child caregivers, the teachers and child care
providers; and (5) the learning environment in the home, teacher and administrator

reports of school resources, and parent-reported measures of neighborhood resources.

The C DS data support the current study whereby a variety of variables critically
related to child development are collected from both the primary caregiver and the child.
It is a unique strength that the responses are from the primary caregiver in the household
instead of simply asking either one of the parents to answer the questions. The primary
caregiver is whoever spends most of the important time with the child. The influence
from the primary caregiver to the child can be profound and critical during the process of
child development. Although, for most pairs of the primary caregiver and the child in the
sample, their relationship is mother-child in the same family, the key point is at least we

know in most cases mothers still play the role of the primary caregivers.

In addition, there are several critical variables collected by the CDS. These variables
consist of child cognitive and behavioral assessments, parenting behaviors, parental
characteristics, social support from outside of the household, family structure, family
income, and family financial strain. The richness of the CDS data facilitates the
integration of social support and social capital theories into the areas of parenting and

child development in the both individual and family levels in this current study.

35

Specifically, social support to the primary caregiver is measured from both sides of
the recipient and the provider; consequently, the pattern of social exchange can be
obtained for investigating the mechanism of resource transformation from parents to
children. The measures of social support are adopted from the 1987-1988 National
Survey of Families and Households (N SF H)(Sweet, Bumpass, and Call 1988). However,
the focus of the NSFH is not on child development, therefore no data can be accessed to
study the effects of parental social support on the parents and their children. The
application of social capital theory is made possible by the usage of the CDS data. The

CDS data are considered to be highly appropriate for this study.

The CDS study is planned to be longitudinal; however, currently only the first wave
of data collections is released. Therefore, this current study is limited to focus only on

cross-sectional analyses based on the first wave.

2. Data Collection and the Instruments

Data collection of the CDS began in January 1997 and was completed in November
1997. In 1997, the Core PSID Family Unit (FU) interview was first completed that
determined the number and ages of children. If there was a child in the family unit
between birth and age 12, the primary caregiver of each eligible child was contacted
about the Child Development survey. For the family units containing more than two
children age twelve and younger, the central office randomly selected two children and
assigned time diary days, and determined a likely primary caregiver. If the primary

caregiver is being interviewed about two children, a separate questionnaire for each child

36

was completed. The primary caregiver of each child was visited for the written
permission to interview the children, obtained child assessments, and child interviews, as
well as a primary caregiver interview and assessment, time use diaries and left self-
administered instruments. In addition to the primary caregiver, the absent father who
lived outside of the home, child caregivers, teachers, and administrators in day care or
school were asked to answer different questionnaires (see Appendix A for more

information in interview procedure).

There are 13 different questionnaires and booklets used for data collection. These
questionnaires and booklets are listed in Table 2 (also see Appendix B for more details).
For this study, only three sets of data are used. These three sets of data are collected by
three instruments that are: (1) Primary Caregiver-Child Booklet, (2) Primary Caregiver-
Household Booklet, and (3) Child Questionnaire. Originally, all data are saved
independently in different files corresponding to the instruments used for data collection.
In addition to the three data files mention above, the demographic file is included for the
demographic information. The demographic data were collected during the PSID main
panel data collection in 1997. This data set is comprised of one record for each child in
the CDS sample. Therefore, respondents from the same household can be recognized by

the family ID.

37

Table 2. Questionnaires and Booklets for Data Collection

 

Primary Caregiver - Child Booklet "

Primary Caregiver - Household Booklet *

Child Questionnaire '

Time Diary

Other Caregiver - Child Booklet

Other Caregiver - Household Booklet

Fathers Outside of the Home - Child Booklet
Fathers Outside of the Home - Household Booklet
Elementary/Middle School - Teacher Booklet

S°P°.\‘.°‘.°‘P.°°N.-‘

—L
.0

Preschool/Daycare - Teacher Booklet

A
A

. Home-Based Care
Elementary/Middle School - Administrator Booklet

—L
N

13. Preschool/Daycare - Administrator Booklet

 

Note: * means that data collected by the instniment are used in this
dissertation.

Primary Caregiver - Child Booklet. The respondents of this booklet are the
primary caregivers. The primary caregivers were asked to report information about each
target child by face to face interview as a preferred mode and telephone interview as
optional. The main information collected by this instrument consists of primary
caregiver’s literacy, children's health history, children's schooling, home environment,
children's behavior, household tasks, parental involvement in school, parental monitoring
of children's activities and friends, child care history, relationship with absent parent (if

applicable), and food availability at home.

38

Primary Caregiver - Household Booklet. The respondents of this booklet are the
primary caregivers. The primary caregivers were asked to report information about the
target child's household and neighborhood. This booklet is administrated in a self-
administered mode, and face to face or telephone interview as optional. Each primary
caregiver completed only one household booklet regardless of the number of children
whom the primary caregiver took care of in the sample. For the households where
contained two primary caregivers, both were asked to complete this booklet. The main
information collected by this instrument consists of characteristics of the family's
neighborhood, social support for the mother/primary caregiver, the distribution of
household tasks, attitudes on childrearing, parenting, attitudes on gender roles in the

household, psychological well-being, family conflict, and work schedules.

Child Questionnaire. The respondents of this booklet are the children who were 3 -
12 years old. The mode of data collection was face to face interview only. Multiple
assessments were administered to children of the appropriate ages. The main information
collected by this instrument consists of children's reading and math ability, children's

memory. self-esteem related to school subjects and to children’s general lives.

3. Sample

The sample for this project was drawn from the 1997 PSID interviews. The PSID
has collected data annually since 1968. As interviews were completed for the 1997 PSID,

households with children who were Family Unit (FU) members under the age of 13 were

39

identified for inclusion in the CDS. Both the PSID core sample and the new PSID
immigrant refresher sample generated Child Development cases. The stratified selection
procedures were applied for sampling. The sample was originally divided into four strata
based on the geographical distance from the interviewing staff. Households outside the

continental US were excluded from interviewing and were coded as non-sample.

The actual number of households eligible for the Child Development Supplement
from the PSID is 2,705. Among these 2,705 households, 2,458 are from the Core sample
and 247 are from the New Immigrant sample. A total of 2,390 households with a total of
3,586 children are interviewed for the Child Development Supplement. The response rate
is 88.2%. Of these 3,586 children identified by CDS, boys and girls are represented in
approximately equal numbers. The race composition is that White children constitute
46.03% of the total child respondents; Black 40.98%; Hispanic 7.49%; Asian 1.76%;
Native American 0.53%; other 2.99% (see Appendix C). The substantial number of
black and other minority families of the PSID sample is the result of an initial
oversampling of low-income families and the addition of a recent sample of immigrant

families.

In order to extract the subsarnple of interests in this current study, a multi-step

procedure of case selection is performed. The procedure of case selection is as follows:

Only children who answered the questions of self-esteem in the Child Questionnaire
are included, and their ages range primarily from 8 to 12. The available child

respondents are 1,088; each child respondent has a correspondent primary caregiver that

40

answered other sets of questionnaires. Both children and their primary caregivers are

included for data analysis.

Among the 1088 children, 328 cases as 164 pairs of siblings who share the same
primary caregiver within each pair of sibling. If all 328 cases are included for data
analysis, the primary caregivers would be overrepresented due to the repeated use of the
household data from the primary caregivers. For avoiding the overrepresentation of the
primary caregivers who had two children in the sample, the older child from the same
family is dropped. The choice of younger children is due the skewed distribution toward
the older age among those children without another sibling answering the questionnaires.
For other demographic variables including sex and race, these two groups (the younger
and the older children from the same household) have almost the same distributions.
After taking out the older children, the valid cases are 924 pairs of primary caregivers and

children (see Table 3).

41

Table 3. Frequency Distribution of Age of Child by Number of
Children Answering Questionnaires from the Same

 

 

 

 

Household
One Child Two Children
Age of Child Answered Answered Total
Older Younger
6 1 0 0 1
7 13 0 10 23
8 137 3 55 195
9 141 15 46 202
10 166 39 31 236
11 131 47 16 194
12 171 60 6 237
Total 760 164 164 1088

 

 

 

 

 

For the purpose of obtaining primary caregiver’s demographic variables (e.g.
primary caregiver’s education, age, sex, employment status), some cases are dropped. If
the relationship between the primary caregiver and the child is mother-to-child or father-
to-child, and the relationship between the head of household and the child is also mother-
to-child or father-to-child, the pair of a primary caregiver and the child is included as
valid cases in this study. Finally, only 875 pairs are included for the data analysis. This
step relates to the process of variable reconstruction in the later section, and more details

will be discussed.

For the 875 pairs of primary caregiver-child cases, there are around 256 missing
cases for variables of the Financial Strain Scale and the social support related variables. It
is a high proportion of missing data for these critical independent variables. They present
two forms of capital. The 256 pairs of cases are dropped out of the analysis. These cases
are with missing data for variables of the Financial Strain Scale and the social support

related variables. Thus, after taking out the cases with missing data, the valid cases are

42

619 pairs of primary caregivers and children. The process of dealing with the missing

data is discussed after the sections of “variable construction” and “measures”.

4. Variable Reconstruction

The demographic variables of the primary caregiver are from the demographic file
that is extracted from the PSID main panel data collection. The respondents to the PSID
annual interviews are usually the male adult head of a household, or sometimes the wife
or cohabitor of the head. This single household respondent provides information about
every member in this family including him/herself. Unfortunately, the demographic
variables of the primary caregiver are not immediately available after merging the data
files together. Therefore, education, age, and sex of the primary caregiver need to be

reconstructed to present the same construct as one single variable.

Four variables are used to identify the role of the primary caregiver in the
household, and they are (1) the relationship of the primary caregiver to the child, (2) the
relationship of the child to the head, (3) the sex of head, and (4) the household type. After
determining the role of the primary caregivers as wives or heads, new demographic
variables are created by retrieving data from corresponding variables. For example, the
primary caregiver’s education needs to be retrieved from either “head education” or “wife
education”. For those who are neither head nor wife, their education levels would be

missing.

For the specific parental behaviors, the similar problem occurs. The same set of

questions is used to ask two different groups of the primary caregivers whose children are

43

aged from 6 to 9 as one group, while the others are 10 and older as the other. The data for
these two groups are saved into two sets of variables. Data for the same question are

combined into one single variable to present each parental behavior.

5. Measures

According to the research theoretical framework of Figure 1 and hypotheses
discussed above, the measures used for the target variables in this study are listed as

follows.

5.1 Social Capital and Support Reciprocity (as the Independent Variable)

According to Portes’ decomposition of social capital, social capital is
operationalized into three levels of measures. The first level is “the sources of social
capital” such as social support for parents. The second level is “the possessor of social
capital” such as parents in a family. The third level is “the resources of social capital”.
Parental behaviors are used as the measures that indicate the level of resources children

received from the primary caregivers.

For the second level, “the possessor of social capital” (who makes the claims), the
focus follows the propositions that parents have long been seen as the primary social
agents to bring and secure resources for their children in the process of socialization. The
possessor of social capital who makes the claims is linked to the primary caregivers or

the parents. The relationship between the child and the primary caregiver is used as the

44

measure. However, most of children’s primary caregivers are mother, stepmother, or
adoptive mother in this data set as mentioned previously. The details of the first and the

third levels will be discussed in the following sections.

Support reciprocity between the primary caregiver (or the parent) and others is used
as the first level of measures, the sources of social capital. The sources here focus on the
parental social ties instead of the children’s social ties. Parental social ties primarily focus
on people outside the household who provide (or receive) social support to (or from) the
primary caregivers. Two different types of social support are measured including
emotional and instrumental support. Furthermore, the people who provided (or received)
support to (or from) the primary caregiver were asked according to their social
relationship. The information on the social ties between the primary caregiver and people
outside the household is critical for identifying the pattern and range of the primary

caregiver’s network. and how different types of social support exchanged or invested.

The measures of support reciprocity are drawn from the National Survey of Families
and Households and from the 1980 Panel Study of Income Dynamics (see Appendix D).
This set of items measures receiving and provision of emotional and instrumental
assistance from (or to) others who are not living with the respondent over the past month.
The question asks if the primary caregiver received any help like child care,
transportation, and repairs to your home or car in the past month. If the answer is
positive, the respondents are asked about those who gave the help. Each respondent can
mention up to five persons including former spouse or partner, own mother/father,
mother/father-in-law, own grandmother/grandfather, spouse’s grandmother/grandfather,

sister/brother, aunt/uncle/ cousin, child’s other parent, friend, counselor/minister/other

45

clergy, members of church or other organization, co-workers, grown child, and other. The
same questions are asked about support provided by primary caregivers to people outside
the household. Therefore, support reciprocity will be identified by using computer
software programming to match respondent’s answers on the support recipients and

support providers.

Specifically, the matching between the support recipients and support providers is
treated by three steps. The first step is to identify those who receive and those who
provide help by what type of support. The second step is to mark respondents into two
sets of four different categories. These two sets are divided by two types of support,
emotional and instrumental support. The four categories are both receiving and
providing, receiving only, providing only, and neither receiving nor providing. The third
step focuses on those who both received and provided support from and to others. Each
respondent who received and provided support answered questions about up to five
relationships that helped most often in the past month. Support recipients and providers
with the same relationship to the respondents will be matched. The matching is to see the
pattern of relationships from support recipients and providers to respondents. The
concern of the matching procedure is on what level of support reciprocity makes a
difference on the primary caregiver’s behaviors. Is it the simple reciprocal support
regardless of the relationship from the recipients and providers to the respondents, or is it

the level of reciprocity that must be fed back between the same relationship?

46

5.2 Parental Behaviors (as the Mediator)

Parental behaviors are used as the third level of measures on social capital, “the
resources of social capital”, that indicate resources children received from the primary
caregivers. Parental Behaviors are measured by (1) parental warmth (see Appendix E);
(2) parental monitoring (see Appendix E); (3) other specific parental behaviors:
grounding, spanking, taking away privileges and allowance, sending the child to a room,
telling positive things about the child to others, praising, and showing physical affection

(see Appendix F).

Parental warmth includes 6 items, asking about how often in the past month the
primary caregivers (I) hugged or showed physical affection to your child, (2) told your
child that you love the child, (3) spent time with your child doing the child’s favorite
activities, (4) joked or played with your child, (5) talked with your child about things the
child is interested in, and (6) expressed appreciation for something the child did.
Responses ranged from not in the past month (1) to every day (5). Each item will be
scored from 1 to 5. The summary scores range 9 to 30. Higher scores indicate higher
parental warmth. The reliability coefficient (Cronbach’s Alpha) of the scale of parental

warth is .80.

Parental monitoring includes 8 items, asking about how things are going in general
in the child’s life including health, friendships, relation to the primary caregiver, etc.
Responses ranged from excellent (1) to poor (4). The negative items are recoded before
computing the average scores of the scale. Each item will be scored from 1 to 4. The
summary scores range 14 to 32. Higher scores indicate higher parental monitoring. The

reliability coefficient is .81.

47

Specific parental behaviors are measured by 8 items including praising, showing
physical affection, telling others positive thing about the child, grounding, spanking,
taking away privileges, taking away allowance, and sending to room. For these 8 specific
parental behaviors, the former three behaviors that are praising, showing physical
affection, and telling others positive things about the child are treated as positive
behaviors to children are treated as positive behaviors to children, and the later five are
treated as negative including grounding, spanking, taking away privileges, taking away

allowance, and sending to room.

These questions are originally asked about how many times in the past week the
caregivers have done these behaviors to the child. In this study, these 8 items are recoded
into dichotomous variables to show whether the primary caregiver did any behavior to
the child. The code of one means the primary caregiver did the behavior, while the code
of zero means the primary caregiver did not do the behavior. The reason for recoding
these parental behaviors is because the distribution of every negative parental behavior is
highly skew toward zero time. For grounding, 73% of the primary caregivers reported
that they had never done this behavior to the child in the past week, 92.2% for spanking,
69.5% for taking away the child’s privilege, 87.1% for taking away the child’s allowance,
and 62.5% for sending the child to a room. The distribution of every positive parean
behavior is highly skew toward between once and one hundred times during the period of
one week. Only 7.4% reported that they had never spoken child’s positive things to
others, 5.8% had never praised the child, and 1.8% had never shown physical affection to

the child.

48

5.3 Human Capital and Other Parental Characteristics (as Independent Variables)

The construct of human capital is measured by the primary caregiver’s education.

Parental education is categorized by years of schooling ranging from 1 to 20 years.

5.4 Financial Capital (as the Independent Variable)

The construct of financial capital is measured by: 1) family income; 2) family
financial strain. The scale of family financial strain is drawn from the work of Glen Elder
and Rand Conger in measuring experiences of economic or financial stress and strain and
practical responses to such financial pressures (Conger & Elder, 1994) (see Appendix G).
This scale includes 16 items including sold possessions or cashed in life insurance,
postponed medical care, postponed major purchases, and etc happened in the last year.
Each item asked the occurrence of each problem. All 16 items are counted into one
variable that indicates the number of economic problems that happened in the last year.
The summary scores range from 0 to 10. The reliability coefficient is .64, which are not

as good as those of other scales used in this study.

5.5 Children’s Self-Esteem (as the dependent variable)

This construct is measured by 8-item scale created by Marsh (1990). Only children
aged 8-12 answered these items (see Appendix H). All items in this scale are positively
stated and responses range from never (1) to always (7). Each item will be scored from 1

to 7. The summary scores range from 17 to 56. Higher scores indicate higher self-esteem.

49

The reliability coefficient (Cronbach’s Alpha) of the scale of children’s self-esteem is

.74. This is an acceptable reliability level.

5.6 Socio-Demographic Variables (as Control Variables)

Socio-demographic variables used in the study include the primary caregiver’s age
and sex. The limitation on race related variables is the lack of primary caregiver’s racial
background. Race of primary caregivers is available only when they are the heads of the
household. However, more than half of the primary caregivers are not the heads of the

household, thus their racial background is unable to be identified.

6. Mean Differences between the Target Subsample and Missing Data

The multi-step procedure of case selection has been performed to extract the
subsample of interests in this study. There are some cases lost in each step of case

selection.

For the first step, only children who answered the questions of self-esteem in the
Child Questionnaire are included. Around 30% (=1 ,088 / 3,586) of the CDS child
respondents are selected for this study. The second step is to avoid overrepresentation on
the primary caregivers (PCG). 164 (=1 ,088-924) cases are taken away. The third step is
to reconstruct demographic variables, and 49 (=924-875) cases could not be identified.

For the last step, 256 primary caregivers reported no information on social support and

50

financial strain related variables. This leads to only 619 (=875-256) cases included for

data analysis.

The differences between the target subsample and the missing data serve as
comparison levels for justifying the limitations of the CDS data used in this study. The
following sections focus on the comparison between the target subsample and the 256
missing cases that result from the last step, missing data on social support and financial

strain related variables.

The purpose of analyzing the differences between the target subsample and missing
data is to confirm that the characteristics of missing group are not significantly different
from the target group. Consequently, the drop of missing data does not make significant

effects on the research results.

The means of socio-demographic variables (including the primary caregiver’s age,
sex, education, and family income) and children’s self-esteem for both the target and the
missing groups are listed in Table 4. The primary caregiver’s mean age of the missing
group tends to be younger than that of the target subsample. Both the primary caregiver’s
mean education and family income of the missing group tend to be lower than those of

the target subsample are. Children’s self-esteem of the missing group is also lower.

51

Table 4. Means of Socio-Demographic Variables and Children’s
Self-Esteem for the Target Subsample and the Missing

 

 

Group
Tagget Group Missing Group
PCG Age Mean 37.63 36.68
Std. Deviation 6.01 6.00
Range 52 (=73-21) 44 (=66-22)
N 615 255
P06 Sex Mean 0.03 0.04
Std. Deviation 0.17 0.19
Range 1 (=1-0) 1 (=1-0)
N 619 256
PCG Education Mean 12.95 12.33
Std. Deviation 2.40 2.65
Range 15 (=17-2) 15 (=17-2)
N 609 254
Family Income Mean 51447.69 45632.33
Std. Deviation 50572.37 97243.92
Range 577000.00 151200000
(=577000-0) (=1512000-0)
N 619 256
Children’s Self- Mean 44.83 44.18
Esteem Std. Deviation 6.70 7.38
Range 39 (=56-17) 48 (=56-8)
N 612 251

 

One-way analysis of variance (ANOVA) is used to test whether the mean responses

of the target subsample and the 256 missing cases are significantly different from each

other. There are significant differences in the primary caregiver’s age and education (see

Table 5).

52

Table 5. ANOVA of Socio-Demographic Variables and Children’s Self-Esteem
between the Target Subsample and the Missing Group

 

 

Sum of df Mean F P
Squares Square Value Value
PCG Age Between Groups 161.07 1 161.07 4.46 0.03
Within Groups 31329.00 868 36.09
Total 31490.07 869
PCG Sex Between Groups 0.01 1 0.01 0.40 0.53
Within Groups 28.03 873 0.03
Total 28.04 874
PCG Education Between Groups 67.67 1 67.67 11.03 0.00
Within Groups 5281.98 861 6.13
Total 5349.65 862
Total Family Between Groups 6.12E+ 1 6.12E 1.34 0.25
Income Within Groups 3.99E+ 873 4.57E
Total 4E+12 874
Children’s Between Groups 74.84 1 74.84 1.57 0.21
Self-Esteem Within Groups 41073.23 861 47.70
Total 41148.08 862

 

For parental warmth, monitoring, and positive parental behaviors (including telling
positive things of the child to others, praising, and showing physical affection), all means

of the target group are slightly higher than those of the missing group (see Table 6).

53

Table 6. Means of Parental Warmth, Monitoring, and Positive Parental
Behaviors for the Target Subsample and the Missing Group

 

 

Tagget Group Missing Group

Parental Wamrth Mean 25.77 25.44

Std. Deviation 3.57 4.23

Range 21 (=30-9) 19 (=30-11)

N 616 252
Parental Monitoring Mean 26.97 26.77

Std. Deviation 3.37 3.56

Range 18 (=32-14) 19 (=32-13)

N 528 223
Telling Positive Mean 0.93 0.91
Things to Others Std. Deviation 0.26 0.29

Range 1 (=1-0) 1 (=1-0)

N 614 247
Praising Mean 0.94 0.91

Std. Deviation 0.24 0.28

Range 1 (=1-0) 1 (=1-0)

N 611 249
Showing Physical Mean 0.98 0.97
Affection Std. Deviation 0.14 0.17

Range 1 (=1-0) 1 (=1-O)

N 593 248

 

There is no significant difference in parental warmth, parental monitoring, and

positive parental behaviors between the target and the missing groups according to one-

way analysis of variance (see Table 7).

54

Table 7. ANOVA of Parental Warmth, Parental Monitoring, and Positive
Parental Behaviors between the Target Subsample and the Missing
Group

 

Sum of df Mean F P
Squares Square Value Value

 

Parental Warmth Between Groups 19.08 1 19.08 1.34 0.25
Within Groups 12336.95 866 14.25
Total 12356.03 867
Parental Monitoring Between Groups 6.29 1 6.29 0.53 0.46
Within Groups 8809.91 749 11.76
Total 8816.2 750
Positive Parental Behaviors
Telling Positive Between Groups 0.04 1 0.04 0.48 0.49
Things to Others Within Groups 62.59 859 0.07
Total 62.63 860
Praising Between Groups 0.15 1 0.15 2.44 0.12
Within Groups 53.94 858 0.06
Total 54.09 859
Showing Physical Between Groups 0.02 1 0.02 0.78 0.38
Affection Within Groups 17.60 839 0.02
Total 17.61 840

 

For negative parental behaviors, the means of “grounding” and “spanking” for the
target group are slightly lower than those of the missing group, but the means of the other

three behaviors for the target group are higher (see Table 8).

55

Table 8. Means of Negative Parental Behaviors for the Target

Subsample and the Missing Group

 

 

Target Group Missing Group
Grounding Mean 0.26 0.27
Std. Deviation 0.44 0.45
Range 1 (=1-0) 1 (=1-0)
N 616 254
Spanking Mean 0.08 0.10
Std. Deviation 0.27 0.30
Range 1 (=1-0) 1 (=1-0)
N 618 254
Taking Away Privilege Mean 0.30 0.28
Std. Deviation 0.46 0.45
Range 1 (=1-0) 1 (=1-0)
N 616 254
Taking Away Allowance Mean 0.08 0.07
Std. Deviation 0.26 0.26
Range 1 (=1-0) 1 (=1-O)
N 583 233
Sending to Room Mean 0.37 0.35
Std. Deviation 0.48 0.48
Range 1 (=1-0) 1 (=1-0)
N 614 253

 

There is no significant difference in any negative parental behavior between the

target and the missing groups according to one-way analysis of variance (see Table 9).

56

Table 9. ANOVA of Negative Parental Behaviors behrveen the Target Subsample

 

 

and the Missing Group
Sum of df Mean F P
Squares Square Value Value
Grounding Between Groups 0.03 1 0.03 0.17 0.68
Within Groups 168.22 868 0.19
Total 168.25 869
Spanking Between Groups 0.12 1 0.12 1.62 0.20
Within Groups 66.76 870 0.08
Total 66.89 871
Taking Away Between Groups 0.06 1 0.06 0.29 0.59
Privilege Within Groups 181.43 868 0.21
Total 181 .49 869
Taking away Between Groups 0.00 1 0.00 0.02 0.90
Allowance Within Groups 56.44 814 0.07
Total 56.44 815
Sending to Room Between Groups 0.06 1 0.06 0.25 0.62
Within Groups 200.77 865 0.23
Total 200.83 866

 

For the primary caregiver’s sex, total family income, parental behaviors and
children’s self-esteem, there are no significant differences between these two groups.
Although the mean differences of age and education exist between these two groups, the
dependent variables (children’s self-esteem and parental behaviors) show no significant
difference. For the consideration of consistency of case inclusion for analysis, the 256

pairs of primary caregivers and children will not be included in data analysis.

57

7. Profile of Respondents in the Target Subsample

The frequency distribution of the demographic variables for the target subsample is
demonstrated in Table 10. As can be seen, 600 (96.9%) of the respondents are female. In

other words. 96.9% of primary caregivers are the child’s mother.

Table 10. Distribution of Socio-Demographic
Variables of the Tflget Subsample

 

 

Variable Category Frequency %
PCG Sex Female 600 96.9
(N=619) Male 19 3.1
PCG Age 21-30 63 10.2
(N=615) 31—40 357 58.0
41-50 186 30.2
51-60 6 1.0
61 -73 3 0.5
PCG Education 211 98 16.1
(N=609) 12 215 35.3
(in Years) 13-14 152 25.0
15-16 103 16.9
17 41 6.7
Child Age 7-8 129 20.9
(N=619) 9 130 21.0
10 135 21.8
11 104 16.8
12 121 19.5
Child Sex Female 317 51.2
(N=619) Male 302 48.8
Child Race White 317 51.4
(N=617) Black 240 38.9
Hispanic 30 4.9
Asian 6 1.0
Native 4 0.6
Other 20 3.2

 

58

Age of primary caregivers ranges from 21 to 73 years old, but the majority (357 or
58%) concentrates between 31 to 40 years old. More than one-third of the primary
caregivers (215 or 35.3%) has finished 12 years of education. As for child age, the range
is from 7 to 12 years old. The sex distribution of children is 48.8% of male and 51.2% of
female. For child race, white and black children are the two largest groups. The former is

51.4% and 38.9%.

59

CHAPTER III
RESULTS

According to the theoretical model (see Figure 1) and research hypotheses, social
capital, parental human capital, and financial capital are the three predictors of parental
behaviors. Parental behaviors consist of parental warmth, parental monitoring, and other
specific parental behaviors including praising, showing physical affection, telling others
positive things about the child, sending children to their room, taking away TV or other
privileges, taking away an allowance, spanking, and grounding. The research hypotheses
are formulated with the attempts to answer the research questions mentioned previously
what types of capital (e. g. social capital, human capital, and financial capital) explain

more variance on parental behaviors.

For the purpose of investigating the effect from each form of capital to each parental
behavior, only one form of capital as the predictor(s) is entered into a regression equation
each time. Although all predictors have been put into regression equations at once, it is
difficult to distinguish the effect from one predictor to another. In addition, the more
predictors a regression equation includes, the more likely multicollinearity occurs.
Multicollinearity is the undesirable situation where the correlations among the
independent variables are strong. The results shown in the following sections focus on the
analysis on the effect from each form of capital to each parental behavior, and on that

from each parental behavior to children’s self-esteem.

60

1. Social Capital and Parental Behaviors

For testing the first hypothesis: “Support reciprocity between parents and people
outside the household (e. g. own mother/father, grandmother/grandfather, friends,
siblings, and co-workers) predicts supportive parental behaviors like the higher degrees
of parental warmth, parental monitoring, telling others positive things about the child,
praising, and showing physical affections”. Social support here is categorized into two
types, instrumental and emotional support. Support reciprocity is classified into four
types: receive and provide, receive only, provide only, and neither receive nor provide.
The analysis on support reciprocity focuses on only the parallel exchange within the same
type of social support. Consequently, the exchange from the instrumental support to the
emotional is not the focus. In other words, the support exchange across different types of
social support is not included. For example, if one respondent provides instrumental
support to a friend and the friend provides emotional support back to that respondent, this

case is excluded in the analysis of this study.

In addition, social tie is another dimension to be considered. The pattern of support
reciprocity may be different from one social tie to another. The following sections are
divided into two parts: the first part focuses on the overall exchange of social support
without consideration of social ties; the second focuses on the exchange within social ties

such as providing support to friends and receiving support from friends, also.

For the continuous dependent variables including parental warmth and parental

monitoring, multiple regression is applied to examine which type(s) of support

61

reciprocity is/are the best predictor(s) of the parental behaviors. The unstandardized
regression coefficients are used in most of the tables of multiple regression. “The
unstandardized coefficient remains fairly stable despite differences in the variances and
the covariances of the variables in different settings or populations” (Pedhazur, 1982,
p.247). The standardized coefficient is “sample-specific and can not be used for the
purpose of generalizations across settings and populations” (Pedhazur, 1982, p.247).
Therefore, the unstandardized coefficient is used in this study instead of the standardized

coefficient.

For the dichotomous variables including the eight specific parental behaviors such
as telling others positive things about the child, praising, and so on, logistic regression is
applied to predict the probabilities of these specific parental behaviors for the primary
caregivers who are characterized by different types of support reciprocity. Among these
eight specific parental behaviors, “telling others positive things about the child” and
“praising the child” are the two major variables that are frequently and significantly
predicted by support reciprocity in general or within certain social tie. Consequently, the
results concerning specific parental behaviors concentrate only on these two supportive
parental behaviors, “telling others positive things about the child” and “praising the

child”.

1.1 Overall Social Support Reciprocity and Parental Behaviors

The following sections focus on the results of data analysis on the prediction from

support reciprocity to parental behaviors. For overall support reciprocity, social support

62

from any social tie is counted no matter whom the primary caregivers receive social
support from. Four categories of support reciprocity (including both receive and provide,
receive only, provide only, and neither receive nor provide) are created for each type of
social support, emotional and instrumental support, by applying the principle of parallel
exchange within the same type of social support to match the support recipients and

support providers.

1.1.1 Overall instrumental support reciprocity and parental behaviors

Table l 1 demonstrates analogous patterns of prediction from instrumental support
reciprocity to parental warmth and to parental monitoring. For parental warmth, the
regression coefficient of the type of “receive and provide” shows that the change from the
group combined all other types together to the group of “receive and provide” predicts an
increase of .70 unit in parental warmth while controlling for the primary caregiver’s age
and sex. Both regression coefficients of the “receive only” and “provide only” types are
negative. Only the “receive and provide” type significantly predicts parental warmth.
Those who receive and provide instrumental support tend to have the significantly

highest degree of parental warmth to their children.

63

Table 11. Regression of Parental Warmth and Parental Monitoring on Overall Social

 

 

Support Reciprocity
Demdent Variable

Indegndent Variable Parental Warmth Parental Monitoring__
Overall Instrumental Support Reciprocity

Receive and provide 70* -- .87* -

Receive only -.35 -- -.54 -

Provide only -.44 - .57 --
Overall Emotional Support Reciprocity

Receive and provide - .68 - .07

Receive only - -.87 -- -.45

Provide only — .66 -- .68
Primary Caregivers age -.02 -.02 .03 .05
Primary Caregivers sex -1.10 -1.10 -.00 .09
Constant 26.42 26.10 25.49 25.15
R2 .02 .02 .03 .01
F 272* 1.80 2.64" 1.08
N 612 602 525 516

 

Note: ** p <.01;*.01s p < .05.
The regression coefficients are unstandardized.

For parental monitoring, the regression coefficient of the type of “receive and
provide” shows that the change from the group combined all other types together to the
group of “receive and provide” predicts an increase of .87 unit in parental monitoring
while controlling for the primary caregiver’s age and sex. The regression coefficient of
the “receive only” type is negative, -.54. The regression coefficient of the “provide only”
type is .57. Only the “receive and provide” type significantly predicts parental
monitoring. Those who receive and provide instrumental support tend to have the

significantly highest degree of parental monitoring to their children.

The results of the multiple regression show that instrumental support reciprocity is a

significant predictor of parental warmth and parental monitoring. There is a positive

64

relationship between instrumental support reciprocity and parental warmth, as well as
between instrumental support reciprocity and parental monitoring controlling for the
primary caregiver’s age and sex. Those who receive and provide instrumental support
have the higher degrees of parental warmth and parental monitoring toward their children

(see Table 1 l).

Controlling for the primary caregiver’s age and sex, the multiple coefficient of
determination (R-square) for parental warmth is .02, and that for parental monitoring is
.03. Only 2 percent of the total variation in parental warmth is explained by the
simultaneous predictive power of instrumental support reciprocity that contains three
dummy-coded independent variables, “receive and provide”, “receive only”, and
“provide only”, controlling for the primary caregiver’s age and sex. Only 3 percent of the

total variation in parental monitoring is explained by the same set of independent

variables (see Table 1 1).

Table 12 demonstrates the results of logistic regression of specific parental
behaviors, “telling others positive things about the child” and “praising the child”, on

overall

65

Table 12. Logistic Regression of Specific Parental Behaviors on
Overall Instrumental Support Reciprocity

Dependent Variable: “Telling Others Positive Things about the Child”

 

 

 

 

 

 

B Exp(B) Sig.
lndegndent Variable
Overall Instrumental Support Reciprocity
Receive and Provide 1.55 4.73 0.00
Receive Only 0.87 2.39 0.12
Provide Only 0.04 1.05 0.90
PCG Age 0.03 1.03 0.33
PCG Sex 0.33 1.39 0.76
Constant 0.98 2.66 0.34
-2 Log likelihood 307.33
N 610

 

 

Dependent Variable: “Praising the Child”

 

 

 

 

 

 

B Exp(B) Sig.
lndegndent Variable
Overall Instrumental Support Reciprocity
Receive and Provide 1.11 3.03 0.01
Receive Only 0.89 2.43 0.17
Provide Only 0.28 1.32 0.54
PCG Age 0.05 1.05 0.13
PCG Sex 007 0.93 0.95
Constant 0.53 1.69 0.65
-2 Log likelihood 263.03
N 607

 

 

instrumental support reciprocity. For specific parental behaviors, instrumental
support reciprocity is significantly and positively associated with the probability of the
two parental behaviors “telling others positive things about the child” and “praising the
child”. With the change from the non-reciprocal group to the reciprocal (receive and

provide) group in exchanging instrumental support in general, the primary caregivers

66

become more likely to tell others positive things about the child and praising the child

(see Table 12).

1.1.2 Overall emotional support reciprocity and parental behaviors

The patterns of prediction from emotional support reciprocity to parental warmth
and to parental monitoring are slightly different from those from instrumental support
reciprocity (see Table 11). For parental warmth, the regression coefficient of the type of
“receive and provide” shows that the change from the group combined all other types
together to the group of “receive and provide” predicts an increase of .68 unit in parental
warmth while controlling for the primary caregiver’s age and sex. The regression
coefficient of the “receive only” type is negative, -.87. The regression coefficient of the
“provide only” type is .66, that is very close to the “receive and provide” type. None of

these regression coefficients significantly predicts parental warmth.

For parental monitoring, the regression coefficient of the type of “receive and
provide” shows that the change from the group combined all other types together to the
group of “receive and provide” predicts an increase of .07 unit in parental monitoring
while controlling for the primary caregiver’s age and sex. The regression coefficient of
the “receive only” type is negative, -.44. The regression coefficient of the “provide only”
type is .68. None of these regression coefficients significantly predicts parental

monitoring (see Table 11).

67

The results of the multiple regression show that emotional support reciprocity is
neither a significant predictor of parental warmth nor one of parental monitoring

controlling for the primary caregiver’s age and sex (see Table 11).

The multiple coefficient of determination (R-square) for parental warmth is .02, and
that for parental monitoring is .03 controlling for the primary caregiver’s age and sex.
Only 2 percent of the total variation in parental warmth is explained by the simultaneous
predictive power of emotional support reciprocity, that contains three dummy variables,
“receive and provide”, “receive only”, and “provide only”, controlling for the primary

caregiver’s age and sex. Only 3 percent of the total variation in parental monitoring is

explained by the same set of independent variables (see Table l 1).

Table 13 demonstrates the results of logistic regression of specific parental
behaviors, “telling others positive things about the child” and “praising the child”, on
overall emotional support reciprocity. The results of emotional support reciprocity are
similar to those of instrumental support reciprocity. Emotional support reciprocity is
significantly associated with the probability of the two supportive parental behaviors
“telling others positive things about the child” and “praising the child”. With the change
from the non-reciprocal group to the reciprocal (receive and provide) group, the primary
caregivers become more likely to tell others positive things about the child as well as

praising the child.

68

Table 13. Logistic Regression of Specific Parental Behaviors on
Overall Emotional Support Reciprocity

Dependent Variable: “Telling Others Positive Things about the Child”

 

 

 

 

 

 

 

B Exp(BL Sig.
Independent Variable
Overall Emotional Support Reciprocity
Receive and Provide 1.45 4.27 0.00
Receive Only 1.42 4.14 0.18
Provide Only 0.60 1.82 0.19
PCG Age 0.02 1.02 0.49
PCG Sex 0.54 1.71 0.62
Constant 0.83 2.29 0.43
-2 Log likelihood 302.14
N 600

 

 

Dependent Variable: “Praisinlthe Child”

 

 

 

 

 

 

B Exp(B) Sig.
Independent Variable
Overall Emotional Support Reciprocity
Receive and Provide 1.27 3.54 0.00
Receive Only 6.26 521.25 0.64
Provide Only 0.28 1.32 0.58
PCG Age 0.04 1.04 0.21
PCG Sex 0.09 1.09 0.94
Constant 0.52 1.67 0.66
-2 Log likelihood 245.97
N 597

 

 

In short, the first hypothesis, support reciprocity between parents and people outside
the household predicts supportive parental behaviors, is partially supported by the CDS
data. Instrumental support reciprocity predicts higher parental warmth and parental
monitoring, while emotional support reciprocity does not significantly predict higher
parental warmth and parental monitoring. Both of instrumental and emotional support

reciprocity significantly predict the higher probability of the two supportive parental

69

behaviors “telling others positive things about the child” and “praising the child”. The
primary caregivers within a reciprocal relationship are more likely to have supportive

parental behaviors.

1.2 Support Reciprocity and Parental Behaviors by Social Tie

After examining the relationship between support reciprocity and parental behaviors
in general, the impact of social ties is of interest. For further understanding the effects of
social ties to the pattern of support reciprocity, the following sections examine the impact
of support reciprocity on parental behaviors by social tie and the type of social support.
The availability of social ties is critical for identifying the pattern of social support
exchange. Social ties here include the primary caregiver’s own mother/father,
mother/father-in-law, own grandmother/ grandfather, spouse’s grandmother/grandfather,
sister/brother, friends, child’s other parent, former spouse or partner, aunt/uncle/cousin,
counselor/minister/other clergy, members of church or other organization, co-workers,

grown child, and others.

Before performing the matching procedure, the principle of parallel support
exchange is also applied here to match the support recipients and support providers.
There are still four categories of support reciprocity (including both receiving and
providing, receiving only, providing only, and neither receiving nor providing). Then, the
connection between a provider and a recipient with the matched social tie is pinpointed.
The data show that the matched connections concentrate on three types of social ties that

are PCG’s own parent, sibling, and friends.

70

1.2.1 Instrumental support reciprocity and parental behaviors by social ties

Table 14 demonstrates the patterns of prediction from instrumental support
reciprocity to parental warmth and to parental monitoring for different types of social
ties. For parental warmth, the “receive and provide” type between PCGs and friends
significantly predicts the higher parental warmth with the regression coefficient, 1.07.
The change from the group combined all other types together to the group of “receive and
provide” between PCGs and friends predicts an increase of 1.07 unit in parental warmth
controlling for the primary caregiver’s age and sex. The “provide only” type between
PCGs and siblings significantly and negatively predicts parental warmth with the
regression coefficient, - .97. The change from the group combined all other types together
to the group of “provide only” between PCGs and siblings predicts a decrease of .97 unit
in parental warmth controlling for the primary caregiver’s age and sex. For parental

monitoring, none of these regression coefficients are significant.

The results of the multiple regression show that instrumental support reciprocity is a
significant and positive predictor of parental warmth, while the social relationship is
friendship. Instrumental support reciprocity is a significant and negative predictor of
parental warmth, while the PCG is an instrumental support provider to his/her sibling.
Instrumental support reciprocity does not significantly predict parental monitoring for

any of the three types of social ties.

For the six equations, the multiple coefficients of determination (R-square) are from

.01 to .02. Only 2 percent of the total variation in parental warmth is explained by the

71

simultaneous predictive power of instrumental support reciprocity controlling for the

primary caregiver’s age and sex, when the social relationship is either friends or siblings.

For the rest four equations, only 1 percent of the total variation in the dependent variable

is explained by the simultaneous predictive power of instrumental support reciprocity

controlling for the primary caregiver’s age and sex (see Table 14).

Table 14. Regression of Parental Warmth and Parental Monitoring on Instrumental
Support Reciprocity by Social Tie

 

Indegndent Variable
Instrumental Support Reciprocity

Between PCG and Own Parent
Receive and provide .53
Receive only -.01
Provide only -.51

Between PCG and Sibling
Receive and provide

Receive only

Provide only --
Between PCG and Friend

Receive and provide --

Receive only -

Provide only -
Primary Caregivers age -.02
Primary Caregivers sex -1.14
Constant 26.45
R2 .01
F 1.15
N 612

.78

-95

497*

-1
26

284*

612

.02
.17
.82

.02

Parental Warmth

1.07"
.10
.44

-.02
-1.35
26.29

.02
234*
612

Degndent Variable

Parental Monitoring

.22
-.59
.49

.03
-.01
25.66

.01
1.12
525

.81
-.79
.07

.01
1.52
525

.25
-.09
.62

.05
25.49

.01
.95
525

 

Note: ** p < .01; * .01 s p < .05. The regression coefficients are unstandardized.

72

Table 15 shows the results of logistic regression of specific parental behaviors,
“telling others positive things about the child” and “praising the child”, on instrumental
support reciprocity between the primary caregiver and his/her own parent. Instrumental
support reciprocity is significantly and positively associated with the probability of
“telling others positive things about the child”, but not with that of “praising the child”.
With the change from the non-reciprocal group to the reciprocal (receive and provide)
group in exchanging instrumental support with own parents, the primary caregivers

become more likely to telling others positive things about the child.

73

Table 15. Logistic Regression of Specific Parental Behaviors on
Instrumental Support Reciprocity between the Primary
Caregiver and His/Her Own Parent

Dgpendent Variable: “Telling Others Positive Thin 5 about the Child”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B EXPIBI 31L
lndegndent Variable
Instrumental Support Reciprocity
Receive and Provide 2.11 8.25 0.04
Receive Only 1.01 2.76 0.10
Provide Only 001 1.00 0.99
PCG Age 0.04 1.04 0.20
PCG Sex 0.22 1.24 0.84
Constant 0.96 2.61 0.35
-2 Log likelihood 313.80
N 610
Dependent Variable: “Praising the Child”

B Exp(B) Sig.
Independent Variable
Instrumental Support Reciprocity
Receive and Provide 0.71 2.04 0.25
Receive Only 1.12 3.06 0.13
Provide Only 0.12 1.13 0.83
PCG Age 0.05 1.05 0.09
PCG Sex 011 0.90 0.92
Constant 0.62 1.86 0.59
-2 Log likelihood 266.63
N 607

 

 

Table 16 shows the results of logistic regression of specific parental behaviors,
“telling others positive things about the child” and “praising the child”, on instrumental
support reciprocity between the primary caregiver and his/her own siblings. Instrumental
support reciprocity with own siblings does not significantly predict the probability of any

of these two behaviors.

74

Table 16. Logistic Regression of Specific Parental Behaviors on
Instrumental Support Reciprocity between the Primary
Caregiver and His/Her Sibling

Dependent Variable: “Telling Others Positive Thin 5 about the Child”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Exp(B) 81L
Independent Variable
Instrumental Support Reciprocity
Receive and Provide 1.67 5.30 0.10
Receive Only 0.28 1.32 0.61
Provide Only 0.08 1.08 0.87
PCG Age 0.03 1.03 0.23
PCG Sex 0.28 1.32 0.79
Constant 1.17 3.22 0.25
-2 Log likelihood 320.37
N 610
Dependent Variable: “Praising the Child”
8 Exp(B) SIL
Independent Variable
Instrumental Support Reciprocity
Receive and Provide 1.43 4.16 0.17
Receive Only 0.34 1.40 0.59
Provide Only 0.05 1.06 0.92
PCG Age 0.05 1.05 0.09
PCG Sex 009 0.92 0.93
Constant 0.71 2.04 0.54
-2 Log likelihood 267.55
N 607

 

 

Table 17 demonstrates the results of logistic regression of “telling others positive
things about the child” and “praising the child” on instrumental support reciprocity

between the primary caregiver and his/her friends. Instrumental support reciprocity is

75

significantly and positively associated with the probability of these two parental

behaviors.

Table 17. Logistic Regression of Specific Parental Behaviors on
Instrumental Support Reciprocity between the Primary
Caregiver and His/Her Friend

Dependent Variable: “Telling Others Positive Thin 5 about the Child”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Exp(B) Sig.
lndegndent Variable
Instrumental Support Reciprocity
Receive and Provide 2.81 16.57 0.01
Receive Only 0.72 2.05 0.25
Provide Only 0.26 1.30 0.49
PCG Age 0.02 1.02 0.38
PCG Sex 002 0.98 0.98
Constant 1.24 3.44 0.22
-2 Log likelihood 305.76
N 610
Dependent Variable: “Praising the Child”
3 Exp(B) sn-
Indeppndent Variable
Instrumental Support Reciprocity
Receive and Provide 1.46 4.28 0.02
Receive Only 0.91 2.48 0.23
Provide Only 0.77 2.16 0.13
PCG Age 0.05 1.05 0.15
PCG Sex 036 0.70 0.73
Constant 0.68 1.98 0.55
-2 Log likelihood 261.37
N 607

 

 

76

With the change from the non-reciprocal group to the reciprocal group in
exchanging instrumental support with friends, the primary caregivers become more likely

to tell others positive things about the child and praising the child.

1.2.2 Emotional support reciprocity and parental behaviors by social tie

Table 18 shows that “receive and provide” of emotional support between PCGs and
their parent, between PCGs and their siblings, and between PCGs and their friends

positively and significantly predicts parental wamrth.

77

Table 18. Regression of Parental Warmth and Parental Monitoring on Emotional
Support Reciprocity by Social Tie

 

Indeppndent Variable Dependent Variable
Instrumental Support Reciprocity

Parental Warmth Parental Monitoring

Between PCG and Own Parent

Receive and provide 108* ~~ -- .50 -- »

Receive only .58 -- - .01 -- --

Provide only .51 -- - -.11 -- --
Between PCG and Sibling

Receive and provide -- .85* - - .21 --

Receive only -- -.18 -- -- -.43 -

Provide only -- .99* - -- .95* --
Between PCG and Friend

Receive and provide -- -— .90“ - -- .20

Receive only -- -- -.86 -- -- -1.65**

Provide only -- -- .47 -- -- .60
Primary Caregivers age —.02 -.02 -.02 .04 .04 .06
Primary Caregiver's sex -1.17 -.94 -1.07 .01 .28 .00
Constant 26.06 26.16 26.17 25.52 25.38 25.45
R2 .02 .02 .03 .01 .02 .03
F 2.46" 2.60“r 3.30“ .83 1.69 3.20”
N 612 612 612 525 525 525

 

Note: ** p < .01; * .01 s p < .05. The regression coefficients are unstandardized.

For the emotional support between PCGs and their siblings, the type of “provide
only” from PCG to the sibling also positively and significantly predicts both parental
warmth and parental monitoring. For the emotional support between PCGs and their
friends, the type of “receive only” from friends to PCGs negatively and significantly
predicts parental monitoring. The multiple coefficients of determination (R-square) for
these equations range from .01 to .03 controlling for the primary caregiver’s age and sex.

Only 1 to 3 percent of the total variation in parental warmth is explained by the

78

simultaneous predictive power of emotional support reciprocity with the parent, sibling,

and friends, respectively.

In sum, controlling for the primary caregiver’s age and sex, emotional support
reciprocity positively and significantly predicts parental warmth for all three types of
social connection. The emotional support providing from PCGs to their siblings
positively and significantly predicts both parental warmth and parental monitoring.
Emotional support receiving from friends negatively and significantly predict parental

monitoring.

The decomposed analysis on support reciprocity by social tie demonstrates different
results from the general analysis on support reciprocity. In general, instrumental support
reciprocity significantly predicts both parental warmth and parental monitoring. When
social ties are broken down, the effect of instrumental support reciprocity is manifest only
in friendship. The effects of emotional support reciprocity are not clear, when all types of
social ties are mixed together. After distinguishing the types of social ties, the results
demonstrate that emotional support reciprocity is a significant predictor for parental
wannth in all three different social ties that includes own mother/father, siblings, and
friends. Social tie is a critical factor to be considered. Different social ties can lead to

different patterns of support reciprocity, and in turn to shape different parental behaviors.

Table 19 shows the results of logistic regression of “telling others positive things
about the child” and “praising the child” on emotional support reciprocity between the

primary caregiver and his/her own parents.

79

Table 19. Logistic Regression of Specific Parental Behaviors on
Emotional Support Reciprocity between the Primary
Caregiver and His/Her Own Parent

Dependent Variable: “Telling Others Positive Things about the Child”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B EXPIBI 51L
Indemdent Variable
Emotional Support Reciprocity
Receive and Provide 1.08 2.94 0.03
Receive Only 0.57 1.77 0.22
Provide Only 0.59 1.81 0.34
PCG Age 0.03 1.03 0.27
PCG Sex 0.21 1.24 0.84
Constant 1.09 2.99 0.28
-2 Log likelihood 318.12
N 610
“Praising the Child”
B Exp(B) SIL
Independent Variable
Emotional Support Reciprocity
Receive and Provide 1.03 2.80 0.06
Receive Only 1.09 2.97 0.08
Provide Only 003 0.97 0.96
PCG Age 0.05 1.05 0.09
PCG Sex 019 0.83 0.86
Constant 0.50 1.65 0.66
-2 Log likelihood 263.50
N 607

 

 

Emotional support reciprocity between the primary caregiver and his/her own
parents is significantly and positively associated with the probability of these two
parental behaviors. With the change from the non-reciprocal group to the reciprocal
group in exchanging emotional support with own parents, the primary caregivers become

more likely to tell others positive things about the child and praising the child.

80

Table 20 shows that emotional support reciprocity between the primary caregiver
and his/her own siblings is significantly associated with the probability of the parental

behavior “telling others positive things about the child”.

Table 20. Logistic Regression of Specific Parental
Behaviors on Emotional Support Reciprocity
between the Primary Caregiver and His/Her

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sibling
“Telling Others Positive Thugs about the Child”

B Exp(B) 81L
Receive and Provide 0.77 2.16 0.08
Receive Only 0.63 1.87 0.31
Provide Only 0.71 2.02 0.16
PCG Age 0.03 1.03 0.34
PCG Sex 0.40 1.49 0.71
Constant 1 .26 3.53 0.20

-2 Log likelihood 319.79
N 610
“Praisingthe Child”

8 Exp(B) aq-
Receive and Provide 0.49 1.63 0.27
Receive Only 0.75 2.12 0.32
Provide Only 0.94 2.56 0.13
PCG Age 0.04 1.05 0.14
PCG Sex 0.04 1.04 0.97
Constant 0.84 2.31 0.45

-2 Log likelihood 266.75
N 607

 

 

The association between instrumental support reciprocity and “telling others

positive things about the child” is positive. With the change from the non-reciprocal

81

group to the reciprocal (receive and provide) group in exchanging emotional support with
own siblings, the primary caregivers become more likely to tell others positive things

about the child.

Table 21 shows that the results of emotional support reciprocity with friends are

similar to those of instrumental support reciprocity with friends.

Table 21. Logistic Regression of Specific Parental
Behaviors on Emotional Support Reciprocity
between the Primary Caregiver and His/Her

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ffiend
“Telligq Others Positive Things about the Child”

B Exp(B) Sig.
Receive and Provide 1.73 5.65 0.00
Receive Only 0.63 1.88 0.26
Provide Only 0.76 2.15 0.07
PCG Age 0.02 1.02 0.42
PCG Sex 0.38 1.46 0.72
Constant 0.93 2.52 0.35

-2 Lpg likelihood 303.45
N 610
“PraisinLthe Child”

B Exp(B) Sig.
Receive and Provide 1.40 4.07 0.00
Receive Only 1.03 2.80 0.18
Provide Only 0.73 2.07 0.14
PCG Age 0.04 1.04 0.18
PCG Sex 001 0.99 0.99
Constant 0.51 1 .67 0.65

-2 Log likelihood 258.24
N 607

 

 

82

Emotional support reciprocity between the primary caregiver and his/her friends is
significantly and positively associated with the probability of the parental behavior
“telling others positive things about the child” and that of “praising the child”. With the
change from the non-reciprocal group to the reciprocal group in exchanging emotional
support with friends, the primary caregivers become more likely to tell others positive

things about the child and praise the child.

2. Financial Capital and Parental Behaviors

Financial capital here is operationalized as family financial strain and family
income. The following sections examine two hypotheses. The first part focuses on the
relationship between family financial strain and parental behaviors. The second

concentrates on that between family income and parental behaviors.

2.1 Family Financial Strain and Parental Behaviors

Table 22 shows that family financial strain is a significant predictor for parental

monitoring, but not for parental warmth.

83

Table 22. Regression of Parental Warmth and Parental Monitoring on
Family Financial Strain

Degndent Variable

 

Parental Warmth Parental Monitoring
Independent Variable B Beta B Beta
Family Financial Strain -.10 -.05 -.41** -.22**
Primary Caregivers Age -.02 -.04 .01 .02
Primary Caregivers Sex 107 -.05 .19 .01
Constant 26.87 27.19
R2 .01 .05
F 1.39 9.04“
N 612 525

 

Note: ** p <.01;‘.01_<. p < .05

Table 23 shows that family financial strain does not significantly predict the
probability of any of these two specific parental behaviors “telling others positive things

about the child” and “praising the child”.

84

Table 23. Logistic Regression of Specific Parental
Behaviors on Family Financial Strain

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent Variable:
“TellinLOthers Positive Things about the Child”
B Exp(B) SL9-
Indemdent Variable
Family Financial Strain 0.08 1.08 0.41
PCG Age 0.03 1.03 0.24
PCG Sex 0.20 1.22 0.85
Constant 1.18 3.25 0.26
-2 Log likelihood 324.26
N 610
Dependent Variable: “Praising the Child”
B Exp(B) Sig.
Independent Variable
Family Financial Strain -0.06 0.94 0.50
PCG Age 0.04 1.05 0.15
PCG Sex 011 0.89 0.91
Constant 1.24 3.46 0.29
-2 Log likelihood 270.25
N 607

 

 

Hypothesis 2.1: “the less family financial strain, the more likely parents have
supportive parental behaviors like the higher degrees of supportive parental behaviors” is
partially supported by the CDS data. Parental monitoring is significantly predicted by
family financial strain. The less the financial strain the higher the parental monitoring on
children; the more the financial strain the lower the parental monitoring on children. Yet,
the prediction of family financial strain is not significant in parental warmth, “telling

others positive things about the child”, and “praising the child”.

85

2.2 Family Income and Parental Behaviors

Table 24 indicates that family income is a significant predictor for both parental
warmth and parental monitoring. Both the regression coefficients of family income on
parental warmth and on parental monitoring are close to zero. Consequently, a one-unit
increase in family income causes a very small increase in parental warmth and parental

monitoring controlling for the primary caregiver’s age and sex.

Table 24. Regression of Parental Warmth and Parental Monitoring on Family

 

Income
Dependent Variable
Parental Warmth Parental Monitoring
Indegndent Variable B Beta B Beta
Family Income 6.060E-06' .09* 1.610E-05“ .24"
Primary Caregivers Age -.03 -.05 -.01 -.01
Primary Caregivers Sex - .98 -.05 .29 .01
Constant 26.68 26.32
R2 .01 .06
F 2.34 10.67“
N 612 525

 

Note: ** p <.01;*.01$ p < .05

The multiple coefficient of determination (R-square) for parental warmth is .01, and
that for parental monitoring is .06 controlling for the primary caregiver’s age and sex.
One percent of the total variation in parental warmth is explained by family income, and
six percent of the total variation in parental monitoring is explained by the simultaneous

predictive power of family income.

86

Family income is significantly and positively associated with the probability of the
parental behavior “telling others positive things about the child” and that of “praising the
child”. With the increase in family income, the primary caregivers become more likely to
tell others positive things about the child and praise the child. Yet, the proportions of the

occurrence of these two behaviors increase very slightly (see Table 25).

Table 25. Logistic Regression of Specific Parental
Behaviors on Family Income

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent Variable:
“Telling Others Positive Thipgs about the Child”
8 Ema) 51L
lndemndent Variables
Family Income 0.00 1.00 0.01
PCG Age 0.01 1.01 0.85
PCG Sex 0.44 1.55 0.68
Constant 1 .74 5.68 0.07
-2 Log likelihood 317.27
N 610
Dependent Variable: “Praisingthe Child”
B Exp(B) Sig.
Independent Varialbes
Family Income 0.00 1.00 0.00
PCG Age 0.01 1.01 0.65
PCG Sex 0.14 1.15 0.90
Constant 1.41 4.10 0.20
-2 Loglikelihood 258.81
N 607

 

 

87

Hypothesis 2.2: “The higher family income, the more likely parents have supportive
parental behaviors,” is supported. Family income significantly predicts both parental
warmth and parental monitoring. The higher the family income is, the higher the parental
warmth is, and so is the higher the parental monitoring. The more likely the behaviors,

“telling others positive things about the child” and “praising the child” occur.

3. Human Capital and Parental Behaviors

Table 26 indicates that the primary caregiver’s education is a very significant
predictor for both parental warmth and parental monitoring. The regression coefficient of
the primary caregiver’s education on parental warmth is .16, and that on parental
monitoring is .29. Consequently, a one-unit increases in the primary caregiver’s
education causes an increase of .16 unit in parental warmth controlling for the primary
caregiver’s age and sex. A one-unit increases in the primary caregiver’s education causes
an increase of .29 unit in parental monitoring controlling for the primary caregiver’s age

and sex.

88

Table 26. Regression of Parental Warmth and Parental Monitoring on Primary

Caregiver’s Education

lndAependent Variable
Primary Caregivers Education

Primary Caregivers Age
Primary Caregivers Sex
Constant

R2
F
N

Degndent Variable

Parental Warmth Parental Monitoring
B Beta 8 Beta
.16“ .11** .29” .21“
-.03 -.05 .00 .00
-1.13 -.06 .19 .01
24.84 23.17
.02 .04
335* 7.58“
606 520

 

Note: ** p <.01;*.01s p < .05

Controlling for the primary caregiver’s age and sex, the multiple coefficient of

determination (R-square) for parental warmth is .02, and that for parental monitoring is

.04. Two percent of the total variation in parental warmth is explained by the

simultaneous predictive power of the primary caregiver’s education, and four percent of

the total variation in parental monitoring is explained by the primary caregiver’s

education.

Parental education is significantly and positively associated with the probability of

the parental behavior “telling others positive things about the child”, but not with that of

“praising the child”. With the increase in the parental education, the primary caregivers

become more likely to tell others positive things about the child (see Table 27).

89

Table 27. Logistic Regression of Specific Parental Behaviors on
the Primary Caregiver’s Education ‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Dependent Variable:
“TellingOthers Positive Things about the Child”
B Exp(B) Sig.
lndemndent Variable
Primary Caregivers Education 0.20 1.22 0.00
PCG Age 0.01 1.01 0.57
PCG Sex 0.18 1.19 0.87
Constant -0.51 0.60 0.65
-2 Log likelihood 317.27
N 610
Dependent Variable: “Praising the Child”
B Exp(B) Sig.
lndegndent Variable
Primary Caregivers Education 0.10 1.11 0.15
PCG Age 0.04 1.04 0.20
PCG Sex 017 0.84 0.87
Constant 0.03 1.03 0.98
-2 Log likelihood 262.44
N 601

 

 

Hypothesis 3: “The higher the parental education is, the more likely parents have
supportive parental behaviors,” is supported. Parental education significantly predicts
both parental warmth and parental monitoring. The higher the parental education is, the

higher the parental warmth is, and so is the higher the parental monitoring.

90

4. Parental Behaviors and Children’s Self-Esteem

The following sections focus on the relationships between parental behaviors and
children’s self-esteem. The parental behaviors to be analyzed including parental warmth,
parental monitoring, praising, showing physical affections, telling others positive things
about the child, grounding, spanking, taking away privilege, taking away allowance, and

sending the child to a room.

Table 28 shows that three parental behaviors (including parental warmth, parental
monitoring, and one supportive parental behavior “telling others positive things about the

child”) positively and significantly predict children’s self-esteem.

The regression coefficient of parental warmth on children’s self-esteem is .16, and
that of parental monitoring on children’s self-esteem is .32. Consequently, a one-unit
increases in parental warmth causes an increase of .16 unit in children’s self-esteem
controlling for the primary caregiver’s age and sex. A one-unit increases in parental
monitoring causes an increase of .32 unit in children’s self-esteem controlling for the

primary caregiver’s age and sex.

The regression coefficient of the supportive parental behavior “telling others
positive things about the child” is 3.02. The coefficient indicates that the change from the
group of “never telling others positive things about the child” to the group of “telling
others positive things about the child” predicts an increase of 3.02 unit in children’s self-

esteem controlling for the primary caregiver’s age and sex.

91

Table 28. Regression of Children’s Self-Esteem on Parental Behaviors

Dependent Variable: Children’s Self-Esteem

 

 

Indepengent Variable
Parental Warmth ,16" -- .. -- .. -- -- -- --

Parental Monitoring -- .32' .- -- -- -- -- -- --

Specific Parental Behaviors
Grounding .- -- -,89 .. .. -- -- -- -- --
Spanking - -. .. -1 ,32 -- -- .. -- -- --
Taking Away Privilege -- -- -- -- -,62 -- .. .. -- --
Taking Away Allowance -- -- -- -- .. , 25 -- -- -- --
Sending to Room -- -- -- -- -- -- -,91 -- .. --
Praising -- .- -- .. .. -- .. 1 .23 -- ..
Showing Physical Affection - - -
Telling Positive Things to
Others

11
'I
I
II
t
II
II
II
II
.-
oz
01
put
0
N
5

Primary Caregivers Age .05 .06 .05 .04 .05 .05 .03 .06 .07 .04
PrimaryCaregiversSex -1.54 -.05 -1.51 -1.78 -1.81 -1.74 -1.74 -1.85 -2.35 -1.73
Constant 38.90 34.08 43.19 43.51 43.39 42.94 43.97 41.33 41.57 40.52

R2 .01 .03 .01 .01 .01 .00 .01 .01 .01 .02
F 2.07 4.91“ 1.30 1.19 1.01 .67 1.45 1.34 1.17 3.44“
N 605 520 605 607 605 574 603 600 582 603

Note: ** p < .01; * .01 s p < .05. The regression coefficients are unstandardized.

The multiple coefficients of determination (R-square) for these equations range
from .01 to .03 controlling for the primary caregiver’s age and sex. Only 1 to 3 percent of
the total variation in children’s self-esteem is explained by the simultaneous predictive
power of each parental behavior, respectively, controlling for the primary caregiver’s age

and sex.

Hypothesis 4: “The higher degree of supportive parental behaviors (including
parental warmth, parental monitoring, telling others positive things about the child,

praising, and showing physical affections), the higher children’s self-esteem,” is

92

supported by three supportive parental behaviors that are parental warmth, parental

monitoring. and of “telling others positive things about the child”.

93

CHAPTER IV
CONCLUSION

This chapter consists of three sections. The first section provides a summary of
research findings related to research intentions and hypotheses. The second section
discusses the implications of this study. Finally, the last section explores the limitations

of this study and provides suggestions for future research.

1. Summary

The research intentions of this study aims to examine the relationship between
parental practice and children’s self-esteem under the broader family context. The
concept of multi-level social context integrated with social capital theory is applied to the
research framework. Basically, there are four levels in the research framework. The first
level focus on children, in particular children’s self-esteem. The second is parents,
precisely parental behaviors. The third is resources inside the family including parental
human capital and family income. The fourth is resources from outside of the family; that
is social support from people outside of the household. According to social capital theory,
the mobilization of resources is critical for utilization. Parents in this case play the key
role to transform and transfer the resources to facilitate the well-being for their young

children.

94

Consequently, the effects of different forms of capital surrounding a family get the
major attention. Since the parent’s role is so critical, parental practice is treated as a
primary factor to shape children’s self-esteem. In the attempt of investigating the effects
of different forms of capital on parental behaviors and what parental behaviors lead to the
higher children’s self-esteem, research hypotheses and data analysis aimed to discover

research evidence for solving the puzzle.

H1: “Support reciprocity between parents and people outside the household predicts
supportive parental behaviors” is partially supported by the CDS data. Overall
instrumental support reciprocity significantly predicts the higher degrees of parental
warmth and parental monitoring, as well as the higher probability of “telling others
positive things about the child” and “praising the child”. Overall emotional support
reciprocity does not significantly predict higher parental warmth and parental monitoring,
but predict the higher probability of “telling others positive things about the child” and

“praising the child”.

For instrumental support reciprocity by social tie, only “telling others positive things
about the child” is significantly and positively predicted by the reciprocal relationship in
instrumental support with own parents. For instrtunental support with siblings, only
parental warmth is significantly and negatively predicted, when instrumental support is
provided by the primary caregivers without being received from siblings. For
instrumental support with friends, the higher degree of parental warmth, and the higher
probability of “telling others positive things about the child” and “praising the child” are

significantly and positively predicted by the reciprocal relationship.

95

For emotional support reciprocity by social tie, parental warmth and “telling others
positive things about the child” are significantly and positively predicted by the
reciprocal relationship with own parents. For emotional support with siblings, parental
warmth is significantly and positively predicted by the reciprocal relationship with
siblings. In addition, both parental warmth and parental monitoring are significantly and
positively predicted when emotional support is provided by the primary caregivers
without being received from siblings. For emotional support with friends, parental
warmth, “telling others positive things about the child”, and “praising the child” are
significantly and positively predicted by the reciprocal relationship with friends. In
addition, parental monitoring is significantly and negatively predicted, when emotional

support is received by the primary caregivers without being provided to friends.

The results of overall support reciprocity are different from the results of support
reciprocity by social tie. In general, only instrumental support reciprocity significantly
predicts both parental warmth and parental monitoring, while both instrumental and
emotional support reciprocity predict both “telling others positive things about the child”
and “praising the child”. When social ties are broken down, the positive effect of
instrumental support reciprocity with friends is particularly manifest on parental warmth,

“telling others positive things”, and “praising”.

The effects of emotional support reciprocity are not clear, when all types of social
ties are mixed together. After distinguishing the types of social ties, the results
demonstrate that emotional support reciprocity is a significant predictor for parental
warmth in all three different social ties that includes own mother/father, siblings, and

friends. Social tie is a critical factor to be considered. The distinction on social ties assists

96

us to further understand the importance of emotional support reciprocity on parental

warmth.

H2. 1. “The less family financial strain, the more likely parents have supportive
parental behaviors” is partially supported by the CDS data. Only parental monitoring is
significantly predicted by family financial strain. The less the family financial strain is,

the higher the parental monitoring is.

H2.2. “The higher family income, the more likely parents have supportive parental
behaviors" is supported. Family income significantly predicts parental warmth, parental

monitoring, “telling others positive things about the child”, and “praising the child”.

H3. ”The higher the parental education is, the more likely parents have supportive
parental behaviors” is partially supported. Parental education significantly predicts

parental warmth, parental monitoring, and “telling others positive things about the child”.

H4. “The higher degree of supportive parental behaviors, the higher children’s self-
esteem” is supported by three supportive parental behaviors that are parental warmth,

parental monitoring, and “telling others positive things about the child”.

2. Implications

This study reveals that social support, family income, and parental human capital
are the important factors to lead to supportive parental behaviors. The pattern of general
reciprocity exchange is different from the pattern of reciprocity exchange within different

social ties. Reciprocity exchange of social support is sustained by different norms and

97

patterns of social ties among individual actors (Portes and Sensenbrenner, 1993).
Reciprocity transaction is regarded as the accumulation of ‘chits’ which actors hope to be
reciprocated (Portes and Sensenbrenner, 1993). Actors may pursue social capital directly
through reciprocity exchange. However, the norms of reciprocity between giving and
receiving still need further investigation in terms of frequency, types of support, social

ties. and psychologically central demand and supply for exchange.

According research finding of this study, the supportive parental behaviors
contribute to the higher children’s self-esteem. It has been argued that the best parenting
style should spare the rod or spare the bug. The childrearing style of conservative
Protestants has been considered to account for the fact that children growing from the
environment of authoritative parental practice are more likely to experience academic
success, and less likely to involve in substance abuse and teen pregnancy (Wilcox, 1998).
Conservative Protestants support the use of corporal punishment to discipline children,
but also emphasize consistent and firm discipline and high levels of parental warmth. No

matter we spare the rod or not, the hug should never be spared.

Concerning family financial capital, the research findings show that higher family
income predicts the higher likelihood of supportive parental behaviors, but family
financial strain neither predicts lower likelihood of supportive parental behaviors nor
does higher likelihood of parental harshness. Financial hardship can be as a source to
decrease children’s self-esteem, when the parent’s modeling effects are apparent. In other
words, parents are negatively influenced by financial hardship that leads to lower parental

competence and the children perceive parental incompetence.

98

Another possibility is peer comparison. Children can learn norms and values
through their friends. Family financial hardship may influence children’s characteristics
such as the living conditions, social activities, learning resources, and the way they look.
Therefore, financial deprivation may lead to children’s negative evaluations toward self
due to the learned norms from their peer group. It has been argued that forms of capital
can be transformed into one another through various institutional mechanisms (Bourdieu,
I985; Fernandez-Kelly, 1995; Portes, 1998). Financial problems are found to be less
psychological central to the youth. Parents’ reflected appraisals through their expression
of warmth, acceptance, and supportiveness to their children are much more
psychologically central to the youth’s self-evaluation than are other family factors
(Whitbeck et al., 1991). In other words, warm and supportive parental behaviors may be

backups of family financial hardship.

The major purpose of this study focuses on child psychological well-being rather
than socially expected behaviors or achievement. The satisfaction toward oneself is the
dynamic of meaningful life, no matter how successful one is. Supportive parental
behaviors are critical in shaping one’s ability of appreciating oneself in human’s early

age. When one can appreciate the self, then he/she can appreciate others.

3. Limitations and Future Research

This study confronts three major limitations: the first one is in the measures of
specific parental behaviors; the second is related to the downside of social capital; the

third is the small values of the R-square of multiple regression and the -2 log likelihood

99

of logistic regression that indicate small effects from each form of capital to parental
behaviors, and from parental behaviors to children’s self-esteem. The small size of R-

squares and -2 log likelihood values can result from the two former limitations.

First, the measures of specific parental behaviors may not catch the real pattern of
daily interactions between the primary caregiver and the child due to the short duration
asked in the questionnaire. The primary caregivers were asked to track back to only one
week before the interview; probably one month may be more appropriate to present the

pattern of one’s behaviors.

Second, the prediction of social support may be constrained within the positive side
of parental behaviors. In other words, social support is usually in light of the positive
effects and its negative effects are frequently neglected. The bi-dimensional characteristic
of social support needs further attention. The downside of social capital like family
conflict, inconsistent childrearing values between two parents or between two generations
may be critical to lead to low parental warmth and harsh parental behaviors. This may be
the reason that the prediction from social support to negative parental behaviors is so

minimal.

Finally, the small values of the R-square of multiple regression and the -2 log
likelihood of logistic regression can be caused by the two former limitations. R-square is
affected “by the variability of a given sample on: (1) the variables being studied; (2) the
variables not included in the study; (3) the errors of measurement of the dependent
variable” (Pedhazur, 1982, p.177). For (1 ), the prediction of social support is limited by

the oversight of its bi-dimensional characteristic. For (2), other variables not included in

100

the study need to be considered for future research and will be discussed in next section.
For (3), the specific parental behaviors might not be adequately measured due to the short
duration of recalling parental behaviors. The variability of a given sample listed above

can also be the causes of the small values of the -2 log likelihood.

For future research, some aspects to be taken into account consist of the parent’s
own socialization experience, intergenerational transmission of parenting, the expression
of parental behaviors related to cultural capital, the stage of child development related to
parental practice, and further classifying parental behaviors into sub-dimensions such as

social, psychological, and physical care.

According to social learning theory, parental behaviors can be learned through
socialization associated with former experience of growing. The parent’s own
socialization experience can influence the conception of parental investment, and lead to
different parenting patterns. Intergenerational transmission of parenting and the
expression of parental behaviors are also crucial factors to shape parental behaviors.
When dealing with both intergenerational transmission and the expression of parental
behaviors, a broader research framework based on cultural capital is suggested. Different
cultures emphasize different childrearing values and different ways to express affections

emotionally and physically, as well as lead to different levels of parental investment.

Concerning the stage of child development related to parental practice, children
have different needs in each developmental stage. Nursing and carrying are necessary for

infants, but affective sharing and different forms of physical contact are needed by

101

toddlers and older children. Consequently, parental-child interactions need to be adjusted

following children’s development.

Finally, current research overwhelmingly focuses on parenting style (e.g.
authoritarian, authoritative, and permissive), and general parental behaviors (e.g. support
and control). It is necessary to classify parental behaviors into sub-dimensions such as
social. psychological, and physical care concerning child development in order to

adequately covering all childrearing typologies.

102

APPENDICES

103

Appendix A. Interview Procedure

The following steps were taken in each household (Hofferth, Davis-Kean, Davis, and
Finkelstein. 1997. Chapter 2):

1.
2.
3.

10.

11.

12.

I3.

14.

15.

Take PSID interview.
If there is an eligible child, a Child Case is generated by Data Control.

Data Control staff print a set of labels, CAI (Computer Assisted Interviewing)
Sample Management System (CSMS) checklist, and an information sheet for the
household and teacher.

The Data Services research assistant prepares and mails the sample line packet
(containing coversheets, labels and information sheets) to the interviewer assigned
to the case.

The interviewer reviews the sample line packet and labels the appropriate
questionnaires.

The interviewer contacts the household, verifies the primary caregiver selection,
explains the study, verifies the mailing address, and sets up an appointment for an
in-person interview.

The interviewer mails an introductory packet to the household containing a study
brochure, introduction letter, and time diaries.

The interviewer visits the household and gets written permission to interview the
child(ren) from the primary caregiver. The interviewer administers the child
assessments and primary caregiver-child questionnaire. The time diaries are
reviewed and edited with the primary caregiver (and the child, if appropriate).

The interviewer gives the appropriate self-administered questionnaires to the
primary caregiver and the other caregiver (if one is in the household).

The interviewer obtains written parental permission to contact teachers and
caregivers for children in school, preschool, or child care.

The interviewer obtains contact information for fathers living outside the home, if
applicable.

After the interview is completed, the interviewer edits the interviews and enters
complete information about the contact and the interview into CSMS.

The interviewer mails the school questionnaires to the appropriate teachers,
caregivers, and school administrators.

The interviewer notifies Ann Arbor via C SMS that the teacher information has
been mailed.

The interviewer mails an introductory letter to the father living outside the home,
if applicable.

104

16.

17.
18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

The interviewer contacts and interviews the father living outside the home, if
applicable.

The interviewer edits the father outside the home questionnaires, if applicable.

The interviewer reports the completed household session to the Team Leader
(TL).

The TL instructs the interviewer to mail the completed questionnaires to Ann
Arbor (if no verification is required) or to the TL (for verification and evaluation).

The interviewer mails completed and edited interviews to Ann Arbor or the TL, as
instructed.

The TL evaluates and verifies the interview, if applicable, gives feedback to the
interviewer, and mails completed interviews to Ann Arbor.

The Ann Arbor staff logs completed questionnaires as they are received from
interviewers, primary caregivers, other caregivers, teachers, childcare providers,
and administrators.

The Survey Services Lab (SSL) staff mails reminder postcards one week after the
interviewer has mailed the materials to the teachers and administrators.

The SSL staff conducts reminder calling for nonresponding teachers, caregivers,
and administrators beginning two weeks after the interviewer has mailed the
materials to the teachers and administrators. A second questionnaire is mailed by
the SSL staff if necessary.

A total of two rounds of 5 calls each will be made before coding the case as final
non-response, if the questionnaires are not returned.

The SSL staff codes and enters the completed time use diaries into the SAS data
entry program created by PSID staff.

The SSL staff enters all completed questionnaires into the SAS data entry
program created by PSID staff.

The SSL conducts check coding on 10% of Home Time Diaries and Home-based
Care Time Diaries, and performs double entry verification on questionnaires for
quality control purposes.

Completed questionnaires are stored by PSID staff once the questionnaires are
coded and processed.

105

Appendix B. Questionnaires for the Child Development Supplement and the
Possible Modes of Interview for Each Instrument during the
Production Phase

 

 

 

 

 

Qnaire Booklet Respondent Number Mode
Child Qnaire Child (3—12) One per child (age 3-12) Face-to—Face
Time Diary Child Two per eligible child (all Mail ahead and Face-to-

ages) Face or phone

Prim Caregiver-Child Primary Caregiver One per child (all ages) Face-to-Face or phone
Qnaire
Prim Caregiver-HH Primary Caregiver One per primary Self-Administered in
Qnaire caregiver home (interviewer

administered if literacy or
language problems)

 

Other Caregiver-Child

Other Caregiver of Child

One per child (for HHs

Self-Administered in

 

 

 

 

 

 

 

 

 

 

 

 

Qnaire with other caregiver) home
Other Caregiver-HH Other Caregiver of Child One per other caregiver Self-Administered in
Qnaire home
Father Outside of the Father Outside of the One per child (for HHs Telephone Interview
Home-Child Qnaire Home with a father living

outside of the home)
Father Outside of the Father Outside of the One per father outside of Telephone Interview
Home-HH Qnaire Home the home
Elem/Middle Elem/MS. English One per child in Elem, Self-Administered mail
School/Home School Teacher/Home school middle or home school survey
Teacher Qnaire teacher
Elementary/Middle Elementary or MS. One per child in Elem or Self-Administered mail
School Administrator School Administrator middle school survey
Qnaire
Preschool/Daycare Preschool/Day Care One per child in Self-Administered mail
Teacher Qnaire Center Teacher preschool or day care survey

center
Preschool/Daycare Preschool/Day Care One per child in Self-Administered mail
Administrator Qnaire Center Administrators preschool or day care survey

center
Home-Based Care In-home day care One per child in home- Self-Administered mail
Qnaire provider based day care survey

 

Source: Hofferth , Sandra, Pamela E. Davis-Kean, Jean Davis, and Jonathan Finkelstein. 1997. Table 2. of
Chapter 2-Data Collecflon in The Child Development Supplement to the Panel Study of Income Dynamics:
1997 User Guide. Survey Research Center, Institute for Social Research, The University of Michigan. Ann

Arbor, MI.

106

 

Appendix C. Race and Gender Composition of the Child Development

 

 

 

 

 

 

 

 

 

 

 

Supplement'I
o .
Total Families Total Children Male Children Female Children A 022:” by

White 1 140 1648 827 821 46.03%
Black 997 1467 772 695 40.98%
Hispanic 158 268 134 134 7.49%
Asian 46 63 31 32 1.76%
Native American 12 19 8 11 0.53%
Other 29 107 50 57 2.99%
NA/DK 7 8 4 4 0.22%
Total 2389 3580 1826 1754 100%
Missing 1 6

 

 

 

 

 

 

Note: a. The numbers in this table are unweighted.
Source: Hofferth , Sandra, Pamela E. Davis-Kean, Jean Davis, and Jonathan Finkelstein. 1997. Table 1. of
Chapter 2-Data Collection in The Child Development Supplement to the Panel Study of Income Dynamics:
1997 User Guide. Survey Research Center, Institute for Social Research, The University of Michigan. Ann

Arbor, MI.

107

 

Appendix D. Social Capital: Measures of Support Reciprocity

We are interested in the help and support that you receive from or give to other people
who are not living with you in this household.

1. In the past month, have you received any help with things like child care,
transportation, repairs to your home or car, or other kinds of work around the house? 1.
Yes 5. No ( GO TO 2.)
1a. Who gave you this help? (Please circle up to 5 relationships. If more than 5
people helped you, choose the five that helped most often.)
F orrner spouse or partner
Own mother, father
Mother-in-law, father-in-law
Own grandmother, grandfather
Spouse's grandmother, grandfather
Sister, brother
Aunt, uncle, cousin
Child's other parent
Friend
10. Counselor, minister, other clergy
1 1. Members of church or other organization
12. Co-workers
13. Grown child
97. Other (Specify):

PWNQVPP’Nf‘

2. Have you provided any help with child care, transportation, repair to home or car, or
other kinds of work around the house to anyone in the past month? 1. Yes 5. No (GO TO
3.)
2a. Who did you help this way? (Please circle up to 5 relationships. If more than 5
people helped you, choose the five that helped most often.)
1. Former spouse or partner
2. Own mother, father
3. Mother-in-law, father-in-law
4. Own grandmother, grandfather
5. Spouse's grandmother, grandfather
6. Sister, brother
7. Aunt, uncle, cousin
8. Child's other parent
9. Friend
10. Counselor, minister, other clergy
11. Members of church or other organization
12. Co-workers
l3. Grown child
97. Other (Specify):

108

3. In the past month, have you received any moral or emotional support such as advice or
encouragement from friends or relatives? 1. Yes 5. No (GO TO 4.)

33. Who gave you this support? (Please circle up to 5 relationships. If more than 5
people helped you, choose the five that helped most often.)

©®N©MPWNr

10.
ll.
12.
13.
97.

Former spouse or partner

Own mother, father
Mother-in-law, father-in-law

Own grandmother, grandfather
Spouse's grandmother, grandfather
Sister, brother

Aunt, uncle, cousin

Child's other parent

Friend

Counselor, minister, other clergy
Members of church or other organization
Co-workers

Grown child

Other (Specify):

4. In the past month, have you provided moral or emotional support such as advice or
encouragement to friends or relatives? 1. Yes 5. No

4a. Who did you support this way? (Please circle up to 5 relationships. If more
than 5 people helped you, choose the five that helped most often.)

1.

PWSP‘MPP’N

10.

12.
13.
97.

Former spouse or partner

Own mother, father
Mother-in-law, father-in-law

Own grandmother, grandfather
Spouse's grandmother, grandfather
Sister, brother

Aunt, uncle, cousin

Child's other parent

Friend

Counselor, minister, other clergy

. Members of church or other organization

Co-workers
Grown child
Other (Specify):

109

Appendix E. Parental Warmth: Measures of Parental Behaviors

About how often in the past month have you:

Would you say not in the past month, one or two times in the past month, about once a
week, several times a week, or every day?

9).”?pr

Hugged or shown physical affection to your child?

Told (CHILD) that you love him/her?

Spent time with (CHILD) doing one of his/her favorite activities?
Joked or played with (CHILD)?

Talked with him/her about thing he/she is especially interested in?
Told (CHILD) you appreciated something he/she did?

Parental Monitoring: Measures of Parental Behaviors

Thing now about how things are going in general in (child)’s life. Please rate each of the
following parts of (child)’s life as either excellent, good, fair, or poor.

9°NP‘WPP’NI‘

(His/Her) health.

(His/Her) friendships.

(His/Her) relationship with you.

(His/Her) feelings about (himself/herself).

(His/Her) prospects for the future.

(His/Her) relationships with brothers, sisters, or other children (he/she) lives with.
(His/Her) relationships with a teacher or caregiver.

(His/Her) relationships with the other parent.

110

Appendix F. Parental Behaviors: Measures of Specific Parental Behaviors

Sometimes kids mind pretty well and sometimes they don't. Sometimes they do things
that make you feel good and sometimes they don't.

How many time in the past week have you...

Grounded (CHILD)?

Spanked (CHILD)?

Taken away TV or other privileges?

Praised (CHILD) for doing something worthwhile?

Taken away (CHILD)’s allowance?

Shown (CHILD) physical affection (kiss, hug, stroke hair, etc.)?

Sent (CHILD) to his/her room?

Told another adult (spouse, friend, co-worker, visitor, relative) something positive
about (CHILD)?

9°N9‘SAPP’NI“

111

Appendix G. Family Financial Strain

In the last year, have you done any of the following or have any of the following
happened as a result of economic problems? (Circle all that apply)

99°N9‘SAPl-“Nt‘

u—Ia—‘h—Ie—dv—di—Au—I
O'NUI-BUJN—‘O

Sold possessions or cashed in life insurance

Postponed major purchases

Postponed medical care

Borrowed money from friends or relatives

Applied for government assistance

Filed for or taken bankruptcy

Fallen behind in paying bills

Obtained a loan to consolidate or pay off debts

Had a creditor call or come to see you to demand payment

. Had your wages attached or garnisheed by a creditor

. Had a lien filed against your property because you could not pay a bill
. Had your home, car or other property repossessed

. Moved to cheaper living quarters

. Moved in with other people

. Sent one or more of your children to live with someone else

. None of the above

112

Appendix H. Children’s Self-Esteem

Now I am going to read some statements. For each, think about whether the statement
never applies to you, sometime applies to you, always applies to you, or is somewhere in
between. Then give me the number from the booklet after I read each sentence.

. 1 do lots of important things. (never, sometimes, always, or somewhere in between?)
. I like being the way I am.

. Overall, I have a lot to be proud of.

. I can do things as well as most people.

. A lot of things about me are good.

. I'm as good as most other people.

. Other people think I am a good person.

. When I do something, I do it well.

OONCNL/t-bwwfl

113

REFERENCES

114

Antonucci, Toni C. 1985. “Social Support: Theoretical Advances, Recent Findings and
Pressing Issues,” Pp.2l-37 in Social Support: Theory, Research, and
Applications, by Irwin G. Sarason and Barbara R. Sarason (eds.). Dordrecht ;
Boston : M. Nihjoff ; Hingharn, MA : Distributors for the US. and Canada,
Kluwer Boston.

1990. “Social support ans social relationships,” Pp. 205-227 in Handbook of Aging
and the Social Sciences, by R. H. Binstock and L.K. Geroge (eds). San Diego:
Academic.

Asher, Herbert B. 1983. Causal Modeling. Newbury Park, California: SAGE
Publications, Inc.

Baker, W. E. 1990. “Market-networks and corporate behavior.” American Journal of
Sociology, 96:589-625.

Barbee, A. P., V. J. Derlega, S. P. Sherburne, A. Grimshaw. 1998. “Helpful and
unhelpful forms of social support for HIV-positive individuals,” Pp. 83-105 in
HIV and Social Interaction, by V. J. Derlega & A. P. Barbee (Eds). Thousand
Oaks, CA: Sage.

Barber, B. K., J. E. Olsen, and S. C. Shagle. 1994. “Associations between parental

psychological and behavioral control and youth internalized and extemalized
behaviors.” Child Development, 65: 1120-1136.

Baumrind, D. 1971. “Current patterns of parental authority.” Developmental Psychology
Monograph, 4 (1, Pt.2): 1-103.

1978. “Parental disciplinary patterns and social competence in children.” Youth
and Society, 9: 239-276.

Becker, W. C. 1964. “Consequences of different kinds of parental discipline,” Pp. 169-
208 in Review of Child Development, by M. L. Hoffman and L. Hoffman (eds).
Chicago: University of Chicago Press.

Belsky, Jay. 1990. “Parental and Nonparental Child Care and Children’s Socioemotional

Development: A Decade in Review.” Journal of Marriage and the Family, 52:
885-903.

Berlin, L., Brooks-Gunn, J ., Spiker, D., and Zaslow, M. 1995. “Examining observational
measures of emotional support and cognitive stimulation in Black and White
mothers of preschoolers.” Journal of Family Issues 16: 664-686.

Bierhoff, Hans W. 1994. “On the Interface Between Social Support and Prosocial
Behavior: Methodological and Theoretical Implications,” Pp. 159-167 in Social
Networks and Social Support in Childhood and Adolescence, by Frank Nestmann
and Klaus Hurrelmann (eds). New York: de Gruyter.

115

Block, J., & R. W. Robins. 1993. “A longitudinal study of consistency and change in self-
esteem from early adolescence to early adulthood.” Child Development, 64, 909-
923.

Bourdieu, P. 1979. “Les trois etats du capital culturel.” Actes de la Recherche en Sciences
Sociales, 3023-6.

1980. “Le capital social: notes provisoires.” Actes de la Recherche en Sciences
Sociales, 31 :2-3.

1985. “The forms of capital,” Pp. 241 -58 in Handbook of Theory and Research for
the Sociology of Education, by J. G. Richardson (eds). New York: Greenwood.

Brenner, Viktor & Robert A. Fox. 1999. “An Empirically Derived Classification of
Parenting Practices.” The Journal of Genetic Psychology, 160 (3): 343-356.

Brody, G., Z. Stoneman & D. Flor. 1995. “Linking family processes and academic
competence among rural African-American youths.” Journal of Marriage and the
Family, 57:567-579.

Burt, Ronald S. 1992. Structural Holes: The Social Structure of Competition. Cambridge,
MA: Harvard University Press.

Carlin, Paul S. 1999. “Economics and the Family,” Pp.525-552 in Handbook of Marriage
and the Family (2nd Edition), by Marvin B. Sussman, Suzanne K. Steinmetz and
Gary W. Peterson (eds). New York, NY: Plenum.

Chesler, M. A., &, O. A. Barbarin. 1984. “Difficulties of providing help in a crisis:
Relationships between parents of children with cancer and their friends.” Journal
of Social Issues, 40, 1 13-134.

Coates, D., C. B. Wortman, & A. Abbey. 1979. “Reactions to victims,” Pp. 21-52 in New
approaches to social problems by I. H. Friez, D. Bartal, & J. E. Singer (Eds). San
Francisco: Jossey-Bass.

Cobb, S. 1976. “Social Support as a Moderator of Life Stress.” Psychosomatic Medicine,
382300-314.

Cochran, M. M., and J. A. Brassard. 1979. “Child Development and Personal Social
Networks.” Child Development, 50: 601-616.

Cotterell, John L. 1986. “Work and Community Influences on the Quality of Child
Rearing.” Child Deveolpment, 57: 362-374.

Cohen, S., and G. McKay. 1984. “Social Support, Stress, and the Buffering Hypothesis:
A theoretical analysis,” Pp. 253-267 in Handbook of Psychology and Health, Vol.
4, by A. Baum, J. E. Singer, and S. E. Taylor (eds). Hillsdale, NJ: Erlbaum.

116

Coleman, J. S. 19883. “Social capital in the creation of human capital.” American Journal
of Sociology, 94 (Supplement), S95-S120.

1988b. “The Creation and Destruction of Social Capital: Implications for the
Law.” Notre Dame Journal of Law, Ethics, and Public Policy, 3(3):375-404.

1990. Foundations of Social Theory. Cambridge, MA: Belknap Press of Harvard
University Press.

Conger, R., & G. H. Elder, Jr. 1994. Families in Troubled Times: Adapting to Changes in
Rural America. New York: Aldine de Gruyter.

Conger, R. D., G. H. Elder, F. O. Lorenz, K. J. Conger, R. L. Simons, L. B. Whitbeck, S.
Huck, & J. N. Melby. 1990. “Linking economic hardship to marital quality and
instability.” Journal of Marriage and the Family, 52: 643-656.

Cooley, Charles Harton. 1902. Human Nature and the Social Order. New York:
Charles Scribner's Son.

Cooper, Sheila McIsaac. 1999. “Historical Analysis of the Family.” Pp.13-37 in
Handbook of Marriage and the Family (2nd Ed.), by Marvin B. Sussman, Suzanne

K. Steinmetz, and Gary W. Peterson (eds.). New York, NY: Plenum.
Coopersmith, Stanley. 1967. Antecedents of Self-Esteem. San F ransico: Freeman.

Cotterell, John L. 1986.”Work and community influences on the quality of child rearing.”
Child Development, 57: 362-374.

Council on Civil Society. 1998. A Call to Civil Society: Why Democracy Needs Moral
Truths. Institute for American Values.

C mic, Keith A., Mark T. Greenberg, Arlene S. Ragozin, Nancy M. Robinson, and Robert
B. Basham. 1983. “Effects of stress and social support on mothers and premature
and full-term infants.” Child Development, 54: 209-217.

C rodkenberg, Susan, and Karen McCluskey. 1986. “Change in maternal behavior during
the baby’s first year of life.” Child Development, 57: 746-753.

DeF rain, John and David H. Olson. 1999. “Contemporary Family Patterns and
Relationships,” Pp.309-326 in Handbook of Marriage and the Family (2"d
Edition), by Marvin B. Sussman, Suzanne K. Steinmetz and Gary W. Peterson
(eds). New York, NY: Plenum.

Demo, David H. & Ritch Savin-Williams. 1983. “Early Adolescent Self-Esteem as a
Function of Social Class: Rosenberg and Pearlin Revisited.” American Journal of
Sociology, 88: 763-764.

117

Dombusch, S., P. Ritter, P. Leiderrnan, D. Roberts, & M. Fraleigh. 1987. “The relation of
parenting style to adolescent school performance.” Child Development, 58: 1244-
1257.

Duncan, G. J., J. Brooks-Gunn, & P. K. Klebanov. 1994. “Economic deprivation and
early childhood development.” Child Development, 65(2), 296-318.

Duncan, G. J ., W.-J. J. Yeung, J. Brooks-Gunn, & J. Smith. 1998. “How much does
childhood poverty affect the life chances of children?” American Sociological
Review, 63, 406-423.

Elder, G., J. Eccles, M. Ardelt, & S. Lord. 1995. “Inner-city parents under economic

pressure: Perspectives on the strategies of parenting.” Journal of Marriage and
the Family, 57: 771-784.

Elshtain, Jean Bethke. 1999. “A Call to Civil Society”. Society, 36(5): 11-19.
Felson, R. B., and M. Reed. 1986. “The effect of parents on the self-appraisals of
children.” Social Psychological Quarterly, 69:302-308.

Fernandez-Kelly, Patricia M. 1995.”Social and cultural capital in the urban ghetto:
Implications for the economic sociology of immigration,” Pp.213-247 in The
Economic Sociology of Immigration: Essays on Networks, Ethnicity, and
Entrepreneurship, by Alejandro Portes (eds). New York: Russell Sage
Foundation.

Florsheim, P, P. Tolan and D. Gorman-Smith. 1996. “Family processes and risk for
externalizing behavior problems among African American and Mexican-
American boys.” Journal of Consulting and Clinical Psychology, 64: 1222-1230.

Franco, Nathalie, and Mary Levitt. 1998. “The Social Ecology of Middel Childhood:
Family Support, Friendship Quality, and Self-Esteem.” Family Relations, 47:315-
321.

Furstenberg, Frank F., Jr. 1999 “Is the modern family a threat to children's health?”
Society: 36(5): 31-37.

Furstenberg, Frank F., Jr and Mary Elizabeth Hughes. 1995. “Social Capital and
Successful Development among At-Risk Youth.” Journal of Marriage and the
Family, 57: 580-92.

Garcia-Coll, C., B. R. Vohr, , J. Hoffman, and W. Oh. 1986. “Maternal and
environmental factors affecting developmental outcome of infants of adolescent
mothers.” Journal of Developmental and Behavavioral Pediatrics, 7: 230-236.

Ge, X., K. Best, R. Conger, and R. L. Simons. 1996. “Parenting behaviors and the
occurrence and co-occurrence of adolescent depressive symptoms and conduct
problems.” Developmental Psychology, 32: 717-731.

118

Gecas, Viktor and Michal L. Schwalbe. 1983. “Beyond the Looking-Glass Self: Social
Structure and Efficacy-Based Self-Esteem.” Social Psychology Quarterly, 46(2):
77-88.

Gold, Steve J. 1995. “Gender and social capital among Israeli immigrants in Los
Angeles.” Diaspora, 42267-301.

Goldsmith, D., & M. R. Parks. 1990. “Communicative strategies for managing the risks
of seeking social support” Pp. 104-121 in Personal relationships and social
support, by S. Duck (Eds). London: Sage Publications.

Greenberger, E., and C. Chen. 1996. “Perceived family relationships and depressed mood
in early and late adolescence: A comparison of European and Asian Americans.”
Developmental Psychology, 32: 707-716.

Grusec, J. E. and J. G. Goodnow. 1994. “Impact of parental discipline methods on the
child’s internalization of values: A reconceptualization of current points of view.”
Developmental Psychology, 30: 4-19.

Hagan J, MacMillan R, Wheaton B. 1996. “New kid in town: social capital and the life
course effects of family migration in children.” American Sociological Review,
61:368-85.

Harter, Susan. 1983. “Developmental perspectives on the self-system,” Pp. 275-3 86 in
Handbook of Child Psychology, Vol. 4 : Socialization, personality, and social
development, by E. M. Hetherington. New York: Wiley.

1985. “Competence as a Dimension of Self-Evaluation: Toward a Comprehensive
Model of Self-Worth,” in The Development of the Self, by R.L. Leahy (eds).
Orlando, FL: Academic Press.

1990. “Self and identity development,” Pp. 352-3 87 in At the threshold: The
developing adolescent, by S. S. F eldman & G. R. Elliott (Eds). Cambridge, MA:
Harvard University Press.

1993. “Causes and Consequences of Low Self-Esteem in Children and
Adolescents,” Pp. in Self-Esteem: The Puzzle of Low Self-Regard, by R. F.
Baumeister (eds). New York, NY: Plenum Press.

1998. “The development of self-representations,” Pp. 553-618 in Handbook of
Child Psychology, Vol. 3: Social, Emotional, and Personality Development, by
W. Damon and N. Eisenberg (eds.). New York: Wiley.

Haveman, R., & B. Wolfe. 1994. Succeeding Generations: 0n the effects of Investments
in Children. New York: Russell Sage.

Haveman, R., & B. Wolfe. 1995. “The determinants of children’s attainments: A review
of methods and findings.” Journal of Economic Literature, 33: 1829-1878.

119

Heller, K. 1979. “The effects of social support: Prevention and treatment implications,”
Pp. 253-3 82 in Maximizing Treatment Gains: Transfer Enhancement in
Psychotherapy. New York: Academic Press.

Heller, K., R. W. Swindle, and L. Dusenbury. 1986. “Component social support
processes: Comments and integration.” Journal of Consulting and Clinical
Psychology, 54: 466-470.

Hertzog, C., and J. R. Nesselroade. 1987. “Beyond autoregressive models: Some
implications of the trait-state distinction for structural modeling of developmental
change.” Child Development, 58: 93-109.

Higgins, E. T. 1991. “Development of self-regulatory and self-evaluative processes:
Costs, benefits, and tradeoffs,” Pp.125-165 in The Minnesota symposia on child

development: Vol. 23. Self processes and development, by M. R. Gunnar and L.
A. Sroufe (eds). Hillsdale, NJ: Erlbaum.

Hofferth, Sandra L., Johanne Boisjoly, and Greg J. Duncan. 1999. “The development of
social capital.” Rationality and Society, 1 1(1): 79-] 10.

Hofferth , Sandra, Pamela E. Davis-Kean, Jean Davis, and Jonathan Finkelstein. 1997.
The Child Development Supplement to the Panel Study of Income Dynamics:
1997 User Guide. Survey Research Center, Institute for Social Research, The
University of Michigan. Ann Arbor, MI.

Hofferth, Sandra L., and John Iceland. 1998. “Social capital in rural and urban
communities.” Rural Sociology, 63(4): 574-598.

Hoffman, M. L. 1994. “Discipline and internalization.” Developmental Psychology,
30:26-28.

House, J. S. 1981. Work Stress and Social Support. Reading, MA: Addison-wesley.

Kaplan, B. H., J. C. Cassel, and S. Gore. 1977. “Social Support and Health.” Medical
Care, 15(Supplement): 47-58.

Kelley, M. L., T. G. Power & D. D. Wimbush. 1992. “Determinants of disciplinary
practices in low-income Black mothers.” Child Development, 63: 573-582.

Kelley, M. L., J. Sanchez-Hucles, & R. Walker. 1993. “Correlates of disciplinary
practices in working-to-middle-class African-American mothers.” Merrill-Palmer
Quarterly, 39, 252-264.

Kessler, R. C. & J. D. McLeod. 1984. “Sex differences in vulnerability to undesirable
life events.” American Sociological Review, 49, 620-631.

120

Kurtz, L., and J. Derevensky. 1994. “Adolescent motherhood: An application of the stress
and coping model to child rearing attitudes and practices.” Canadian Journal of
Community Mental Health, 1 3: 5-24.

La Gaipa, J. J. 1990. “The negative effects of informal support” Pp. 122-139 in Personal
relationships and social support, by S. Duck (Eds). London: Sage Publications.

Lepper, M. R. 1981. “Socail control processes, attributions of motivation, and the
internalization of social values,” Pp. 294-330 in Social Cognition and Social
Behavior: Developmental Perspectives, by E. T. Higgins, D. N. Ruble, and W. W.
Hartup (eds). Cambridge, MA: Cambridge University Press.

Levitt, M. J., N. Guacci-Franco, and J. L. Levitt. 1993. “Convoys of social support in
childhood and early adolescence: Structure and function.” Developmental
Psychology, 29:81 1-818.

Lewis. C. C. 1981. “The effects of firm control: a reinterpretation of findings.”
Psychological Bulletin, 90: 547-563.

Lin. Nan. 1990. “Social Resources and Social Mobility: A Structural Theory of Status
Attainment,” Pp. 247-271 in Social Mobility and Social Structure, by Ronald L.
Breiger. Cambridge, England: Cambridge University Press.

Lin, Nan, W. M. Ensel, R. S. Simeone, and W. Kuo. 1979. “Social Support, Stressful Life
Events, and Illness: A model and empirical test.” Journal of Health and Social
Behavior, 20: 108-119.

Light, Ivan, and Steven J. Gold. 2000. Ethnic economies. San Diego, CA: Academic
Press.

Lu, Luo. 1997. “Social support, reciprocity, and well-being.” Journal of Social
Psychology, 137 :618-628.

Lu, Luo. & M. Argyle. 1992. “Receiving and giving support: Effects on relationships and
well-being.” Counselling Psychology Quarterly, 5, 123-133.

Maccoby, E., and J. Martin. 1983. “Socialization in the context of the family: parent-
child interaction,” Pp. 1-101 in Handbook of Child Psychology, Vol. 4:
Socialization, personality, and social development, by E. M. Hetherington (eds).
New York: Wiley.

Marsh, H. 1990. Self-Description Questionnaire Manual. Macarthur, Australia:
University of Western Sydney.

Mayhew, Kathryn P. and Jacques D. Lempers. 1998. “The relation among financial
strain, parenting, parent self-esteem, and adolescent self-esteem.” Journal of
Early Adolescence, 18(2): 145-172.

121

McLanahan, S., G. Sandefur. 1994. Growing Up with a Single Parent: What Hurts, What
Helps. Cambridge, MA: Harvard Univ. Press.

McLeroy, K. R., R. DeVellis, B. DeVellis, B. Kaplan & J. Toole. 1984. “Social support
and physical recovery in a stroke population.” Journal of Social and Personal
Relationships, 1, 395-413.

McLoyd, V., T. G. Jayaratne, R. Ceballo, & J. Borquez. 1994. “Unemployment and work
interruption among African-American single mothers: Effect on parenting and
adolescent socioemotional functioning.” Child Development, 65: 562-589.

Mead, G. H. 1934. Mind, Self and Society. Chicago: University of Chicago Press.

Miller, G. R., & M. Steinberg. 1975. Between people: A new analysis of interpersonal
communication. Chicago: Science Research Associates.

Mirowsky, John & Catherine E. Ross. 1986. “Social Patterns of Distress.” Annual Review
of Sociology, 12: 23-45.

Mruk, Christopher J. 1995. Self-Esteem: Research, Theory, and Practice. New York,
NY: Springer Publishing Company.

National Institutes of Health (NIH) and National Institute of Child Health and Human
Development (NICHD). 2000. URL: http://\\’\\'w.nih.go\ix’ncxxr's/prt’ianZ()(f)(,)/nichtl-
31 .htm Released on Monday, January 31, 2000 6:00 pm. EST or
http://www.mhbc.ed.ac.uk/hbsc

Nestmann, F. and K. Hurrelmann. 1994. Social Networks and Social Support in
Childhood and Adolescence, pp. 305-322. New York: de Gruyter.

Newcomb, Michael D. 1990. “Social support and personal characteristics: A
developmental and interactional perspective.” Journal of Social and Clinical
Psychology, 9: 54-68.

Newcomb, Michael D. and Keunho Keefe. 1997. “Social support, self-esteem, social
conformity, and gregariousness: Developmental patterns across 12 years,” Pp.
303-333 in Sourcebook of Social Support and Personality, by Gregory R. Pierce,
Brian Lakey, Irwin G. Sarason, and Barbara R. Sarason (eds). New York:
Plenum.

Openshaw, D. K. and D. L. Thomas. 1986. “The adolescent self and the family,” Pp.104-
129 in Adolescents in families, by G. K. Leigh and G. W. Peterson (eds).
Cincinnati, OH: South-westem Publishing Co.

Panel Study Income Dynamics-Child Development Supplement (PSID-CDS). 1997.
URL: http://www.isr.umich.edu/src/child-development/home.html

122

Parcel, Toby L. and Elizabeth G. Menaghan. 1993. “Family Social Capital and Children's
Behavior Problems.” Social Psychology Quarterly, 56 (2)2120-135.

Parcel, Toby L. and Elizabeth G. Menaghan. 1994. “Early Parental Work, Family Social
Capital and Early Childhood Outcomes.” American Journal of Sociology, 99
(4):972-1009.

Parke, R.. and R. Buriel. 1998. “Socialization in the family: Ethnic and ecological
perspectives,” Pp. 463-552 in Handbook of Child Psychology, Vol. 3: Social,
Emotional, and Personality Development, by W. Damon and N. Eisenberg (eds).
New York: Wiley.

Pascoe, J. M., M. Loda, U. Jeffries, and J. Earp. 1981. “The Association between
Mothers’ Social Support and Provision of Stimulation to Their Children.”
Development and Behavior Pediatrics, 2: 15-19.

Patterson, G. R. and D. M. Capaldi. 1991. “Antisocial parents: Unskilled and vulnerable,”
Pp.195-218 in Family Transitions, by P. A. Cowan and E. M. Hetherington (eds).
Hillsdale, NJ: Erlbaum.

Pearlin, L. 1., M. A. Lieberman, E. G. Menaghan, J. T. Mullan. 1981. “The Stress
Process.” Journal of Health and Social Behavior, 22: 337-356.

Pedhazur, Elazar J. 1982. Multiple Regression in Behavior Research: Explanation and
Prediction. Forth Worth, TX: Holt, Rinehart and Winston, Inc.

Peters-Golden, H. 1982. “Breast cancer: Varied perceptions of social support in the
illness experience.” Social Science and Medicine, 6: 483-491.

Peterson, Gary W. and Della Hann. 1999. “Socializing Children and Parents in Families,”
Pp. 327-3 70 in Handbook ofMarriage and the Family (2nd Ed.), by Marvin B.
Sussman, Suzanne K. Steinmetz, and Gary W. Peterson (eds). New York, NY:
Plenum.

Peterson, G. W. and B. C. Rollins. 1987. “Parent-child socialization,” Pp.471-507 in
Handbook of Marriage and the Family, by M. B. Sussman and S. K. Steinmetz
(eds). New York: Plenum.

Peterson, G. W., B. C. Rollins, and D. L. Thomas. 1985. “Parental influence and
adolescent conformity: Complicance and internalization.” Youth and Society, 16:
397-420.

Portes, Alejandro. 1998. “Social capital: its origins and applications in modern
sociology.” Annual Review of Sociology, 24: 1-24.

Portes, Alejandro, and Patricia Landolt. 1996. “The Downside of Social Capital.”
American Prospect; 26:18-21 .

123

Portes, Alejandro, and Julia Sensenbrenner. 1993. “Embeddedness and immigration:
notes on the social determinants of economic action.” American Journal of
Sociologv, 98:1320-50.

Putallaz, M., and A. H. Heflin. 1990. “Parent-child interaction,” Pp.189-216 in Peer
Rejection in Childhood, by S. R. Asher and J. D. Coie (eds). New York:
Cambridge University Press.

Raikkonen, K., & L. Keltikangas-Jarvinen. 1992. “Mothers with hostile, Type
Apredisposing child-rearing practices.” Journal of Genetic Psychology, 153, 243-
354.

Rhodes, J ., and M. Woods. 1995. “Comfort and conflict in the relationships of pregnant,
minority adolescents: Social support as a moderator of social strain.” Journal of
Community Psychology, 23: 74-84.

Rohner, R. P. 1986. The Warmth Dimension: F oudation of Parental Acceptance-
Rejection Theory. Beverly Hills, CA: Sage.

Rollins, B. C., and D. L. Thomas. 1979. “Parental support, power, and control techniques
in the socialization of children,” Pp.317-364 in Contemporary Theories about the
Family: Vol. I, by W. R. Burr, R. Hill, F. I. Nye, and I. R. Reiss. New York: Free

Press.

Rook, K. S. 1984. “The negative side of social interaction: Impact on psychological well-
being.” Journal of Personality and Social Psychology, 46: 1097-1 108.

1992. “Detrimental aspects of social relationships: Taking stock of an emerging
literature,” Pp. 157-169 in The Meaning and Measurement of Social Support, by
H. O. F. Veiel and A. Baumann (eds). New York: Hemisphere Publishing
Corporation.

Rosenberg, Monis J. 1965. "When dissonance fails: On eliminating evaluation
apprehension from attitude measurement." Journal of Personality and Social
Psychology, 1:28-42.

1979. Conceiving the Self. Malabar, Florida: Basic Books.

1986. “Self-concept from middle childhood through adolescence,” Pp. 107-136 in
Psychological Perspectives on the Self, vol. 3, by J. Suls and A. G. Greenwald
(eds). Hillsdale, NJ: Lawrence Erlbaum.

Rosenberg, Morris, C. Schooler, C. Schoenbach, and F. Rosenberg. 1995. "Global Self-
Esteem and Specific Self-Esteem: Different Concept, Different Outcomes."
American Sociological Review, 60: 141-156.

Sandefur, Rebecca L. and Edward O. Laumann. 1998. “A Paradigm for Social Capital.”
Rationality and Society, 10(4): 481-501.

124

Sandler, I., P. Miller, J. Short, and S. Wolchik. 1989. “Social support as a protective
factor for children in stress, “ Pp.277-307 in Children’s Social Networks and
Social Supports, by D. Belle (eds). NewYork: Wiley.

Schaefer, E. S. 1959. “A circumplex model for maternal behavior.” Journal of A bnormal
and Social Psychology, 59: 226-235.

Schiff. M. 1992. “Social capital, labor mobility, and welfare.” Rationality and Society,
4: 157-75.

Schultz, K. (1991). “Women's adult development: The importance of friendship.”
Journal of Independent Social Work, 5(2), 19-30.

Shelton. Marisa M. 1990. The Impact of Others on a Child 's Self-Esteem. Master's
Thesis. Research paper for Master of Education degree, Midwestern State
University, Texas.

Shrauger, J. S. and T. J. Schoeneman. 1979. “The symbolic interactionists' view of self-
concept: through the looking-glass darkly.” Psychological Bullitin, 86:549-573.

Shumow, Lee, D. L. Vandell & J. K. Posner. 1998. “Harsh, firm, and permissive
parenting in low-income families: Relations to children's academic achievement
and behavioral adjustment.” Journal of Family Issues, 19: 483-507.

Siegelman, E. 1965. “College student personality correlates of early parent-child
relationships.” Journal of Consulting Psychology, 29: 558-564.

Simons, Ronald L., Jay Beaman, Rand D. Conger, & Wei Chao. 1993. “Stress, Support,
and Antisocial Behavior Trait as Determinants of Emotional Well-Being and
Parenting Practices among Single Mothers.” Journal of Marriage and the Family,
55: 385-398.

Smith, Mark, Lionel Beaulieu and Glenn Israel. 1992. “Effects of Human Capital and
Social Capital on Dropping Out of High School in the South.” Journal of
Research in Rural Education, 8: 75-87.

Stafford, L., and C. L. Bayer. 1993. Interaction between Parents and Children. Newbury
Park, CA: Sage.

Sweet, J., L. Bumpass, & V. Call. 1988. The Design and Content of the National Survey
of Families and Households. NSFH Working Paper No. 1. Madison, WI: Center
for Demography and Ecology, University of Wisconsin.

Takeuchi, D. T., D. R. Williams & R. K. Adair. 1991. “Economic stress in the family and
children's emotional and behavior problems.” Journal of Marriage and the
Family, 53: 1031-1041.

125

Teachman, Jay, Karen Carver and Kathleen Paasch. 1996. “Social Capital and Dropping
Out of School Early.” Journal of Marriage and the Family, 58: 773-83.

Thoits, Peggy A. 1985. “Social Support and Psychological Well-Being: Theoretical
Possibilities,” Pp.51-72 in Social Support: Theory, Research, and Applications,
by Irwin G. Sarason and Barbara R. Sarason (eds). Dordrecht ; Boston : M.
Nihjoff ; Hingham, MA : Distributors for the US. and Canada, Kluwer Boston.

Tracey, A., K. M. Revenson, S. Schiafino, D. Majerovitz, & A. Gibofsky. 1991. “Social
support as a double-edged sword: The relation of positive and problematic

support to depression among rheumatoid arthritis patients.” Social Sciences
Medicine, 33, 801-813.

Turner, H. A., and D. Finkelhor. 1996. “Corporal punishment as stressor among youth.”
Journal of Marriage and the Family, 58: 155-166.

Turner, R. J. 1983. “Direct, indirect, and moderating effects of social support upon
psychological distress and associated conditions,” in Psychosocial Stress: Trends
in theory and research, by H. B. Kaplan (eds). New York: Academic Press.

US. Bureau of the Census. 1994. Statistical Abstract of the United States (114th ed.).
Washington, DC: US. Government Printing Ofrfice.

van Aken, M. A. G., and J. B. Asendorpf. 1997. “Support by parents, classmates, friends,
and siblings in preadolescence: Covariation and compensation across
relationships.” Journal of Social and Personal Relationships, 14: 79-93.

Waldinger, Roger. 1995. “The Other Side of Embeddedness.” Ethnic and Racial Studies,
18(3): 555-580.

Weinraub, Marsha, and Barbara M. Wolf. 1983. “Effects of stress and social supports on

mother-child interactions in single- and two-parent families.” Child Development,
54:1297-1311.

Whitbeck, Les 8., Ronald L. Simons, Rand D. Conger, Frederick O. Lorenz, Shirley
Huck, Glenn H. Elder, Jr. 1991. “Family Economic Hardship, Parental Support,
and Adolescent Self-Esteem.” Social Psychology Quarterly, 54(4): 353-363.

Wilcox, Bradford. W. 1998. “Conservative Protestant Parenting: Authoritarian or
Authoritative?” American Sociological Review, 63:796-809.

Williams, K. B.. & K. D. Williams. 1983. “Social inhibition and asking for help: The
effects of number, strength, and immediacy of potential help givers.” Journal of
Personality and Social Psychology, 44, 67-77.

Wolchik, S. A., J. Beals, and I. N. Sandler. 1989. “Mapping children’s support networks:
Conceptual and methodological issues,” Pp. 191-220 in Children’s Socical
Networks and Social Supports, by D. Belle (eds). New York: Wiley.

126

Wong, Raymond Sin-Kwok. 1998. “Multidimensional Influences of Family Environment
in Education: The Case of Socialist Czechoslovakia.” Sociology of Education, 71:
1-22.

Woolcock, Michael. 1998. “Social Capital and Economic Development: Toward a
Theoretical Synthesis and Policy Framework.” Theory and Society, 27(2): 1 5 1 -
208.

Wortman, C. N. 1984. “Social support and the cancer patient: Conceptual and
methodological issues.” Cancer, 53: 2339-2360.

Wortman, C. B., & C. Dunkel-Schetter. (1979). “Interpersonal relationships and cancer.”
Journal of Social Issues, 35, 120-155.

Wortman, C. B., & Lehman, D. R. (1985). “Reactions to victims of life crisis: Support
attempts that fail,” Pp. 463-489 in Social support: Theory, research and
applications by I. G. Sarason & B. R. Sarason (Eds). Dordrecht, The
Netherlands: Martinus Nijhoff.

Zarling, Cynthia, L. Hirsch, and Susan Landry. 1988. “Maternal social networks and
mother-infant interactions in full-time and very low birthweight, preterm infants.
Child Development, 59: 178-185.

,9

Zhou, Min and Carl L. Bankston. 1998. Growing Up American: How Vietnamese
Children Adapt to Life in the United States. New York: Russell Sage.

127

IGAN 8

11111111

    

1111111111111111111111

2314 5943

1111111111

 

312930