ABSTRACT
Title of Dissertation: CHRONIC ILLNESS AND BEHAVIOR PROBLEMS 
IN CHILDREN: MEDIATING AND MODERATING 
INFLUENCES
Laura Ruth Hoard, Doctor of Philosophy, 2004
Dissertation directed by:
Professor Sandra Hofferth
Family Studies 
Research indicates that children living with a chronic illness have a higher 
level of behavior problems than children not living with an illness.  However, 
mediating factors must be examined in order to create a clearer picture of the 
influence of a chronic health condition on children.  Therefore, this research 
examined the mediating influence of economic strain, child's health stress, parental 
psychological distress, and parenting behaviors on child behavior, as well as the 
moderating impact of social support on all previously mentioned variables.  
Structural equation modeling was used to model each of these relationships.  Data 
came from responses to the Panel Study of Income Dynamics Child Development 
Supplement that included both children not living with an illness (n=806) and 
children living with a chronic illness (n=693) who were between the ages of 6 and 13 
and their primary caregivers.  The chronic illnesses included in the sample are 
anemia (n=120), asthma (n=157), diabetes (n=3), and chronic otitis media (n=522).  
Comorbidity was present with 102 children living with two illnesses and four 
children living with three illnesses.  The findings indicate that the presence of a 
chronic illness has a relationship with higher levels of child behavior problems, but 
health stress and parental psychological distress contribute more to child behavior 
problems than the presence of illness.  Social support was found to buffer the 
relationships between the presence of illness and economic strain and the presence of 
illness and health stress.  Findings suggest that researchers, practitioners, and 
policymakers should focus on the alleviation of economic strain, health stress, and 
parental psychological distress because of their relationship with poorer parenting 
and higher levels of child behavior problems.  Possible directions for future research 
were also discussed.     
CHRONIC ILLNESS AND BEHAVIOR PROBLEMS IN CHILDREN: 
MEDIATING AND MODERATING INFLUENCES 
by
Laura Ruth Hoard
Dissertation proposal submitted to the Faculty of the Graduate School of 
the University of Maryland, College Park in partial fulfillment 
of the requirements for the degree of 
Doctor of Philosophy 
2004
Advisory Committee:
Professor Sandra Hofferth, Chair
Professor Elaine A. Anderson
Professor Norman B. Epstein
Professor Gregory Hancock
Professor Sally A. Koblinsky
?Copyright by
Laura Ruth Hoard
2004
ii
Acknowledgements
I would like to thank Dr. Sandra Hofferth for the many ways she has assisted 
in the direction and completion of this research.  Sandy's mentorship has provided 
professional growth in conference presentations and grant funding.  I also wish to 
thank Dr. Gregory Hancock for opening my eyes to the wonderful world of structural 
equation modeling.  His expertise and passion for the subject has "worn off" on me, 
and I hope to continue to improve my statistical skills.  Thank you to Dr. Elaine 
Anderson for her continued mentorship and for her willingness to discuss my 
proposal drafts.  Lastly, thank you to Dr. Sally Koblinsky and Dr. Norman Epstein 
for their invaluable insights and support for my research and ideas throughout the 
years.  I will be a better family researcher for having known and worked with this 
committee.  
Lastly, I would like to thank the people who assisted me personally through 
this dissertation/Ph.D. journey.  Thank you to my grandfather for always 
encouraging my continued education, and constantly reminding me to publish, 
publish, publish.  My parents have been supportive and always willing to lend an 
inspirational word.  During my dissertation journey I was lucky enough to marry the 
most wonderful person, Matthew Welsh, who has been with me throughout my 
college career.  He has listened to all of my successes and challenges, read many of 
my early and later writings, and most importantly has provided not only emotional 
encouragement, but instrumental support with housecleaning, cooking, and now 
puppy watching.  My life would not be the same without him.
iii
TABLE OF CONTENTS
List of Tables .................................................................................................   x
List of Figures................................................................................................  xi
Chapter I: Introduction...................................................................................   1
Chapter II: Literature Review ........................................................................   7 
Defining chronic illness .....................................................................   7  
Chronic illness and child behavior problems.....................................   9  
Ecological model ...............................................................................  11
Theoretical research model....................................................  16
Economic strain .................................................................................  18
Economic strain and chronic illness ......................................  18
Economic strain and child health stress.................................  20
Economic strain and parental psychological distress.............  22
Economic strain and parenting behaviors ..............................  22
Economic strain and child behavior problems.......................  24
Child health stress and chronic illness...............................................  25
Child health stress and economic strain.................................  27
Child health stress and parental psychological distress .........  28
Child health stress and child behavior problems ...................  30
Parental psychological distress ..........................................................  32
Parental psychological distress and chronic illness ...............  33
Parental psychological distress and parenting behavior ........  35
iv
Parental psychological distress and child behavior 
problems..................................................................... 36
Parenting behaviors............................................................................  37
Parenting behaviors and chronic illness.................................  37
Parental cognitive stimulation................................................  38
Parental emotional support ....................................................  39
Parental warmth .....................................................................  40
Parental involvement .............................................................  41
Social support.....................................................................................  42
Social support and buffering of stress....................................  43
Social support and chronic illness..........................................  44
Social support and economic strain .......................................  48
Social support and parental psychological distress................  48
Social support and parenting behaviors .................................  49
Social support, chronic illness, and children's adjustment.....  49
Definitions..........................................................................................  50
Hypotheses.........................................................................................  51
Chapter III: Method .......................................................................................  54
Participants.........................................................................................  54
Procedure ...........................................................................................  59
Instruments.........................................................................................  60
Economic strain ....................................................................  61
v
Child health stress..................................................................  63
Parental psychological distress ..............................................  64
Parental cognitive stimulation ...............................................  66
Parental emotional support ....................................................  66
Parental warmth .....................................................................  70
Parental involvement .............................................................  70
Child behavior problems........................................................  71
Social support.........................................................................  72
Weights ..............................................................................................  74
Missing data.......................................................................................  75
Statistics .............................................................................................  75
Descriptive analysis ...............................................................  75
Multivariate analysis..............................................................  76
Chapter IV: Results........................................................................................  83
Demographic characteristics..............................................................  83
Participants.............................................................................  83
Measures ................................................................................  90
Structural equation model for the total sample..................................  92
Multigroup structural equation model ...............................................  95
Hypothesis thirteen ................................................................  99
Hypothesis one....................................................................... 101
Hypothesis one social support moderation ................ 101
vi
Hypothesis two....................................................................... 101
Hypothesis two social support moderation................ 102
Hypothesis three..................................................................... 102
Hypothesis three social support moderation.............. 102
Hypothesis four...................................................................... 103
Hypothesis four social support moderation ............... 103
Hypothesis five ...................................................................... 103
Hypothesis five social support moderation................ 104
Hypothesis six........................................................................ 104
Hypothesis six social support moderation ................. 104
Hypothesis seven ................................................................... 105
Hypothesis seven social support moderation............. 105
Hypothesis eight..................................................................... 105
Hypothesis eight social support moderation.............. 106
Hypothesis nine...................................................................... 106
Hypothesis nine social support moderation ............... 106
Hypothesis ten........................................................................ 107
Hypothesis ten social support moderation ................. 107
Parental psychological distress and 
     parental cognitive stimulation............................... 107
Parental psychological distress and 
     parental emotional support.................................... 108
vii
Parental psychological distress and 
     parental warmth .................................................... 108
Parental psychological distress and 
     parental involvement............................................. 108
Hypothesis eleven .................................................................. 109
Hypothesis eleven social support moderation............ 109
Hypothesis twelve.................................................................. 110
Hypothesis twelve social support moderation ........... 110
Parental cognitive stimulation and 
     child behavior problems........................................ 110
Parental emotional support and 
 child behavior problems....................................... 111
Parental warmth and child behavior problems........... 111
Parental involvement and 
     child behavior problems........................................ 111
Summary of hypothesis thirteen ............................................ 112
Indirect effects of the presence of illness........................................... 113
Indirect effects of the presence of illness and social support............. 115
Chapter V: Discussion ................................................................................... 118
Chronic illness and economic strain .................................................. 118
Chronic illness and child health stress............................................... 119
Chronic illness and parental psychological distress........................... 121
viii
Chronic illness and child behavior problems..................................... 123
Economic strain and parental psychological distress......................... 125
Child health stress and parental psychological distress ..................... 126
Economic strain and parental cognitive stimulation.......................... 128
Economic strain and parental involvement........................................ 129
Child health stress and child behavior problems .............................. 130
Parental psychological distress and parenting behaviors................... 131
Parental psychological distress and child behavior problems............ 133
Parenting behaviors and child behavior problems ............................. 134
Chronic illness and child behavior problems..................................... 136
Social support......................... ........................................................... 138
Ecological theory ............................................................................... 140
Family stress theory of economic hardship ....................................... 142
Implications for Researchers, Practitioners, and Policy Makers ....... 142
Limitations ......................................................................................... 146
Directions for future research ............................................................ 150
Recommendations.............................................................................. 153
Conclusions........................................................................................ 154
Appendix A: Introductory Letter to Participants ........................................... 155
Appendix B: Economic Strain Index ............................................................. 157
Appendix C: Child Health Stress Index......................................................... 159
Appendix D: Non-Specific Psychological Distress Scale ............................. 160
ix
Appendix E: Home Observation for Measurement of the Environment 
(children aged six to nine)...................................................... 161
Appendix F: Home Observation for Measurement of the Environment  
(children aged ten and older) ................................................. 165
Appendix G: Parental Warmth Scale............................................................. 169
Appendix H: Parental Involvement Scale...................................................... 170
Appendix I: Behavior Problems Index .......................................................... 171
Appendix J: Social Support Index ................................................................. 173
Appendix K: Correlation Matrix Total Sample ............................................. 175
Appendix L: Correlation Matrix Low Levels of Social Support ................... 176
Appendix M: Correlation Matrix Mid to High Levels of Social Support...... 177
Appendix N: Error Corrections for Total Sample, and 
Low and Mid to High Levels of Social Support.................... 178
References ................................................................................................ 179
x
List of Tables
Table 1: Sample Rate of Chronic Illness ......................................................  57
Table 2: Child Demographics ........................................................................  84
Table 3: Parent Demographics.......................................................................  86
Table 4: Household Demographics................................................................  89
Table 5: Measure Demographics ...................................................................  91
Table 6: Standardized Direct, Indirect, and Total Effects for Total Sample..  94
Table 7: Standardized Direct, Indirect, and Total Effects for 
Levels of Social Support........................................................  98
Table 8: Indirect Effects of the Presence of Illness on Child Behavior 
Problems ................................................................................114
Table 9: Indirect Effects of the Presence of Illness on Child Behavior 
Problems for Low Levels of Social Support..........................116
Table 10: Indirect Effects of the Presence of Illness on Child Behavior
Problems for Mid to High Levels of Social Support .............117
xi
List of Figures
Figure 1: Depiction of the Ecological Model ................................................  13
Figure 2: Theoretical Model ..........................................................................  17
Figure 3: Structural Equation Model .............................................................  78
Figure 4: Transformation of Observed Variables to Latent Variables .........  80
Figure 5: Total Sample SEM Standardized Solutions ...................................  93
Figure 6: Low Levels of Social Support SEM Standardized Solutions.........  96
Figure 7: Mid to High Levels of Social Support SEM 
Standardized Solutions........................................................... 97
Figure 8: SEM Low and Mid to High Levels of Social Support
 with Constraints Applied ..................................................................100
1
Chapter I: Introduction
Today, 18 percent of American children (12.6 million) 18 years old and 
younger have a chronic condition that requires some health service beyond routine 
child medical care (Newacheck, Strickland, Shonkoff, Perrin, McPherson, 
McManus, Lauver, Fox, & Arango, 1998).  Alarmingly, the number of children 
being diagnosed with a chronic condition continues to rise.  For instance, asthma 
among American children has increased from three percent in 1981 to five percent in 
1998 (Benson, 2001).  Otitis media, or ear infections, have increased from 34.8 
percent of children between 1989 and 1991 to 41.1 percent between 1991 and 1994 
(Auinger, Lanphear, Kalkwarf, & Mansour, 2003).  
Despite these increased rates of chronic health conditions, in recent decades 
medical technologies have increased chronic illness survival rates (Blum, Garell, 
Hodgman, & Slap, 1993).  For example, prior to the discovery of insulin, children 
who were diagnosed with diabetes were expected to die within a year (Krall, 1978).  
Insulin for diabetes; inhalers and peak flow monitors for asthma; iron supplements, 
blood transfusions, and bone marrow transplants, for anemia; and antibiotics and ear 
tubes for otitis media are all treatments that have been discovered in recent years.  
These medical advances are saving thousands of individuals from disability and 
some from an otherwise inevitable death.  
Chronic illness does not discriminate based on race, age, socioeconomic 
status, or gender.  Although some children -- African-Americans, boys, older 
children, and children in low income and single-parent households -- are more likely 
2
to be living with a chronic condition (Newacheck, et al., 1998), all families are 
vulnerable to having a child diagnosed with a chronic illness.  The presence of a 
chronic illness can influence both child and family functioning.  
Research in the family health arena has historically focused on the deficit 
model, in particular the burden of chronic illness on the family and individual.  
Caregivers were considered to be burdened, passive agents and the individual living 
with the illness was simply considered "the burden? (Zarit, 1989).  Further, the mere 
presence of a chronic illness was directly linked to pathology and negative effects for 
children and families (Bedell, Giordani, Amour, Tavormina, & Boll, 1977; Olkin, 
1999).  
Deficit model research often examines pathology, targets clinical samples that 
are thought to exhibit the pathology in question, and creates or uses instruments to 
detect pathology within the sample (Olkin, 1999).  Consequently, the research finds 
that individuals living with a chronic illness exhibit some form of pathology and a 
discussion of the burden of the "sick" individual on families ensues.  Clearly, the 
deficit literature equates the individual and the disability, creating an asthmatic or a 
diabetic rather than a person living with asthma or diabetes.  
Just as the social sciences utilized the deficit model, so too did the medical 
field.  The medical community not only reduced the individual to an illness, but 
further reduced the individual to the cellular level (Olkin, 1999).  These reductions 
lead to the examination of deficient genes, cells, or organs instead of focusing on the 
individual living with a chronic illness.  
3
In recent years there has been a paradigm shift from the deficit model to a 
socioecological
1
 or ecological model in the social sciences and medical fields 
(Pledger, 2003; Schor, Billingsley, Golden, McMillan, & Meloy, 2003; Tate & 
Pledger, 2003; Wertlieb, 2003).  The ecological model examines the relational nature 
of the illness and the environment (Pledger, 2003).  More specifically, the ecological 
model emphasizes the intersection of environmental, family, and individual 
characteristics that shape the experience of living with a chronic illness (Pledger, 
2003; Tate & Pledger, 2003).  For example, medical professionals are shifting focus 
from illness and cells to examining how families function in relation to health and 
how to treat and maintain health within the family context (Wertlieb, 2003).  
The ecological model tends to examine the protective factors and risks in the 
environment that may influence individual and family outcomes.  For instance, 
research explores the influence of the disability or illness, separate from the person, 
on individual and family functioning (Olkin, 1999).  Illness is not equated with the 
individual, and no one pattern of response to chronic illness is expected; instead 
unique protective factors and vulnerabilities are explored (Brown & DuPaul, 1999).  
Vulnerabilities or risks might include economic strain and child health stress; 
whereas protective factors might include parental well-being, positive parenting, and 
social support.  Research that focuses solely on the risks involved in families living 
with a chronic illness may miss the more important mediating factors such as 
1
 The socioecological model "maintains that disability is a product of the intersection of individual 
characteristics (e. g. , conditions or impairments, functional status, or personal and socioeconomic 
qualities) and characteristics of the natural, built, cultural, and social environments" (National Institute 
on Disability and Rehabilitation Research, 1999, p. 2).
4
economic strain, child health stress, parental mental health, and parenting behaviors 
(Brown & DuPaul, 1999) as well as the moderating factor of social support.  The use 
of the socioecological model that examines risks and protective factors will provide 
useful information in the study of families and the true environment in which they 
live (Wertlieb, 2003).  Protective factors and risks have been examined previously in 
research on child behavior problems through the use of mediating and moderating 
variables.  
Previous research has examined the relationships between economic strain, 
parental well-being, parenting behavior, and child behavior problems, however none 
of the research has focused on chronic illness (Conger, Conger, Elder, Lorenz, 
Simons, & Whitbeck, 1992; Jackson, Brooks-Gunn, Huang, & Glassman, 2000; 
Yeung, Linver, & Brooks-Gunn, 2002).  Additionally, the participants involved in 
the research often represent particular population groups.  Conger et al. (1992) 
focused on rural, Midwest Caucasians, and Jackson et al. (2000) examined urban 
African American single mothers.  In contrast, Yeung et al. (2002) utilized a 
nationally representative data set, the Panel Study of Income Dynamics Child 
Development Supplement, to research three to five year olds.  The current research 
expands upon this previous research to examine a nationally representative sample of 
school age children, ages six to thirteen, who are either living with a chronic illness 
or not.  
Two methodological issues that are present in the current literature on 
chronic illness -- the use of clinical versus non-clinical samples and the use of 
5
matched control groups -- will also be addressed in the proposed research.  Few of 
the present research studies examine nationally representative samples.  Instead 
smaller samples are drawn from clinical populations (Anthony, Gil, & Schanberg, 
2003; Silver, Bauman, & Ireys, 1995).  Clinical samples are more likely to be 
experiencing stress or difficulties in daily life.  Additionally, many individuals living 
with a chronic illness are not involved with hospitals or specialty care.
The research on chronic illness generally does not have randomly sampled 
control groups.  Instead much of the research relies on matched control groups.  
Matching variables have included gender, race, age, and school grade (Gerhardt, 
Vannatta, McKellop, Zeller, Taylor, Passo, & Noll, 2003; Holmes, Yu, & Frentz, 
1999).  However, matched control groups may be problematic because important 
variables may not be included in the matching criteria (Babbie, 2001).  To address 
these methodological issues, this research uses a nationally representative sample of 
non-clinical families comparing children living with a chronic illness to those who 
are not living with an illness.     
In order to address the existing gaps in the literature on families living with a 
chronic illness, the present study uses data collected for the Panel Study of Income 
Dynamics Child Development Supplement (PSID-CDS) of 1997.  The PSID-CDS 
was constructed to examine economic status and time use of families, parental 
resources, and the subsequent development of children over time.  Using the PSID-
CDS data provides not only a nationally representative sample of children, but also a 
sample that is not selected on the basis of clinical status.  Utilizing an ecological 
6
model, this study uses structural equation modeling to examine the influence of 
chronic illness, economic strain, and child health stress on parental psychological 
distress; parenting behaviors including warmth, involvement, cognitive stimulation, 
and emotional support; and child behavior problems.  Social support was also 
examined as a possible moderator or buffer to the previously mentioned relations.
7
Chapter II: Literature Review
Defining chronic illness
Adult definitions and measures of chronic illness are insufficient for use with 
children (Westbrook, Silver, & Stein, 1998).  When adult measures of chronic illness 
are used with children, 90 percent of childhood issues are missed (Brooks, 2001).  
Children cannot be held to the same standards as adults since the goals of childhood 
are to play, learn, and grow (Westbrook, et al., 1998).   To combat these issues, there 
are numerous definitions of chronic illness that have been created for research, 
programmatic, and policy considerations (Westbrook, et al., 1998).  Various 
definitions found in the literature include: 
Chronic illness is generally defined as a physical or mental condition that is
persistent or recurring and can range from mild to severely disabling (Stark, 
2000).
?Chronic illness includes those diseases involving a protracted course 
that may be fatal or result in compromised mental, cognitive and/or 
physical functioning and that are often characterized by acute 
complications, which may result in hospitalizations or other forms of 
intensive treatment? (Brown & DuPaul, 1999, p. 175).
Chronic illness is the "lived experience, involving permanent deviation 
from the normal, caused by unalterable pathological changes" (Cameron 
& Gregor, 1987, p. 672).  
Due to the differing definitions, there have been fluctuating reported 
prevalence rates.  Prevalence rates for children living with a chronic condition vary 
between 5 and 31 percent, with the former reflecting only the most severe conditions 
and the latter reflecting conditions with modest effects on children's activities 
8
(Newacheck & Taylor, 1992; Newacheck, et al., 1998). To address this concern the 
Maternal and Child Health Bureau's Division of Services for Children with Special 
Health Care Needs has created a definition of children with special health care needs.  
"Children with special health care needs are those who have or are at increased risk 
for chronic physical, developmental, behavioral, or emotional conditions and who 
also require health and related services of a type or amount beyond that required by 
children generally" (McPherson, Arango, Fox, McManus, Newacheck, Perrin, 
Shonkoff, & Strickland, 1998, p. 138).  Using this new definition, research indicates 
that approximately 18 percent or 12.6 million American children are living with a 
chronic health condition (Newacheck, et al., 1998).
The Maternal and Child Health Bureau (MCHB) has proposed two categories 
for operationalizing the new MCHB definition (Neff, Sharp, Muldoon, Graham,  
Popalisky, & Gay, 2002).  The categorical approach is based on the presence of a 
specific condition a child is living with, such as asthma or anemia (Neff, et al., 
2002).  The noncategorical approach classifies individuals based on the 
characteristics related to needs the illness creates such as frequency of service use or 
functional status (Neff, et al., 2002).  The Panel Study of Income Dynamics Child 
Development Supplement does not provide information relating to the 
noncategorical approach, therefore the categorical approach that is based on the 
presence of a specific condition was used in the current research.  In the remainder of 
the chapter, the relationship between chronic illness and child behavior problems 
addressed in existing literature is described.  
9
Chronic illness and child behavior problems
Historically chronic illness in children was viewed from the deficit perspective, 
namely individuals are a burden reflecting some pathology, and from this perspective 
disastrous consequences for children were thought to follow (Bedell, et al., 1977; 
Olkin, 1999).  The deficit perspective often found that children living with a chronic 
illness experienced greater internalizing and externalizing behavior problems than 
children living without an illness. For example, Gillaspy et al. (2002) found that 
high-risk adolescents living with asthma have significantly higher levels of anxiety, 
depression, and global psychological distress when compared with a matched sample 
of high-risk individuals not living with asthma.  However, the research did not assess 
the type of risk the adolescents were living with and how the risk contributed to their 
anxiety and depression.  Instead the researchers attributed the difference in 
psychological distress to the presence of an illness, thus supporting the deficit model.  
However, some research does not support the deficit perspective that suggests a 
positive relationship between chronic illness and child behavior problems.  For 
example, child behavior problems have not been found to be related to the presence 
of hemophilia (Minden, 1999).  Additionally, when children living with a chronic 
illness are compared with healthy controls, the differences are minimal.  For 
example, one study found that children living with cancer did not differ on measures 
of depression, anxiety, loneliness, or self-concept when compared with controls 
(Noll, Gartstein, Vannatta, Correll, Bukowski, & Davies, 1999).  Similarly, no 
differences were found between "healthy" children and children living with chronic 
10
asthma in terms of their emotional regulation, with poor regulation defined by 
disruptive negative affect and low frustration tolerance (Klinnert, McQuaid, 
McCormick, Adinoff, & Bryant, 2000).
Further, the deficit perspective has largely ignored family strains or limited 
social support in families living with a chronic illness that might account for the 
increase in child behavior problems.  The failure to include these additional variables 
creates an artificial explanation for children's behavior problems and the relationship 
with chronic illness.  Kazak and Christakis conclude "that there is no 
incontrovertible evidence for a direct, linear, cause-and-effect relationship between 
illness burden and maladjustment implies that mediating variables must exist" (1994, 
p. 342).  Therefore, research has begun to examine possible mediating and 
moderating factors that could account for the relationship between chronic illness 
and behavior problems (Drotar, 1997; Ellerton, Stewart, Ritchie, & Hirth, 1996; 
Holmes, et al., 1999; Klinnert, et al., 2000; Paradise, Feldman, Colborn, Campbell, 
Dollaghan, Rockette, Janosky, Kurs-Lasky, Bernard, & Smith, 1999; Perrin, 
MacLean, & Perrin, 1989; Rogers & Floyd, 2003; Wallander & Varni, 1989).  
Research is now examining the context or environment of children living with a 
chronic illness.  
Holmes et al. (1999) found that the presence of a chronic illness such as 
diabetes is associated with mild, nonspecific behavioral problems.  However, these 
behavior problems are synonymous with children's reactions to generalized distress.  
Additionally, the research points to negative life events and the family environment 
11
influencing the relationship between the presence of a chronic illness and child 
behavior problems.  
There has been some research that does find a relationship between chronic 
illness and behavior problems.  For example, Haffner and associates (Haffner, Ester, 
Munch, Parzer, Raue, Steen, Klett, & Resch, 2002) found that the presence of a 
chronic illness was a predictor of behavior problems in children.  However the 
research also showed that chronic illness did not explain a majority of the variance; 
instead family problems accounted for 10 percent of the variance in child behavior 
problems.  Gender, chronic illness, adverse living conditions, the family 
composition, and the amount of television the child watched accounted for another 
10 percent of the variance in child behavior problems.  Therefore, chronic illness 
explained only a small portion of child behavior problems.   
To fully examine the relationship between chronic illness and child behavior 
problems the context in which the child lives must be examined.  The systems within 
the environment -- family, friends, schools, and employers -- have reciprocal effects 
on the child and in turn the child on these systems.  Because each of these systems 
impacts one another, the study of chronic illness in children is best examined from 
the ecological perspective (Bronfenbrenner, 1986).     
Ecological model
Many researchers have used the ecological theoretical model in relation to 
families with a child living with a chronic illness because this model best addresses 
the contextual factors associated with a chronic illness and how the illness affects the 
12
individual and the family (Kazak & Christakis, 1994; Vinson, 2002).  The ecological 
model is useful for examining the reciprocal impact of chronic illness on individuals 
and internal and external systems to the family (Kazak & Christakis, 1994).  Thus, 
the model explains that events external to the family unit are just as influential as the 
activities inside the family unit and inside the individual family member 
(Bronfenbrenner, 1986).  For example, a child's behavior problems are not solely 
based on the presence of a chronic illness, but are influenced by stresses within the 
family and the functioning of external factors such as health insurance policies and 
social supports received from family, friends, and medical professionals.  The 
ecological model acknowledges that external barriers -- such as the flexibility of 
one's health insurance -- are associated with living with a chronic illness and 
considers how these factors can affect individual and family functioning.  
According to ecological theory, family and individual functioning are 
embedded in the following four systems: the microsystem, mesosystem, exosystem, 
and macrosystem (see Figure 1).  The ecological model views these four different 
systems as constantly interacting and influencing each other (Bronfenbrenner, 1986).  
The microsystem is the individual level that includes the settings that the individual 
experiences and creates as a day-to-day reality.  This level includes such factors as 
the physical and mental health of the individual, the people with whom the 
individual living with a chronic illness resides, and the financial viability of the 
family.
13
Figure 1: Depiction of the Ecological Model
The mesosystem is the relationship between many other microsystems 
(Bronfenbrenner, 1986).  This level includes the extended family, friends, and other 
areas of emotional or instrumental support for the family and individual living with a 
chronic illness.  The mesosystem also includes the interaction of doctors, nurses, 
pharmacists, employers, and schools in the lives of the individual and family.  
Families living with a chronic illness may have more interaction with the 
mesosystem than families not living with illness, due to the demands of the illness.  
For instance, parents may need to take additional leave from work to care for a child 
(Ex. Social support)
Micro
(Ex. Family)
Exo
(Ex. Employers)
Macro
(Ex. Culture)
Meso
14
experiencing an acute illness episode; this episode may also necessitate interaction 
with medical professionals and schoolteachers.  
The exosystem and macrosystem are further removed from daily family 
interaction, but continue to have great influence over family functioning.  The 
exosystem has power over individuals? lives, but individuals possess little control 
over that system (Bronfenbrenner, 1986).  For instance, the policies towards 
employee leave and health insurance in the workplace are included in the exosystem.  
Additionally, Medicare, Medicaid, and health insurance companies are in the 
exosystem and create policies that delineate the cost, availability, and procedures for 
obtaining medical treatment for those living with a chronic illness.  
The macrosystem represents the broad ideological and institutional settings of 
a particular culture or subculture.  The macrosystem includes societal and cultural 
prejudices and stigma towards individuals living with a chronic illness (Dowling & 
Dolan, 2001).  Historically individuals living with an illness or disability were 
considered to be defective and generally lacking in ability to function in society 
(Disability Social History Project, 2001; Morgan, 2001; Russell, 1998).  Although 
these beliefs have been changing in recent years, their historical legacy continues 
(Dowling & Dolan, 2001).  The influence of the macrosystem on the other systems 
and its reciprocal relationship to the ecological model is displayed in the following 
example.  The stigma and stereotypes of illness in society (macrosystem) may affect 
the employer?s choice of medical insurance (exosystem), the presence of extended 
15
familial social support (mesosystem), and the individual's personal beliefs about the 
illness s/he must live with (microsystem).  
The ecological framework also addresses risks and protective factors that are 
present in the four systems.  Risk to the individual and family may be directly related 
to the presence of a chronic illness (Cadman, Boyle, Szatrnari, & Offord, 1987).  
Additional risks might include external issues such as societal stigma, federal policy 
regulations, and types of health coverage, as well as internal aspects such as 
economic strain, health stress, parental psychological distress, and child behavior 
problems.  Unfortunately, often the relationships with health care providers, 
insurance companies, and government agencies are among the most stressful for 
individuals and families living with a chronic illness (Doherty, McDaniel, & 
Hepworth, 1994).  
In contrast, protective factors may help to remove risk from families and 
individuals.  Creating a shared family view of the illness is often a protective factor.  
For example, the impact of the illness on the family tends to decrease when the 
parents of a child living with a chronic illness have a shared view of the illness, its 
management, and its impact on the family (Knafl & Zoeller, 2000).  Protective 
factors might also include attending support groups, having helpful extended family 
and community supports, and/or medical professionals who are willing to educate 
and work together with the family.  Finally, positive parenting may act as a 
protective factor and help to reduce the risk of child behavior problems.
16
Theoretical research model.
The ecological model and previous research helps to shape the theoretical 
model that was employed in the current research (see Figure 2).  As shown, the 
presence of a chronic illness is hypothesized to affect economic and child health 
stressors, parental psychological distress, and child behavior problems directly.  The 
model does not imply that the economic strain, parental psychological distress, and 
child behaviors are created solely by the presence of illness, just that the illness has 
some impact upon these variables.  According to the model, the influence of illness 
on parenting behavior is mediated by economic strain, child health stress, and 
parental psychological distress.  The economic and child health stressors directly 
influence parental psychological distress and indirectly influence parenting behavior 
and child behavior.  Economic strain directly influences the parenting behavior of 
cognitive stimulation.  Child health stress also directly impacts child behavior 
problems.  Parental psychological distress directly influences parenting behaviors 
and child behavior problems.  Finally, parenting behaviors directly influence child 
behavior problems.  The model posits that social support will moderate each of the 
previously mentioned relations.  However mid to high levels of social support, as 
reported by the primary caregivers, will be a more successful moderator than low 
levels of social support. The following sections examine the literature on the 
previously mentioned relations.  
Child health 
stress
Figure 2: Theoretical Model
Note: Social support will moderate each of the above relations.  The two levels of social support, low and mid to high, 
will be compared for each relation.
Economic strain Cognitive stimulation
Involvement
Social support
Presence of 
chronic illness
Economic strain
Parental 
Psychological 
Distress
Parenting Behaviors

 Cognitive 
stimulation

 Emotional 
support

 Warmth

 Involvement 
Child 
behavior 
problems
17
18
Economic strain
?Economic strain refers to the subjective evaluation of one?s financial 
situation? (Voydanoff & Donnelly, 1998, p. 346) and may be marked by a lack of 
financial resources such as not having enough money at the end of the month, 
postponing family purchases, and not having health insurance.  The 
operationalization of the economic strain concept is broad enough to capture any 
family that experiences financial shortfalls.  Under this definition it is possible that 
middle and upper class families as well as families from lower socioeconomic 
classes experience economic strain.  Therefore, economic strain is not equated with 
poverty (McLoyd, 1990).  However, economic strain is not a concept that is used in 
federal statistics.  Therefore, at times the inclusion of research on poverty illustrates 
some of the problems families living under economic strain may face.
Economic strain and chronic illness.
A search of the social science literature on economic strain, chronic illness, and 
children produced very limited results.  Currently, there is research on children in 
poverty who are living with a chronic illness, but as mentioned previously economic 
strain and poverty are two very different concepts (McLoyd, 1990).  The financial 
burden due to the presence of asthma was examined in a study of families living in 
East Harlem, New York with children (23 African American and Hispanic) living 
with asthma (Mailick & Holden, 1994).  The research found large, significant 
correlations between financial strain, social isolation, and personal strain (Mailick & 
19
Holden, 1994).  However the study used a convenience sample of 23 parents 
providing little generalizability (Mailick & Holden, 1994).  
The resources, both financial and emotional, that families have may be more 
important when a child is living with a chronic illness.  The research shows that 
children who are living with a chronic illness and have good general functioning are 
more likely to have more resources than children who are living with a chronic 
illness but have lower general functioning (Jessop & Stein, 1985).  Research also 
indicates that even if the family has low income, as long as basic needs are met, the 
lack of income does not contribute to negative psychosocial outcomes in families 
living with a child with sickle cell anemia (Barbarin, Whitten, Bond, & Conner-
Warren, 1999).  Although the research is limited, one could hypothesize that having 
a chronic illness may either create economic strain or increase levels of current strain 
due to the financial requirements of disease care. 
Individual health is often compromised due to economic strain, although 
causation is not explained with cross-sectional data (Chen, Matthews, & Boyce, 
2002).  For example, is it the demands of the illness that prevent parents from 
seeking or maintaining employment when their child is ill (Dowling & Dolan, 2001; 
Kazak & Christakis, 1994; Smith, Romero, Wood, Wampler, Chavkin, & Wise,  
2002) or does the lack of financial resources lead to decreased physical well-being 
and the presence of chronic illness in children?  Nevertheless, low income children 
are more likely than higher income children to experience fair or poor health 
(Vandivere, Moore & Brown, 2000) and to live with a chronic condition (Wise, 
20
Wampler, Chavkin, & Romero, 2002).  Specifically, low levels of income have been 
found to be related to the prevalence of asthma (Miller, 2000), otitis media, and 
anemia (Egbuonu & Starfield, 1982) in children.  However, again causation is not 
clear.
     Economic strain and child health stress.
Family economic strain may also influence the child's health stress created by 
the illness, as evidenced by poor health ratings, physical limitations, and severity of 
symptoms.  Children living below the poverty line are less likely than children of 
higher incomes to be in good or excellent health as rated by their parents (Benson,
2001).  Similar to parental ratings, children with a lower socioeconomic status tend 
to have more physical limitations due to a chronic condition.
One national study found that 11 percent of children living in poverty, 
between the ages of 5 and 17, are physically limited due to a chronic condition 
versus 7 percent of children in families above the poverty line (Benson, 2001). Low 
income may also be associated with complications created by the chronic illness, 
possibly due to the family's inability to afford adequate health care.  For example, 
low income has been related to hearing impairments caused by chronic otitis media 
(Egbuonu & Starfield, 1982).  
The financial costs associated with maintaining the health of children living 
with a chronic illness often add additional economic strain.  The estimated cost for 
treating otitis media in the United States is $2.9 billion in direct and $2.1 billion in 
indirect (e.g., lost wages) costs (Kirschstein, 2000).  In general, individuals living 
21
with a chronic illness can expect to pay an average of $369 per year in out-of-pocket 
expenses, whereas those without a chronic condition spend $182 per year (National 
Chronic Care Consortium, 2003).  Yearly treatment of asthma can range between 
$326 to $1,315 per person (National Institutes of Health, 1996).  These additional 
costs can influence the financial allocation of funds in families.  For example, the 
cost of medications for children with chronic otitis media may cause parents to have 
to weigh paying for medical costs against needed household items (Asmussen, 
Olson, & Sullivan, 1999).  Many families who are seeking health care express 
concerns that they will be financially unable to afford the care (Barlow, 1997). 
Often many of the financial costs of living with a chronic illness are 
reimbursed through health insurance.  In 1999, 86 percent of children had health 
insurance; however, 14 percent did not (Benson, 2001).  Poor children are at the 
greatest risk for inadequate or absent health insurance (Wise, et al., 2002). 
Uninsured children are more likely to have poorer quality of medical care than their 
insured counterparts (Ferris, Crain, Oken, Wang, Clark, & Camargo, 2001), possibly 
leading to a higher rate of hospital and emergency room usage in low income 
children (Miller, 2000).  
Gaps in coverage may create unmet healthcare needs.  Near-poor and poor 
children are three times more likely to have an unmet health care need (e.g., medical, 
dental, and vision care; prescription medications) than non-poor children 
(Newacheck, Hughes, Hung, Wong, & Stoddard, 2000).  Uninsured children are 
three times more likely to have an unmet healthcare need than insured children 
22
(Newacheck, et al., 2000). General economic strain coupled with the financial costs 
of health insurance, illness maintenance, and unmet healthcare needs may create 
mental health stress in parents.
Economic strain and parental psychological distress.
Economic strain may be a constant stressor for parents.  This stressor can lead 
to feelings of tension and sadness (Voydanoff & Donnelly, 1998), or other forms of 
psychological distress such as depression (Belle, 1990; Conger, et al., 1992; Elder, 
1998; Jackson, et al., 2000; Kwon & Rueter, 2003; Lorenz, Conger, & Montague, 
1994; Simons, Lorenz, Wu, & Conger, 1993; Vinokur, Price, & Caplan, 1996; 
Voydanoff & Donnelly, 1998).  Resource loss, which may accompany economic 
strain, has also been linked to increased emotional distress including depression 
(Hobfoll, Johnson, Ennis, & Jackson, 2003).  
Economic strain and parenting behavior.
Economic strain indirectly influences parenting practices through the 
psychological health of the parent.  The relationship between economic strain and 
parenting quality has been found to be mediated by the parent?s emotional well being 
(Conger, et al., 1992; Jackson, et al., 2000; McLoyd, 1990; McLoyd & Wilson, 
1990; Simons, et al., 1993).  The mediation effect has found that higher levels of 
economic strain are related to parenting practices such as decreased involvement and 
warmth (Elder, 1998; McLoyd, 1990; Pinderhughes, Nix, Foster, & Jones, 2001) and 
increased punitive parenting (Elder, 1974).  For instance, mothers experiencing 
psychological distress have reported greater difficulty in parenting and are less 
23
nurturing than mothers who are not experiencing psychological distress (McLoyd & 
Wilson, 1990).  
Economic strain may directly influence parents' ability to cognitively stimulate 
their children (Guo & Harris, 2000; Miller & Davis, 1997). Higher income families 
are financially better able to provide a cognitively stimulating environment (e.g., 
books, puzzles) than lower income families (Garrett, Ferron, Ng'andu, Bryant, & 
Harbin, 1994).  Thus, higher income has been associated with better learning 
environments, including the presence of educational toys and encouragement from 
parents to learn (Klebanov, Brooks-Gunn, & Duncan, 1994).  Cognitive stimulation 
and other parenting behaviors have been found to be directly related to child 
behavior.   
Similarly, economic strain can negatively influence parental involvement in 
their child's life (Simons, Lorenz, Conger, & Wu, 1992).  Financial strains may 
require parents to obtain additional employment, which then detracts from the 
possible time spent with their children.  However, research has found that the mere 
presence of economic strain is negatively related to the time parents and adolescents 
spend engaging in activities together such as playing a game or sport (Voydanoff & 
Donnelly, 1998).  In sum, economic strain increases the level of parental 
psychological distress; parental psychological distress in turn negatively affects 
parenting practices; which may ultimately create negative effects on children 
(Conger, et al., 1992; Conger, Patterson, & Ge, 1995; Conger, Rueter, & Conger, 
2000; McLoyd, 1990; Yeung, et al., 2002).
24
     Economic strain and child behavior problems.
Economic strain can also influence children's behavior problems indirectly 
through parental psychological well-being.  Parents living with economic strain and 
who report psychological distress tend to have children that are at increased risk for 
behavior problems (Coiro, 1997) and poor emotional well-being (Kwon & Rueter, 
2003).  Parents' well-being not only directly influences child behavior, but also 
indirectly affects behavior through parenting behaviors.  When parental 
psychological health is compromised, research indicates that the quality of parenting 
practices decreases.    
Economic strain can also indirectly influence child behavior problems (Elder, 
1998; Takeuchi, Williams, & Adair, 1991; Vandivere, et al., 2000).  Research has 
shown children from lower levels of socioeconomic status, whether they are living 
with a chronic illness (e.g., diabetes, otitis media) or not, have higher levels of 
behavior problems than children who are more economically advantaged (Holmes, et 
al., 1999; Paradise, et al., 1999).  In a recent study, the influence of family income on 
child outcomes was examined using mediator variables in the Panel Study of Income 
Dynamics Child Development Supplement (Yeung, et al., 2002).  The research 
indicates that family income is associated with child behavior problems, in children 
aged three to five, through the mediating variables of maternal psychological distress 
and parenting practices (Yeung, et al., 2002).  Thus, the previous literature points to 
the connection between economic strain and parent psychological distress, parenting 
behaviors, and child behavior problems.  
25
Child health stress and chronic illness
Both children living with a chronic condition and those that are not may 
experience health stress.  However, children living with a chronic illness have a 
much higher chance of incurring health stress due to the illness than those without a 
chronic condition.  Health stresses may be experienced by the child in the form of 
activity limitations, school absences, schoolwork difficulties, or a general poor rating 
of health.
 In 1992, Newacheck and Taylor (1992) reported that 31 percent of children 
were living with chronic conditions.  Of those children, 66 percent had mild 
conditions with little or no activity limitation and the illness created little or no 
bother in their lives.  Twenty-nine percent of those with chronic conditions reported 
moderate severity of the illness with some bother or activity limitation.  The 
remaining five percent experienced severe chronic conditions, with frequent bother 
and limitation.  Less specifically, a 1998 study found that chronic health conditions 
limited the activities of seven percent of America's children aged five to seventeen 
(Benson, 2001).  Furthermore, chronic health conditions account for 66 million days 
of restricted activity per year for children (Newacheck & Halfon, 1998).  
Activity limitation can range from a restriction in play to limitations in the 
ability to conduct major activities, such as personal care (Newacheck & Halfon, 
1998). For example, the presence of recurrent otitis media has been found to limit a 
child's ability to play (Asmussen, et al., 1999).  Children living with a chronic illness 
have reported that restriction from activities due to illness is a major illness stressor 
26
(Ellerton, et al., 1996).  Although all children can experience activity limitation, 
those living with a chronic condition are more likely to face such limitations.  
Msall and associates (Msall, Avery, Tremont, Lima, Rogers, & Hogan, 2003) 
examined the functional and school limitations of 41,300 school-age children.  Of 
children living with a chronic illness (e.g, diabetes), episodic impairment (e.g., 
recurrent otitis media), or asthma, 15 to 18 percent were found to have a severe 
functional disability such as difficulty with mobility, self-care, communication, or 
learning-behavior.  Conversely, only 4.7 percent of children who do not experience 
medical impairments had such limitations (Msall, et al., 2003).  Thus children living 
with a chronic illness are more likely to experience functional and school limitations 
than healthy children. 
 School limitations may be in the form of difficulty completing schoolwork or 
missed school days.  Children living with a chronic illness may have many absences 
from school (Klerman, 1988; Rodder, Kroonenberg, & Boekaerts, 2003).  When 
compared with non-ill children, children living with a chronic illness have three 
times as many school absences as other children (Newacheck, et al., 1998; Msall, et 
al., 2003).  It is estimated that 24 million days of school are missed each year due to 
chronic health conditions (Newacheck & Halfon, 1998).  These school absences may 
be a function of unmet health care needs (Weitzman, Klerman, Lamb, Menary, & 
Alpert, 1982) or an indication of acute medical phases.  Unfortunately, absences 
from school are linked to poor school performance (Weitzman, et al., 1982), and 
children living with a chronic illness who miss school have lower school 
27
achievement than the general absentee population (Schwarz & Lui, 2000).  Thus, 
children living with a chronic illness may have difficulty completing schoolwork 
while absent and may be unable to effectively complete schoolwork once back in 
school.
Children living with a chronic illness may face more activity, school, and 
schoolwork limitations than children not living with an illness.  Furthermore, these 
limitations, coupled with the parental perception of the severity of the illness (e.g., 
number of acute phases) may cause the parent to rate the child as being in fair or 
poor health.  The presence of these limitations and perceived rating of health may act 
as stressors for both the parents and child.  
Child health stress and economic strain.
As previously discussed, economic strain is related to child health stress, with 
unclear causality.  The presence of economic strain may create barriers to attending 
medical appointments, purchasing medical equipment or medication, or maintaining 
appropriate dietary requirements.  To prevent acute medical phases, medical 
attention must be sought early, and compliance with medical advice is necessary 
(Dosa, Boeing & Kanter, 2001).  Thus, economic strain may increase children's 
health stresses.  Conversely, children's health stress may cause parents to be unable 
to hold steady employment due to missed school days and additional time required 
for medical treatments.  Nonetheless, child health stress is a drain on both parents 
and children.
28
Health stress and parental psychological distress.
When families are living with both economic strain and child health stress, 
parents may experience higher levels of psychological distress.  In general, higher 
levels of psychological distress have been found to be related to a pile-up of 
stressors, such as unemployment, family conflict, health problems, and low self-
esteem (Gerard & Buehler, 1999; Hoard & Anderson, 2004; Silver, et al., 1995).  
Whether parents are experiencing their child's health stress alone or a pile-up of 
health and other stressors, their psychological well-being can be influenced by the 
strain. 
The diagnosis of a chronic illness creates new challenges and losses that must 
be addressed by both parents and children (Mullins, et al., 1995).  Research indicates 
that both parents and children show mild symptoms of psychological distress after 
the illness has been diagnosed (Northam, Anderson, Adler, & Werther, 1996).  
However, with time the symptoms fade, with the exception of those who experience 
more severe psychological symptoms at the onset (Thernlund, Dahlquist, Ivarsson, & 
Ludvigsson, 1996).  The presence of child health stress may prolong the period of 
time it takes for symptoms of psychological distress to recede.
Parents may experience psychological distress in the form of anxiety or 
frustration due to their inability to directly control the illness and its course.  Fisher 
(2001) examined the literature on the needs of parents living with a child with a 
chronic illness and found that parents need normality and certainty.  However, 
normality and certainty are difficult to obtain when the child is living with health 
29
stress. For example, parents of children living with recurrent otitis media report 
frustration over the inability to predict future cases of otitis media (Asmussen, et al., 
1999).  In working with parents of children living with diabetes, a father reported to 
me that each time his child's blood sugar was outside of the desired range he felt like 
he was a bad father.  This lack of control may lead parents to question their efficacy 
as a parent, thus creating psychological distress.  
The severity of the chronic illness, as perceived by the parent, may also 
influence parental well-being.  Parents of children living with severe sickle-cell 
anemia reported more strain, anxiety, and anger creating overall poorer parental 
adjustment (Barbarin, et al., 1999).  Conversely, disease severity has been found to 
not be predictive of parental psychological distress in families living with juvenile 
rheumatoid arthritis (Manuel, 2001).  Some research does indicate that functional 
health stresses (e.g., activity limitations) are more significantly related to parental 
psychological distress than the severity of the illness (Canning, Harris, & Kelleher, 
1996; Manuel, 2001; Timko, Stovel, & Moos, 1992; Wallander, Varni, Babani, 
Banis, & Wilcox, 1989). 
Whether it is parental perception of illness or functional limitations, parental 
reaction to the presence of the chronic illness may affect the child living with the 
illness.   When parents believe the illness is severe and that their child is vulnerable, 
there has been found to be an increase in social anxiety in children even when 
disease severity and child's age are controlled (Anthony, et al., 2003).  This 
30
relationship may point to the transmission of parental belief of child health stress to 
children.
     Child health stress and child behavior problems.
Children do directly experience the effects of health stress.  The presence of 
health stressors may lead children to become frustrated with the illness and the limits 
that it places on their activities (Ellerton, et al., 1996). For example, children with 
diabetes often have periods of hypoglycemia or low blood sugar where they must 
stop what they are doing and receive treatment (e.g., orange juice or glucose tablets).  
Treatment for the hypoglycemia is not instantaneous, and thus the child must sit or 
remain still until the treatment takes effect.  Episodes of hypoglycemia can occur at 
any time, but may be more frequent during active periods such as sporting events.  In 
this case, the illness is causing the child to sit out and not participate.  This 
frustration, caused by the illness, may then lead to frustration and ultimately an 
increase in behavior problems.  
Similarly, children may become frustrated because they do not feel well 
physically.  Physical feelings/symptoms of the illness may indicate the severity of 
the illness, which may be reflected in parental ratings of the child's health.  Research 
has found that children living with severe asthma have difficulties with emotion 
regulation, behavior problems, interpersonal difficulties (Klinnert, et al., 2000), and 
social adjustment, evidenced by lower levels of social competence and ability to 
meet new challenges (Perrin, et al., 1989).  
31
Although chronic illness is typically invisible, children living with a chronic 
illness may be teased for simply living with the illness or for complying with a 
required medical regimen (Barlow, Harrison, & Shaw, 1998).  Teasing may occur 
because of special meals, trips to the nurse's office for medication, not being able to 
fully participate in gym class, or the inability to eat sugary foods.  The teasing may 
be due to the "healthy" peers' negative impression of chronic illnesses such as 
diabetes (Underwood, 1997) or unspecified illnesses (Bukowski & Anita, 2000).  
This teasing may create internalizing (anxiety, withdrawal) and externalizing (anger, 
aggression) behavior problems in the child living with a chronic illness.  
Transitions in the family may be more difficult due to the presence of a 
chronic illness (Kazak & Christakis, 1994) and the child's health stress.  Parents may 
prevent children living with a chronic illness from passing through traditional social 
developmental milestones.  For example, parental involvement in the child's life 
especially if health stresses exist, may prevent a child from becoming independent 
(Barlow, et al., 1998).  Parents may also limit the type and extent of interaction with 
peers, which can negatively impact peer relationships (Nassau & Drotar, 1995).  
Parents may be fearful of allowing their child who is experiencing health stresses to 
behave as "healthy" do children.  This overprotection may cause children to act out 
as a way to exert independence.  Thus, child health stressors can influence children 
and parental well-being as well as the parent-child relationship.
32
Parental psychological distress
Parental psychological distress may be thought of as the presence of 
depression and anxiety.  Approximately 32.6 to 35.1 million American adults (12.6 
percent) experience a major depressive disorder in their lifetime (Kessler, Berglund, 
Demler, Jin, Koretz, Merikangas, Rush, Walters, & Wang, 2003b).  In a 12-month 
period, approximately 6.6 percent or 13.1 to 14.2 million adults experience an 
episode of major depressive disorder with the average episode lasting 16 weeks 
(Kessler, et al., 2003b).  Of those cases, 10.4 percent are mild depression, 38.6 
percent are moderate levels of depression, 38.0 percent are severe depression, and 
12.9 percent are very severe levels of depression (Kessler, et al., 2003b).  Of 
individuals with severe or very severe levels of depression, 59.3 percent experience 
role impairment (Kessler, et al., 2003b).  Depressive symptomatology often presents 
as feelings of sadness or hopelessness, changes in sleep, cognitive processing, and 
decreased energy (American Psychiatric Association, 2000).      
Anxiety is another form of psychological distress.  In a 12-month period, 
three percent of the adult population experience some form of anxiety.  Lifetime 
prevalence rates for anxiety in adults are five percent (American Psychiatric 
Association, 2000).  Anxiety is often thought of as "a diffuse emotional reaction that 
is out of proportion to threats from the environment" (Oltmanns & Emery, 1998, p. 
676).  Although anxiety and depression represent differing aspects of psychological 
distress, research indicates that role impairment of anxiety disorders can be 
compared with that of depression (Kessler, DuPont, Berglund, & Wittchen, 2000).  
33
Parental psychological distress and chronic illness.
Some research finds that the presence of a chronic illness can increase 
parental psychological distress both during initial diagnosis (Northam, et al., 1996; 
Thernlund, et al., 1996) and throughout the illness course (Barlow, et al., 1998; 
Cadman, Rosenbaum, Boyle, & Offord, 1991).  Mild symptoms of psychological 
distress have been found in parents after the diagnosis of diabetes in their children, 
but these symptoms tend to fade with time (Northam, et al., 1996; Thernlund, et al., 
1996).  However, parents may experience psychological distress as a reaction to the 
presence of a chronic illness in their childs' lives.
For example, parents of children living with juvenile chronic arthritis have 
reported anxiety about the future and guilt about the possibility of heredity and the 
inability to provide children with reassurance about the future (Barlow, et al., 1998). 
Anxiety and guilt were the primary emotions parents felt towards the illness.  
Families not living with a chronic illness may not experience such anxiety and guilt.  
Research has found that parents of children living with a chronic illness tend 
to have higher levels of psychological distress when compared with "healthy" 
controls.  For instance, mothers of children living with a chronic illness have 
increased negative affect scores in comparison to mothers of children not living with 
an illness (Cadman, et al., 1991).  These rates of psychological distress may be 
related to additional stressors families face.  Fathers of children with a chronic 
illness, when compared with fathers of healthy children, have been found to 
experience a greater number of stressful life events such as financial and work 
34
stresses (Katz & Krulik, 1999).  Daily hassles created by the illness may also add to 
psychological distress.  For instance, chronic otitis media has been found to disrupt 
parents' daily lives in areas like sleep, work schedules, and childcare (Asmussen, et 
al., 1999).  
Conversely, some research has indicated that the presence of a chronic illness 
does not lead to negative effects on parental well-being.  Contrary to expectations, 
the caregiving role was not found to be deleterious for parental well-being in a 
sample of predominantly Caucasian (n=5,782) parents caring for either a child, 
spouse, or parent living with a chronic condition (Marks, 1998).  Families of children 
living with juvenile rheumatoid arthritis generally reported similar levels of parental 
psychological distress, family functioning, and social support when compared with 
"healthy" controls (Gerhardt, et al., 2003).  It is possible that the examination of 
mediating variables, such as family functioning, lessens the effects of the illness and 
presents a realistic picture of the family environment. 
Higher levels of parental well-being in families living with a chronic illness 
may be mediated by parents seeking therapeutic assistance.  Parents of children 
living with a chronic illness may be better able to seek help for psychological 
distress because of their continued involvement in the health care setting.  In a 
community sample, parents of children living with chronic illness were more likely 
to report mental health treatment than healthy parents (Cadman, et al., 1991).  
Treatment for psychological distress may act as a protective factor.  Whether parents 
35
seek help or not, psychological distress can affect all other aspects of life such as 
parenting and relationships with loved ones.  
Parental psychological distress and parenting behaviors.
In a report to Congress on effective parenting, the National Council on 
Family Relations (2003) states that effective parenting such as involvement with the 
child, cognitive stimulation, emotional support, and warmth, is only possible when 
the mother experiences emotional, psychological, and physical well-being.  For 
instance, when a parent experiences emotional or psychological distress, parents may 
provide less emotional support to their children and use more harsh discipline (Elder, 
1974; Richters & Pellegrini, 1989).  Thus, anxiety or depression can affect parenting 
practices.  
The symptomatology associated with psychological distress may prevent 
parents from using positive parenting behaviors.  For instance, fatigue or a decrease 
in energy are symptoms of both depression and generalized anxiety disorder and may 
prevent parents from interacting with their children.  Parents living with 
psychological distress may isolate themselves from the family and the child, thus 
reducing cognitive stimulation, emotional support, and warmth.
Parenting practices have been found to be affected by the presence of 
depression in parents, with higher levels of depressive symptoms directly and 
negatively linked to parenting quality (Jackson, et al., 2000).  Parents living with 
depression may reduce the effort they put into interaction with their children 
(Downey & Coyne, 1990), and be less nurturing and involved with their child 
36
(Conger, et al., 1992; Simons, et al., 1992; Voydanoff & Donnelly, 1998).  Parental 
depression has also been found to be related to lower levels of cognitive stimulation 
and emotional support, and a poorer physical home environment, as evidenced by a 
dark/monotonous interior or unsafe play area (Jackson, et al., 2000; Klebanov, et al., 
1994).  Low levels of cognitive stimulation, emotional support, and poor physical 
home environment have been linked to children?s behavior problems (Jackson, et al., 
2000).
 Parental psychological distress and child behavior problems.
In general, parental psychological distress directly and significantly 
influences children (Jackson, et al., 2000).  Children may become troubled and begin 
to act out when their parents are experiencing psychological distress.  For example, 
school-aged children living with a parent who is experiencing depressive 
symptomatology have been found to show higher levels of behavior problems, both 
externalizing and internalizing, when compared with children whose parents are not 
experiencing depressive symptomatology (Lee & Gotlib, 1989; Richters & 
Pellegrini, 1989).  Similarly, parental psychological distress, particularly in mothers, 
has been found to be related to poor child adjustment in "healthy" children (Downey 
& Coyne, 1990) and children living with a chronic illness (Drotar, 1997).  In families 
where a child is living with either asthma or diabetes, higher levels of parental 
distress are associated with internalizing problems for the children living with the 
illness in comparison to healthy controls (Rogers & Floyd, 2003).  Children's 
37
behavior problems are clearly associated with parental well-being, but are also linked 
to parenting behaviors.
Parenting behaviors
Parenting behaviors are directly linked to child outcomes.  Positive parenting 
behaviors such as cognitive stimulation, emotional support, warmth, and 
involvement have been found to decrease child behavior problems (Conger, et al., 
1992; Gerard & Buehler, 1999).  As discussed previously the relationship between 
parenting behaviors and child behavior has been found to be influenced by additional 
variables such as economic strain and parental psychological distress.  Luckily 
supportive parenting has been found to buffer effects of family adversity, such as 
economic strain, on children (Pettit, Bates, & Dodge, 1997).  
Parenting behaviors and chronic illness
The literature is extremely limited in the area of parenting behaviors and 
chronic illness.  Although there may be additional strains involved in living with a 
chronic illness, parents may use the same parenting behaviors for children living 
with and without a chronic illness.   On the other hand, parents may become 
overprotective of children living with a chronic illness restricting their ability to 
become independent (Barlow, et al., 1998).  Parents may also spend more time with 
the child living with a chronic illness due to the demands of the medical regimen.  In 
total, these behaviors may create a unique environment for the child living with a 
chronic illness.
38
Parental cognitive stimulation.
Parental cognitive stimulation can be defined as the family and 
environmental factors that promote children's learning and understanding of the 
world around them.  Parents may encourage hobbies, reading, or take the child to 
cultural events as a way of providing cognitive stimulation.  Economic strain may 
negatively influence the ability of parents to provide adequate cognitive stimulation 
because of the financial and time requirements of cognitive stimulation
There is a relationship between socioeconomic status and level of parental 
cognitive stimulation with lower income families providing lower levels of cognitive 
stimulation than families of higher incomes.  Low income children tend to read 
infrequently, have less access to books and computers, and parents are less involved 
in their school activities (Evans, 2004).  However, families are resilient and may care 
or provide for their children in resourceful ways.  Thus previous research suggests 
that the cognitive stimulation can mediate the influence of economic strain on 
children (Cosica, Christensen, Henry, Wallston, Radcliffe, & Rutstein, 2001; Linver, 
Brooks-Gunn, & Kohen, 2002; Menaghan & Parcel, 1991; Parcel & Menaghan, 
1990).
Cognitive stimulation has been found to be related to positive outcomes for 
children.  With greater cognitive stimulation, there are fewer behavior problems in 
children (Jackson, et al., 2000), and the stimulation acts as a protective factor against 
future behavior problems (Parcel & Menaghan, 1993).  More specifically, greater 
cognitive stimulation is related to lower levels of childhood aggression and social 
39
withdrawal (Saltaris, Serbin, Stack, Karp, Schwartzman, & Ledingham, 2004).  
Children who are raised in strong home environments characterized by higher levels 
of cognitive stimulation have been found to have fewer behavior problems than 
children not raised in such environments (Rogers, Parcel, & Menaghan, 1991).  
Parental emotional support.
Parental emotional support is the interpersonal interaction (e.g., time spent 
together, verbal communication, and discipline) between family members, 
particularly the parent and the child.  Research has found that when emotional 
support is lacking, children can experience negative outcomes.  McLoyd and Smith 
(2002) recently examined the relationship between emotional support and behavior 
problems in 1,039 Caucasian, 550 African American, and 401 Hispanic children.  
Their findings indicate that children who receive low levels of parental emotional 
support have more behavior problems than children who receive higher levels of 
support.  Conversely, positive outcomes occur when emotional support is present.  
For example, emotional support has been found to increase adherence to medical 
regimens for children living with diabetes (La Greca & Bearman, 2002).
The emotional support that parents provide remains with children into their 
adulthood.  For example, in a nationally representative sample of adults reporting on 
their childhood, researchers found that the lack of parental emotional support during 
childhood was related to higher levels of depressive symptoms in adulthood (Shaw, 
Krause, Chatters, Connell, & Ingersoll-Dayton, 2004).  Clearly, parental emotional 
40
support influences not only childhood outcomes, but can also affect later adult 
functioning. 
     Parental warmth.
Similar to parental cognitive stimulation and emotional support, parental 
warmth influences children's functioning.  Parental warmth towards a child is 
operationalized as the demonstrative interaction, physical affection, and time spent in 
positive discussions with a child (Peterson & Zill, 1986).  The warmth a parent 
provides to a child has been associated with lower levels of both externalizing  
(Eisenberg, Losoya, Guthrie, Murphy, Shepard, Padgett, Fabes, Poulin, & Reiser, 
2001; Moris, Meesters, & van den Berg, 2003; Saltaris, et al., 2004; Yeung, et al., 
2002; Zhou, Eisenberg, Losoya, Fabes, Reiser, Guthrie, Murphy, Cumberland, & 
Shepard, 2002) and internalizing (Minden, 1999; Moris, et al., 2003; Saltaris, et al., 
2004) behavior problems.  Conversely, parenting behaviors that are lacking in 
warmth have been associated with greater behavior problems in children.  For 
example, maternal criticism has been found to be related to externalizing behavior 
problems in both healthy children and children living with a chronic illness (Minden, 
1999).  
Parental warmth appears to have similar influences on children living with a 
chronic illness and children that are not living with a chronic illness.  Low levels of 
maternal warmth have been found to be related to internalizing behavior problems in 
both healthy children and children living with a chronic illness (Minden, 1999).  
41
There has been some support for the buffering of parental warmth on behavior 
problems in children living with recurrent otitis media (Lopez, 1998).
Parental warmth is not only linked to a lower incidence of child behavior 
problems, but is also associated with additional positive outcomes in children.
Parental warmth shown towards a child may help to promote emotional security 
(Eisenberg, et al., 2001) and appropriate emotional expression by the child 
(Eisenberg, et al., 2001; Zhou, et al., 2002).  Parental warmth also may decrease 
levels of depression in adolescents (Gil-Rivas, Greenberger, Chen, & Lopez-Lena, 
2003), and may be related to better school grades and more positive child beliefs 
about their competence (Juang & Silbereisen, 2002). 
     Parental involvement. 
Parental involvement in activities with children is vital to educate and 
socialize the child to the world (Moore, Chalk, Scarpa, & Vandivere, 2002).
Additionally, the parent-child interaction is satisfying to both child and parent alike 
(Moore, et al., 2002).   This satisfying interaction has been found to translate into 
better behavior in children (Mosley & Thomson, 1995; Thomson, Hanson, & 
McLanahan, 1994).
Many national data sets have examined parental involvement and children's 
behavior.  Positive parental involvement, such as private talks and engaging in 
activities together, has been found to be significantly associated with fewer child 
behavior problems in the National Survey of Families and Household data set 
(Amato & Rivera, 1999).  Additionally, data from the National Longitudinal Survey 
42
of Youth indicates that adolescents who have fun with their family (e.g., playing 
games, going to sporting events) have fewer behavior problems (Moore, et al., 2002).
Longitudinal research also points to the positive effects of parental 
involvement.  In a longitudinal multi-site study, mothers (n=585; 80 percent 
Caucasian, 12 percent African American, 2 percent other; predominantly middle 
class) completed measures about their children in kindergarten and again in sixth 
grade (Pettit, et al., 1997).  The results indicate that supportive parenting, which 
includes involvement and warmth, has been found to be predictive of children?s 
behavioral, social, and academic adjustment  (Pettit, et al. 1997).
Research on parental involvement and chronic illness focuses on time spent 
managing the illness, not on parent-child activities (McCubbin, 1988).  Although the 
literature is absent, two alternative hypotheses about parental involvement 
concerning activities may be viable.  One, parents may have less time to participate 
in activities with children due to the time demands of the medical regimen.  Two, 
parents may spend more time involved in parent-child activities to normalize the 
experience of the child living with a chronic illness.  Clearly, further research that 
separates parents' activity involvement and time spent on the chronic illness regimen 
is needed to fill this gap in the literature.    
Social Support 
There are a number of concepts of social support, including perceived social 
support.  Barrera (1986) defined perceived social support as an individual?s cognitive 
appraisal of his or her connection to others.  Thus perceived social support refers to 
43
the perception of available or provided social resources an individual accesses in 
interpersonal relationships in both informal and formal relationships (Cohen, 
Gottlieb, & Underwood, 2001; Cooke, Rossman, McCubbin, & Patterson, 1988).  
Social support has been found to act as a buffer against life stress, especially when 
the social support is operationalized as perceived availability of support (Jackson, 
1992). 
Social support and buffering of stress.
Cohen and Wills (1985) found evidence for the presence of the buffering 
effect of social support if the measure used to assess social support taps the 
individual's perceived availability of resources that are needed to cope with a 
stressful event.  Thus if perceived social support is available and regarded as helpful 
to an individual, the support can buffer or moderate the negative effects of the 
stressor.  The social support buffering process may prevent individuals from 
perceiving a situation as stressful or may deter individuals from reacting in a 
maladaptive fashion (Stets, 1991).  
Examples of the buffering effect of social support have been found in 
previous research.  Higher levels of social support have been shown to act as a 
buffer/moderator against depression for older adults (Hyers, 1995) and in individuals 
faced with severe illness-related disabilities, such as blindness resulting from 
complications of diabetes (Littlefield, Rodin, Murray, & Craven, 1990).  Research 
also indicates that social support networks involving friends, neighbors, and 
extended family directly and indirectly influence parenting and act as a 
44
buffer/moderator against parental psychological distress (Belsky, 1984; Belsky & 
Vondra, 1989).  
Conversely, social support has not always been found to buffer against the 
influence of stressful life events.  For example, formal and informal supports did not 
buffer the negative effects of economic strain, single-parent families, long work 
hours, and neighborhood problems on parental well-being for 929 nationally 
representative families who participated in the National Commission on Children 
1990 survey (Voydanoff & Donnelly, 1998).  The failure of social support to buffer 
the stress may be due to the type of stressor and support received.  Emotional support 
from friends will not decrease long work hours or improve neighborhood problems.  
Likewise, instrumental support provided by others may not be enough to decrease 
severe economic strains.     
Social support and chronic illness.
Whether social support buffers stress or not, research has shown that social 
support is a resiliency factor for children, adults, and families living with and without 
a chronic illness (Garwick & Millar, 1996; Jacelon, 1997; Mandleco & Peery, 2000; 
McCubbin & McCubbin, 1996; Tak & McCubbin, 2002).  For example, social 
support is a commonly used coping strategy for parents living with a child with 
hemophilia (Miller, Sabin, Goldman, Clemente, Sadowski, Taylor, & Lee, 2000).  
Parents want and need support for dealing with the uncertainty of the condition's 
trajectory (Asmussen, et al., 1999; Swallow & Jacoby, 2001).  Therefore, living 
without social support may prevent the family from coping successfully with the 
45
illness.  The social support for families living with a chronic illness may come from 
within the family system through spousal support or from the extended family.  
Medical professionals, friends, and the community can also provide social support 
for such families.
In general spousal support can help families get through tough times.  In 
families living with a chronic illness, spousal support has been found to be very 
important (Barbarin, Hughes & Chesler, 1985).  For example, spousal social support 
has been found to moderate the influence of economic strain on the positive 
parenting of mothers (Simons, et al., 1992).  However, some mothers have reported 
that the social support provided by grandparents, friends, and families living with a 
similar condition are more helpful than father support (Tannila, Jarvelin, & 
Kokkonen, 1999).
Social support from extended family has been found to be helpful for both 
parents and children.  Support provided by grandparents of children living with 
chronic conditions has been found to be helpful to all in the family (Schilmoeller & 
Baranowski, 2003).  Parents living with a child with blood disorders (sickle cell 
anemia or thalassaemia) have reported the importance of various dimensions of 
social support -- practical, emotional, and financial -- from family members in 
helping them to deal with the challenges created by the chronic illness (Atkin & 
Ahmad, 2000).  In research on childhood cancer, social support from extended 
family included sibling care, transportation, parent respite, and emotional supports 
(McCubbin, Balling, Possin, Frierdich & Bryne, 2002).  These supports have helped 
46
the parents to cope with the care needed for their children (Atkin & Ahmad, 2000). 
The support that families experience from others may moderate the relationship 
between illness presence and other individual processes such as challenges created 
by child health stress or parental psychological distress.  
Medical professionals may also provide social support to families and 
individuals.  Parents want and need to be educated about the illnesses their children 
are diagnosed with (Asmussen, et al., 1999).  The information doctors provide is 
often a form of instrumental support.  Patterson (1991) found that creating a 
collaborative relationship, a form of support, with medical professionals acts as a 
resiliency factor.  For example, medical professionals have helped parents of 
children with sickle cell disorder or thalassaemia to reduce stress and improve 
coping by offering the parents? information, financial help, and emotional support 
(Atkin & Ahmad, 2000). Similarly, oncology teams that provided families with 
support in the form of reassurance, hope, information, and assistance improved 
family adaptation and coping (McCubbin, et al., 2002).  The support that comes from 
the medical profession may help to buffer or moderate the influence of child health 
stresses on parental psychological distress or child behavior problems.
Friends have been found to provide support to families living with a chronic 
illness.  Of particular assistance to families is the support provided by other families 
living with the same condition (Tannila, et al., 1999).  Unfortunately, some parents 
find that friends not living with a child with an illness are not understanding of the 
challenges that are created due to the illness (Asmussen, et al., 1999).  Perceived 
47
social support that is provided by friends may help to moderate the stresses created 
by the illness.
The workplace and community in which the family lives may also be sources 
of social support.  Fellow employees, employers, and employment policies may be 
forms of support.  In families with children living with cancer, parents found flexible 
schedules, time off, and job assurance to be supportive factors (McCubbin, et al., 
2002).  Communities can also provide support to families living with a chronic 
illness through resources such as schools, programs for children, and health care 
centers (Mandleco & Peery, 2000).  In addition, members of the community may 
provide childcare, financial aid, assistance in home maintenance, and emotional 
support (McCubbin, et al., 2002).  The presence of these supports may buffer 
institutional stresses that can be created by the presence of illness (e.g., daycare 
centers that do not accept children who are living with a chronic illness).  
Research has also examined the differences between families living with and 
without a chronic illness in relation to social support levels.  The research is mixed, 
with some finding similar levels of support in both controls and families living with 
a child with a chronic illness (Cadman, et al., 1991; Gerhardt, et al., 2003), and other 
research finding significant differences between the social support levels of the two 
groups (Ellerton, et al., 1996; Holroyd & Guthrie, 1986).  Sheer numbers may not 
tell the whole story; it is possible that families living with a chronic illness may have 
a smaller, but more effective support group for their needs.  The presence of supports 
48
for families living with a chronic illness may help to buffer the possible negative 
influences of the presence of illness on family and individual processes.    
Social support and economic strain.
Families living with economic strain and low incomes have been found to 
have less perceived social support than families with higher incomes (Klebanov, et 
al., 1994).  This relationship is unfortunate because social support has been found to 
have a beneficial impact on parenting for economically disadvantaged mothers 
(Ceballo & McLoyd, 2002).  Simons et al. (1993) found that support from a spouse 
buffers or moderates the relationship between economic strain and depression, and 
the relationship between depression and parenting behaviors. 
Social support and parental psychological distress.
Research has indicated a relationship between social support and parental 
psychological distress.  Lower levels of support are related to higher levels of 
depressive symptomatology (Dean, Lin, & Ensel, 1981; Reinherz, Giaconia, 
Wasserman, Silverman, & Burton, 1999; Turner, 1981; Zimmerman, Ramirez-
Valles, Zapert, & Maton, 2000).   Parents who are satisfied with their current level of 
support have higher levels of well-being (King, King, Rosenbaum, & Goffin, 1999).  
Thus, support may moderate the relationship between stress and well-being with 
higher levels of support creating a buffer against the negative influence of stress on 
well-being.  
49
Social support and parenting behaviors.
The presence of social support has also been found to positively influence 
parenting behaviors.  Emotional forms of social support have been found to be 
related to nurturing behaviors in parent-child relationships, whereas instrumental 
social support can lead to less punitive parenting (Ceballo & McLoyd, 2002).  
Specifically, spousal support is an important influence on positive parenting 
behaviors (Simons, et al., 1993; Simons, et al., 1992).  Higher levels of social 
support may buffer external stresses that would decrease positive parenting 
behaviors.  When lower levels of social support are present, the external stressors 
may decrease positive parenting behaviors.  Positive parenting behaviors have been 
linked to a decrease in child behavior problems and thus social support may also 
indirectly promote positive functioning in children through the improvement of 
parental well-being (Ceballo & McLoyd, 2002).  
Social support, chronic illness, and children's adjustment.
Parents may receive the majority of support, but children have also been 
found to benefit from social support.  Social support for children living with a 
chronic illness has been received by friends both prior to and post diagnosis 
(Tannila, et al., 1999).  The support that is received has been found to lead to better 
adjustment (Amer, 1999; Wallander & Varni, 1989) and fewer behavior problems if 
the children have social support from both family and peers (Wallander & Varni, 
1989).  Social support may buffer or moderate the stresses that are created by the 
presence of illness in children leading to better adjustment and fewer behavior 
50
problems.  Unfortunately, some research indicates that children living with a chronic 
illness have smaller support and peer networks than healthy children (Ellerton, et al., 
1996).  Nevertheless, any support provided to the child and family may assist with 
adaptation and buffering or moderating stress.  
Definitions
The following are definitions of concepts used in the current research.  
Chronic illness: "includes those diseases involving a protracted course that 
may be fatal or result in compromised mental, cognitive and/or physical functioning 
and that are often characterized by acute complications, which may result in 
hospitalizations or other forms of intensive treatment? (Brown & DuPaul, 1999, p. 
175).
Economic strain: ?refers to the subjective evaluation of one?s financial 
situation? (Voydanoff & Donnelly, 1998, p. 346) and may be marked by a lack of 
financial resources such as not having enough money at the end of the month, 
postponing family purchases, and not having health insurance.
Health stress: refers to the child's activity limitations, school absences, school 
work difficulties created by a physical or mental limitation and/or a general poor 
rating of health by the child's parent.
Parental psychological distress: refers the presence of depression and/or 
anxiety.
51
Parental cognitive stimulation: refers to the parental, family and 
environmental factors (e.g., involvement) that promote children's learning and 
understanding of the world around them as well as greater achievement in school.   
Parental emotional support: refers to the interpersonal interaction (e.g., 
discipline, time spent together, and verbal responses) between family members, 
particularly the parent, and the child.  
Parental warmth: is the parent's demonstrative interaction, physical affection, 
and time spent in positive discussions with a child (Peterson & Zill, 1986). 
Parental involvement: are the behavioral parent-child interactions both within 
and outside the home (e.g., household chores, sports, shopping).  
Child behavior problems: are behaviors (e.g., impulsivity, disobedience, and 
cheating) or emotional expressions (e.g., anxious, tense, and feels worthless) that are 
disruptive to the child and/or family. 
Social support: refers to the perception of available or provided social 
resources an individual can access in both informal and formal interpersonal 
relationships (Cohen, et al., 2001; Cooke, et al., 1988).  
Hypotheses 
Based on the previous research, 13 hypotheses were created to describe the 
predicted relationships between the presence of a chronic illness, economic strain, 
child health stress, parental psychological distress, parental cognitive stimulation, 
parental emotional support, parental warmth, parental involvement, child behavior 
problems, and social support.
52
1. The presence of a chronic illness has a positive influence on economic strain.
2. The presence of a chronic illness has a positive influence on child's health 
stress.
3. The presence of a chronic illness has a direct and positive effect on parental 
psychological distress.
4. The presence of a chronic illness has a direct and positive effect on child 
behavior problems.  
5. Higher levels of economic strain increase parental psychological distress.
6. Higher levels of child health stress increase parental psychological distress.
7. Economic strain leads to a decrease in parental cognitive stimulation.
8. Economic strain leads to a decrease in parental involvement.
9. Higher levels of child health stress lead to an increase in child behavior 
problems.
10. Higher levels of parental psychological distress negatively influence all 
parenting behaviors: parental cognitive stimulation, parental emotional 
support, parental warmth, and parental involvement.
11. Higher levels of parental psychological distress increase child behavior 
problems.
12. Higher levels of parental involvement, warmth, emotional support, and 
cognitive stimulation lead to a decrease in child behavior problems.
13. Mid to high levels of social support moderates each of the following 
relations: illness to economic stress, child health stress, parental distress, and 
53
child behavior problems; economic strain to parental well-being, parental 
cognitive stimulation, and parental involvement; child health stress to 
parental psychological distress and child behavior problems; parental 
psychological distress to parental cognitive stimulation, parental emotional 
support, parental warmth, parental involvement, and child behavior problems; 
parental cognitive stimulation to child behavior problems; parental emotional 
support to child behavior problems; parental warmth to child behavior 
problems; and parental involvement to child behavior problems.    
54
Chapter III: Method
Participants
The Panel Study of Income Dynamics (PSID) Child Development 
Supplement is the data source for the current research.  The PSID is conducted by 
the University of Michigan, Survey Research Center, Institute for Social Research 
and has been funded over the years by government agencies, foundations, and 
organizations (Hofferth, Davis-Kean, Davis, & Finkelstein, 1997). The original PSID 
sample was created from the combination of the Survey Research Center and Survey 
of Economic Opportunity participants.  The Survey Research Center sample used 
equal probability sampling of households from the 48 contiguous states and had 
3,000 participants.  The Bureau of the Census for the Office of Economic 
Opportunity conducted the Survey of Economic Opportunity from which the PSID 
included approximately 2,000 low income, primarily African-American families 
with heads of household under the age of 60.  
In 1968 the PSID was begun with a representative sample of 5,000 families 
to obtain information about family economics and structure; marriage, divorce, birth, 
and death rates; and health (Hofferth, et al., 1997).  By 1996 the sample had grown 
to 8,700 families due to the expansion of the initial sample and the subsequent 
families that formed from aging children.  In 1997, as a cost saving measure the 
PSID reduced this core sample by eliminating participants from the Survey of 
Economic Opportunity sample by two-thirds.  However, due to the introduction of 
additional funding, 609 African-American families from the Survey of Economic 
55
Opportunity were able to be retained (Hofferth, et al., 1997).  The refreshed sample 
also included a new sample of post 1968 immigrant families (n=441).  The addition 
of the immigrant sample and the reduction of the original core sample created a 
sample that is representative of the United States population.  
The PSID survey was conducted yearly until 1997 when it began to be 
collected biennially (Hofferth, et al., 1997).  In 1997, the PSID included the Child 
Development Supplement, which expanded the survey to include questions aimed at 
understanding the growth and development of children (Hofferth, 1998). Families 
with children below the age of 13 were asked to participate (Hofferth, 1998).  There 
were 2,705 families eligible for participation; of that group, 2,394 families that 
included 3,563 children completed the supplement (Institute for Social Research, 
2002).  The response rate for the supplement was 88.2 percent (Institute for Social 
Research, 2002).   
The sample for this research was limited to families with children six years 
and older, thus excluding 1,610 children from the initial sample.  Of the remaining 
1,953 children, 16 were eliminated from analysis because their primary caregivers 
stated that doctors had reported that their child had either autism (n = 6) or 
retardation (n = 10).  Both autism and retardation were not included in the sample 
because of the additional caregiving requirements necessary when a child is living 
with cognitive difficulties.  
Further sibling reductions of the sample were also conducted.  The sample 
consists of 438 sibling pairs.  One of the siblings was eliminated to avoid double 
56
counting.  Three sibling pair types were present, two children not living with a 
chronic illness (n=214), one child with and one child living without a chronic illness 
(n=141), and two children living with a chronic illness (n=83).  Siblings in 
homogeneous pairs were alternately sampled (e.g., youngest then oldest and so on).  
Sibling pairs that had one child living with and one child living without an illness did 
not follow the previous sampling methods.  Instead the child with the chronic illness 
was retained in order to maintain an adequate chronic illness sample size.
After deletions, the final sample that was analyzed consisted of 1,500 
children and is a representative sample of children living in the United States in 
1997, ages six to twelve, who do not have autism or retardation.  However when the 
normalized weight was applied to the sample a case was deleted from analysis.  This 
is most likely a result of rounding error.  If the weight was taken out to additional 
decimal places, making the weight more exact, the case might not have been deleted.  
Nonetheless the final sample included 1,499 children.
The entire sample (n=1,499) was split into two groups, families that report a 
chronic illness (asthma, anemia, diabetes, and otitis media; n= 693) and those that do 
not (n=806).  Previous research has combined similar illnesses into one group for 
analysis (Dadds, Stein, & Silver, 1995).  In the current research the presence of 
chronic illness is indicated by the reports of primary caregivers to the question; ?Has 
your doctor or health professional ever said that (CHILD) had?asthma, diabetes, 
more than three ear infections in a year, or anemia or iron deficiency??  If the 
caregiver answered affirmatively to any of the illnesses, the family was included in 
57
the chronic illness group.  These physical health questions have been used previously 
in the National Longitudinal Survey of Children and Youth, National Longitudinal 
Survey of Youth, and National Health Interview Survey (Hofferth, et al., 1997).  
In the PSID Child Development Supplement there are a total of 693 children 
that are included in the chronic illness sample.  Of those children, 120 are living with 
anemia (8 percent of the total sample), 157 are living with asthma (10.5 percent), 3 
are living with diabetes (0.2 percent), and 522 are living with chronic otitis media 
(34.9 percent). 
In the current sample comorbidity of chronic illnesses is present.  Of the 693 
children, 588 reported only one chronic illness (39.2 percent of total sample and 84.7 
percent of chronic sample), 102 reported two chronic illnesses (6.8 percent of total 
sample and 14.7 percent of chronic sample), and 4 reported three chronic illnesses 
(0.3 percent of total sample and 0.6 percent of chronic sample).  Chronic otitis media 
has the greatest comorbidity with 106 children living with chronic otitis media and at 
least one other illness.  
Table 1
Sample Rate of Chronic Illness 
Chronic illness Number of 
children
Percentage of chronic 
illness sample (n=693)
Percentage of total 
sample   (N=1,499)
Anemia 120 17.3 8.0
Asthma 157 22.6 10.5
Diabetes 3 0.4 0.2
Otitis Media 522 75.2 34.9
58
National prevalence rates differ only slightly from the rates found in the 
current data. Between 1999 and 2000, approximately four percent of children 
between the ages of 6 and 11 were living with anemia (Centers for Disease Control 
and Prevention, 2002).  Currently there are nine million children (12 percent), under 
the age of 18, who have been diagnosed with asthma (Blackwell, Vickerie & 
Wondima, 2003).  Juvenile onset diabetes affects 151,000 children (0.19 percent) 
under the age of 20 or about one in every 400 to 500 children (National Institutes of 
Diabetes & Digestive & Kidney Disorders, 2003). Ear infections affect 24.6 percent 
of children under the age of 18 (Coiro, Zill & Bloom, 1994).  The current sample has 
more children with chronic ear infections than the national percentage.  This may be 
due to the age range included in the national sample.  Adolescent children may be 
less likely to have ear infections than younger school age children.
Asthma, anemia, diabetes, and chronic otitis media are the illnesses that are 
examined based on shared attributes.  First, each of the illnesses can manifest during 
childhood.  Second, each of the illnesses is invisible to individuals outside of the 
family.  Thus there are no stigmatizing changes in appearance, providing the family 
and child with discretion regarding whom to inform about the presence of the illness.  
Third, each of the illnesses can have acute episodes followed by chronic periods that 
require regular monitoring.  Fourth, medication is required to treat each of the 
illnesses.  Fifth, each of the illnesses can cause permanent injury to the child if not
monitored appropriately.  
59
Procedure
Families first completed the traditional PSID measures.  From these measures 
researchers determined the eligibility of family participation for the Child 
Development Supplement based on the ages of the children.  Eligible families were 
told by the PSID interviewer that they would be contacted in the future about the 
Child Development Supplement.  Families with one or more children age 12 and 
younger were eligible to participate.  If the family had more than two children within 
the desired age range, the central office of the PSID randomly selected two children 
for the measures to be given, and determined the likely primary caregiver.  Children 
living with or without a chronic illness all had equal chances of being included in the 
total PSID-CDS sample.
Field researchers after family's eligibility was determined and the Child 
Development Supplement was explained, the researchers identified the primary 
caregiver, scheduled a time for the interview, and mailed an introductory letter (see 
Appendix A).  The researcher met the primary caregiver at the household to obtain 
written permission for the child interview, assess the child using standardized 
achievement tests and, if old enough (eight years and older), to complete child 
interviews.  Also while in the home, the researcher conducted the primary caregiver 
child interview and assessment, collected the time use diaries, and provided 
additional self-administered instruments for the primary and other caregivers to 
complete.  The interviewer also completed observations of the primary caregivers 
interaction with the child and the home environment for the HOME measure.  After 
60
completion of the interview, primary caregivers were given a small token of 
appreciation for their participation and the children were given a small gift.  
Completed interviews and self-administered questionnaires were mailed to the 
Survey Research Center, Institute for Social Research central office for coding and 
entered into a database by the SRC Survey Services Lab.
The questionnaires used in the PSID Child Development Supplement were 
developed over 11 months and pretested twice (Institute for Social Research, 2002).  
The first pretest was conducted by the Survey Research Center in June 1996 for the 
household questionnaire.  The second pretest in October 1996 was a rehearsal of the 
entire PSID and Child Development Supplement, including all interviews and 
mailings.  In January 1997, the pretests were concluded and the data collection 
begun.
Instruments
The instruments used in the current research come from two questionnaires 
that were given to participants in the Child Development Supplement (Hofferth, et 
al., 1997).  The first questionnaire is the Primary Caregiver Child Questionnaire that 
was completed by the primary caregiver either with the interviewer face to face or 
over the phone.  The second is the Primary Caregiver Household Questionnaire that 
was also completed by the primary caregiver, but was self-administered in the home.  
However, if literacy or language problems were present, the interviewer 
administered the questionnaire.
61
Economic strain.
The economic strain index (see Appendix B) focuses on subjective financial 
stress and how the individual or family responds to those financial pressures (Conger 
& Elder, 1994).  Participants were asked to indicate the financial stresses they 
experienced within the last year.  The measure has 15 different responses to financial 
stress ranging from ?postponed major purchases? to ?sent one or more of your 
children to live with someone else.?  Individuals not experiencing financial stresses 
indicated "none of the above."  
Two additional questions that were constructed by the Institute for Social 
Research (2002) were added to the economic strain index by the author.  The first 
additional question assesses the money families have at the end of the month.  The 
questions asks, "At the end of the month, do you end up with some money left over, 
just enough to make ends meet, or not enough money to make ends meet?"  
Participants may respond that they have either some money left over, just enough to 
make ends meet, or not enough to make ends meet.  The responses were 
dichotomized to reflect the yes/no response set of the economic strain measure.  
Therefore, the responses "either some money left over or just enough to make ends 
meet" were coded no or zero, and "not enough to make ends meet" was coded yes or 
one indicating economic strain.  This question was added to the economic strain 
index because it may capture families that are beginning to experience economic 
strain.  To ensure that the question is related to the economic strain index, a 
62
Pearson?s Product-Moment correlation was conducted revealing a statistically 
significant relationship (r=.46, p<.001). 
The second question added to the economic strain index assesses the presence of 
health insurance.  Primary caregivers answered the question, ?How many of the past 
12 months was (CHILD) covered by health insurance or any other kind of health 
care plan?  Include health insurance obtained through employment or purchased 
directly as well as government programs like Medicare and Medicaid.?  If a child 
was not covered by health insurance during any of the past 12 months a score of one 
was assigned, indicating the presence of economic strain.  If a child was covered by 
insurance for the entire previous year a score of zero was given.  This question was 
added to tap the economic strain families might experience if their child did not have 
health insurance.  To insure that the health insurance question is related to the 
economic strain index a Pearson?s Product-Moment correlation was conducted 
revealing a statistically significant relationship (r=.17, p<.001). 
Participants will have a total score ranging between 0 and 17, with higher scores 
indicating a higher level of economic strain.  The index has a high level of face 
validity and has a Cronbach alpha of 0.70 with the additional questions.  If the 
"money at the end of the month" question is removed, the Cronbach alpha falls to 
0.66; and if the health insurance question is removed the Cronbach alpha falls to 
0.69.  Cronbach alphas were also conducted for children living with a chronic illness 
(0.70) and children not living with an illness (0.69).  
63
No previous research has used the economic strain index in its entirety.  
However, Conger and associates used a modified economic strain index and included 
two additional questions referring to money left over at the end of the month and 
difficulty in paying bills (Conger, Elder, Lorenz, Conger, Simons, Whitbeck, Huck, 
& Melby, 1990).  The modified index in that research yielded a Cronbach alpha of 
.84 (Conger, et al., 1990).  
     Child health stress.
The child health stress index was created for the current research project 
using questions from the Child Development Supplement (see Appendix C).  There 
are four items that make up the index.  The first question asks the parent to rate the 
child?s health as excellent, very good, good, fair, or poor.  Children that are rated by 
their parents to be in fair or poor health may be an indicator of health stresses.  The 
majority of children (97.4 percent) were rated as having either good, very good, or 
excellent health.  Therefore to separate the most extreme scores the question was 
dichotomized with excellent, very good, and good given a score of zero or no health 
stress and children rated as fair or poor given a score of one indicating the possibility 
of health stress.
The next three questions examine activity limitations.  Previous research has 
used activity limitations during the last 12 months and school days missed as a 
measure of the illness condition (Chernoff, List, DeVet, & Ireys, 2001) and as a 
measure of response to chronic pain in sickle cell disease (Gil, Carson, Porter, 
Ready, Valrie, Redding-Lallinger, & Daeschner, 2003).  The three questions used the 
64
same stem; does your child ?currently have any physical or mental conditions that 
would limit or prevent the child?s ability to??  The primary caregiver then answers 
the three questions (1) ?do usual childhood activities such as play, or participate in 
games or sports,? (2) ?attend school (preschool or day care) regularly,? and (3) ?do 
regular school work.?  The caregiver was asked to respond to each question with 
either a "yes" or a "no."  Yes responses were given a score of one indicating health 
stress. 
For the child health stress index each item was dichotomized, yes or no, with 
yes answers indicating a stressful health circumstance and given a score of one.  
"No" responses are given a score of zero or no health stress.  The responses to the 
index are summed, creating a possible range of index scores from zero to four.  
Higher scores indicate a higher level of child health stress.  The Cronbach alpha for 
the index is 0.59.  Additional analysis revealed that the Cronbach alpha for children 
not living with a chronic illness was 0.74 and for children living with a chronic 
illness was 0.50.  
Parental psychological distress.
Non-specific psychological distress was measured using a ten-item scale 
referred to as the K10 (see Appendix D) that was developed for the National Health 
Interview Survey (Kessler, Andrews, Colpe, Hiripi, Mroczek, Normand, Walters, & 
Zaslavsky, 2002).  During the creation of the K10, clinical and community pilot tests 
were conducted to screen possible questions for inclusion (Kessler, et al., 2002).  
65
Item response theory was used to select the 10 items that are included in the K10 
(Kessler, et al., 2002).  
The K10 assesses various symptoms of non-specific psychological distress 
over the past month using a response scale of all of the time (one) to none of the time 
(five).  Items include "during the past 30 days, how often did you feel tired out for no 
good reason" and "how often did you feel so sad nothing could cheer you up?"  Each 
of the ten questions is reverse coded so that higher item scores indicate more non-
specific psychological distress.  Scores can range from 10 to 50, and a higher total 
score indicates greater non-specific psychological distress.  
Initial cut-off scores for the K10 have been developed by the Clinical 
Research Unit for Anxiety and Depression, School of Psychiatry, University of New 
South Wales (Department of Human Services Center for Population Studies in 
Epidemiology, 2002).  Scores of 10 to 15 indicate low or no risk, 16 to 29 indicate a 
medium risk, and 30 to 50 indicate a high risk of psychological distress (Department 
of Human Services Center for Population Studies in Epidemiology, 2002).  
However, additional research is needed to confirm these cut-off scores.
The K10 has been found to be a valid and reliable measure of psychological 
distress (Department of Human Services Center for Population Studies in 
Epidemiology, 2002; Furukawa, Kessler, Slade, & Andrews, 2003; Kessler, et al., 
2002; Kessler, Barker, Colpe, Epstein, Gfroerer, Hiripi, Howes, Normand, 
Manderscheid, Walters, & Zaslavsky, 2003a).  The measure is able to detect DSM-
IV depressive and anxiety disorders in individuals (Furukawa, et al., 2003) and has 
66
consistent severity values across sociodemographic subsamples (Kessler, et al., 
2002).  The K10 has also been found to have excellent internal consistency for a 
nationally representative telephone sample (Cronbach alpha = .92, Kessler, et al. 
2002) and a community interview sample (Cronbach alpha = .93, Kessler, et al. 
2003a).  For the current sample the Cronbach alpha is 0.91.  The index's reliability 
coefficient for children living without a chronic illness was 0.92 and 0.89 for 
children living with a chronic illness.  
     Parental cognitive stimulation and parental emotional support.
The Home Observation for Measurement of the Environment (HOME, see 
Appendix E and F) scale has been used frequently in research to assess the 
relationship between parent and child, and the resources that are available to the 
child for play and learning (Caldwell & Bradley, 1984).  The scale assesses both 
parental cognitive stimulation and emotional support that is provided to children.
The PSID Child Development Supplement used a shortened form of the 
HOME to permit comparison with the National Longitudinal Survey of Youth.   The 
modified HOME inventory items were selected by Bradley and based on reliability 
and validity coefficients, discrimination indices, and factor loadings from previous 
research (Baker & Mott, 1989).  The shortened form has been found to capture the 
breadth of the original larger set of questions (Parcel & Menaghan, 1989) and to 
have a stable factor structure across time (Menaghan & Parcel, 1995).    
The measure collects both the primary caregivers' report of available 
activities to the child and the child?s contact with others, and the interviewer's 
67
observations of the home and the mother-child interaction.  The questions in the 
modified HOME differ slightly based on the child?s age in order to tap children's 
stages of development.  Therefore, there are separate forms for children aged 6 
through 9 and 10 and older.  However, the HOME items and scales are generally 
comparable across ages (Baker & Mott, 1989).   
In the original and modified HOME scale there are three items that discuss 
contact with a father or father figure.  The items state, "about how often does 
(CHILD) spend time with (his/her) father;" "about how often does (CHILD) spend
time with (his/her) father in outdoor activities;" and "how often does (CHILD eat a 
meal with both mother and father?"  These items are excluded from analysis to avoid 
possible family composition bias.  
The two subscales of the modified HOME, parental cognitive stimulation and 
emotional support are analyzed separately in this research.   The parental cognitive 
stimulation subscale taps the family and environmental factors that may contribute to 
the cognitive stimulation of the child.  There are 14 questions in the scale for 6 
through 9 year olds and 13 questions in the 10 and older scale.  Examples of items 
include "how often do you read to your child," and "does your family encourage your 
child to start and keep doing hobbies?"  Also, there are questions that the interviewer 
answers about the home environment such as, "is the interior of the home dark or 
perceptually monotonous," and "is the child's play environment safe with no 
potentially dangerous health or structural hazards within the child's range." 
68
The emotional support subscale examines the interpersonal interaction 
between the family members, particularly the parent and the child.  Items include 
"most children get so angry at their parents that they say things like 'I hate you' or 
swear in a tantrum.  Please look at this list and tell me which actions you would take 
if this happened;" and "about how often does our whole family get together with 
friends or relatives."  Interviewers also answered questions pertaining to the primary 
caregiver's voice quality and whether it conveyed positive feeling toward the child, 
and if the caregiver spontaneously spoke or conversed with the child.  There are 9 
questions in the scale for 6 through 9 year olds and 10 questions in the 10 and older 
scale.  The modified HOME is scored by translating each of the items into 
dichotomous, zero-one variables.  The items are then summed.  
Scores for the cognitive stimulation subscale range between 0 and 14 for 
children between the ages of 6 and 9, and 0 and 13 for children 10 and older.  Higher 
scores indicate a higher level of cognitive stimulation provided to the child.  Scores 
on the emotional support scale range between 0 and 9 (6 through 9 year olds) or 10 
(10 and older), with higher scores indicating more emotional support.  
The HOME has been used previously in many national research studies 
(Miller & Davis, 1997), and has been used successfully with different ethnic groups 
within the United States including Caucasians, African Americans, and Mexican 
Americans (Bradley, et al., 1989).  The internal consistency reliabilities for the 
global measure of the modified HOME in the NLSY data set were 0.69 in 1986 and 
0.72 in 1988 (Menaghan & Parcel, 1995).  
69
Internal consistency reliability for the cognitive stimulation subscale has been 
found to range between 0.50 and 0.72 for children three and older (Baker, Keck, 
Mott & Quinlan, 1993; Baker & Mott, 1989).  The NLSY data reveal internal 
consistency reliabilities for children 6 to 9 years old as 0.67 and for children 10 and 
older, 0.62 (Baker, et al., 1993).  For the current total sample the Cronbach alpha for 
children 6 through 9 is 0.58 and 0.56 for children 10 and older.  Children living with 
a chronic illness aged six through nine had a Cronbach alpha of 0.60, and children 
living without a chronic illness had a Cronbach alpha of 0.64.  Children aged 10 and 
older had a 0.47 Cronbach alpha for children living with a chronic illness, and a 0.66 
Cronbach alpha for children not living with a chronic illness.         
The internal consistency reliability for the emotional support subscale, for 
children six and older, has been reported to range between 0.35 and 0.70 (Baker, et 
al., 1993; Baker & Mott, 1989).  The NLSY data indicate internal consistency 
reliabilities for children 6 to 9 years old as 0.61 and for children 10 and older, 0.58 
(Baker, et al., 1993).  In the current total sample the Cronbach alpha for children six 
through nine is 0.56 and 0.53 for children 10 and older.  Children living with a 
chronic illness aged six through nine had a Cronbach alpha of 0.66, and children 
living without a chronic illness had a Cronbach alpha of 0.55.  Children aged 10 and 
older had a 0.65 Cronbach alpha for children living with a chronic illness, and a 0.76 
Cronbach alpha for children not living with a chronic illness.        
70
     Parental warmth.
The parental warmth measure (see Appendix G) was created for the National 
Survey of Children (Peterson & Zill, 1986) and then modified for the PSID-CDS 
(Hofferth, et al., 1997).  The scale has six items that assess the relationship warmth 
between the caregiver and child.  Participants are asked how often they have 
completed some activity in the past month, such as "told your child you loved them" 
or hugged and shown affection, "either every day (five), several times a week (four), 
about once a month (three), one or two times in the past month (two), or not in the 
past month (one)."  Total scores can range between 6 and 30, with higher total scores 
indicating greater relationship warmth.  The Cronbach alpha for parental warmth 
measure in the current sample is 0.79.  The internal reliability for children living 
with a chronic illness was 0.77, and 0.81 for children not living with a chronic 
illness.
Parental involvement.
The household activities measure (see Appendix H) was used to assess 
parental involvement (Sweet, Bumpass & Call, 1988).  The measure consists of 13 
items that assess parent-child interaction both within and outside the home within the 
past month.  The items include questions that focus on both interactions that do and 
do not require additional toys or equipment.  For instance, questions ask have you 
"washed or folded clothes together" or "played sports or did outdoor activities 
together?"  Primary caregivers may respond to each of the 13 items with one of five 
responses: ?did not do it in the past month with (CHILD)(one), did it one or two 
71
times in the past month (two), about once a week (three), several times a week 
(four), or every day (five).? The total scores for the household task measure range 
from 13 to 65, with higher scores indicating higher parent-child interaction.  The 
Cronbach alpha for the current total sample is 0.76.  The chronic illness subgroup 
has a Cronbach alpha of 0.76 and the nonchronic illness subgroup has a Cronbach 
alpha of 0.75.   
     Child behavior problems.
The behavior problems index (Peterson & Zill, 1986) measures the 
occurrence and severity of child behavior problems when the child is over the age of 
four (see Appendix I).  The index was adapted fr om the longer Achenbach Behavior 
Problems Checklist (Achenbach & Edelbrock, 1981).  Peterson and Zill (1986) had 
three criteria for the items that were pulled from the original Achenbach checklist, 
(1) reliability, (2) high factor loadings on a subscale, and (3) the flexibility of 
interview sites.  Four additional items were added to the Peterson and Zill scale 
(1986) by the National Longitudinal Survey of Youth staff to gain additional 
information for the withdrawn, internalizing behavior subscale (Hofferth, et al., 
1997).  
The behavior problems index asked primary caregivers to decide whether 30 
different behaviors or emotional expressions were often true (one), sometimes true 
(two), or not true about their child (three).  Items include, "(he/she) is restless or 
overly active, cannot sit still" and "(he/she) is withdrawn, does not get involved with 
others."  The items are reverse coded so higher numbers coincide with greater 
72
numbers of behavior problems.  The total score for the index can range between 30 
and 90, with higher scores indicating higher levels of behavior problems.  
The 26 items included in the original Peterson and Zill (1986) index have 
been found to have good test-retest reliability and discriminant validity (Rogers, 
Parcel, & Menaghan, 1991).  These items have also have been found to have a high 
internal consistency with Cronbach alphas ranging between .88 and .90  (Rogers, et 
al., 1991; Parcel & Menaghan, 1993).  The Cronbach alpha for the total 30 item 
index based on the current total sample is 0.90.  The chronic illness subgroup has a 
Cronbach alpha of 0.91 and the nonchronic illness subgroup has a Cronbach alpha of 
0.88.
     Social support.
Items for the social support index (see Appendix J) were taken from the 
National Survey on Families and Households and from the 1980 Panel Study of 
Income Dynamics (Hofferth, et al., 1997).  The index taps both instrumental (e.g., 
transportation, child care, help around the house) and emotional support.  Two 
contingency questions ask participants if they have received instrumental or 
emotional support in the past month.  Participants who respond affirmatively are then 
asked to choose five people from whom they have received help.  Individuals who 
have had support from more than five people in the previous month are asked to 
choose those that have helped them the most.  Participants may choose from a list of 
13 helpers, ranging from relatives (own mother, father; mother-in-law, father-in-law, 
73
grandparents, sister, brother, aunt, uncle, cousin) to friends, clergy, and co-workers. 
The "other" category was also included.  
An individual social support score can range between 0 and 10.  A higher 
index score indicates higher levels of support.  Individuals with scores of zero are 
considered to be without perceived social support within the past month.  The social 
support mean for the entire sample was 3.60; the nonchronic group had a mean score 
of 3.35 and the chronic group a mean score of 3.91.  The difference between the 
groups was significant (t(1391) = -3.86, p<.001) with parents of children living with 
a chronic illness perceiving higher levels of support than parents of children not 
living with an illness.
The social support variable was dichotomized into two groups for analysis in 
SEM (Bollen, 1989).  The current literature on social support did not provide a clear 
cut-score for dichotomizing the groups into low and mid to high.   Much of the 
support literature uses Likert scales to assess the intensity and direction of support 
(Charron-Prochownik & Kovacs, 2000; Dunst, Trivette, & Cross, 1986; Jackson, 
Gyamfi, Brooks-Gunn, & Blake, 1998; Kaplan & Hartwell, 1987; Wanamaker & 
Glenwick, 1998).  Additionally these studies did not dichotomize, but use social 
support as a continuous variable.  
When no clear theoretical cut-score is present, dichotomizing the data using a 
median split has been suggested.  Previous research has also examined the 
moderating influence of low and high levels of social support by using a median split 
74
(Simons, et al., 1992; Simons, et al., 1993). Additionally, a median split helps to 
ensure that the sample size in each group is similar.    
The median score for the entire sample and the subsample of no illness was 
three.  The median score for the chronic illness group was four.  Therefore, three was 
chosen as the median cut-score.  Thus, individuals with lower levels of perceived 
social support have scores ranging from 0 to 3, and higher levels of perceived social 
support range from 4 to 10.
Weights
PSID staff created a weight to be applied to the Child Development 
Supplement data file that adjusts for the oversampling of African American families, 
to represent the national population of children.  The weight also adjusts for primary 
caregivers who failed to complete the primary household booklet.  The PSID staff 
created a nonresponse adjustment for the data by ?multiplying each weight by the 
ratio of the sum of weights for the total eligible by the sum of weights for the total 
interviewed.  This nonresponse adjustment was computed separately for each post-
stratification cell? (Institute for Social Research, 1999, p. 1).
In order to maintain actual sample sizes the weight must be normalized to 
avoid bias.  The weight was normalized by dividing the weight by its mean 
(11.07627).  This new weight was then applied to the data.  Data analysis was 
completed with the weight applied.  
75
Missing data
The weight that was applied to the data adjusts for instances where the 
primary caregiver failed to complete entire scales.  However, when individual items 
are missing within a scale, the items that were completed by that individual were 
averaged and the resulting mean replaced the missing item (Allison, 2002).  If 
individuals are missing more than half of the items on a particular scale, no 
substitution is applied and the items remain missing for that individual
Statistics
Descriptive analysis.
Descriptive statistics, such as frequencies, means, and standard deviations 
were used to summarize the demographic information of families, using the SPSS 
11.0 statistical package.  The demographic variables that were analyzed include age, 
race, and gender of child; age, race, gender, education, and current employment 
status of the head of the household; the age, race, gender, education, and current 
employment status of the spouse; the number of children in the household under the 
age of 18; the family's total income; the family's poverty level status; and the family 
structure.  Demographic information was further analyzed by comparing 
characteristics of the two groups, chronic illness and no illness, using independent t-
tests for continuous data and Chi-square tests for nominal data analysis.  
Additionally, Pearson?s Product-Moment correlation coefficients were computed to 
examine the relationships among all variables. The education of the head of the 
76
household and the number of children under the age of 18 living in the household 
were controlled for during multivariate analysis.
Multivariate analysis. 
Structural equation modeling (SEM) is the chosen method for examining the 
current data and was chosen for many reasons.  First, SEM has more flexible 
assumptions, such as permitting multicollinearity, than multiple regression (Garson, 
2003).  Second, SEM has the ability to test the entire model simultaneously rather 
than each of the coefficients individually, as in regression (Garson, 2003; Peyrot, 
1996).   Third, SEM can test models with multiple dependent variables (Garson, 
2003).  Fourth, SEM permits the modeling of multiple mediating variables (Garson, 
2003).  This modeling allows for examination of mediating variables in a causal, 
multilevel additive fashion.  Fifth, SEM has the ability to examine the effect of 
moderator variables for each of the variable paths (Garson, 2003).  Sixth, error terms 
are modeled, measured, and can be correlated in SEM (Chen & Land, 1986; Garson, 
2003).  SEM allows for disattenuation of measurement error either by having 
multiple indicators of factors or by building in reliability corrections for single 
indicators.  Seventh, SEM expands analysis to include indirect relationships (Peyrot, 
1996).  Indirect relationships can be assessed by examining "the sum of the 
compound paths" that connect the two variables (Loehlin, 1998, p. 9).  If SEM is not 
utilized, significant indirect effects may be missed (Peyrot, 1996).  Eighth and 
finally, SEM permits the testing of coefficients across multiple between-subject 
groups (Garson, 2003).  
77
Structural equation models specify the relationship among variables that are 
based on theory and previous research (Lavee, 1988).  The arrows or paths between 
variables explain which variables are directly and indirectly related (Loehlin, 1998).  
Double-headed arrows indicate covariation paths.  The path coefficients that are 
created through EQS indicate the strength and direction of the relationship between 
the variables (Loehlin, 1998).   
Figure 3 illustrates the model evaluated in this research.  The only exogenous 
variable in the model is the presence of chronic illness.  The remaining variables are 
endogenous and, with the exception of child behavior problems, are mediating as 
well.  The moderating effects of social support are analyzed by examining the effect 
of low versus mid to high levels of social support on each of the above paths.  Each 
of the paths present in the model is tested across the two groups to assess if the level 
of support moderates the relationship between factors.  
Parental  
Psychological 
Distress
D1
D2
Presence 
of Illness
Child
Health 
Stress
Parental 
Emotional 
Support
Parental 
Warmth
Parental 
Involvement
Economic 
Strain
Child 
Behavior 
Problems
Parental 
Cognitive 
Stimulation
D5
D7
D8
Figure 3: Structural Equation Model
Disturbance path covariation
D1 
 D2   D4 
 D6   D5 
 D6
D4 
 D5   D4 
 D7   D5 
 D7
D6 
 D7
D4
D3
D6
78
79
The model indicates that the presence of a chronic illness directly influences 
economic strain, child health stress, parental psychological distress, and child 
behavior problems.  Parenting behaviors are indirectly influenced by the presence of 
chronic illness through the mediating variables of economic strain, child health 
stress, and parental psychological distress.  Economic strain is directly linked to 
parental psychological distress, parental cognitive stimulation, and parental 
involvement.  Covariation is only acceptable in independent variables in SEM; 
therefore to indicate the possible correlation between economic strain and child 
health stress the disturbance paths were made to covary (Loehlin, 1998).  Economic 
strain is indirectly linked to parental emotional support and warmth, and child 
behavior problems.  Child health stress is directly related to parental psychological 
distress and child behavior problems. The parenting variables are indirectly linked to 
health stress.  Parental psychological distress is directly linked to all four parenting 
variables and child behavior problems.  The four parenting variables (parental 
cognitive stimulation, parental emotional support, parental warmth, and parental 
involvement) are all directly related to child behavior problems.  To indicate the 
possible correlation between parenting behaviors, all of the disturbance paths of the 
parenting behaviors are made to covary.  
Participant index or scale totals were calculated for economic strain, child 
health stress, family conflict, parental well- being, parental warmth, parental 
cognitive stimulation, parental emotional support, parental warmth, parental 
involvement, and child behavior problems.  Each of the scale totals acts as observed 
80
indicators of the given variable and "are assumed to correspond exactly with their 
underlying constructs" (Garrett, et al., 1994, p. 152).  For example, the scale score 
for child health stress serves as the observed measurement of the child health stress 
variable.  Unfortunately, single indicators of a variable express no measurement 
error, but in reality may be error ridden (Kline, 1998).  To correct for this problem, 
error variance was estimated by subtracting the scale's Cronbach alpha from 1 and 
then multiplying by the scale's total variance ((1-)(variance)).  The inclusion of the 
estimated error variance then creates a latent variable from the observed indicator.  
See Figure 4 for a pictorial transformation of the variable.   
Figure 4: Transformation of Observed Variables to Latent Variables 
This error correction was completed for economic strain, health stress, parental 
psychological distress, parental cognitive stimulation, parental emotional support, 
parental warmth, parental involvement, and child behavior problems.  
New latent 
variable
Scale total
(observed 
variable)
Error, whose 
variance that is 
being estimated
New latent variable 
due to creation of 
estimated error 
New estimated error 
created by 
((1-)(variance))  
81
In order to obtain error variances for parental cognitive stimulation and 
emotional support one alpha that collapsed the age groups had to be created.  To do 
this each of the items were analyzed for similarity in wording and scoring across the 
two age groups.  Of the items, only two were specific to the two age groups.  One 
item, "how often do you read to (CHILD)?" applied only to the six through nine year 
old group.  The second item, "How often is (CHILD) expected to do each of the 
following: pick up after (himself/herself)?" pertained only to the 10 and older group.  
Therefore, these two items were removed from the analysis of Cronbach alphas in 
order to create one error variance for both groups.
A structural equation model with an observed variable (presence of chronic 
illness) and latent factors was analyzed using maximum likelihood estimation.  
Maximum likelihood estimation was used because it has been found to be robust 
when normality assumptions are violated (Newcomb & Bentler, 1988).  The latent 
factors were run simultaneously with the structural equation model in the EQS 
structural equation program (Bentler, 1997).  Each of the variables has only one 
indicator; therefore the measurement model would have fit perfectly and was not run.  
Only the structural model was run.  
The structural model, as seen previously in Figure 3, was applied to the data.  
Next the model fit was assessed in order to retain the appropriate model and reject an 
unsuitable model.  Fit was assessed using Hu and Bentler's (1999) joint fit criteria.  
Hu and Bentler (1999) recommended assessing model fit on two criteria 
simultaneously, such as Bentler's Comparative Fit Index (CFI) of 0.96 or larger and 
82
Standardized Root Mean-square Residual (SRMR) of 0.10 or smaller; or the Steiger 
and Lind Root Mean-Square Error of Approximation (RMSEA) of .06 or smaller and 
the previously mentioned CFI may also be used for joint fit criteria.  
After the structural model was run for the entire sample, the moderation of 
social support was explored using multigroup structural equation models to analyze 
the two groups of low and mid to high levels of social support.  Model fit was 
assessed for each model (low and mid-high levels of social support).  Next each path 
was constrained individually (e.g., made to have the same coefficients for both 
groups).  If the constrained path was significant at the .05 level in the Lagrange 
Multiplier Test for Constraints, the constraint was considered not tenable due to 
group differences.  If the constrained path was not significant, the constraint was 
considered tenable for both groups.  Each of the paths in the model was run in this 
manner.  Thus, the final structural equation model has paths that are constrained and 
similar across groups, and/or paths that differ for the two groups with separate 
loadings or paths for each.    
83
Chapter 4: Results
Demographic characteristics
   Participants.
The sample consisted of a total of 1,499 children, 693 (46.2 percent) living 
with a chronic illness and 806 (53.8 percent) not living with a chronic illness.  Of the 
693 children living with a chronic illness, 587 reported only one chronic illness (39.2 
percent of total sample and 84.7 percent of chronic sample), 102 reported two 
chronic illnesses (6.8 percent and 14.7 percent), and 4 reported three chronic 
illnesses (0.3 percent and 0.6 percent).  The numbers of children living with 
individual illnesses include 120 living with anemia (8 percent of the total sample and 
17.3 percent of chronic sample), 157 living with asthma (10.5 percent and 22.6 
percent), 3 living with diabetes (0.2 percent and 0.4 percent), and 522 living with 
chronic otitis media (34.9 percent and 75.3 percent).  Chronic otitis media has the 
greatest comorbidity with 106 children living with chronic otitis media and at least 
one other illness.  
The demographic characteristics of the children are presented in Table 2.  
The children ranged in age from 6 to 13.6 with a mean age of 9.5 years.  There are 
709 (47.3%) male and 790 (52.7%) female children in the sample.  The majority of 
the children are Caucasian (n=1062, 70.9%) followed by African American (n=181, 
12.1%), Hispanic (n=150, 10.0%), Asian (n=50, 3.4%), Native American (n=14, 
0.9%), and finally other (n=42, 2.8%).  
84
Table 2: Child Demographics (N = 1,499)
Total             Chronic         Nonchronic 
Mean (SD)     Range        Mean (SD)   Mean (SD)
Age     9.5 (2.2)         6-13.6     9.4 (2.1)            9.6 (2.2)
N (%)     N (%)      N(%)
Gender
  Male 709  (47.3%) 369 (53.2%)          341 (42.3%)
  Female  790  (52.7%) 324 (46.8%)          465 (57.7%)
Race
  Caucasian            1062   (70.9%) 486 (70.1%)   576 (71.5%)
  African American 181   (12.1%)   97 (14.0%)             84 (10.4%)
  Hispanic   150   (10.0%)           69 (10.0%)             80   (9.9%)
  Asian         50     (3.4%)           13   (1.9%)             38   (4.7%)
  Native American   14     (0.9%)     7   (0.1%) 7    (0.9%)
  Other                           42     (2.8%)   21   (3.0%)     21    (2.6%)
85
The demographic characteristics of the parents are presented in Table 3.  The 
majority of households were headed by men (n=1430, 95.6%) and have two parents 
(n=1410, 94.0%).  Eighty- nine (6.0%) families are considered "other,? or not a two-
parent family.  The age of the head of household ranges between 22 and 77 with a 
mean of 40.  The household head has an average of 12.9 years of schooling with a 
range of 1 to 17 years.  The age of the spouse ranges between 23 and 73 with a mean 
of 38.1.  The wife has an average of 13.1 years of schooling with a range of 2 to 17 
years.  The majority of the head of household are Caucasian (n=1090, 72.7%), 
followed by African American (n=174, 11.6%), Hispanic (n=164, 10.9%), Asian 
(n=53, 3.6%), Native American (n=9, 0.6%), and other (n=9, 0.6%).
  The individuals who completed the questionnaires are considered to be the 
primary caregivers.  The majority of the primary caregivers for the children in the 
current study are mothers (stepmother/adoptive mother, n=1459, 97.4%), followed 
by fathers (stepfather/adoptive father, n=23, 1.5%), legal guardians (n=16, 1.1%), 
and another adult in the household (n=1, 0.1%).
86
Table 3: Parent Demographics (N=1,499)
Total      Range       Chronic         Nonchronic 
Mean (SD)         Mean (SD)   Mean (SD)
Age of head 40   (8.0) 22-77         40.5 (8.6)     39.6 (7.4)
Age of spouse             38.1(6.9)  23-73           38.4 (7.3)      37.9 (6.5)
Education 
   Head of 
      household            12.9 (3.2)          1-17          13.0 (3.2)                12.8 (3.2) 
   Spouse            13.1 (2.9)          2-17          13.1 (3.0)      13.0 (2.8) 
N (%)     N (%)      N(%)
Sex of head of household
   Male         1430 (95.4%)                     664 (95.8%)   766  (95.0%)
   Female             69   (4.6%)           29   (4.2%)     39    (4.8%)
Family structure
   Two parent 1410 (94.0%)        650 (93.8%)   759  (94.2%)
   Other             89   (6.0%)          43   (6.2%)     47    (5.8%)
Race of head of household
  Caucasian              1090(72.7%)                   511 (73.7%)   579  (71.8%)
  African American   174 (11.6%)         88 (12.7%)     86  (10.7%)
  Hispanic                 164 (10.9%)         68   (9.8%)     96  (11.9%)
  Asian                        53   (3.6%)         16   (2.3%)     38    (4.7%)
  Native American        9   (0.6%)           5   (0.7%)       5    (0.6%)
  Other                          9   (0.6%)           6   (0.9%)       3    (0.4%)
Child's responding parent
   Mother        1459 (97.4%)      677  (97.7%)  782  (97.0%)
   Father            23   (1.5%)   7    (1.0%)    15    (1.9%)
   Legal guardian        16   (1.1%)   8    (1.2%)      8    (1.0%)
   Other adult in 
      household 1    (0.1%)           1   (0.1%) --
Note: Mother includes stepmother and adoptive mother. Father includes stepfather 
and adoptive father.
87
The demographic characteristics of the household are presented in Table 4.    
The number of individuals in the family unit ranges from 2 to 12 (mean 4.6), and the 
number of children younger than 18 in the household ranges between one and nine 
(mean 2.44).  The majority of families were above the poverty line (n=1367, 91.2%) 
with an average total family income of $60,817.80 (range 0 to 1,512,000).  The 
average total family income may be high because of the inclusion of the family 
income of 1.512 million.  Therefore, examining the median household income is 
necessary.  The median income for the sample is 46,700.  In 2000, the median 
national household income was 41,994 (U.S. Census Bureau, 2000).  Thus, the 
median for the sample is a clearer picture of household income than the average. 
T-tests and chi square tests were used to assess demographic differences 
between children living with a chronic illness and children not living with a chronic 
illness.  There were no significant differences between the demographic profiles of 
the children in the chronic and nonchronic groups based on the child's age; sex, race 
and education of the household head; the age and education of the wife; the 
participant's relationship to the child; the household type (dual parent or other); the 
number of people in the family unit; the family income; and whether the family is at 
or below the poverty level.  However, differences were found between the groups for 
the sex of the child (more female nonchronic, male chronic; 
2 
(1) = 17.9, p <.001), 
the race of the child (more Caucasian, Hispanic, and Asian nonchronic; more African 
American chronic; 
2 
(5) = 13.4, p < .02) the number of children in the household 
88
(more children in the nonchronic group; t(1486.98) = 3.02, p < .002), and the age of 
the household head (older in chronic group; t(1497) = -2.31, p < .02).
89
Table 4: Household demographics (N=1,499)
Total             Range            Chronic         Nonchronic 
Mean (SD)         Mean (SD)   Mean (SD)
Number of people
  in family unit           4.6  (1.3) 2-12 4.5 (1.3)                4.6 (1.2)
Number of children
  under 18                   2.4  (1.2)          1-9             2.4 (1.1)       2.5 (1.2)
Total family 
income($)   60,818 (74,650)    0-1.512M   62,266 (52,772)    59,573 (89,294) 
N (%)     N (%)      N(%)
At or below poverty level
   No     1367  (91.2%)         635 (91.6%)    733 (90.9%)
   Yes       132    (8.8%)           58   (8.4%)      73   (9.5%)
90
    Measures.
Table 5 provides the means, standard deviations, and range of scores for each 
of the measures used in the current study for the sample total, chronic, nonchronic, 
low social support, and mid to high social support groups.  T-tests were run for each 
of the groups, chronic/nonchronic and low/mid-high social support.  Children living 
with a chronic illness experience significantly more child health stress (t(1246.68) = 
-4.34, p <.001), more parental cognitive stimulation (t(1492.59) = -2.06, p < .04), 
more parental warmth (t(1490) = -2.41, p < .02), and more child behavior problems 
(t(1320.81) = -3.58, p < .001) than children not living with a chronic illness.  
However, primary caregivers of children not living with a chronic illness experience 
significantly greater levels of psychological distress (t(1422.37) = 1.93, p < .05) than 
caregivers of children living with a chronic illness.  Mid to high levels of social 
support were also found to be significantly related to higher levels of economic strain 
(t(1181.17) = -4.74, p < .01), child health stress (t(1131.71) = -2.79, p < .01), 
parental cognitive stimulation (t(1391) = -7.89, p < .01), parental emotional support 
(t(1391) = -1.99, p <  .05), and parental warmth (t(1328.13) = -3.15, p < .01) than 
low levels of social support.
Table 5: Measure Demographics
   Total          C            NC                   LSS       HSS                   C        NC      LSS    HSS
Measure Mean SD R Mean Mean t-test Mean Mean t-test SD SD SD SD
Economic strain   1.60 1.92   0-11  1.56   1.64   1.36   1.87 *** 1.88 1.95 1.71 2.09
Child health stress   0.10 0.42   0-3  0.15   0.06 ***   0.07   0.14 ** 0.48 0.35 0.33 0.51
Parental 
psychological distress
16.26 5.80 10-45 15.94 16.53 * 16.15 16.44 5.37 6.14 6.10 5.60
Parental cognitive 
stimulation
10.23 1.84   4-14 10.33 10.14   9.87 10.64 *** 1.74 1.91 1.85 1.79
Parental emotional 
support
 9.15 1.19   2-11  9.18  9.12   9.13   9.26 * 1.09 1.28 1.18 1.19
Parental warmth 26.10 3.32   9-30 26.32 25.91 * 25.86 26.42 *** 3.19 3.41 3.72 2.83
Parental involvement 34.14 7.58 15-60 34.36 33.96 33.90 34.45 7.70 7.47 7.16 8.02
Child behavior 
problems
39.98 8.24 30-75 40.82 39.26 *** 39.64 40.16 8.99 7.46 7.99 8.42
 Note: C = Children living with a chronic illness.  NC = Children not living with a chronic illness.  LSS = Low levels of 
social support.  HSS = Mid to high levels of social support.  M = Mean.  SD = Standard deviation.  R = Range.
t-test = *p<.05.  **p<.01.  ***p<.001. 
91
92
Structural equation model for the total sample
The fit of the model was good (model 
2  
= 58.185, df=11, CFI = 0.964; 
SRMR = 0.027; RMSEA = 0.054, RMSEA confidence interval 0.040 to 0.067) 
suggesting that analysis of individual paths is acceptable.  Figure 5 provides a 
pictorial representation of the individual direct paths.  Direct, indirect, and total 
effects of the relationship between variables can be seen in Table 6.
-.067*
.301***
.127***
-.072*
-.125***
-.270***
.326***
-.041
-.095***
-.038
-.190***
-.008
-.311***
.171***
Note: *p<.05.  **p<.01.  ***p<.001.
93
.992   D2
Presence 
of Illness
Child
Health 
Stress
Parental 
Emotional 
Support
Parental 
Warmth
Parental 
Involvement
Economic 
Strain
Parental  
Psychological 
Distress
Child 
Behavior 
Problems
Parental 
Cognitive 
Stimulation
.963 D5
.998 D7
.810 
D8
Figure 5: Total Sample SEM Standardized Solution
D1
D2      .072**
D4
D5      .201***
D4
D6    1.626***
D4
D7   4.604***
D5
D6    -.155
D5
D7     .602**
D6
D7 12.283***
.974 D4
.950 D3
.995 D6
-.013
1   D1
.079**
.306***
-.156***
94
Table 6: Standardized Direct, Indirect and Total Effects for Total Sample
Predictor Dependent Variable     Direct Effect     Indirect Effect    Total Effect
Illness presence
Economic strain -.013 -- -.013
Child health stress .127*** --  .127
Psychological distress -.067*                 -.013 -.080
Cognitive stimulation -- .012?  .012
Emotional support -- .022**  .022
Warmth -- .008*  .008
Involvement -- .004  .004
Child behavior problems      .079** .009  .088
Economic strain
Psychological distress      .301*** --   .301
Cognitive stimulation -.156*** -.037***        -.193
Emotional support -- -.081*** -.081
Warmth -- -.029** -.029
Involvement -.041 -.011 -.053
Child behavior problems      -- .135***  .135
Child health stress
Psychological distress -.072* -- -.072
Cognitive stimulation -- .009?  .009
Emotional support -- .019*  .019 
Warmth -- .007?  .007
Involvement -- .003  .003
Child behavior problems      .306***            -.027* .279
Parental psychological distress
Cognitive stimulation -.125*** -- -.125
Emotional support -.270*** -- -.270
Warmth -.095*** -- -.095
Involvement -.038 -- -.038
Child behavior problems      .326*** .049**  .375
Parental cognitive stimulation
Child behavior problems -.190*** -- -.190
Parental emotional support
Child behavior problems -.008 -- -.008
Parental warmth
Child behavior problems -.311*** -- -.311
Parental involvement
Child behavior problems      .171*** -- .171
Note: ?p<.10. *p<.05.  **p<.01.  ***p<.001.
95
Multigroup structural equation model 
The fit for the low levels of social support model (model 
2  
= 52.916, df = 
11, CFI = 0.927; SRMR = 0.033; RMSEA = 0.073, RMSEA confidence interval 
0.054 to 0.093) and the mid to high levels of social support model (model 
2  
= 
68.449, df = 11, CFI = 0.942; SRMR = 0.046; RMSEA = 0.088, RMSEA confidence 
interval 0.068 to 0.108) did not meet the joint criteria for good model fit; CFI  0.96 
and SRMR  0.10 or RMSEA  0.06 and SRMR  0.10 (Hu & Bentler, 1999).  
However, when the two models were run simultaneously, without constraints, a 
multisample adjustment was conducted in the fit indices by EQS.  After the 
adjustment the fit is good using joint criteria (model 
2  
= 121.364, df = 22, CFI = 
0.937; SRMR = 0.040; RMSEA = 0.057, RMSEA confidence interval 0.047 to 
0.067).  Figures 6 and 7 provide pictorial representations of the individual direct 
paths for the low and mid to high social support models.  Standardized direct, 
indirect, and total effects of the relationship between variables can be seen in Table 
7.  The figures and tables show standardized solutions, but the unstandardized 
solutions are the paths that are compared.  Standardized solutions are provided for 
interpretability.  
.085?
-.132**
.159**
.302***
-.063
-.031
-.359***
.188***
-.101*
-.041
-.066*
-.202**
-.093
-.172**
.049
Note:  ?p<.10. *p<.05.  **p<.01.  ***p<.001.
.953 D2
Presence 
of Illness
Child
Health 
Stress
Parental 
Emotional 
Support
Parental 
Warmth
Parental 
Involvement
Economic 
Strain
Parental  
Psychological 
Distress
Child 
Behavior 
Problems
Parental 
Cognitive 
Stimulation
.933 D5
.992 D7
.916 
 D8
Figure 6: Low Levels of Social Support SEM Standardized Solution
D1
D2      .080***
D4
D5      .369***
D4
D6    1.515***
D4
D7   4.185***
D5
D6    -.494**
D5
D7     .177
D6
D7 12.616***
.949 D4
.979 D3
.999 D6
.996 D1
.078?
.102?
-.311***
96
-.132**
.022
.416***
.030
-.059
-.304***
-.155**
.416***
-.052
-.202***
-.028
-.000
.040
-.726***
.439***
Note: *p<.05.  **p<.01.  ***p<.001.
1 D2
Presence 
of Illness
Parental 
Emotional 
Support
Parental 
Warmth
Parental 
Involvement
Economic 
Strain
Parental  
Psychological 
Distress
Child 
Behavior 
Problems
Parental 
Cognitive 
Stimulation
.988 D5
.998 D7
.495 D8
Figure 7: Mid to High Levels of Social Support SEM Standardized Solution
D1
D2      .012
D4
D5      .005
D4
D6    1.670***
D4
D7   4.659***
D5
D6     .106
D5
D7     .992**
D6
D7 12.330***
.938 D4
.909 D3
.979 D6
.991 D1
.097**
.489***
-.081
Child
Health 
Stress
97
98
Table 7: Standardized Direct, Indirect, and Total Effects for Levels of Social Support
  Low social support Mid-high social support
           Direct    Indirect    Total  Direct    Indirect   Total
Predictor     Dependent var.   Effect   Effect      Effect  Effect     Effect     Effect
Presence of illness
     Economic strain .085? --  .085          -.132** -- -.132
    Child health stress .302***  --  .302 .030 --     .030
     Psychological distress        -.132** -.006    -.137 .022 -.057** -.034
     Cognitive stimulation -- -.022    -.022 --  .021     .021
     Emotional support --   .049**  .049 --  .005     .005
     Warmth --   .006  .006 --  .007     .007
     Involvement --   .001  .001 --  .008     .008
     Child behavior prob. .078?     .004  .082 .097** -.001     .096
Economic strain
     Psychological distress .159** --  .159          .416*** --     .416
     Cognitive stimulation         -.311***-.005     -.316         -.081 -.127*** -.208
     Emotional support -- -.057** -.057 -- -.065** -.065
     Warmth -- -.006     -.006 -- -.084***   -.084
     Involvement                -.101* -.011     -.112         -.052 -.012 -.064
     Child behavior prob. --   .095*** .095 --  .204***    .204
Child health stress
     Psychological distress         -.063 -- -.063         -.059 -- -.059
     Cognitive stimulation --   .002  .002 -- .018     .018
     Emotional support --   .023  .023 -- .009     .009
     Warmth --   .003      .003 -- .012     .012
     Involvement --   .004  .004 -- .002     .002
     Child behavior prob. .102? -.015  .087 .489*** -.032     .456
Psychological distress
     Cognitive stimulation         -.031 -- -.031         -.304*** -- -.304
     Emotional support             -.359***   -- -.359         -.155** -- -.155
     Warmth                              -.041   -- -.041         -.202*** -- -.202
     Involvement                       -.066*       -- -.066         -.028 -- -.028
     Child behavior prob.           .188***  .043?  .232 .416***   .128***   .544
Cognitive stimulation
    Child behavior prob.           -.202** -- -.202         -.000 -- -.000
Emotional support
     Child behavior prob.           -.093 -- -.093 .040 --             .040
Warmth
     Child behavior prob.           -.172**    -- -.172      -.726*** -- -.726
Involvement
     Child behavior prob.      .049       --           .049 .439*** --             .439
Note: ?p<.10. *p<.05.  **p<.01.  ***p<.001.
99
Hypothesis thirteen.
Hypothesis 13 was analyzed using multigroup structural equation modeling.  
Hypothesis 13 states that mid to high levels of social support moderate all of the 
relations in the model.  In order to more clearly assess this hypothesis each of the 
relationships between variables is examined after each of the corresponding 
hypotheses.  Each of the paths was individually constrained to be the same across 
groups, if the groups differed significantly; the constraint was released allowing the 
group's paths to vary.  If the constraint was not statistically significant, the group's 
paths are considered equal, indicating no difference between groups.  To assess 
constraint significance, the difference between the model chi-square and degrees of 
freedom for the constrained and unconstrained models were examined.  The 
difference between the model chi-squares and significance are included for each 
relationship below.  Figure 8 provides a pictorial representation of the structural 
equation model both with and without constraints.  Both unstandardized and 
standardized path coefficients are included.  
-.057  -.027
(-.359***  -.155**)
.236  -.470
(.085?  -.132*)
.137  .025
(.302***  .030)
-.023  -.092
(-.041  -.202***)
-.059  -.059
(-.056  -.043)
-.852  -.852
(-.033  -.067)
-1.550  .238
(-.132**  .022)
.674  1.243
(.159**  .416***)
-.314  -.314
(-.071*  -.078*)
-.008  -.080
(-.031  -.304***)
-1.051  -.003
(-.202**  -.000)
-.032  -.032
(-.004  .004)
-.384  -2.412
(-.172**  -.726***)
.060  .495
(.049  .439***)
-.326  -.064
(-.311***  -.081)
3.428  9.426
(.102?  .489***)
1.426  1.426
(.094***  .089***)
Note: Top numbers unstandardized coefficients, low/high social support.  Bottom numbers standardized coefficients, 
low/high social support.    = statistically significant (.05 or smaller) difference between group path coefficients.   = no 
statistically significant difference between group path coefficients.  ?p<.10. *p<.05.  **p<.01.  ***p<.001.
.243  .632
(.188***  .416***)
Presence 
of illness
Economic 
strain
Psychological 
distress
Involvement
Emotional 
support
Health 
stress
Warmth
Cognitive 
stimulation
Child 
behavior 
problems
Figure 8: SEM Low and Mid to High Levels of Social Support with Constraints Applied
100
101
Hypothesis one.
The first hypothesis predicted that the presence of a chronic illness positively 
influences economic strain.  The path from the presence of chronic illness to 
economic strain was negative and not significant (	 = -.013, p > .05).  This indicates 
that there is not support for a relationship between the presence of a chronic illness 
and economic strain.  Thus, hypothesis one was not supported.   
Hypothesis one social support moderation. 
The Lagrange Multiplier Test for Constraints revealed that the constraint for 
the path coefficient between presence of illness and economic strain does not hold 
for the low and mid to high levels of social support groups (
2 
= 11.827, p <.001).  
The unstandardized paths are .236 (standardized .085, p < .10) for low levels of 
social support and -.470 (standardized -.132, p < .001) for high levels of social 
support.  These coefficients indicate that under conditions of low levels of social 
support, illness presence is marginally positively related to economic strain.  In 
contrast, under conditions of mid to high levels of social support, the presence of a 
chronic illness is associated with decreased economic strain.  This suggests that 
higher levels of social support act as a buffer against possible effects of illness 
presence on economic strain.  
     Hypothesis two.
Hypothesis two states that the presence of a chronic illness positively 
influences child health stress.  The path from the presence of chronic illness to child 
health stress was positive and statistically significant (	 = .127, p < .001).  As 
102
expected children living with a chronic illness are more likely to have health stress 
than children not living with an illness.  Hypothesis two was supported.  
Hypothesis two social support moderation. 
The constraint for the relationship between the presence of chronic illness 
and child health stress does not hold for the low and mid to high levels of social 
support groups (
2 
= 5.753, p < .02).  The unstandardized paths are .137 
(standardized .302, p < .001) for low levels of social support and .025 (standardized 
.030, p > .05) for high levels of social support.  These coefficients indicate that mid 
to high levels of social support moderate or buffer the relationship between the 
presence of chronic illness and child health stress.    
     Hypothesis three.
The third hypothesis asserts that the presence of a chronic illness has a direct 
and positive effect on parental psychological distress.  The path between illness 
presence and parental psychological distress is negative and statistically significant 
(	 = -.067, p < .05).  This indicates that the presence of a chronic illness significantly 
decreases parental psychological distress.  These findings do not support the third 
hypothesis.
Hypothesis three social support moderation. 
The constraints were not tenable for the relationship between the presence of 
illness and parental psychological distress (
2 
= 7.424, p < .01).  Once the constraints 
were released the unstandardized paths were -1.550 (standardized -.132, p < .01) for 
low levels of social support and .238 (standardized .022, p > .05) for high levels of 
103
social support.  Under conditions of low levels of social support the presence of a 
chronic illness significantly decreases parental psychological distress.  There is no 
relationship between illness presence and psychological distress when there are mid 
to high levels of support.  Therefore there is no buffering effect of social support in 
this relationship.  Instead there appears to be reverse buffering, under conditions of 
lower levels of social support there is an association with a decrease in parental 
psychological distress when a chronic illness is present.
     Hypothesis four.
The fourth hypothesis states that the presence of a chronic illness has a direct 
and positive effect on child behavior problems.  Chronic illness has a positive and 
significant relationship with child behavior problems (	 = .079, p < .01).  Hypothesis 
four is supported.
Hypothesis four social support moderation. 
The path coefficients between the presence of chronic illness and child 
behavior problems are constrained for both groups (
2 
= .219, p > .05).  The 
unstandardized paths for both low and mid to high levels of social support is 1.426, p 
< .001 (standardized .094 for low and .089 for mid to high levels of social support).  
Social support does not moderate the relationship between the presence of a chronic 
illness and child behavior problems.
     Hypothesis five.
Hypothesis five predicts that higher levels of economic strain increase 
parental psychological distress.  The path between economic strain and parental 
104
psychological distress is positive and statistically significant (	 = .301, p < .001).  
Thus, the fifth hypothesis is supported.
Hypothesis five social support moderation.
The constraint for the relationship between economic strain and parental 
psychological distress was not tenable for both groups (
2 
= 4.681, p <.05).  The 
unstandardized paths are .674 (standardized .159 , p < .01) for low levels of social 
support and 1.243 (standardized .416, p < .001) for high levels of social support.  
These coefficients indicate that social support does not buffer the relationship 
between economic strain and parental psychological distress.  Instead there appears 
to be reverse buffering, and lower levels of social support are associated with a 
smaller increase in parental psychological distress when economic strain is present.  
     Hypothesis six.
The sixth hypothesis states that higher levels of child health stress increase 
parental psychological distress.  Child health stress does have a significant, but 
negative influence on parental psychological distress (	 = -.072, p < .05).  The 
direction of this relationship is contrary to the hypothesis, thus the sixth hypothesis is 
not supported.
Hypothesis six social support moderation.
The path coefficients between child health stress and parental psychological 
distress are constrained for both groups (
2 
= .24, p > .05).  The unstandardized paths 
for both low and mid to high levels of social support is -.852 (standardized -.033 for 
low and -.067 for mid to high levels of social support).  There is no moderation of 
105
social support for the relationship between child health stress and parental 
psychological distress.
     Hypothesis seven.
Hypothesis seven asserts that economic strain leads to a decrease in parental 
cognitive stimulation.  The data indicates that economic strain has a significant 
negative effect on parental cognitive stimulation (	 = -.156, p < .001).  The seventh 
hypothesis is supported.
Hypothesis seven social support moderation.
The constraint for the relationship between economic strain and parental 
cognitive stimulation was not tenable for both groups (
2 
= 11.881, p <.001).  The 
unstandardized paths are -.326 (standardized -.311, p < .001) for low levels of social 
support and -.064 (standardized -.081, p > .05) for high levels of social support.  
Economic strain more negatively influences parental cognitive stimulation under 
conditions of lower levels of social support.  Under conditions of mid to high levels 
of support the relationship between economic strain and cognitive stimulation 
disappears.  This indicates that social support acts as a buffer between these two 
variables.
     Hypothesis eight.
The eighth hypothesis states that economic strain leads to a decrease in 
parental involvement.  Economic strain is negatively associated with parental 
involvement, but not significantly (	 = -.041, p > .05).  Therefore, the eighth 
hypothesis is not supported.
106
Hypothesis eight social support moderation.
The path coefficients between economic strain and parental involvement are 
constrained for both groups (
2 
= .702, p > .05).  The unstandardized paths for both 
low and mid to high levels of social support is -.314, p  < .05 (standardized -.071 for 
low and -.078 for mid to high levels of social support).  There is no moderation of 
social support for the relationship between economic strain and parental 
involvement.
Hypothesis nine.
Hypothesis nine maintains that higher levels of child health stress lead to an 
increase in child behavior problems.  Child health stress positively and significantly 
increases child behavior problems (	 = .306, p < .001).  The ninth hypothesis is 
supported.
Hypothesis nine social support moderation.
The constraints for the relationship between child health stress and child 
behavior problems were not tenable for both groups (
2 
= 6.823, p <.01).  The 
unstandardized paths are 3.428 (standardized .102, p  < .10) for low levels of social 
support and 9.426 (standardized .489, p  < .001) for high levels of social support.   
Social support does not act as a buffer for the relationship between child health stress 
and child behavior problems.  Instead there appears to be reverse buffering, where 
child health stress, under conditions of lower levels of social support, is associated 
with fewer child behavior problems than when higher levels of social support are 
present.
107
Hypothesis ten.
The tenth hypothesis states that higher levels of parental psychological 
distress negatively influence all parenting behaviors; parental cognitive stimulation, 
parental emotional support, parental warmth, and parental involvement.  Parental 
psychological distress was negatively and significantly related to parental cognitive 
simulation (	 = -.125, p < .001), emotional support (	 = -.270, p < .001), and warmth 
(	 = -.095, p < .001).  Parental involvement is negatively associated with parental 
psychological distress in this sample, but not significantly (	 = -.038, p > .05).  
Hypothesis ten is partially supported.
Hypothesis ten social support moderation. 
Parental psychological distress and parental cognitive stimulation.
The constraints were not tenable for the relationship between parental 
psychological distress and parental cognitive stimulation (
2 
= 13.781, p < .001).  
The unstandardized paths are -.008 (standardized -.301, p > .05) for low levels of 
social support and -.080 (standardized -.304, p < .001) for high levels of social 
support.  Social support does not buffer the relationship of parental psychological 
distress and parental cognitive stimulation.  Instead there appears to be reverse 
buffering.  Under conditions of low levels of social support there is no relationship 
between parental psychological distress and parental cognitive stimulation, 
conversely under conditions of mid to high levels of support parental psychological 
distress significantly decreases parental cognitive stimulation.  
108
Parental psychological distress and parental emotional support.
The constraints for the relationship between parental psychological distress 
and parental emotional support were not tenable for both groups (
2 
= 6.306, p <.02).  
The unstandardized paths are -.057 (standardized -.359, p< .001) for low levels of 
social support and -.027 (standardized -.155, p < .01) for high levels of social 
support.  Under conditions of mid to high levels of social support the magnitude of 
the relationship between parental psychological distress and parental emotional 
support is smaller than under conditions of low levels of social support.  This 
suggests a reverse buffering effect.  
Parental psychological distress and parental warmth.
The constraints for parental psychological distress and parental warmth were 
released (
2 
= 4.411, p <.05).  The unstandardized paths are -.023 (standardized         
-.041) for low levels of social support and -.092 (standardized -.202, p < .001) for 
high levels of social support.  Higher levels of social support do not buffer the 
influence of parental psychological distress on parental warmth.  Instead there 
appears to be reverse buffering.  Under conditions of low levels of social support 
there is no relationship between parental psychological distress and parental warmth, 
whereas under conditions of mid to high levels of support there is parental 
psychological distress significantly decreases parental warmth.   
Parental psychological distress and parental involvement.
The path coefficients between parental psychological distress and parental 
involvement are constrained for both groups (
2 
= .149, p >.05).  The unstandardized 
109
paths for both low and mid to high levels of social support is -.059 (standardized -
.056 for low and -.043 for mid to high levels of social support).  Social support does 
not buffer the relationship between parental psychological distress and parental 
involvement.
Hypothesis eleven.
Hypothesis eleven asserts that higher levels of parental psychological distress 
increase child behavior problems.  The direct path between parental psychological 
distress and child behavior problems is positive and significant (	 = .326, p < .001), 
indicating that as a parent's psychological distress increases so does the child's 
behavior problems.  Thus hypothesis eleven is supported.
Hypothesis eleven social support moderation.
The EQS program could not estimate the relationship between parental 
psychological distress and child behavior problems in a constrained model; therefore 
the constrained chi-square was calculated manually
2
.  The levels of social support 
differed statistically significantly for the relationship between parental psychological 
distress and child behavior problems (
2 
= 20.313, p <.001). The unstandardized 
paths are .243 (standardized .188, p < .001) for low levels of social support and .632 
(standardized .416, p < .001) for high levels of social support.  Social support does 
not buffer the relationship between parental psychological distress and child 
behavior problems.  Instead there appears to be reverse buffering.  Under conditions 
2
 SE(P
1
? P
2
) = 
[SE(P
1
)]
2
 + [SE(P
2
)]
2
     z = P
1
? P
2
/SE(P
1
? P
2
)        z
2
 = 
2 
df1
 SE = Standard error. P
1 
= Low social support unconstrained path. P
2 
= Mid to high 
unconstrained path.
110
of low levels of social support the magnitude of the path coefficient is smaller from 
parental psychological distress to child behavior problems than when under 
conditions of mid to high levels of social support. 
Hypothesis twelve.
The twelfth hypothesis states that higher levels of parental cognitive 
stimulation, emotional support, warmth, and involvement lead to a decrease in child 
behavior problems.  Parental cognitive stimulation (	 = -.190, p < .001) and warmth 
(	 = -.311, p < .001) are associated with a decrease in child behavior problems.  
Emotional support does have a negative relationship with child behavior problems, 
but not a significant one (	 = -.008, p > .05).  Unexpectedly, parental involvement is 
significantly associated with increased child behavior problems (	 = .171, p < .001).  
Therefore, the twelfth hypothesis is only partially supported.
Hypothesis twelve social support moderation.
Parental cognitive stimulation and child behavior problems.
The constraints for the relationship between parental cognitive stimulation 
and child behavior problems were not tenable for both groups (
2 
= 4.298, p <.05).  
The unstandardized paths are -1.051 (standardized -.202, p < .01) for low levels of 
social support and -.003 (standardized -.000, p > .05) for high levels of social 
support.  Under conditions of low levels of social support, behavior problems are 
significantly decreased by the presence of parental cognitive stimulation.  Contrary 
to expectation, higher levels of social support do not enhance this relationship 
between parental cognitive stimulation and child behavior problems.  
111
Parental emotional support and child behavior problems.
The path coefficients between parental emotional support and child behavior 
problems are constrained for both groups (
2 
= 2.929, p <.10).  The unstandardized 
paths for both low and mid to high levels of social support is -.032 (standardized -
.004, p > .05).  There is no moderation of social support for the relationship between 
the parental emotional support and child behavior problems.
Parental warmth and child behavior problems.
The constraints were released for the relationship between parental warmth 
and child behavior problems (
2 
= 54.612, p <.001).  The unstandardized paths are     
-.384 (standardized -.172, p < .01) for low levels of social support and -2.412 
(standardized -.726, p < .001) for high levels of social support.  Parental warmth has 
a negative relationship with child behavior problems for both low and mid to high 
levels of social support.  However under conditions of mid to high levels of social 
support, the magnitude of the relationship between parental warmth has a more 
negative influence on child behavior problems than under conditions of low levels of 
social support.  Therefore mid to high levels of social support buffer the relationship 
between parental warmth and child behavior problems. 
Parental involvement and child behavior problems.
The path coefficient between parental involvement and child behavior 
problems varies based on individual levels of support (
2 
= 12.139, p <.001).  The 
unstandardized paths are .060 (standardized .049, p > .05) for low levels of social 
support and .495 (standardized .439, p < .001) for high levels of social support.  
112
Contrary to expectation, parental involvement has a statistically significant, positive 
relationship with child behavior problems under conditions of mid to high levels of 
social support.  This relationship indicates that higher levels of social support do not 
buffer the relationship between parental involvement and child behavior problems.  
Instead there appears to be reverse buffering.  Under conditions of low levels of 
social support parental involvement does not influence child behavior problems, but 
under conditions of mid to high levels of social support parental involvement is 
significantly associated with increased child behavior problems.      
Summary of hypothesis thirteen.
There are five paths that mid to high levels of social support buffer or 
moderate the relationship: presence of illness and economic strain, presence of 
illness and child health stress, economic strain and parental cognitive stimulation, 
parental psychological distress and parental emotional support, and parental warmth 
and child behavior problems.  There are four paths that are the same across the social 
support groups: presence of illness and child behavior problems, child health stress 
and parental psychological distress, economic strain and parental involvement, and 
parental emotional support and child behavior problems.  Contrary to expectations 
eight paths have a reverse buffer with lower levels of social support having more 
favorable path coefficients than mid to high levels of support.  These paths include: 
presence of illness and parental psychological distress, economic strain and parental 
psychological distress, child health stress and child behavior problems, parental 
psychological distress and parental cognitive stimulation, parental psychological 
113
distress and parental warmth, parental psychological distress and child behavior 
problems, parental cognitive stimulation and child behavior problems, and parental 
involvement and child behavior problems. 
Indirect effects of the presence of illness
The indirect effects of the presence of illness on child behavior problems 
were examined for the entire model.  The total indirect effect of illness on behavior 
problems is 	 = .009.  However, the indirect sequence from the presence of illness to 
child health stress to child behavior problems yields a coefficient of 	 = .03886.  The 
magnitude coefficient is lost when the remaining indirect sequences, which are 
mainly negative, are summed to create the overall indirect effect of the presence of 
illness on child behavior problems.  Table 8 provides the indirect sequences from the 
presence of illness to child behavior problems through economic strain and child 
health stress.  
Table 8: Indirect Effects of the Presence of Illness on Child Behavior Problems
Through economic strain and: 
Parental psychological distress (but not parenting)
Economic strain  Parental psychological distress  Behavior problems -.00127
Parental psych distress and parenting (Conger/Elder model)
Economic strain  Parental cognitive stimulation  Behavior -.00039
Economic strain  Parental involvement  Behavior -.00009
Economic strain  Parental psychological distress  Parental cognitive stimulation  Behavior -.00009
Economic strain  Parental psychological distress  Parental emotional support  Behavior -.00000
Economic strain  Parental psychological distress  Parental warmth  Behavior -.00012
Economic strain  Parental psychological distress  Parental involvement  Behavior  .00003
Through child health stress and:
Direct
Child health stress  behavior problems  .03886
Parental psychological distress (but not parenting)
Child health stress  Parental psychological distress  Behavior problems -.00298
Parental psychological distress and parenting (Conger/Elder model)   .
Child health stress  Parental psychological distress  Parental cognitive stimulation  Behavior -.00022
Child health stress  Parental psychological distress  Parental emotional support  Behavior -.00002
Child health stress  Parental psychological distress  Parental warmth  Behavior -.00270
Child health stress  Parental psychological distress  Parental involvement  Behavior  .00005
114
115
Indirect effects of the presence of illness and social support
The indirect effects of the presence of illness on child behavior problems 
were also examined for low and mid to high levels of social support.  The total 
indirect effect of illness on behavior problems under conditions of low social support 
is 	 = .004 and under conditions of mid to high levels of social support is 	 = -.001.  
The sequence from the presence of illness through child health stress to child 
behavior problems is the largest sequence under low levels of social support (	 = 
.03084 for low levels of social support).  The sequence from child health stress to 
behavior problems was 	 = .01467 for mid to high levels of social support.  The 
largest sequence for mid to high levels of support was through economic strain and 
parental psychological distress to child behavior problems (	 = -.02284).  The 
magnitude of these coefficients were lost when the remaining indirect sequences, 
which many are negative, are summed to create the overall indirect effect of the 
presence of illness on child behavior problems.  Table 9 and 10 provide the indirect 
sequences from the presence of illness to child behavior problems through economic 
strain and child health stress.
Table 9: Indirect Effects of the Presence of Illness on Child Behavior Problems for Low Levels of Social Support
Through economic strain and: 
Parental psychological distress (but not parenting)
Economic strain  Parental psychological distress  Behavior problems  .00254
Parental psych distress and parenting (Conger/Elder model)
Economic strain  Parental cognitive stimulation  Behavior  .00534
Economic strain  Parental involvement  Behavior -.00042
Economic strain  Parental psychological distress  Parental cognitive stimulation  Behavior  .00008
Economic strain  Parental psychological distress  Parental emotional support  Behavior  .00045
Economic strain  Parental psychological distress  Parental warmth  Behavior  .00010
Economic strain  Parental psychological distress  Parental involvement  Behavior -.00004
Through child health stress and:
Direct
Child health stress  behavior problems  .03080
Parental psychological distress (but not parenting)
Child health stress  Parental psychological distress  Behavior problems -.00358
Parental psychological distress and parenting (Conger/Elder model)   .
Child health stress  Parental psychological distress  Parental cognitive stimulation  Behavior -.00012
Child health stress  Parental psychological distress  Parental emotional support  Behavior -.00064
Child health stress  Parental psychological distress  Parental warmth  Behavior -.00013
Child health stress  Parental psychological distress  Parental involvement  Behavior  .00006
116
Table 10: Indirect Effects of the Presence of Illness on Child Behavior Problems for Mid to High Levels of Social Support
Through economic strain and: 
Parental psychological distress (but not parenting)
Economic strain  Parental psychological distress  Behavior problems -.02284
Parental psych distress and parenting (Conger/Elder model)
Economic strain  Parental cognitive stimulation  Behavior 0
Economic strain  Parental involvement  Behavior  .00301
Economic strain  Parental psychological distress  Parental cognitive stimulation  Behavior 0
Economic strain  Parental psychological distress  Parental emotional support  Behavior  .00034
Economic strain  Parental psychological distress  Parental warmth  Behavior -.00805
Economic strain  Parental psychological distress  Parental involvement  Behavior -.00067
Through child health stress and:
Direct
Child health stress  behavior problems  .01467
Parental psychological distress (but not parenting)
Child health stress  Parental psychological distress  Behavior problems -.00074
Parental psychological distress and parenting (Conger/Elder model)   .
Child health stress  Parental psychological distress  Parental cognitive stimulation  Behavior 0
Child health stress  Parental psychological distress  Parental emotional support  Behavior  .00011
Child health stress  Parental psychological distress  Parental warmth  Behavior -.00026
Child health stress  Parental psychological distress  Parental involvement  Behavior  .00002
117
118
Chapter 5: Discussion
The current study examined the relationship between chronic illness in 
childhood and child behavior problems, including the mediating influences of 
economic strain, child health stress, parental psychological stress, and parenting 
behaviors and the moderating influence of social support.  
Chronic illness and economic strain
The study's findings suggest that the presence of a chronic illness is not 
related to economic strain.  These results are surprising considering the additional 
financial costs that are associated with daily illness care (Asmussen, et al., 1999; 
Barlow, 1997; Kirschstein, 2000; National Chronic Care Consortium, 2003; National 
Institutes of Health, 1996).  However, it is possible that medical insurance covers the 
additional costs of illness care, such as doctor visits and medications.  Children 
living with a chronic illness had an average of 11.0 months of health insurance 
coverage and children not living with an illness had 10.7 months of coverage 
(t(1493.028) = -1.483, p = .143).  These extended periods of health insurance 
coverage may have relieved some economic strain for families.  
Currently, the literature on economic strain and chronic illness is virtually 
nonexistent.  Research has found a relationship between the presence of chronic 
illness and poverty (Egbuonu & Starfield, 1982; Miller, 2000; Vandivere, et al., 
2000; Wise, et al., 2002).  However, as previously mentioned economic strain is not 
equated with poverty (McLoyd, 1990). Therefore, it is possible that there is no 
119
significant relationship between the presence of chronic illness in childhood and 
economic strain. 
When the moderation of social support was examined it was found that mid 
to high levels of social support moderate the relationship between the presence of 
illness and economic strain.  Under conditions of low levels of social support there is 
a marginal positive relationship between the presence of illness and economic strain, 
whereas, under mid to high levels of support this relationship is actually negative.  
This moderation is important because of the positioning of economic strain in the 
model.  Under mid to high levels of social support the link between the presence of 
illness and increased economic strain is severed, preventing the presence of illness 
from influencing variables later in the model.  It is important to note, however, that 
for the total sample there is no relationship between the presence of illness and 
economic strain as positive and negative effects cancel each other out.
Parents that have mid to high levels of social support also reported 
significantly higher economic strain than parents living with lower levels of social 
support.  Parents may seek out assistance (e.g., health insurance for their child, 
employment with better benefits) due to the presence of economic strain.  If the 
parent receives the desired support for economic strain, the parent is creating a buffer 
between the presence of illness and economic strain.  
Chronic illness and child health stress
Chronic illness statistically significantly increased child health stress.  This 
result is as expected.  Children might be at a greater risk of having health stresses if 
120
their health is somehow compromised by a chronic illness.  Previous research has 
found that children living with a chronic illness may face moderate to severe 
limitations and bother caused by the illness (Newacheck & Taylor, 1992).  
Additionally, children living with a chronic illness have a better chance of having 
their physical activities limited due to illness (Benson, 2001; Newacheck & Halfon, 
1998) and having schoolwork and school attendance problems (Msall, et al., 2003; 
Newacheck, et al., 1998; Schwarz & Lui, 2000) than children without an illness. 
Under conditions of mid to high levels of social support, the relationship 
between the presence of illness and child health stress was buffered.  When families 
are living with low levels of support there is a positive significant relationship 
between the presence of illness and child health stress; whereas when mid to high 
levels of support are present there is no relationship between the presence of illness 
and child health stress.  This moderation is important because of the positioning of 
child health stress in the model.  As with economic strain, mid to high levels of 
social support break the link between the presence of illness and child health stress, 
preventing the presence of illness from influencing mediating variables later in the 
model.  
Parents who have higher levels of social support reported significantly higher 
levels of child health stress than people with lower levels of social support.  It is 
possible that because of these stresses parents sought specific assistance with these 
problem areas.  If assistance is sought and received for a particular problem the 
parents may be creating a buffer for their stress.  Or it may be that families living 
121
with an illness are more able or willing to seek help (Cadman, et al., 1991) and when 
they do get the support they use it to decrease stress.
Chronic illness and parental psychological distress 
Contrary to expectations, the presence of a chronic illness did not contribute 
to parental psychological distress.  Instead, illness presence was associated with a 
significant decrease in parental psychological distress.  Some previous research has 
also found that the presence of illness is not related to an increase in parental 
psychological distress (Gerhardt, et al., 2003; Marks, 1998).  Gerhardt and fellow 
researchers (2003) examined families living with juvenile rheumatoid arthritis and 
matched controls to assess parental well-being.  Their results indicate no differences 
between families living with an illness and families not living with an illness 
(Gerhardt, et al., 2003).  Similarly, Marks (1998) examined the effects of caring for 
disabled children, spouses, parents and other kin on individuals who were middle 
age, predominately Caucasian, and high school graduates.  Marks (1998) found that 
global assessment of the caregiving role did not harm caregiver well-being.  It may 
be that, although the presence of an illness is stressful, it does not translate into 
parental psychological distress.   
Additionally, it is possible that parents experienced psychological distress 
after the initial diagnosis of the illness, but adjusted with time.  Researchers have 
noted that mild symptoms of parental psychological distress were found immediately 
after diagnosis of childhood diabetes (Northam, et al., 1996; Thernlund, et al., 1996).  
However, these symptoms had typically been resolved at a 12-month follow up 
122
(Northam, et al., 1996).  An exception was that individuals who had an intense 
emotional reaction at the time of the diagnosis did not resolve symptoms as quickly 
(Thernlund, et al., 1996).  The current research may be capturing parents after they 
have already adjusted to the presence of the chronic illness.
Another possible explanation for the current relationship may have to do with 
help-seeking behaviors.  Parents of children living with a chronic illness may seek 
therapeutic assistance when they are experiencing psychological distress, whereas 
parents not living with a child?s illness may not (Cadman, et al., 1991).  The 
continued involvement with the medical community that is necessary for parents 
living with a child with a chronic illness may assist in the help-seeking behavior or 
may even act as a therapeutic resource.  Additionally, some research suggests that 
families living with a child with a chronic illness have a greater number and broader 
range of coping responses than other families, which may or may not have been 
learned while seeking therapeutic assistance (Spence, 1993).
Help-seeking behavior by parents of children living with a chronic illness 
may expand beyond therapeutic services.  Parents may seek out additional social 
supports to assist them with instrumental tasks and emotional needs.  In the current 
sample, parents of children living with a chronic illness had an average of 3.91 
supports whereas parents without a child living with a chronic illness had an average 
of 3.35 supports (t(1391) = -3.86, p < .001).  These additional supports that parents 
living with a child with a chronic illness have may assist in decreasing their levels of 
psychological distress.  
123
The moderation of social support was found to be in the opposite direction 
expected.  Under conditions of mid to high levels of social support there was no 
relationship between the presence of illness and parental psychological distress, and 
under conditions of low levels of social support the presence of illness was 
significantly negatively associated with parental psychological distress.  It is possible 
that the social support that was accessed or received by the parent was for assistance 
with illness issues and does not translate into support for parental psychological 
distress.  Or it is possible that parents living with low levels of social support have 
already sought help for psychological distress and the results of that assistance are 
being seen here.  
Chronic illness and child behavior problems
The presence of a chronic illness significantly increases the chance of a child 
having behavior problems.  The current findings are similar to a large 
epidemiological German study (n= 4,363) of six year olds that found chronic illness 
to be a significant indicator of child behavior problems (Haffner, et al., 2002).  
Similarly Klinnert and associates (2000) found that parents of children living with 
asthma reported more behavior problems than parents of children not living with 
asthma.  Although research points to the relationship between the presence of a 
chronic illness and child behavior problems, the cross-sectional nature of the current 
data prevents discussion of causation.  Did the children already have behavior 
problems before the diagnosis of illness or did the illness create the behavior 
problems?  
124
Although there was a relationship found between the presence of chronic 
illness and child behavior problems, it is important to examine the index scores for 
children living with and without a chronic illness.  The Child Behavior Problem 
index has a possible score range from 30 to 90. Children living with a chronic illness 
had an average score on the child behavior index of 40.8 and children not living with 
an illness had an average score of 39.3 (t(1320.8)=-3.6, p <.00).  Although these two 
means differ significantly statistically, this difference may not be truly meaningful.
The data do not provide information about how and why illness increases 
child behavior problems.  Do the children complain about their medical regimen, 
pout or have temper tantrums when they have doctor appointments or are prevented 
from engaging in activities with peers?  One could imagine that the child is having a 
normal reaction to adverse stimuli, but parents may interpret these behaviors as a 
form of disobedience or as behavior problems.  Similarly, noncompliance issues 
(e.g., not following a proscribed diet, not taking medicine in a timely fashion) may 
also be considered by parents to be a form of behavior problems (Gartstein, Noll & 
Vannatta, 2000). 
There was no moderation of social support and no significant difference 
between low and mid to high levels of support in the relationship between the 
presence of chronic illness and child behavior problems.  It is possible that if the 
child's rather than the parent?s perceived levels of social support were assessed and 
used in the examination of social support moderation a difference would be found 
among groups.  The social support that was analyzed was perceived parental support.  
125
This support may not "trickle down" to decrease the influence of the presence of 
illness on child behavior problems.  
Economic strain and parental psychological distress
Parents experiencing economic strain have a significantly higher chance of 
also having parental psychological distress.  Economic strain may be a constant 
stressor that is omnipresent.   This constant stress can then lead to psychological 
distress.  There is strong support for this relationship in the literature (Belle, 1990; 
Conger, et al., 1992; Conger, Ebert, Sun, Simons, McLoyd, & Brody, 2002; Elder, 
1998; Jackson, et al., 2000; Kwon & Rueter, 2003; Lorenz, et al., 1994; Simons, et 
al., 1993; Vinokur, et al., 1996; Voydanoff & Donnelly, 1998).  The association 
between economic strain and parental psychological distress has been found in many 
different studies.  For example, Simons, et al. (1993) looked at 451 two-parent 
families from Iowa who were all Caucasians  from rural areas.  The average income 
for these families was $29,642 (Simons, et al., 193).  Conger, et al. (2002) examined 
the influence of economic strain on parental distress in 422 African American 
families (two caregivers in household, 39 percent married, remaining percentage 
were different dyads) from Iowa and Georgia.  The average income for these 
families ranged between $8,000 and $9,000 depending on locality (Conger, et al., 
2002).  Because the current data are cross-sectional, it is impossible to know which 
came first, the strain or the distress; however, the results are consistent with a large 
body of previous longitudinal research.
126
The moderation of social support was contrary to expectations.  Economic 
strain has a larger increase on parental psychological distress under conditions of 
mid to high levels of social support than under conditions of low levels of social 
support.  Parents reporting mid to high levels of social support have significantly 
higher levels of economic strain than parents living with low levels of social support.  
This difference in index scores may create the larger path magnitude for mid to high 
levels of social support.  Additionally it is possible that higher levels of social 
support are more costly in time and money than lower levels of social support.  For 
example, reciprocating birthday presents or meals when the family is already 
experiencing economic strain may add to the parental psychological distress.  
Child health stress and parental psychological distress
Higher levels of child health stress significantly decrease the chances of 
parental psychological distress.  This finding is contrary to expectations.  It is 
possible that parents may seek or receive additional supports when their child is 
experiencing health stress.  The data indicate that parents with higher levels of social 
support also have children living with more health stress.  This additional support 
may prevent the child's health stress from negatively influencing parental 
psychological distress.
Another possible explanation for the relationship between child health stress 
and parental psychological distress may be the items that make up the child health 
stress index.  It is possible that the items were not severe enough to create true 
psychological distress.  More severe child health stresses might include the number 
127
of hospitalizations and emergency room or urgent care center visits that occur within 
a year due to the illness or questions about the complications the illness may create. 
Alternatively, the items comprising the child health stress variable may be more 
related to the child's personal experience of health stress.  This personal experience 
of stress may not be shared with the parent.  Items that might tap child health stress 
that could influence parental psychological distress may include the daily hassles 
created by the presence of illness or the amount of time that is required weekly to 
maintain the illness.    
If the child's health stress is severe it could be that the parent may not have 
the time or ability to have psychological distress.  Similar relationships have been 
observed in people living in war zones (Friedman, 1995).  For example, doctors 
found that individuals living in Beirut in the early 1980s, who were also living with 
depression, had improvements in their mental health during the war, but their mental 
health returned to prewar levels after the war had ended (Friedman, 1995).  The 
individuals living with depression stated that they had to focus on survival and not 
on their mental status (Friedman, 1995).  
There was no moderation of social support and no significant difference 
between low and mid to high levels of support in the relationship between child 
health stress and parental psychological distress.  These results are interesting 
because in the total model there is a negative, significant relationship between the 
two variables, but when social support is examined the relationship disappears.  It 
128
may be that any form of support, whether it be low or mid to high levels, removes 
the negative influence of child health stress on parental psychological distress.
Economic strain and parental cognitive stimulation
Economic strain is significantly related to lower levels of parental cognitive 
stimulation. These findings are consistent with the existing literature (Evans, 2004; 
Guo & Harris, 2000; Miller & Davis, 1997). Economic strains may cause families to 
adjust their discretionary spending.  Therefore, families that are experiencing 
economic strain may not have the ability to provide a cognitively stimulating 
environment, such as books, musical instruments, equipment for hobbies, or museum 
visits (Evans, 2004; Garrett, et al., 1994; Klebanov, et al., 1994).    
Under conditions of mid to high levels of social support the relationship 
between economic strain and parental cognitive stimulation was buffered.  Findings 
reveal that parents with mid to high levels of social support provide significantly 
more cognitive stimulation to their children than parents living with low levels of 
social support. When parents have low levels of social support and economic strain 
they may not have the time or ability to provide their child with cognitive 
stimulation.  Parents with mid to high levels of social support may utilize the support 
system to supply cognitive stimulation to their child.  For example, individuals in the 
system may take the child to a museum, encourage hobbies, or even provide 
cognitive materials (e.g., books, craft materials).
129
Economic strain and parental involvement
Parental involvement is not directly decreased by the presence of economic 
strain.  Thus, the mere presence of economic strain does not influence how involved 
parents are with their children.  The parental involvement scale includes six home-
based items (13 items total) that do not require resources and are based on household 
maintenance (e.g., washed or folded clothes together and cleaned the house 
together).  Parents may continue to do chores and find time to participate in 
additional "fun" activities with their children even with the presence of economic 
strain.  This indicates that parents may be successful at preventing financial strains 
from directly interfering in their time spent with their children.  
Parental involvement may be influenced by other factors that are related to 
economic strain, but were not examined in the current study.  For example, parents 
may need to work overtime or take additional jobs to handle economic strains.  
These hours away from the home may significantly decrease the level of 
involvement parents have with their children.   
There was no moderation of social support and no significant difference 
between low and mid to high levels of support in the relationship between economic 
strain and parental involvement.  The path between the two variables is significant 
for both low and mid to high levels of support.  This is interesting because there is no 
relationship observed in the total model.  It may be that when economic strain is 
present individuals behave similarly (e.g., work more hours) whether they have low 
or mid to high levels of social support.
130
Child health stress and child behavior problems
Health stress in children's lives was found to be related to significantly more 
behavior problems.  Living with a chronic illness may already be difficult for a child, 
but with the addition of health stresses (e.g., activity limitations, and schoolwork and 
attendance problems) a child may become frustrated and act out (Ellerton, et al., 
1996) or become withdrawn and quiet (Rodder, et al., 2003).  For example, if a child 
is unable to participate in physical activities, classmates may tease him/her 
(Bukowski & Anita, 2000; Underwood, 1997).  If the child is absent for long periods 
of time due to health problems his/her relationship with other children may not 
develop and the child living with the illness may become withdrawn or considered an 
outcast by fellow peers.  
The moderation of social support was contrary to expectations.  Under 
conditions of mid to high levels of social support child health stress is positively and 
significantly related to child behavior problems, whereas under conditions of low 
levels of social support there is only a marginal positive relationship between the 
variables.  Parents who reported higher levels of social support also reported higher 
levels of child health stress.  It is possible that the support system activated by the 
family to assist with higher levels of child health stress have not yet begun to work.  
Or it is possible that parental perceived support does not successfully moderate the 
relationship between child health stress and behavior problems, but higher levels of 
perceived social support received by the child would moderate the relationship.
131
Parental psychological distress and parenting behaviors
When parents are living with psychological distress their levels of positive 
parenting significantly decrease.  Specifically higher levels of parental psychological 
distress were associated with less cognitive stimulation, emotional support, and 
warmth.  It may be difficult for parents to nurture their children when their own lives 
are in emotional turmoil (Conger, et al., 1992; Simons, et al., 1992; Voydanoff & 
Donnelly, 1998).  These findings are consistent with the existing literature on 
parental psychological distress and parenting behaviors (Conger, et al., 1992; 
Conger, et al., 2002; Downey & Coyne, 1990; Jackson, et al., 2000; Simons, et al., 
1992; Voydanoff & Donnelly, 1998).  The literature that supports these findings has 
drawn from samples that are nationally representative (Voydanoff & Donnelly, 
1998), African American (Conger, et al., 2002), and Caucasian (Simons, et al, 1992).  
Unexpectedly, there was no relationship between parental psychological 
distress and parental involvement.  It is possible that parental involvement may not 
require nurturing behaviors, which are diminished when psychological distress is 
present (Conger, et al., 1992; Simons, et al., 1992; Voydanoff & Donnelly, 1998).  
Nurturing may not be included in many of the involvement items included in the 
study measure, such as household chores and errands.  These activities are necessary 
for family functioning and cannot stop if psychological distress is present.  
Additionally this form of involvement may place minimal additional stress on the 
parent.  
132
The moderation of social support on the relationship between parental 
psychological distress and parenting behaviors is mixed.  The relationship between 
parental psychological distress and parental cognitive stimulation and parental 
psychological distress and parental warmth indicates that there is no relationship 
between the variables under conditions of low levels of social support, but under 
conditions of mid to high levels of support there is a significant negative relationship 
between the variables.  It may be that when parents have high levels of social support 
they must expend energy to that network which may detract from their positive 
parenting behaviors.  Also, it may be that individuals that have lower levels of social 
support actually have better quality supports for their problems, which may then 
prevent their problems from influencing parenting behaviors.
The relationship between parental psychological distress and parental 
emotional support is negative and significant for both support groups.  However, 
under conditions of low levels of support the magnitude of the negative relationship 
between psychological distress and emotional support is greater than under 
conditions of mid to high levels of social support.  The emotional support scale taps 
discipline, chores, and how the primary caregiver speaks with the child.  It is 
possible that under conditions of mid to high levels of social support these more 
basic parenting behaviors are not disturbed as much as under conditions of low levels 
of social support.  
There was no moderation of social support and no significant difference 
between low and mid to high levels of support in the relationship between parental 
133
psychological distress and parental involvement.  This relationship between parental 
psychological distress and parental involvement was also observed in the total 
model.  Thus, the moderation of social support for the relationship between parental 
psychological distress and parenting behaviors is mixed, with one expected 
relationship and three unexpected relationships.
Parental psychological distress and child behavior problems
Parental psychological distress was found to significantly increase child 
behavior problems.  These findings are consistent with the current literature on 
parental psychological distress and behavior problems in children (Conger, et al., 
1995; Conger, et al., 2002; Downey & Coyne, 1990; Drotar, 1997; Jackson, et al., 
2000; Lee & Gotlib, 1989; Richters & Pellegrini, 1989; Rogers & Floyd, 2003).  The 
samples from this literature vary from single, low income, African American 
mothers (Jackson, et al., 2000) to predominately two-parent, working class, 
Caucasian families (Conger, et al., 1995).  Parents clearly have a direct effect on 
their children's behavior, and also have indirect influences through their parenting 
behavior.  Current findings indicate that parental psychological distress also 
significantly influences child's behavior problems indirectly through parenting 
behaviors.
The moderation of social support on the relationship between parental 
psychological distress and child behavior problems was contrary to expectations.  
There is a larger relationship between the variables under conditions of mid to high 
levels of social support than under conditions of low levels of social support.  This 
134
relationship may speak to the type of support sought and received by the parent.  It is 
possible that the parent seeks assistance for dealing with the presence of illness, but 
has not sought help for their individual well-being.  This may help to explain why 
higher levels of support are associated with a larger relationship between parental 
psychological distress and child behavior problems.  
Parenting behaviors and child behavior problems
In the current study parental cognitive stimulation and warmth significantly 
decreased child behavior problems.   These findings are consistent with the current 
research on parental cognitive stimulation and warmth and their relation to behavior 
problems (Eisenberg, et al., 2001; Jackson, et al., 2000; Minden, 1999; Moris, et al., 
2003; Parcel & Menaghan, 1993; Rogers, et al., 1991; Yeung, et al., 2002; Zhou, et 
al., 2002).  The samples from this literature range from predominantly middle class, 
Caucasian families (Eisenberg, et al., 2001) to single, African American mothers 
(Jackson, et al., 2000).  
Surprisingly, parental involvement was related to significantly higher levels 
of child behavior problems.  Based on the current data it is impossible to sort out 
causation.  Are the parents increasing their involvement with their child because the 
child has a behavior problem to begin with or is the involvement of the parent 
creating the behavior problems?  It is also possible that a third factor that was not 
assessed is contributing to both parental involvement and child behavior problems.  
The moderation of social support on the relationship between parenting 
behaviors and child psychological distress is mixed.  The relationship between 
135
parental cognitive stimulation and child behavior problems is contrary to 
expectations.  Under conditions of low levels of social support cognitive stimulation 
is associated with a significant decrease in child behavior problems.  However, under 
mid to high levels of support there is no relationship between the variables.  This 
relationship is surprising because parents receiving mid to high levels of social 
support also report providing significantly more cognitive stimulation than parents 
receiving low levels of social support.  It is possible that although more cognitive 
stimulation is provided it may not be considered quality or desired stimulation by the 
child.  For example, it could be that when mid to high levels of social support are 
present those supports are interacting with the child in cognitively simulating ways.  
Whereas when the parent receives low levels of support, the parent may be the one 
providing the stimulation.  It is possible that parental simulation may be more helpful 
for children than when others provide the cognitive stimulation.
There was no moderation of social support and no significant difference 
between low and mid to high levels of support in the relationship between parental 
emotional support and child behavior problems.  This relationship is consistent with 
observations in the total model.  
The relationship between parental warmth and child behavior problems is 
negative and significant for both support groups.  However, under conditions of low 
levels of support the magnitude of the negative relationship between warmth and 
behavior problems is greater than under conditions of mid to high levels of social 
support.  Parents may be receiving emotional support from their social network, this 
136
support may then parlay into their ability to provide more parental warmth.  Findings 
indicate that there are significantly higher levels of parental warmth when mid to 
high levels of social support are present than low levels of social support.  
The relationship between parental involvement and child behavior problems 
is contrary to expectations.  Under conditions of mid to high levels of social support 
parental involvement is positively and significantly related to child behavior 
problems, while under conditions of low levels of social support there is no 
relationship between the variables.  It is possible that parents receiving higher levels 
of support are activating their support networks to assist in dealing with child 
behavior problems.  This larger network may then allow parents to be more involved 
with their child.  However causation is unclear.  Did the involvement come first or 
the behavior problems?  Thus, the moderation of social support for the relationship 
between parenting behaviors and child behavior problems is mixed, with one 
expected relationship and three unexpected relationships.
Chronic illness and child behavior problems
The deficit model says that the mere presence of a chronic illness creates 
devastating effects for children.  The current research did not find full support for 
this assumption.  Although there is a significant link between the presence of a 
chronic illness and child behavior problems, the magnitude of that relationship is 
smaller than other variables.  The standardized coefficient for the path between the 
presence of a chronic illness and child behavior problems is 	 = .079, whereas the 
coefficient between child health stress and child behavior problems is 	 = .306 and 
137
parental psychological distress and child behavior problems 	 = .326.  Therefore, 
when child health stress and parental psychological distress are present they 
contribute more to child behavior problems than the presence of a chronic illness.  
This speaks to the importance of mediating factors and the need to examine the 
magnitude of paths as well as the significance level.  
The effect size of the presence of illness on child behavior problems is 
important to examine.  Cohen (1965) defines effect size as the extent to which a 
relationship exists.  A small effect size is considered to be .25 (Cohen, 1965).  The 
effect size for the direct relationship between the presence of illness and child 
behavior problems is very small (.165).  Therefore, it could be argued that the 
relationship between the presence of illness and child behavior problems is also very 
small.
The indirect effects of chronic illness on child behavior problems were 
explored and are not significant.  This is because the effects of illness on economic 
strain are practically zero and the effects of illness and child health stress on 
psychological distress are the reverse of that expected.
Overall the current research finds that there is a connection between the 
presence of illness and child behavior problems, but, based on the small effect size 
and small, nonsignificant indirect effects, I am not convinced that illness presence is 
as detrimental to children as was once thought.  The presence of illness in childhood 
does bring additional challenges and strains, but families and children are resilient 
and find ways to prevent the illness from creating negative outcomes.  The current 
138
research points to the need for additional research to further assess the relationship 
between chronic illness and child behaviors. 
Social support
Social support was thought to moderate or buffer each of the relationships in 
the current research.  The findings reveal that this was not the case.  Instead 5 of the 
18 possible relationships were successfully buffered by social support.  Eight of the 
18 relationships show a reverse buffering with lower levels of social support being 
more beneficial.  However, the fact that buffering was observed is significant 
considering the limitations of the measure and the use of a median split.
The social support measure that was used in this research had many 
limitations.  The social support index was based on two items.  These two items, one 
tapping advice and emotional support and the other physical help and assistance, 
clearly did not tap all of the dimensions of social support.  Along with the limited 
number of items, participants were limited in the number of individuals they could 
include in their support network, which decreased variability.  The measure also did 
not provide intensity information, such as how helpful these people are to the 
participant.  Participants may have 10 people in their support network, but these 
individuals may not actually provide necessary assistance.
The social support index was split into two groups using a median split.  It is 
possible that splitting the participants in a different way would yield more buffering 
affects of social support.  For example, the groups might be split into no support (a 
zero score on the index) and some support (a score of one or more on the index).
139
Utilizing a more comprehensive measure and creating a more meaningful 
group split may assist in forming a more clear picture of the moderating influence of 
social support, but it would also be important to understand the reason parents are 
contacting social supports.  For example, could it be that people seek out help for 
particular problems or do family members and friends insist on helping parents 
through tough times even if no assistance was asked for?  These supports may 
intrude on family life, disrupting family patterns as well as providing support.  The 
support that is provided, whether sought for or given, may create additional stresses 
for parents because of the costs associated with reciprocity (Taylor, Chatters, Tucker 
& Lewis, 1990).  It may be stressful for parents already living with strain to 
reciprocate the support they are given.  
Finally, it may be that advice and emotional support buffer some of the 
relationships in the model and actual help and assistance buffer other relationships.  
In this study both types of support were considered together.  Future research should 
examine the potential buffering effect of the two different forms of support and other 
potential forms as well.
Even with these limitations social support was found to buffer two of the 
most important relationships in the model, presence of illness and economic strain 
and presence of illness and child health stress.  These relations are the most 
important because they are part of the basis for the model.  With the relationship 
between presence of illness and economic strain and presence of illness and child 
health stress cut, the presence of illness does not influence additional variables in the 
140
model.  Therefore, social support is important for all families, but particularly for 
families living with a chronic illness.  If the family is able to receive sufficient 
supports the chronic illness may not negatively influence later family and individual 
processes such as psychological well-being and parenting behaviors.    
It is important to mention that there are problems with interpreting the 
buffering or moderating influence of social support.  Because the data are cross-
sectional there is speculation as to which came first, the stresses or the social 
support.  Did the parents have the support and then the stress entered their life or did 
the stress occur and the parents sought out support for assistance?  Future research 
that examines longitudinal data will help to answer this question.
Ecological theory
The ecological theoretical model was used in the current research because it 
best addresses the contextual factors associated with chronic illness and how the 
illness affects the individual and the family (Kazak & Christakis, 1994; Vinson 
2002).  The ecological model states that the environment and families are mutually 
shaping systems, each changing with time, and adapting to changes in the other 
(Bronfenbrenner, 1986).  These systems interact and influence each other in a 
reciprocal fashion.  This interaction was observed in the direct and indirect effects in 
the current research.  Both the direct and indirect effects point to the interaction of 
microsystems.  For example, children?s health stress is associated with parents? 
psychological distress and the parents? distress is then related to child behavior 
141
problems.  The interaction observed in the current research lends support to the 
ecological model.
The ecological model explains that events external to the family unit are just 
as influential as the activities inside the family unit and inside the individual family 
member (Bronfenbrenner, 1986).  The model acknowledges there are barriers to 
optimal family functioning both on the individual level (e.g., mental health) and the 
employer and government level (e.g., health insurance coverage, family leave).  The 
interaction between the family and their various environments produces both 
opportunities and risks.   
There are opportunities in the microsystem (e.g., parenting behaviors) and 
mesosystem (e.g., social support).  These opportunities may help to decrease or 
lessen the influence of risk.  Risks were examined in the microsystem and include 
child health stress and parental psychological distress.  The ecological model 
recognizes and the current research supports that systems interact allowing risks and 
opportunities from one system to influence the other (Bronfenbrenner, 1986).    
In addition to the microsystem and mesosystem, there are opportunities and 
risks in the exosystem and macrosystem even though they were not examined in the 
current research.  For example, flexible work schedules (exosystem) might be 
considered opportunities because they may help parents to schedule doctor visits or 
tend to their child when they become ill.  Risks might include stigma and stereotypes 
(macrosystem) about illness that are present in society.  Clearly all of the systems 
142
(micro, mesos, exo, macro) and opportunities and risks interact in the current 
research, thus further lending support to the ecological model. 
Family stress theory of economic hardship
The family stress theory of economic hardship posits that economic strain 
directly influences parental psychological distress and indirectly influences parental 
conflict, positive parenting, and finally child behavior (Conger, Rueter, & Conger, 
2000).  Researchers have found strong support for this model (Conger, et al. 2000; 
Conger, et al., 2002; Jackson, et al., 2000; Simons, et al., 1993; Vinokur, et al., 1996; 
Voydanoff & Donnelly, 1998).  Parental conflict was not included in the model that 
was analyzed in the current study, but each of the other relationships suggested in the 
family stress theory was supported.  This research adds to the family stress theory 
with the addition of child chronic illness and child health stresses as potential sources 
of family strain.  
Implications for Researchers, Practitioners, and Policy Makers
The findings from this study have implications for researchers, practitioners, 
and policymakers.  Researchers should continue to investigate mediating and 
moderating variables when examining the influence of illness on individuals and 
families.  The use of structural equation modeling to conduct these investigations 
may assist in finding previously unexplored relationships and allow for testing of 
alternative models.  
Based on the current research, children living with a chronic illness have a 
greater chance of experiencing behavior problems through the mere presence of 
143
illness and child health stress than those without a chronic illness.  To reduce the 
negative effects of health stress on children, practitioners might use individual or 
group therapy to discuss the stresses and strains of living with a chronic condition.  
Preferably, support groups with other children living with the same or similar illness 
could be formed.  Interaction with others who are living with a similar illness can 
help children to not feel alone.  Practitioners might begin the interaction with other 
children who are currently living with the illness at the time of diagnosis or shortly 
thereafter.  Additionally, teaching the children problem solving skills such as 
emotion regulation (LeBlanc, Goldsmith & Patel, 2003) and disease self-
management (Hill-Briggs, 2003) may assist children to decrease stresses and lead to 
a prevention of acting out behaviors.
Practitioners should assess developmental aspects of a child living with a 
chronic illness before attributing behavior problems to the illness alone.  Many 
parents do not understand that some forms of acting out are normative to the child's 
developmental age.  The practitioner should normalize the child's behavior if at all 
possible.  
Compliance with medical regimens may be arduous for both children and 
parents.  Parent-child conflict can occur when the child refuses or acts out when a 
treatment is necessary.  Practitioners should discuss with both children and parents 
how to make compliance issues easier.  For example, children may feel babied if 
they are unable to assume control of pieces of their own medical regimen.  
144
Practitioners could discuss age-appropriate periods when parents can transfer 
medical regimen responsibilities to the child.  
Practitioners also need to be aware of the positive influence social support 
can have for families living with a chronic illness.  Families should be encouraged to 
join local affiliations of the national nonprofits that correspond with the illness they 
live with (e.g., Juvenile Diabetes Foundation, The Allergy and Asthma 
Network/Mothers of Asthmatics).  Practitioners also might teach families how to ask 
for help effectively and how to eliminate social supports that are more draining than 
supportive. 
This research points to the influence of parental psychological distress on 
child behavior problems.  Practitioners should be aware of this relationship and 
assess parental psychological distress when a child presents with behavior problems.  
If parental psychological distress is found, the parent should be encouraged to 
receive treatment. 
Parenting behaviors were found to be related to a decrease in child behavior 
problems.  Practitioners might create group classes to teach effective parenting 
behaviors.  For example, parenting classes might focus on how to cognitively 
stimulate a child.  The class might suggest parents set aside time each day to read a 
book to their child or the class might teach how to plan a successful trip to a local 
museum. 
Parental psychological distress was found to be related to economic strains.  
Program planners might create short-term, community support groups to help parents 
145
when there are economic strains.  These programs could provide emotional as well 
as instrumental support (e.g., how to apply for the Earned Income Tax Credit or 
where to go for low cost health care).  Childcare and transportation would need to be 
provided for such a program.  The supports that would be provided might assist in 
decreasing economic strain and eventually parental psychological distress.  
Policymakers might also concentrate on economic strain or health insurance 
issues.  Policymakers may decrease economic strain by subsidizing and locating 
housing, increasing and making more frequent payments of the Earned Income Tax 
Credit, and/or providing health insurance to low income families.  Health insurance 
that is provided to families often does not include a mental health component.  
Mental health treatment coverage should be included in health insurance policies for 
both children and their parents.    
Based on the current research, economic strain is significantly associated 
with increases in parental psychological distress, which in turn significantly 
increases child behavior problems.  Child behavior problems might lead to 
educational difficulties (Hofferth & Curtin, 2004) that can eventually lead to 
difficulties later in life.  Additionally, adults that had behavior problems in childhood 
tend to provide less cognitive stimulation to their children than adults who did not 
experience behavior problems in childhood (Saltaris, et al., 2004).  Therefore, 
prevention of the possible downward spiral created by economic strain would be 
advisable.  
146
Researchers, practitioners, and policymakers should also focus on the 
resiliency of families.  This research points to positive parenting behaviors such as 
parental cognitive stimulation and warmth decreasing behavior problems.  
Additionally, some individual and family strains (e.g., economic strain, child health 
stress) were buffered by the presence of social support.  These forms of resiliency 
should be encouraged and expanded through research, training of practitioners to use 
culturally competent models (Olkin, 1999), and the implementation of family 
friendly policies.  Robert Louis Stevenson, who lived with pulmonary tuberculosis, 
might have explained resiliency in chronic illness best, "'life is not a matter of 
holding good cards, but of playing a poor hand well'" (Isaacs & Sewell, 2003, p. 
235).
Limitations
In interpreting this study's findings, there are several limitations that should 
be addressed.  First, the data are cross-sectional, which prevents conclusions about 
causality.  This is particularly important with the moderation of social support.  The 
perennial problem of which came first, the social support or the stress cannot be 
definitively answered.  Likewise, some parents may experience constant trait like 
psychological distress such as a generalized anxiety disorder or dysthymia (low 
grade depression).  However, because the data are one point in time, the direction of 
relationships involving traits like psychological distress cannot be analyzed. 
Second, the current sample has more two-parent households than the United 
States population did in 1997.  Ninety-four percent of the current sample were in two 
147
parent households in 1997, leaving six percent of "other" or single parent families.  
In 1997, 24 percent of U.S. households were single parent families (Bryson & 
Casper, 1998).  Single parent families may differ from dual parent families in their 
level of economic strain, parental psychological distress, parenting behaviors, and 
levels of social support.  The education of the head of household and number of 
children in the household were controlled for during analysis.  Nonetheless, the 
results of the current study should be applied to two-parent families.   
Third, the gender of the child was not controlled for during analysis.  
Findings indicate that there were more females in the nonchronic sample and more 
males in the chronic sample.  It is possible that there are differences in child health 
stress based on the gender of the child.  For example, activity limitation for boys may 
cause them to act out more.  This acting out may be a way for boys to expel 
additional energy that would normally be used during activities.  Child behavior 
problems may be different for boys and girls with boys using externalizing behaviors 
and girls using internalizing behaviors.  However because externalizing and 
internalizing behavior problems were combined in the child behavior problems 
measure, gender differences may not be seen.  
Fourth, the chronic illnesses that are assessed in the current research, anemia, 
asthma, diabetes, and otitis media do not tap all of the chronic illnesses in childhood.  
The illnesses that were analyzed may be more benign than other unexamined 
illnesses.  For example, recent years have seen changes in the illnesses that are 
considered chronic.  Leukemia in children and HIV may now be considered chronic 
148
conditions due to the creation of medications and treatment protocols.  The inclusion 
of these illnesses that may be considered more chronic might change the outcome of 
the current research.
Fifth, the use of a preexisting data set limits the items and measures available 
for use in analysis.  The PSID-CDS was not created to fully explore chronic illness 
in children.  Therefore there are important pieces of information about diagnosis, 
time since diagnosis, daily hassles of illness, compliance, and the general attitude 
toward the illness that are not present.  For example, families that have been living 
with the illness for many years may have adapted to the illness.  This adaptation over 
time may then create less parental psychological distress.  If more information on the 
chronic illness were available, the research would provide a more vivid depiction of 
child health stress and the influence of illness on family and individual well-being.    
Further, when an instrument is present in a preexisting data set that examines 
a variable of interest, the measure may not be sufficient for the proposed research.  
For example, the social support measure restricts respondents from reporting more 
than five supports for instrumental and emotional support.  This restriction limits the 
variability of the index and the index response set prevents participants from 
reporting how helpful that support is to them.  Possibly the Family Support Scale 
(Dunst, Jenkins & Trivette, 1984) would be a more appropriate measure of both 
emotional and instrumental support for raising children.  Support parents receive for 
caring for their children may not transfer to personal support.  Therefore an 
additional measure, such as the Norbeck Social Support Questionnaire (Norbeck, 
149
Lindsey & Carrieri, 1981), which examines both the number and helpfulness of the 
support, might be used to assess parents' personal experience of perceived levels of 
social support.  
Additionally, with the use of preexisting data there was a reliance on existing 
indices.  Although they are adequate, they do not provide intensity information (e.g., 
how stressful or how much bother you experience with economic strain).  For 
example, the economic strain index might have asked first if you had the strains and 
then how stressful the strain is for the family.  Some individuals may face economic 
strains, but have mechanisms (e.g., coping) that prevent the strains from becoming 
detrimental.  On the other hand there are individuals who may lack these 
mechanisms and find the strains overwhelming.  This perception of the degree of 
strain, rather than merely the presence of strain may provide a more clear 
representation of individual and family functioning.   
Sixth, the social support and parental involvement measures both combine 
two different dimensions of the variable.  The social support index collapses both 
instrumental and emotional support into one score.  However, the different types of 
support may moderate paths differently based on support type.  For example, 
instrumental support may influence the path between economic strain and 
psychological distress more than emotional support.  Similarly, the parental 
involvement measure assesses both fun activities and chores.  Chores may create 
arguments between parent and child and may cause children to act out in order to 
avoid completing chores.  On the other hand, parents that are involved in fun 
150
activities with their children (e.g., playing games, reading) may cause children to act 
out less.  Therefore, interpretability of social support and parental involvement is 
limited due to the combination of dimensions.
Directions for future research
The current study addressed issues of the presence of a chronic illness and 
child behavior problems while examining the mediating influences of economic 
strain, child health stress, parental psychological distress, parental cognitive 
stimulation, parental emotional support, parental warmth, and parental involvement.  
Additionally the influence of low and mid to high levels of social support were 
explored.  However, further research is needed to investigate the aforementioned 
variables more thoroughly and to include additional variables that may influence the 
relationship between the presence of illness and child behavior problems.   
First, future research must expand the examination of illness to include more 
than the mere presence of illness.  Illness should be assessed through measures of 
severity based on doctor, parent, and child ratings, and child health stresses such as 
hospitalizations, compliance/management issues, daily hassles, and attitude toward 
illness.  The time since diagnosis should also be assessed.  By providing additional 
measures of illness, the explanation for the connection between the presence of 
illness and child behavior problems may become clearer.  
Second, future research should examine the differences or similarities between 
children living with an illness under the age of six.  Younger children may be 
diagnosed with chronic conditions but the adjustment may be different for families 
151
when the child does not have full language capabilities, as opposed to school age 
children. 
Third, economic strain should be examined from both a perceived and actual 
experience of economic hardships.  Some families may have reported "none of the 
above" on the economic strain index, but perceive strain.  It is also possible that 
families who experience actual economic strain do not perceive themselves to have 
economic strain.  Future research should use a combination of perceived and actual 
experience of economic hardship to more fully examine the possible detrimental 
affects of strain on parental psychological distress and parenting behaviors.
Fourth, future research should further examine parental involvement.  In the 
current research, parental involvement with the child included both chores and fun 
activities.  Chores, although done together, may be met with tantrums or fights 
between parent and child; whereas, fun activities such as reading a book or playing a 
game may be enjoyable for all involved.  Future research should separate the two 
types of involvement to better examine the relationship between parental 
involvement and child behavior problems.
Fifth, research should obtain multiple reports of behavior problems and not 
simply rely on parental reports.  Research has found that mothers report more 
aggressive behaviors in children living with a chronic illness than teachers and peers 
(Gartstein, et al., 2000).  This may be due to medial compliance issues (Gartstein, et 
al., 2000) or environmental factors.  Nonetheless, observational or multiple 
respondents would help to insure the reliability of parental responses. Additionally, 
152
future research might explore both internalizing and externalizing behavior problems 
because it is possible that parental psychological distress and parenting behaviors 
influence these two behavior problems in differing ways.   
Sixth, the measure of social support should be expanded to include both the 
support perceived by the parent and the perceived support the child receives.  
Parental support should be broken down into perceived assistance in raising the child 
and personal perceived support.  The support the child receives should also be 
assessed since it has been found to improve feelings of acceptance by peers (La 
Greca, et al., 1995) and adherence to medical regimens (Hauser, et al., 1990; La 
Greca, 1992).  
Additionally, instrumental and emotional support might be split into two 
separate measures of support.  In the current research both supports were collapsed 
into a measure of global social support.  However, each of these support types may 
influence paths differently.  For instance, instrumental support may act as a better 
buffer against economic strain than emotional support.  Further, it would be 
important for future research to assess how helpful that support is to the individual 
and family.
Seventh, future research should continue to examine mediating and moderating 
variables that influence the relationship between the presence of illness and child 
behavior problems.  Such variables might include children's school performance, 
peer and sibling relations, neighborhood measures, and family conflict.  For 
example, family conflict should also be considered when examining illness and child 
153
behavior problems.  Family conflict has been found to increase childhood anxiety 
and depression (Jekielek, 1998), hospitalizations in children living with asthma 
(Chen, Bloomberg, Fisher & Strunk, 2003), and has been a risk factor for behavior 
problems for children in general (Vandewater & Lansford, 1998), and in children 
living with sickle cell disease in specific (Thompson, et al., 1999; Thompson, et al., 
2003).  Additionally, parenting behaviors are negatively influenced by family 
conflict (Gerard & Krishnakumar, 2003; Klausli & Owen, 2003).  
Eighth, longitudinal data should be collected in future research.  Longitudinal 
data would assist in answering questions of causation.  Fortunately, a second wave of 
data were collected in 2002, which will permit longitudinal analysis of this study in 
the future.
Recommendations
 Children living with a chronic illness may need assistance with adjusting to and 
coping with their illness.  This assistance may prevent child behavior problems.
 Children living with a chronic illness may need assistance in responding to their 
own health stress.
 Parents need to be aware of their own mental health and seek assistance when 
necessary.  
 Parents living with psychological distress need assistance to improve their mental 
health and prevent their distress from interfering with positive parenting 
behaviors.
154
 When a child is having behavior problems, parental psychological distress should 
be assessed.
 Families living with a child living with a chronic illness who have low levels of 
social support may need assistance to increase their number of supports in order 
to prevent the presence of illness from influencing economic strain and child 
health stress. 
Conclusions
The current study adds to the literature on families and children living with a 
chronic illness.  Some previous research has focused specifically on the relationship 
between chronic illness in childhood and increased behavior problems.  While the 
current research did find a relationship between illness and behavior problems, the 
additional mediating factors of child health stress and psychological distress were 
found to contribute more to child behavior problems than the mere presence of 
illness.  
The buffering effect of social support was also examined, finding that under 
conditions of mid to high levels of social support the relationship between illness and 
economic strain and illness and child health stress are buffered.  This buffering effect 
early in the model can indirectly funnel down to other processes, such as parenting 
behavior.  Clearly, mediating and moderating family and individual influences must 
continue to be examined to understand the relationship between chronic illness and 
child behavior problems.
155
Appendix A: Introductory Letter to Participants
Spring 1997
Dear________________________,
Thank you for agreeing to participate in the Child Development Supplement study.  
This study will help researchers learn how the nation's children are faring over time.  
This project is funded by the national Institute of Child and Health and Human 
Development to gather information on how financial and social resources in the 
home, community and school affect children's health and achievement.  We have 
enclosed a brochure containing additional information about the study.
One the date recorded below, the interviewer you recently talked with will come to 
your home to interview both you and your child/ren.  The name of each child chosen 
for an interview is listed below.  It is important that each child listed is present on the 
day of the interview at the scheduled time.  The interview will take approximately 
one and a half hours per child.  As part of this interview, the interviewer will be 
administering a standard achievement assessment.  We want to make sure that we are 
able to test each child as accurately as possible.  Because distractions tend to lower 
test scores, it would be best if the interview could be done in a quiet place with as 
few distractions as possible.
The Time Diary your interviewer mentioned is enclosed with this letter.  Please note 
that there are two diaries for each child.  Each diary should be completed on the day 
specified on the front cover.  We appreciate your taking the time to carefully 
complete the diaries on each day specified.  Your interviewer will pick them up at 
the time of your child's interview. 
In appreciation for your participation in our study, we will be sending you a check 
for $25.00 and each child who participates will receive a small gift.  Your 
interviewer has pledged to protect your privacy.  Any information given in your 
interview and in your time diaries will be treated completely confidentially.  If you 
have further questions, please don't hesitate to call us at our toll free number, 1-800-
759-7947.
Sincerely, 
Sandra Hofferth, Ph.D. Pamela Davis-Kean, Ph.D.
Co-Director Project Manager, Child Development Supplement
Family Economics Study Family Economics Study
University of Michigan University of Michigan
156
Your interviewer will be: ________________________________________
Interview Date and Time:________________________________________
157
Appendix B:  Economic Strain Index
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)
1. Sold possessions or cashed in life insurance
2. Postponed major purchases
3. Postponed medical care
4. Borrowed money from friends or relatives
5. Applied for government assistance
6. Filed for or taken bankruptcy
7. Fallen behind in paying bills
8. Obtained a loan to consolidate or pay off debts
9. Had a creditor call or come to see you to demand payment
10. Had your wages attached or garnished by a creditor
11. Had a lien filed against your property because you could not pay a 
bill
12. Had your home, car or other property repossessed
13. Moved to cheaper living quarters
14. Moved in with other people
15. Sent one or more of your children to live with someone else
16. None of the above
158
At the end of the month, do you end up with some money left over, just enough to 
make ends meet, or not enough money to make ends meet?
How many of the past 12 months was (CHILD) covered by health insurance or any 
other kind of health care plan?
159
Appendix C: Child Health Stress Index
Does (CHILD) currently have any physical or mental condition that would limit or 
prevent (his/her) ability to?(yes/no)
1. do usual childhood activities such as play, or participate in games or sports?
2. attend school (preschool or day care) regularly?
3. do regular school work?
In general, would you say (CHILD'S) health is excellent, very good, good, fair, or 
poor?
160
Appendix D: Non-Specific Psychological Distress Scale
During the past 30 days, how often did you?all of the time (1), most of the time 
(2), some of the time (3), a little of the time (4), none of the time (5)
1. Feel tired out for no good reason?
2. Feel nervous?
3. Feel so nervous that nothing could calm you down?
4. Feel hopeless?
5. Feel restless or fidgety?
6. Feel so restless you could not sit still?
7. Feel depressed?
8. Feel that everything was an effort?
9. Feel so sad nothing could cheer you up?
10. Feel worthless?
161
Appendix E: Home Observation for Measurement of the Environment 
(children aged six to nine)
1.  About how many books does (CHILD) have?
None (1), One or two (2), Three to nine (3), 10 to 19 (4), 20 or more (5)
2.  How often do you read to (CHILD)?  Would you say never (1), several times a 
year (2), several times a month (3), about once a week (4), a few times a week (5), 
every day (6)
3.  How often is (CHILD) expected to do each of the following?  Would you say 
almost never (1), less than ? the time (2), about ? the time (3), more than ? the time 
(4), almost always (5)?
Make (his/her) own bed?
Clean (his/her) own room?
Clean up after spills?
Bathe (himself/herself)?
Pick up after (himself/herself)?
4.  Is there a musical instrument (for example, piano, drum, guitar, etc.) that 
(CHILD) can use at home? (Yes/No)
5.  Does your family get a daily newspaper? (Yes/No)
6.  About how often does (CHILD) read for enjoyment? Would you say every day 
(1), several times a week (2), several times a month (3), several times a year (4), 
never (5)? ( [if volunteered] child cannot read (6))
162
7.  Does your family encourage (CHILD) to start and keep doing hobbies? (Yes/No)
8.  Does (CHILD) participate in any extracurricular activities such as gymnastics, 
scouts, music lessons, a sports team, or a boys' or girls' club? (Yes/No)
9.  How often has a family member taken or arranged to take (CHILD) to any type of 
museum (children's scientific, art, historical, etc.) within the last year? Would you 
say never (1), once or twice (2), several times (3), about once a month (4), more than 
once a month (5)?
10.  How often has a family member taken or arranged to take (CHILD) to any type 
of musical or theatrical performance within the past year? (never (1), once or twice 
(2), several times (3), about once a month (4), more than once a month (5))
11.  About how often does our whole family get together with friends or relatives?  
Would you say once a year or less (1), a few times a year (2), about once a month 
(3), two or three times a month (4), or about once a week or more (5)?
12.  When your family watches TV together, do you (or (CHILD'S) other parent) 
discuss TV programs with (him/her)? (Yes/No/If volunteered, do not have a TV)
13.  Most children get so angry at their parents that they say things like "I hate you" 
or swear in a tantrum.  Please look at this list and tell me which actions you would 
take if this happened. [Check all that apply.]
A. Ground child
B. Spank child
C. Talk with child
D. Give (him/her) household chore
163
E. Ignore it
F. Send to room from more than one hour
G. Take away (his/her) allowance
H. Take away TV, phone, or other privileges
I. Put child in short "time out"
J. Other (specify) 
14. 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 times in the 
past week have you spanked (CHILD)? 
(Interview observation of home environment)
1.  Primary caregiver responded verbally to (CHILD'S) speech, questions or request:
Often Sometimes  Never
   1 2         3 4 5
2.  If child is older than three years of age: Primary caregiver introduced interviewer 
to (CHILD) by name. (Yes/No)
3.  Primary caregiver's voice conveyed positive feeling about this (CHILD):
Often Sometimes  Never
   1 2         3 4 5
4.  Primary caregiver spontaneously spoke or conversed with (CHILD) (excluding 
scolding or suspicious comments):
Often Sometimes  Never
   1 2         3 4 5
164
5.   Interior of the home is dark or perceptually monotonous.
Very Monotonous       Somewhat Not at all
   1 2         3 4 5
6.  All visible rooms in the (house/apartment) are:
Very clean Somewhat clean Not at all clean
   1 2         3 4 5
7.  All visible rooms in the (house/apartment) are:
Very clean Somewhat cluttered Not at all cluttered
       1 2 3 4 5
8.  (CHILD's) play environment is safe (no potentially dangerous health or structural 
hazards within child's range).  EXAMPLES: Falling plaster, peeling paint, rodents, 
glass, poisons and cleaning materials, flames and heat, frayed electrical wires.) 
(Yes/No/Not observed)
165
Appendix F: Home Observation for Measurement of the Environment 
(children aged ten and older)
1.  About how many books does (CHILD) have?
None (1), One or two (2), Three to nine (3), 10 to 19 (4), 20 or more (5)
2.  How often is (CHILD) expected to do each of the following?  Would you say 
almost never (1), less than ? the time (2), about ? the time (3), more than ? the time 
(4), or almost always (5)?
Make (his/her) own bed?
Clean (his/her) own room?
Help keep shared living areas clean and straight?
Do routine chores such as mow the lawn, help with dinner, wash dishes, etc.?
Help manage (his/her) own time (get up on time, be ready for school, etc.)?
Pick up after (himself/herself)?
3.  Is there a musical instrument (for example, piano, drum guitar, etc.) that (CHILD) 
can use at home? (Yes/No)
4.  Does your family get a daily newspaper? (Yes/No)
5.  About how often does (CHILD) read for enjoyment? Would you say every day 
(1), several times a week (2), several times a month (3), several times a year (4), 
never (5)? ( [if volunteered] child cannot read (6))
6.  Does your family encourage (CHILD) to start and keep doing hobbies? (Yes/No)
166
7.  Does (CHILD) participate in any extracurricular activities such as gymnastics, 
scouts, music lessons, a sports team, or a boys' or girls' club? (Yes/No)
8.  How often has a family member taken or arranged to take (CHILD) to any type of 
museum (children's scientific, art, historical, etc.) within the last year? Would you 
say never (1), once or twice (2), several times (3), about once a month (4), more than 
once a month (5)?
9.  How often has a family member taken or arranged to take (CHILD) to any type of 
musical or theatrical performance within the past year? (never (1), once or twice (2), 
several times (3), about once a month (4), more than once a month (5))
10.  About how often does our whole family get together with friends or relatives?  
Would you say once a year or less (1), a few times a year (2), about once a month 
(3), two or three times a month (4), or about once a week or more (5)?
11.  When your family watches TV together, do you (or (CHILD'S) other parent) 
discuss TV programs with (him/her)? (Yes/No/If volunteered, do not have a TV)
12.  Most children get so angry at their parents that they say things like "I hate you" 
or swear in a tantrum.  Please look at this list and tell me which actions you would 
take if this happened. [Check all that apply.]
Spank child
13.  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 times in the 
past week have you spanked (CHILD)? 
167
(Interview observation of home environment)
1.  Primary caregiver responded verbally to (CHILD'S) speech, questions or request:
Often Sometimes  Never
   1 2         3 4 5
2.  If child is older than three years of age: Primary caregiver introduced interviewer 
to (CHILD) by name. (Yes/No)
3.  Primary caregiver's voice conveyed positive feeling about this (CHILD):
Often Sometimes  Never
   1 2         3 4 5
4.  Primary caregiver spontaneously spoke or conversed with (CHILD) (excluding 
scolding or suspicious comments):
Often Sometimes  Never
   1 2         3 4 5
5.   Interior of the home is dark or perceptually monotonous.
Very Monotonous       Somewhat Not at all
   1 2         3 4 5
6.  All visible rooms in the (house/apartment) are:
Very clean Somewhat clean Not at all clean
   1 2         3 4 5
7.  All visible rooms in the (house/apartment) are:
Very clean Somewhat cluttered Not at all cluttered
       1 2 3 4 5
168
8.  (CHILD's) play environment is safe (no potentially dangerous health or structural 
hazards within child's range).  EXAMPLES: Falling plaster, peeling paint, rodents, 
glass, poisons and cleaning materials, flames and heat, frayed electrical wires.) 
(Yes/No/Not observed)
169
Appendix G: Parental Warmth Scale
About how often in the past month have you:
1. Hugged or shown physical affection to your child?  Would you say not in the 
past month (1) one or two times in the past month (2), about once a week (3), 
several times a week (4), or every day (5)? 
2. Told (CHILD) that you love (him/her)?
3. Spent time with (CHILD) doing one of (his/her) favorite activities?
4. Joked or played with (CHILD)?
5. Talked with (him/her) about things (he/she) is especially interested in?
6. Told (CHILD) you appreciated something (he/she) did?
170
Appendix H: Parental Involvement Scale
Now I'd like to ask about things you and (CHILD) did together in the past month.  
These things might be done together anywhere, they don't have to be done at home.  
For each please tell me if you did not do it in the past month with (CHILD), did it 
one or two times in the past month, about once a week, several times a week, or 
every day.
a. Washed or folded clothes together?
b. Done dishes together?
c. Gone to the store with (CHILD)?
d. Looked at books or read stories with (him/her)?
e. Talk to (him/her) about your family?
f. Prepared food together?
g. Done arts and crafts together?
h. Played sports or did outdoor activities together?
i. Cleaned the house together?
j. Built or repaired something together?
k. Worked or played on a computer or played video games with (CHILD)?
l. Worked on homework with (him/her)?
m. Played a board game or card game or done puzzles with (him/her)?
171
Appendix I: Behavior Problems Index
For the next set of statements, decide whether they are often true, sometimes true, or 
not true according to (CHILD'S) behavior.
a. (He/She) has sudden changes in mood or feelings.
b. (He/She) feels or complains that no one loves him/her.
c. (He/She) is rather high strung, tense and nervous.
d. (He/She) cheats or tells lies.
e. (He/She) is too fearful or anxious.
f. (He/She) argues too much.
g. (He/She) has difficulty concentrating, cannot pay attention for long. 
h. (He/She) is easily confused, seems to be in a fog.
i. (He/She) bullies or is cruel or mean to others.
j. (He/She) is disobedient.
k. (He/She) does not seem to feel sorry after (he/she) misbehaves.
l. (He/She) has trouble getting along with other children.
m. (He/She) is impulsive, or acts without thinking.
n. (He/She) feels worthless or inferior.
o. (He/She) is not liked by other children.
p. (He/She) has a lot of difficulty getting (his/her) mind off certain thoughts. 
(IF NEC: has obsessions)
q. (He/She) is restless or overly active, cannot sit still.
172
r. (He/She) is stubborn, sullen, or irritable.
s. (He/She) has a very strong temper and loses it easily.
t. (He/She) is unhappy, sad or depressed.
u. (He/She) is withdrawn, does not get involved with others.
v. (He/She) breaks things on purpose or deliberately destroys (his/her) own 
or another's things.
w. (He/She) clings to adults.
x. (He/She) cries too much
y. (He/She) demands a lot of attention
z. (He/She) is too dependent on others.
aa. (He/She) feels others are out to get (him/her).
bb. (He/She) hangs around with kids who get into trouble.
cc. (He/She) is secretive, keeps things to (himself/herself).
dd. (He/She) worries too much.
173
 Appendix J: Social Support Index
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
2. No
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.)
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
13. Grown child
14. Other
174
In the past month, have you received any moral or emotional support such as advice 
or encouragement from friends or relatives?
 1.  Yes
2. No
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.)
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
13. Grown child
14. Other
175
Appendix K: Correlation Matrix Total Sample
          1              2             3         4                5       6        7          8           9
1.    ---
2.  -.012          ---
3.   .102***   .067* ---
4.  -.074**    .228***  -.032 ---
5.   .018       -.094***  -.070* -.130***     ---
6.   .059*     -.090**      .055 -.190***    .127***      ---
7.   .081**    .083**      .017 -.082**      .270***    -.025        ---
8.   .028        .002   .013 -.042           .332***     .080**   .493***   ---
9.   .067*     .177***     .232*** .324***    -.224***    -.057*   -.244*** -.107***  ---
I.
Note: 1=Presence of illness.  2=Economic strain.  3=Health stress.  4=Parental 
psychological distress.  5=Parental cognitive stimulation.  6=Parental emotional 
support.  7=Parental warmth.  8=Parental involvement.  9=Child behavior problems.
N= 1214. *p<.05.  **p<.01.  ***p<.001. (2-tailed)
176
Appendix L: Correlation Matrix Low Levels of Social Support
       1             2  3  4        5           6     7             8   9
1.    ---
2.  .057        ---
3.  .205*** .147***   ---
4. -.130*** .097*      -.036      ---
5. -.102*   -.181***   -.021 -.055        ---
6.   .061     -.083*       .027 -.268***  .197***     ---
7.   .037      .065         -.026 -.036      .211***    -.103*        ---
8.   .044     -.060 .056 -.064     .336***      .044     .472***  ---
9.   .083*    .241***    .051       .200*** -.191***    -.149***  -.169*** -.138***    ---
II.
Note: 1=Presence of illness.  2=Economic strain.  3=Health stress.  4=Parental 
psychological distress.  5=Parental cognitive stimulation.  6=Parental emotional 
support.  7=Parental warmth.  8=Parental involvement.  9=Child behavior problems.
N= 718. *p<.05.  **p<.01.  ***p<.001 (2-tailed)
177
Appendix M: Correlation Matrix Mid to High Levels of Social Support
       1            2              3      4           5             6               7           8           9
1.   ---
2.  -.112**  ---
3.    .025     -.002         ---
4.   -.027      .333*** -.043         ---
5.    .123** -.087*     -.132** -.253***     ---
6.    .056     -.110**    .073 -.111**    .038          ---
7.    .132*** .078        .040 -.170***  .370***   .053           ---
8.    .028       .015       -.019 -.045       .334***   .112**   .546***     ---
9.    .043       .119**     .352***  .459***-.282***   .026        -.375***   -.087*       ---
III.
Note: 1=Presence of illness.  2=Economic strain.  3 = Health stress.  4=Parental 
psychological distress.  5=Parental cognitive stimulation.  6=Parental emotional 
support.  7=Parental warmth.  8=Parental involvement.  9=Child behavior problems.
N= 675. *p<.05.  **p<.01.  ***p<.001 (2-tailed)
178
Appendix N: Error Corrections for Total Sample, and Low and Mid to High Levels 
of Social Support
Total sample
Error correction with estimated error variances for economic strain (1.12), 
child health stress (0.07), parental psychological distress (3.18), parental cognitive 
stimulation (1.26), parental emotional support (0.56), parental warmth (2.32), 
parental involvement (13.99), and child behavior problems (6.98).  
Low levels of social support 
Error correction with estimated error variances for low levels of social 
support for economic strain (1.01), child health stress (0.06), parental psychological 
distress (3.08), parental cognitive stimulation (1.28), parental emotional support 
(0.54), parental warmth (2.44), parental involvement (13.95), and child behavior 
problems (6.75).
Mid to high levels of social support
Error correction with estimated error variances for mid to high levels of 
social support for economic strain (1.21), child health stress (0.09), parental 
psychological distress (3.25), parental cognitive stimulation (1.26), parental 
emotional support (0.55), parental warmth (2.13), parental involvement (13.47), and 
child behavior problems (6.99).
179
References
Achenbach, T., & Edelbrock, C. (1981). Behavioral problems and competencies 
reported by parents of normal and disturbed children aged four though 
sixteen. Monographs for the Society for Research in Child Development, 
46(1, Serial No. 188).
Allison, P. D. (2002). Missing data. Thousand Oaks, California: Sage University 
Paper.
Amato, P. R., & Rivera, F. (1999). Parental involvement and children's behavior 
problems. Journal of Marriage and the Family, 61(2), 375-384.
Amer, K. S. (1999). Children's adaptation to insulin dependent diabetes mellitus: A 
critical review of the literature. Pediatric Nursing, 25(6), 627-639.
American Psychiatric Association. (2000). Diagnostic and statistical manual of 
mental disorders (4th ed., text revision). Washington D.C.: Author.
Anthony, K. K., Gil, K. M., & Schanberg, L. E. (2003). Brief report: Parental 
perceptions of child vulnerability in children with chronic illness. Journal of 
Pediatric Psychology, 28(3), 185-190.
Asmussen, L., Olson, L. M., & Sullivan, S. A. (1999). 'You have to live it to 
understand it..'  Family experiences with chronic otitis media in children. 
Ambulatory Child Health, 5, 303-312.
Atkin, K., & Ahmad, W. I. U. (2000). Family care-giving and chronic illness: How 
parents cope with a child with a sickle cell disorder or thalassaemia. Health 
and Social Care in the Community, 8(1), 57-69.
180
Auinger, P., Lanphear, B. P., Kalkwarf, H. J., & Mansour, M. E. (2003). Trends in 
otitis media among children in the United States. Pediatrics, 112(3), 514-520.
Babbie, E. (2001). The practice of social research (9th ed.). Belmont, CA: 
Wadsworth/Thompson Learning.
Baker, P. C., Keck, C. K., Mott, F. L., & Quinlan, S. V. (1993). NLSY child 
handbook revised edition: A guide to the 1986-1990 National Longitudinal 
Survey of Youth Child Data. Columbus, Ohio: Center for Human Resource 
Research, The Ohio State University.
Baker, P. C., & Mott, F. L. (1989, June). NLSY child handbook 1989: A guide and 
resource document for the National Longitudinal Survey of Youth 1986 Child 
Data (Center for Human Resource Research, pp. 1-463). The Ohio State 
University.
Barbarin, O. A., Hughes, D., & Chesler, M. A. (1985). Stress, coping, and marital 
functioning among parents of children with cancer. Journal of Marriage and 
the Family, 47(2), 473-480.
Barbarin, O. A., Whitten, C. F., Bond, S., & Conner-Warren, R. (1999). The social 
and cultural context of coping with sickle cell disease: II. The role of 
financial hardship in adjustment to sickle cell disease. Journal of Black 
Psychology, 25(3), 294-315.
Barlow, J., Harrison, K., & Shaw, K. (1998). The experience of parenting in the 
context of juvenile chronic arthritis. Clinical Child Psychology and 
Psychiatry, 3(3), 1359-1045.
181
Barlow, T. (1997, October). Managed care for children with disabilities and the 
interface of legislative policies. In PREL Briefing Paper (Pacific Resources 
for Education and Learning, pp. 1-9).  Honolulu, HI.
Barrera, M. (1986). Distinctions between social support concepts, measures, and 
models. American Journal of Community Psychology, 14(4), 413-443.
Bedell, J. R., Giordani, B., Amour, J. L., Tavormina, J., & Boll, T. (1977). Life stress 
and the psychological and medical adjustment of chronically ill children. 
Journal of Psychosomatic Research, 21, 237-242.
Belle, D. (1990). Poverty and women's mental health. American Psychologist, 45(3), 
385-389.
Belsky, J. (1984). The determinants of parenting: A process model. Child 
Development, 55, 83-96.
Belsky, J., & Vondra, J. (1989). Lessons from child abuse: The determinants of 
parenting. In D. Cicchetti & V. Carlson (Eds.), Child maltreatment: Theory 
and research on the causes and consequences of child abuse and neglect (pp. 
153-202). New York: Cambridge University Press.
Benson, C. (2001, July). America's children: Key national indicators of well-being, 
2001. Washington D.C.: Federal  Interagency Forum on Child and Family 
Statistics.
Bentler, P. M. (1997). EQS 5.7. Encino, California: Multivariate Software, Inc.
182
Blackwell, D. L., Vickerie, J. L., & Wondima, E. A. (2003, October). Summary 
health statistics for US children: National Health Interview Survey 2000
(Vital and Health Statistics: Series 10, Number 213). Retrieved October 14, 
2003, from http://www.cdc.gov/nchs/data/series/sr_10/sr10_213.pdf.
Blum, R. W., Garell, D., Hodgman, C. H., & Slap, G. B. (1993). Transition from 
child-centered to adult health-care systems for adolescents with chronic 
conditions: A position paper of the Society for Adolescent Medicine. Journal 
of Adolescent Health, 14, 570-576.
Bollen, K. (1989). Structural equations with latent variables. New York: Wiley.
Bradley, R. H., Caldwell, B. M., Rock, S. L., Ramey, C. T., Barnard, K. E., Gray, C., 
Hammond, M. A., Mitchell, S., Gottfried, A. W., Siegel, L., & Johnson, D. L.  
(1989). Home environment and cognitive development in the first three years 
of life: A collaborative study involving six sites and three ethnic groups in 
North America. Developmental Psychology, 25(2), 217-235.
Bronfenbrenner, U. (1986). Ecology and the family as context for human 
development:. Developmental Psychology, 22(6), 723-742.
Brooks, D. (2001). Bridging the gaps in chronic illness care for children. Disease 
Management Advisor, 7(5), 65-69.
Brown, R. T., & DuPaul, G. J. (1999). Introduction to the mini-series: Promoting 
school success in children with chronic medical conditions. School 
Psychology Review, 28(2), 175-181. 
183
Bryson, K., & Casper, L. M.  (1998, April).  Household and family characteristics: 
March 1997.  Current population reports: Population characteristics.  
Retrieved May 24, 2004, from http://www.census.gov/prod/3/98pubs/p20-
509.pdf
Bukowski, E., & Anita, L. (2000). Children's perceptions of chronically ill peers: An 
investigation of salient illness characteristics and social context. Age 
Differences, Human Sex Differences, Psychosocial Factors Dissertation 
Abstract. (From  PsycInfo).
Cadman, D., Boyle, M., Szatrnari, P., & Offord, D. (1987). Chronic illness, 
disability, and mental and social well-being: Findings of the Ontario Child 
Health Study. Pediatrics, 79(5), 805-814.
Cadman, D., Rosenbaum, P., Boyle, M., & Offord, D. R. (1991). Children with 
chronic illness: Family and parent demographic characteristics and 
psychosocial adjustment. Pediatrics, 87(6), 884-889.
Caldwell, B. M., & Bradley, R. H. (1984). Home observation for measurement of the 
environment. Little Rock, AR: University of Arkansas at Little Rock.
Cameron, K., & Gregor, F. (1987). Chronic illness and compliance. Journal of 
Advanced Nursing, 12, 671-676.
Canning, R. D., Harris, E. S., & Kelleher, K. J. (1996). Factors predicting distress 
among caregivers to children with chronic medical conditions. Journal of 
Pediatric Psychology, 21, 735-749.
184
Ceballo, R., & McLoyd, V. C. (2002). Social support and parenting in poor, 
dangerous neighborhoods. Child Development, 73(4), 1310-1321.
Centers for Disease Control and Prevention. (2002, October 11). Iron Deficiency -
United States, 1999-2000. Morbidity and Mortality Weekly Report, 51(40), 
897-899. Retrieved October 14, 2003, from 
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5140a1.htm#tab2.
Charron-Prochownik, D., & Kovacs, M. (2000). Maternal health-related coping 
patterns and health and adjustment outcomes in children with type one 
diabetes. Children's Health Care, 29(1), 37-45.
Chen, E., Matthews, K. A., & Boyce, W. T. (2002). Socioeconomic differences in 
children's health: How and why do these relationships change with age? 
Psychological Bulletin, 128(2), 295-329.
Chen, E., Bloomberg, G. R., Fisher, E. B. J., & Strunk, R. C. (2003). Predictors of 
repeat hospitalization in children with asthma: The role of psychosocial and 
socioenvironmental factors. Health Psychology, 22(1), 12-18.
Chen, M., & Land, K. C. (1986). Testing the health belief model: LISREL analysis 
of alternative models of causal relationships between health beliefs and 
preventative health behavior. Social Psychology Quarterly, 49, 45-69.
Chernoff, R. G., List, D. G., DeVet, K. A., & Ireys, H. T. (2001). Maternal reports of 
raising children with chronic illnesses: The prevalence of positive thinking. 
Ambulatory Pediatrics, 1(2), 104-107.
185
Cohen, J.  (1965).  Some statistical issues in psychological research.  In B. B. 
Wolman (Ed).  Handbook of clinical psychology, pp. 95-121.  New York: 
McGraw-Hill
Cohen, S., Gottlieb, B. H., & Underwood, L. G. (2001). Social relationships and 
health: Challenges for measurement and intervention. Advances in Mind-
Body Medicine, 17(2), 129-142.
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering 
hypothesis. Psychological Bulletin, 98(2), 310-357.
Coiro, M. J. (1997, April). Maternal depressive symptoms as a risk factor for the 
development of children poverty. Presented at the Biennial Meeting of the 
Society for Research in Child Development, Washington, DC.
Coiro, M. J., Zill, N., & Bloom, B. (1994, December). Health of our nation's 
children (Vital and Health Statistics: Series 10, Number 191). Retrieved 
October 14, 2003, from 
http://www.cdc.gov/nchs/data/series/sr_10/sr10_191.pdf.
Conger, R.D., Ebert, C., Sun, L, Simons, R. L., McLoyd, V. C., & Brody, G. H. 
(2002).  Economic pressure in African American families: A replication and 
extension of the Family Stress Model.  Developmental Psychology, 38(2), 
179-193.
186
Conger, K. J., Rueter, M. A., & Conger, R. D. (2000). The role of economic pressure 
in the lives of parents and their adolescents: The Family Stress Model. In L. 
J. Crockett & R. K. Silbereisen (Eds.), Negotiating adolescence in times of 
social change (pp. 201-223). New York: Cambridge University Press.
Conger, R. D., Conger, K. J., Elder, G. E., Lorenz, F. O., Simons, R. L., & Whitbeck, 
L. B. (1992). A family process model of economic hardship and adjustment 
of early adolescent boys. Child Development, 63, 526-541.
Conger, R. D., & Elder, G. H. (1994). Families in troubled times: Adapting to 
change in rural America. Hawthorne, NY: Aldine de Gruyter.
Conger, R. D., Patterson, G. R., & Ge, X. (1995). It takes two to replicate: A 
mediational model for the impact of parents' stress on adolescent adjustment. 
Child Development, 66, 80-97.
Conger, R. D., Elder, G. H., Lorenz, F. O., Conger, K. J., Simons, R. L., Whitbeck, 
L. B., Huck, S., & Melby, J. N.  (1990). Linking economic hardship to 
marital quality and instability. Journal of Marriage and the Family, 52, 643-
656.
Cooke, B. D., Rossman, M. M., McCubbin, H. I., & Patterson, J. M. (1988). 
Examining the definition and assessment of social support: A resource for 
individuals and families. Family Relations, 37, 211-216.
187
Cosica, J. M., Christensen, B. K., Henry, R. R., Wallston, K., Radcliffe, J., & 
Rutstein, R. (2001). Effects of home environment, socioeconomic status, and 
health status on cognitive functioning in children with HIV-1 infection. 
Journal of Pediatric Psychology, 26(6), 321-329.
Dadds, M. R., Stein, R. E. K., & Silver, E. J. (1995). The role of maternal 
psychological adjustment in the measurement of children's functional status. 
Journal of Pediatric Psychology, 20(4), 527-544.
Dean, A., Lin, N., & Ensel, W. M. (1981). The epidemiological significance of social 
support systems in depression. Research in Community and Mental Health, 2,
77-109.
Department of Human Services Center for Population Studies in Epidemiology. 
(2002). The Kessler Psychological Distress Scale (K10). Brief Reports(14). 
Retrieved from www.dhs.sa.gov.au/pehs/PROS/br-kessler-scale02-14.pdf.
Disability Social History Project. (2001). Significant dates in disability history. 
Retrieved May 13, 2001, from http://disabilityhistory.org/timelist.html#BC.
Doherty, W. J., McDaniel, S. H., & Hepworth, J. (1994). Medical family therapy: An 
emerging arena for family therapy. Journal of Family Therapy, 16(1), 31-46.
Dosa, N. P., Boeing, N. M., & Kanter, R. K. (2001, March). Excess risk to severe 
acute illness in children with chronic health conditions. Pediatrics, 107(3), 
499-504.
Dowling, M., & Dolan, L. (2001). Families with children with disabilities -
Inequalities and the social model. Disability & Society, 16(1), 21-35.
188
Downey, G., & Coyne, J. C. (1990). Children of depressed parents: An integrative 
review. Psychological Bulletin, 108(1), 50-76.
Drotar, D. (1997). Relating parent and family functioning to the psychological 
adjustment of children with chronic health conditions: What have we 
learned? What do we need to know? Journal of Pediatric Psychology, 22(2), 
149-165.
Dunst, C. J., Jenkins, V., & Trivette, C. (1984). The family support scale: Reliability 
and validity. Journal of Individual, Family, and Community Wellness, 1(4), 
45-52.
Dunst, C. J., Trivette, C. M., & Cross, A. H. (1986). Mediating influences of social 
support: Personal, family, and child outcomes. American Journal of Mental 
Deficiency, 90(4), 403-417.
Egbuonu, L., & Starfield, B. (1982). Child health and social status. Pediatrics, 69(5), 
550-557.
Eisenberg, N., Losoya, S., Guthrie, I. K., Murphy, B., Shepard, S. A., Padgett, S. J., 
Fabes, R. A., Poulin, R., & Reiser, M. (2001). Parental socialization of 
children's disregulated expression of emotion and externalizing problems. 
Journal of Family Psychology, 15(2), 183-205.
Elder, G. H. (1974). Children of the great depression. Chicago: University of 
Chicago Press.
Elder, G. H. (1998). The life course as development theory. Child Development, 
69(1), 1-12.
189
Ellerton, M., Stewart, M. J., Ritchie, J. A., & Hirth, A. M. (1996). Social support in 
children with a chronic condition. Canadian Journal of Nursing Research, 
28(4), 15-36.
Evans, G. W. (2004). The environment of childhood poverty. American 
Psychologist, 59(2), 77-92.
Ferris, T. G., Crain, E. F., Oken, E., Wang, L., Clark, S., & Camargo, C. A. (2001). 
Insurance and quality of care for children with acute asthma. Ambulatory 
Pediatrics, 1(5), 267-274.
Fisher, H. R. (2001). The needs of parents with chronically sick children: A literature 
review. Journal of Advanced Nursing, 36(4), 600-607.
Friedman, T. L. (1995). From Beirut to Jerusalem. New York: Anchor Books.
Furukawa, T. A., Kessler, R. C., Slade, T., & Andrews, G. (2003). The performance 
of the K6 and K10 screening scales for psychological distress in the 
Australian National Survey of Mental Health and Well-being. Psychological 
Medicine, 33, 357-362.
Garrett, P., Ferron, J., Ng'andu, N., Bryant, D., & & Harbin, G. (1994). A structural 
model for the developmental status of young children. Journal of Marriage 
and the Family, 56, 147-163.
Garson, G. D. (2003). Structural equation modeling. Retrieved June 12, 2003, from 
http://www2.chass.ncsu.edu/garson/pa765/structur.htm.
190
Gartstein, M. A., Noll, R. B., & Vannatta, K. (2000). Childhood aggression and 
chronic illness: Possible protective mechanisms. Journal of Applied 
Developmental Psychology, 21(3), 315-333.
Garwick, A. E., & Millar, H. E. C. (1996). Promoting resilience in youth with 
chronic conditions and their families. Washington D.C. Health Resources 
and Services Administration (National Maternal and Child Health 
Clearninghouse).
Gerard, J. M., & Buehler, C. (1999). Multiple risk factors in the family environment 
and youth problem behaviors. Journal of Marriage and the Family, 61, 343-
361.
Gerard, J. M., & Krishnakumar, A. (2003, April). Marital conflict, parenting quality, 
and youth maladjustment: A longitudinal investigation of the spillover 
hypothesis. Poster Session presented at SRCD Annual Conference, Tampa, 
Florida.
Gerhardt, C. A., Vannatta, K., McKellop, J. M., Zeller, M., Taylor, J., Passo, M., & 
Noll, R.  (2003). Comparing parental distress, family functioning, and the 
role of social support for caregivers with and without a child with juvenile 
rheumatoid arthritis. Journal of Pediatric Psychology, 28(1), 5-15.
Gil, K. M., Carson, J. W., Porter, L. S., Ready, J., Valrie, C., Redding-Lallinger, R., 
& Daeschner, C. (2003). Daily stress and mood and their association with 
pain, health-care use, and school activity in adolescents with sickle cell 
disease. Journal of Pediatric Psychology, 28(5), 363-373.
191
Gil-Rivas, V., Greenberger, E., Chen, C., & Lopez-Lena, M. M. (2003). 
Understanding depressed mood in the context of a family-oriented culture. 
Adolescence, 38(149), 93-110.
Gillaspy, S., R., Hoff, A. L., Mullins, L. L., & Van Pelt, J. C., & Chaney. (2002). 
Psychological distress in high-risk youth with asthma. Journal of Pediatric 
Psychology, 27(4), 363-371.
Guo, G., & Harris, K. M. (2000). The mechanisms mediating the effects of poverty 
on children's intellectual development. Demography, 37(4), 431-447.
Haffner, J., Ester, C., Munch, H., Parzer, P., Raue, B., Steen, R., Klett, M., & Resch, 
F.  (2002). Parent-reported problems of six year old pre-school children: 
Prevalence and risk factors in an epidemiological study. Praxis der 
Kinderpsychologie und Kinderpsychiatrie. (From Psyc Info: 0032 -7034).
Hauser, S. T., Jacobson, A. M., Lavori, P., Wolfsdorf, J. I., Herskowitz, R. D., 
Milley, J. E., Bliss, R., Wertlieb, D., & Stein, J.  (1990). Adherence among 
children and adolescents with insulin-dependent diabetes mellitus over a 
four-year longitudinal follow-up: II. Immediate and long-term linkages with 
the family milieu. Journal of Pediatric Psychology, 15(4), 527-542.
Hill-Briggs, F. (2003). Problem solving in diabetes self-management: A model of 
chronic illness self-management behavior. Annals of Behavioral Medicine, 
25(3), 182-193.
192
Hoard, L. R., & Anderson, E. A. (2004). Factors related to depression in rural and 
urban, non-custodial fathers. Journal of Community Psychology, 32(1), 103-
119.
Hobfoll, S. E., Johnson, R. J., Ennis, N., & Jackson, A. P. (2003). Resource loss, 
resource gain, and emotional outcomes among inner city women. Journal of 
Personality and Social Psychology, 84(3), 632-643.
Hofferth, S. L., & Curtin, S. (2004, April). The association between family income 
and educational attainment in middle adulthood. Presented at the Population 
Association of America, Boston.
Hofferth, S. (1998). Healthy environments, healthy children: Children in families. A 
report on the 1997 Panel Study of Income Dynamics Child Development 
Supplement. Ann Arbor, Michigan: University of Michigan.
Hofferth, S., Davis-Kean, P. E., Davis, J., & Finkelstein, J. (1997). The Child 
Development Supplement of the Panel Study of Income Dynamics: 1997 user 
guide. Retrieved August 29, 2002, from http://www.isr.umich.edu/src/child-
development/usergd.html.
Holmes, C. S., Yu, Z., & Frentz, J. (1999). Chronic and discrete stress as predictors 
of children?s adjustment. Journal of Consulting and Clinical Psychology, 
67(3), 411-419.
Holroyd, J., & Guthrie, D. (1986). Family stress with chronic childhood illness: 
Cystic fibrosis, neuromuscular disease, and renal disease. Journal of Clinical 
Psychology, 42(4), 552-561.
193
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure 
analysis: Conventional criteria versus new alternatives. Structural Equation 
Modeling: A Multidisciplinary Journal, 6(1), 1-55.
Hyers, D. A. (1995). A structural analysis of relationships among stress, social 
support, dysfunctional attitudes, and depression in older adults [Abstract]. 
Dissertation Abstracts International, 56(6-A), 2117.
Institute for Social Research. (1999, October 27). Description of the 1997 PSID 
Child Supplement weights. Retrieved August 1, 2003, from 
http://psidonline.isr.umich.edu/CDS/weightsdoc.html.
Institute for Social Research. (2002). The PSID: Family income data 1997: Code 
book. Retrieved June 5, 2002, from University of Michigan: 
http://www.isr.umich.edu/src/psid/income/faminc/famcbk97.htm.
Isaacs, D., & Sewell, J. R. (2003). Children with chronic conditions.  The Medical 
Journal of Australia, 179(5), 235-236.
Jacelon, C. S. (1997). The trait and process of resilience. Journal of Advanced 
Nursing, 25, 123-129.
Jackson, A. P., Brooks-Gunn, J., Huang, C.-C., & Glassman, M. (2000). Single 
mothers in low-wage jobs: Financial strain, parenting, and preschoolers' 
outcomes. Child Development, 71(5), 1409-1423.
194
Jackson, A. P., Gyamfi, P., Brooks-Gunn, J., & Blake, M. (1998). Employment 
status, psychological well-being, social support, and physical discipline 
practices of single black mothers. Journal of Marriage and the Family, 60,
894-902.
Jackson, P. B. (1992). Specifying the buffering hypothesis: Support, strain, and 
depression. Social Psychology Quarterly, 55(4), 365-378.
Jekielek, S. M. (1998). Parental conflict, marital disruption, and children's emotional 
well-being. Social Forces, 76(3), 905-936.
Jessop, D. J., & Stein, R. E. K. (1985). Uncertainty and its relation to the 
psychological and social correlates of chronic illness in children. Social 
Science Medicine, 20(10), 993-999.
Juang, L. P., & Silbereisen, R. (2002). The relationship between adolescent academic 
capability beliefs, parenting and school grades. Journal of Adolescence, 
25(1), 3-18.
Kaplan, R. M., & Hartwell, S. L. (1987). Differential effects of social support and 
social network on physiological and social outcomes in men and women with 
type two diabetes mellitus. Health Psychology, 6(5), 387-398.
Katz, S., & Krulik, T. (1999). Fathers of children with chronic illness: Do they differ 
from fathers of healthy children? Journal of Family Nursing, 5(3), 292-315.
195
Kazak, A. E., & Christakis, D. A. (1994). Caregiving issues in families of children 
with chronic medical conditions. In E. Kahana, D. E. Biegel & M. L. Wykle 
(Eds.), Family caregiver applications series: Volume 4; Family caregiving 
across the lifespan (pp. 331-355). Thousand Oaks, California: Sage.
Kessler, R. C., Andrews, G., Colpe, L. J., Hiripi, E., Mroczek, D. K., Normand, S.-L. 
T., Walters, E. E., & Zaslavsky, A. M.  (2002). Short screening scales to 
monitor population prevalences and trends in non-specific psychological 
distress. Psychological Medicine, 32, 959-976.
Kessler, R. C., Barker, P. R., Colpe, L. J., Epstein, J. F., Gfroerer, J. C., Hiripi, E., 
Howes, M. J., Normand, S. T., Manderscheid, R. W., Walters, E. E., & 
Zaslavsky, A. M.  (2003a). Screening for serious mental illness in the general 
population. Archive of General Psychiatry, 60,  184-189.
Kessler, R., Demler, O.; Jin, R.; Koretz, D.; Merikangas, K.R.; Rush, A.J.; Walters, 
E.E.; & Wang, P.S.  (2003b). The epidemiology of major depressive disorder: 
Results from the National Comorbidity Survey Replication. Journal of the 
American Medical Association, 289(23), 3095-3105.
Kessler, R., DuPont, R., Berglund, P., & Wittchen, H. (2000). Impairment in pure 
and comorbid generalized anxiety disorder and major depression at 12 
months in two national surveys. American Journal of Psychiatry, 157(12), 
2060-2061.
196
King, G., King, S., Rosenbaum, P., & Goffin, R. (1999). Family-centered caregiving 
and well-being of parents of children with disabilities: Linking process with 
outcome. Journal of Pediatric Psychology, 24(1), 41-53.
Kirschstein, R. (2000, February). Disease-specific estimates of direct and indirect 
costs of illness and NIH support. Retrieved October 31, 2003, from 
http://www1.od.nih.gov/osp/ospp/ecostudies/COIreportweb.htm.
Klausli, J. F., & Owen, M. T. (2003, April). Relations between marriage and 
parenting for fathers and mothers. In Poster session. Society for Research in 
Child Development, Tampa, Florida.
Klebanov, P. K., Brooks-Gunn, J., & Duncan, G. J. (1994). Does neighborhood and 
family poverty affect mothers' parenting, mental health, and social support? 
Journal of Marriage and the Family, 56, 441-455.
Klerman, L. V. (1988). School absence: A health perspective. The Pediatric Clinics 
of North America, 35(6), 1253-1269.
Kline, R. B. (1998). Principles and practice of structural equation modeling. New 
York: The Guildford Press.
Klinnert, M. D., McQuaid, E. L., McCormick, D., Adinoff, A. D., & Bryant, N. E. 
(2000). A multimethod assessment of behavioral and emotional adjustment in 
children with asthma. Journal of Pediatric Psychology, 25(1), 35-46.
Knafl, K., & Zoeller, L. (2000). Childhood chronic illness: A comparison of mothers' 
and fathers' experiences. Journal of Family Nursing, 6(3), 287-302.
197
Krall, L. P. (1978). Joslin Diabetes Manual (11th ed.). Philadelphia, PA: Lea & 
Sebiger.
Kwon, H.-K., & Rueter, M. A. (2003, November 22). Protecting adolescents from 
the long-term effects of economic pressure: The moderating role of 
parenting. Presented at the National Council on Family Relations, 
Vancouver, BC.
La Greca, A. M. (1992). Peer influences in pediatric chronic illness: An update. 
Journal of Pediatric Psychology, 17(6), 775-784.
La Greca, A. M., Auslander, W. F., Grecco, P., Spetter, D., Fisher, E. B., & 
Santiago, J. V. (1995). I get by with a little help from my family and friends: 
Adolescent's support for diabetes care. Journal of Pediatric Psychology, 
20(4), 449-476.
La Greca, A., & Bearman, K. J. (2002). The diabetes social support questionnaire-
family version: Evaluating adolescents' diabetes-specific support from family 
members. Journal of Pediatric Psychology, 27(8), 665-676.
Lavee, Y. (1988). Linear structural relations (LISREL) in family research. Journal of 
Marriage and the Family, 50, 937-948.
LeBlanc, L. A., Goldsmith, T., & Patel, D. (2003).  Behavioral aspects of chronic 
illness in children and adolescents. Pediatric Clinics of North America, 50(4), 
859-878.
Lee, C., & Gotlib, I. (1989). Maternal depression and child adjustment: A 
longitudinal analysis. Journal of Abnormal Psychology, 98, 78-85.
198
Linver, M. R., Brooks-Gunn, J., & Kohen, D. E. (2002). Family processes as 
pathways from income to young children's development. Developmental 
Psychology, 38(5), 719-734.
Littlefield, C. H., Rodin, G. M., Murray, M. A., & Craven, J. L. (1990). Influence of 
functional impairment and social support on depressive symptoms in persons 
with diabetes. Health Psychology, 9(6), 737-749.
Loehlin, J. C. (1998). Latent variable models: An introduction to factor, path, and 
structural analysis (3rd ed.). Mahwah, New Jersey: Lawrence Erlbaum 
Associates.
Lopez, M. L. (1998, March). The relationship between recurrent otitis media and 
parent-child interactions. Dissertation Abstracts International: Section B: 
The Sciences & Engineering, 58(9-B), 5164. In.
Lorenz, F. O., Conger, R. D., & Montague, R. (1994). Doing worse and feeling 
worse: Psychological consequences of economic hardship. In R. D. Conger & 
G. H. Elder Jr. (Eds.), Families in troubled times: Adapting to change in 
rural America (pp. 167-185). New York: Aldine de Gruyter.
Mailick, M. D., & Holden, G. (1994). Coping with childhood asthma: Caretakers' 
views. Health & Social Work, 19(2), 103-112.
Mandleco, B. L., & Peery, J. C. (2000). An organizational framework for 
conceptualizing resilience in children. Journal of Child & Adolescent 
Psychiatric Nursing, 13(3), 99-112.
199
Manuel, J. C. (2001). Risk and resistance factors in the adaptation in mothers of 
children with juvenile rheumatoid arthritis. Journal of Pediatric Psychology, 
26(4), 237-246.
Marks, N. (1998). Does it hurt to care? Caregiving, work family conflict, and midlife 
well-being. Journal of Marriage and the Family, 60, 951-966.
McCubbin, M. A. (1988). Family stress, resources, and family types: Chronic illness 
in children. Family Relations, 37, 203-210.
McCubbin, M., Balling, K., Possin, P., Frierdich, S., & Bryne, B. (2002). Family 
resiliency in childhood cancer. Family Relations, 51(2), 103-112.
McCubbin, M., & McCubbin, H. (1996). Resiliency in families: A conceptual model 
of family adjustment and adaptation in response to stress and crisis. In H. 
McCubbin, A. Thompson & M. McCubbin (Eds.), Family assessment: 
Resiliency, coping and adaptation - inventories for research and practice
(pp. 1-64). Madison, Wisconsin: University of Wisconsin System.
McLoyd, V. C. (1990). The impact of economic hardship on black families and 
children: Psychological distress, parenting, and socioemotional development. 
Child Development, 61, 311-346.
McLoyd, V., & Wilson, L. (1990). Maternal behavior, social support, and economic 
conditions as predictors of distress in children. In V. C. McLoyd & C. 
Flanagan (Eds.), New directions for child development: No. 46.  Economic 
stress: Effects on family life and child development (pp. 49-69). San 
Francisco: Jossey-Bass.
200
McPherson, M., Arango, P., Fox, H., McManus, M., Newacheck, P. W., Perrin, J. 
M., Shonkoff, J. P., & Strickland, B.  (1998). A new definition of children 
with special health care needs. Pediatrics, 102(1), 137-140.
Menaghan, E. G., & Parcel, T. L. (1991). Determining children's home 
environments: The impact of maternal characteristics and current 
occupational and family conditions. Journal of Marriage and the Family, 
53(2), 417-431.
Menaghan, E. G., & Parcel, T. L. (1995). Social sources of change in children's 
home environments: The effects of parental occupational experiences and 
family conditions. Journal of Marriage and the Family, 57(1), 69-84.
Miller, J. E. (2000). The effects of race/ethnicity and income on early childhood 
asthma prevalence and health care use. American Journal of Public Health, 
90(3), 428-430.
Miller, J. E., & Davis, D. (1997). Poverty history, marital history, and quality of 
children's home environments. Journal of Marriage and the Family, 59, 996-
1007.
Miller, R., Sabin, C. A., Goldman, E., Clemente, C., Sadowski, H., Taylor, B., & 
Lee, C. A.  (2000). Coping styles in families with hemophilia. Psychology, 
Health, and Medicine, 5(1), 3-12.
Minden, D. J. (1999). Vulnerability and resilience in children with chronic illness: 
An exploration of the role of the family. Dissertation Abstracts International: 
The Sciences and Engineering, 60(1-B3), 371.
201
Moore, K. A., Chalk, R., Scarpa, J., & Vandivere, S. (2002). Preliminary research 
on family strengths (A KIDS COUNT working paper). Child Trends.
Morgan, D. (2001, February 14). U.S. eugenics like Nazi policy study: Forced 
sterilizations carried out longer than thought. Retrieved February 20, 2001, 
from ABC News.com: 
abcnews.go.com/sections/science/DailyNews/eugenics_000214.html.
Moris, P., Meesters, C., & van den Berg, S. (2003). Internalizing and externalizing 
problems as correlates of self-reported attachment style and perceived 
parental rearing in normal adolescents. Journal of Child & Family Studies, 
12(2), 171-183.
Mosley, J., & Thomson, E. (1995). Fathering behavior and child outcomes: The role 
of race and poverty. In  W. Marsiglio (Ed.), Fatherhood: Contemporary 
theory, research, and social policy (pp. 148-165). Thousand Oaks, CA: Sage.
Msall, M. E., Avery, R. C., Tremont, M. R., Lima, J. C., Rogers, M. L., & Hogan, D. 
P. (2003). Functional disability and school activity limitations in 41,300 
school age children: Relationship to medical impairments. Pediatrics, 111(3), 
548-553.
Mullins, L. L., Chaney, J. M., Hartman, V. L., Olson, R. A., Youll, L. K., Reyes, S., 
& Blackett, P.  (1995). Child and maternal adaptation to cystic fibrosis and 
insulin-dependent diabetes mellitus: Differential patterns across disease 
states. Journal of Pediatric Psychology, 20(2), 173-186.
202
Nassau, J. H., & Drotar, D. (1995). Social competence in children with IDDM and 
asthma: Child, teacher, and parent reports of children's social adjustment, 
social performance, and social skills. Journal of Pediatric Psychology, 20(2), 
187-204.
National Chronic Care Consortium. (2003). The prevalence and cost of chronic 
conditions. Retrieved October 30, 2003, from Partnership for solutions: 
Better lives for people with chronic conditions: 
http://www.nccconline.org/pdf/PrevalenceandCostFacts.pdf.
National Council on Family Relations. (2003). Promoting effective mothering in low 
income and working-poor families. Minneapolis, Minnesota: Author.
National Institute on Disability and Rehabilitation Research. (1999). Long-range 
plan 1999-2003. Washington, DC: Author.
National Institutes of Diabetes & Digestive & Kidney Disorders. (2003, May). 
National diabetes statistics. In NIH Publication No. 03-3892. Retrieved from 
http://diabetes.niddk.nih.gov/dm/pubs/statistics/index.htm#8.
National Institutes of Health. (1996, September). National asthma education and 
prevention program task force on the cost effectiveness, quality of care, and 
financing of asthma care (NO. 55-807). Retrieved October 30, 2003, from 
National Institutes of Health: National Heart, Lung, and Blood Institute: 
http://www.nhlbi.nih.gov/health/prof/lung/asthma/ast_cost.pdf.
203
Neff, J. M., Sharp, V. L., Muldoon, J., Graham, J., Popalisky, J., & Gay, J. C. (2002). 
Identifying and classifying children with chronic conditions using 
administrative data with the clinical risk group classification system.
Ambulatory Pediatrics, 2(1), 71-79.
Newacheck, P. W., & Halfon, N. (1998). Prevalence and impact of disabling chronic 
conditions in childhood. American Journal of Public Health, 88(4), 610-617.
Newacheck, P. W., Hughes, D. C., Hung, Y.-Y., Wong, S., & Stoddard, J. J. (2000). 
The unmet health needs of America's children. Pediatrics, 105(4), 989-997.
Newacheck, P. W., & Taylor, W. R. (1992). Childhood chronic illness: Prevalence, 
severity, and impact. American Journal of Public Health, 82(3), 364-371.
Newacheck, P. W., Strickland, B., Shonkoff, J. P., Perrin, J. M., McPherson, M., 
McManus, M., Lauver, C., Fox, H., & Arango, P.  (1998, July). An 
epidemiological profile of children with special health care needs. Pediatrics, 
102(1), 117-123.
Newcomb, M. D., & Bentler, P. M. (1988). Impact of adolescent drug use and social 
support on problems of young adults: A longitudinal study. Journal of 
Abnormal Psychology, 97(1), 64-75.
Noll, R. B., Gartstein, M. A., Vannatta, K., Correll, J., Bukowski, W. M., & Davies, 
W. H.  Social, emotional, and behavioral functioning of children with cancer.  
Pediatrics, 103(1), 71-78.
Norbeck, J., Lindsey, A., & Carrieri, V. (1981). The development of an instrument to 
measure social support. Nursing Research, 30, 264-269.
204
Northam, E., Anderson, P., Adler, R., & Werther, G. (1996). Psychosocial and 
family functioning in children with insulin-dependent diabetes at diagnosis 
and one year later. Journal of Pediatric Psychology, 21(5), 699-717.
Olkin, R. (1999). What psychotherapists should know about disability. New York: 
The Guilford Press.
Oltmanns, T. F., & Emery, R. E. (1998). Abnormal psychology (2nd ed.). New 
Jersey: Prentice Hall.
Paradise, J. L., Feldman, H. M., Colborn, D. K., Campbell, T. F., Dollaghan, C. A., 
Rockette, H. E., Janosky, J. E., Kurs-Lasky, M., Bernard, B. S., & Smith, C. 
G.  (1999). Parental stress and parent-rated child behavior in relation to otitis 
media in the first three years of life. Pediatrics, 104(6), 1264-1273.
Parcel, T. L., & Menaghan, E. G. (1989). Child home environment as a mediating 
construct between SES and child outcomes., The Ohio State University, 
Columbus, Ohio.
Parcel, T. L., & Menaghan, E. G. (1990). Maternal working conditions and children's 
verbal facility: Studying the intergenerational transmission of inequality from 
mothers to young children. Social Psychology Quarterly, 53, 132-147.
Parcel, T. L., & Menaghan, E. G. (1993). Family social capital and children's 
behavior problems. Social Psychology Quarterly, 56(2), 120-135.
Patterson, J. (1991). Family resilience to the challenge of a child's disability. 
Pediatric Annals, 20(8), 491-499.
205
Perrin, J. M., MacLean, W. E., & Perrin, E. C. (1989). Parental perceptions of health 
status and psychological adjustment of children with asthma. Pediatrics, 
83(1), 26-30.
Peterson, J. L., & Zill, N. (1986). Marital disruption, parent-child relationships, and 
behavior problems in children. Journal of Marriage and the Family, 48(2), 
295-307.
Pettit, G. S., Bates, J. E., & Dodge, K. A.  (1997). Supportive parenting, ecological 
context, and children's adjustment: A seven year longitudinal study. Child 
Development, 68(5), 908-924.
Peyrot, M. (1996). Causal analysis: Theory and application. Journal of Pediatric 
Psychology, 21(1), 3-24.
Pinderhughes, E. E., Nix, R., Foster, E. M., & Jones, D. (2001). Parenting in context: 
Impact of neighborhood poverty, residential stability, public services, social 
networks, and danger on parental behaviors. Journal of Marriage and 
Family, 63, 941-953.
Pledger, C. (2003). Discourse on disability and rehabilitation issues. American 
Psychologist, 58(4), 279-284.
Reinherz, H. Z., Giaconia, R. M., Wasserman, M. S., Silverman, A. B., & Burton, L. 
(1999). Coming of age in the 1990's: Influences of contemporary stressors on 
major depression in young adults. In P. Cohen & C. Slomkowski (Eds.), 
Historical and geographical influences on psychopathology (pp. 141-161). 
Mahwah, NJ: Lawrence Erlbaum Associates.
206
Richters, J., & Pellegrini, D. (1989). Depressed mothers' judgments about their 
children: An examination of the depression-distortion hypothesis. Child 
Development, 60, 1068-1075.
Rodder, I., Kroonenberg, P.M., & Boekaerts, M. (2003). Psychosocial functioning 
and stress-processing of children with asthma in the school context: 
Differences and similarities with children without asthma. Journal of Asthma, 
40(7), 777-787.
Rogers, C. H., & Floyd, F. J. (2003, April). The effects of maternal stress on child 
internalizing problems for children with chronic illness. In Poster session.
Society for Research in Child Development, Tampa, Florida.
Rogers, S. J., Parcel, T. L., & Menaghan, E. G. (1991). The effects of maternal 
working conditions and mastery on child behavior problems: Studying the 
intergenerational transmission of social control. Journal of Health and Social 
Behavior, 32(2), 145-164.
Russell, M. (1998). Beyond ramps: Disability at the end of the social contract.
Maine: Common Courage Press.
Saltaris, C., Serbin, L. A., Stack, D. M., Karp, J. A., Schwartzman, A. E., & 
Ledingham, J. E. (2004). Nurturing cognitive competence in preschoolers: A 
longitudinal study of intergenerational continuity and risk. International 
Journal of Behavioral Development, 28(2), 105-115.
207
Schilmoeller, G. L., & Baranowski, M. D. (2003, November 21). Grandparent 
involvement and support in families with children with disabilities. Poster 
session presented at the National Council on Family Relations, Vancouver 
BC.
Schor, E. L., Billingsley, M. M., Golden, A. L., McMillan, J. A., & Meloy, L. D. 
(2003). Family pediatrics: Report of the task force on the family. Pediatrics, 
111(6), 1541-1569.
Schwarz, C., & Lui, E. (2000). The link between school performance and health 
insurance: Current research. San Francisco, California: Consumers Union.
Shaw, B. A., Krause, N., Chatters, L. M., Connell, C. M., &. Ingersoll-Dayton, B. 
(2004). Emotional support from parents early in life, aging, and health. 
Psychology and Aging, 19(1), 4-12.
Silver, E. J., Bauman, L. J., & Ireys, H. T. (1995). Relationships of self-esteem and 
efficacy to psychological distress in mothers of children with chronic 
physical illness. Health Psychology, 14(4), 333-340.
Simons, R. L., Lorenz, F. O., Conger, R. D., & Wu, C.-I. (1992). Support from 
spouse as mediator and moderator of the disruptive influence of economic 
strain on parenting. Child Development, 63, 1282-1301.
Simons, R. L., Lorenz, F. O., Wu, C.-I., & Conger, R. D. (1993). Soical network and 
marital support as mediators and moderators of the impact of stress and 
depression on parental behavior. Developmental Psychology, 29(2), 368-381.
208
Smith, L. A., Romero, D., Wood, P. R., Wampler, N. S., Chavkin, W., & Wise, P. H. 
(2002). Employment barriers among welfare recipients and applicants with 
chronically ill children. American Journal of Public Health, 92(9), 1453-
1457.
Spence, L. J. (1993). Family adaptation to chronic childhood illness (Doctoral 
dissertation, Michigan State University). Dissertation Abstracts 
International, 53, 4969-B.
Stark, P. (2000, July 11). Introduction of the Chronic Illness Care Improvement Act 
of 2000 (Statement to House of Representatives). Retrieved November 20, 
2001, from http://www.house.gov/stark/documents/106th/hr4981intro.html.
Stets, J. E. (1991). Cohabitating and marital aggression: The role of social isolation. 
Journal of Marriage and the Family, 53(3), 669-680.
Swallow, V. M., & Jacoby, A. (2001). Mothers' coping in chronic childhood illness: 
The effect of presymptomatic diagnosis of versicoureteric reflux. Journal of 
Advanced Nursing, 33(1), 69-79.
Sweet, J., Bumpass, L., & Call, V. (1988). The design and content of the National 
Survey of Families and Households. Madison, WI: Center for Demography 
and Ecology, University of Wisconsin.
Tak, Y. R., & McCubbin, M. M. (2002). Family stress, perceived social support and 
coping following the diagnosis of a child's congenital heart disease. Journal 
of Advanced Nursing, 39(2), 190-198.
209
Takeuchi, D. T., Williams, D. R., & Adair, R. K. (1991). Economic stress in the 
family and children's emotional and behavioral problems. Journal of 
Marriage and the Family, 53, 1031-1041.
Tannila, A., Jarvelin, M., & Kokkonen, J. (1999). Cohesion and parents' social 
relations in families with a child with disability or chronic illness. 
International Journal of Rehabilitation Research, 22(2), 101-109.
Tate, D. G., & Pledger, C. (2003). An integrative conceptual framework of disability. 
American Psychologist, 58(4), 289-295.
Taylor, R. J., Chatters, L. M., Tucker, M. B., & Lewis, E. (1990). Developments in 
research on black families: A decade review. Journal of Marriage and the 
Family, 52, 993-1014.
Thernlund, G., Dahlquist, G., Ivarsson, S. A., & Ludvigsson, J. (1996). The initial 
psychological reactions of the parents at the onset of insulin-dependent 
diabetes mellitus in children and short-time adaptation. Nordic Journal of 
Psychiatry, 50(3), 199-206.
Thompson, R. J., Armstrong, F. D., Link, C. L., Pegelow, C., H., Moser, F., & 
Wang, W. C. (2003). A prospective study of the relationship over time of 
behavior problems, intellectual functioning, and family functioning in 
children with sickle cell disease: A report from the cooperative study of 
sickle cell disease. Journal of Pediatric Psychology, 28(1), 59-65.
210
Thompson, R. T., Armstrong, F. D., Kronenberger, W. G., Scott, D., McCabe, M. A., 
Smith, B., Radcliffe, J., Colangelo, L., Gallagher, D., Islam, S., & Wright, E. 
(1999). Family functioning, neurocognitive functioning, and behavior 
problems in children with sickle cell disease. Journal of Pediatric 
Psychology, 24(6), 491-498.
Thomson, E., Hanson, T. L., & McLanahan, S. S. (1994). Family structure and child 
well-being: Economic resources vs. parental behaviors. Social Forces, 73(1), 
221-242.
Timko, C., Stovel, K. W., & Moos, R. H. (1992). Functioning among mothers and 
fathers of children with juvenile rheumatic disease: A longitudinal study. 
Journal of Pediatric Psychology, 17, 705-724.
Turner, R. J. (1981). Social support as a contingency in psychological well-being. 
Journal of Health and Social Behavior, 22, 357-367.
Underwood, T. H. (1997). IDDM youths' perceptions of well-youths' attitudes 
[Abstract]. Dissertation Abstracts International: Section B: The Sciences and 
Engineering, 57(11-B), 7239.
U.S. Census Bureau.  (2000).  Profile of selected economic characteristics: 2000.  
Retrieved May 24, 2004, from http://censtats.census.gov/data/US/01000.pdf
Vandewater, E. A., & Lansford, J. E. (1998). Influences of family structure and 
parental conflict on children's well-being. Family Relations, 47, 323-330.
211
Vandivere, S., Moore, K. A., & Brown, B. (2000). Child well-being at the outset of 
welfare reform: An overview of the Nation and 13 states.  New federalism: 
National survey of America's families, B-23.  Retrieved September 19, 2003, 
from the Urban Institute: 
http://www.urban.org/Template.cfm?NavMenuID=24&template=/TaggedCo
ntent/ViewPublication.cfm&PublicationID=7037
Vinokur, A. D., Price, R. H., & Caplan, R. D. (1996). Hard times and hurtful 
partners: How financial strain affects depression and relationship satisfaction 
of unemployed persons and their spouses. Journal of Personality and Social 
Psychology, 71(1), 166-179.
Vinson, J. A. (2002). Children with asthma: Initial development of the child 
resilience model. Pediatric Nursing, 28(2), 149-159.
Voydanoff, P., & Donnelly, B. W. (1998). Parents' risk and protective factors as 
predictors of parental well-being and behavior. Journal of Marriage and the 
Family, 60, 344-355.
Wallander, J. L., & Varni, J. W. (1989). Social support and adjustment in chronically 
ill and handicapped children. American Journal of Community Psychology, 
17(2), 185-201.
Wallander, J. L., Varni, J. W., Babani, L., Banis, H. T., & Wilcox, K. T. (1989). 
Family resources as resistance factors for psychological maladjustment in 
chronically ill and handicapped children. Journal of Pediatric Psychology, 
14(2), 157-173.
212
Wanamaker, C. E., & Glenwick, D. S. (1998). Stress, coping, and perceptions of 
child behavior in parents of preschoolers with cerebral palsy. Rehabilitation 
Psychology, 43(4), 297-312.
Weitzman, M., Klerman, L. V., Lamb, G., Menary, J., & Alpert, J. J. (1982). School 
absence: A problem for the pediatrician. Pediatrics, 69(6), 739-746.
Wertlieb, D. (2003). Converging trends in family research and pediatrics: Recent 
findings for the American Academy of Pediatrics Task Force on the Family. 
Pediatrics, 111(6), 1572-1587.
Westbrook, L. E., Silver, E. J., & Stein, R. E. K. (1998). Implications for estimates 
of disability in children: A comparison of definitional components. 
Pediatrics, 101(6), 1025-1031.
Wise, P. H., Wampler, N. S., Chavkin, W., & Romero, D. (2002, September). 
Chronic illness among poor children enrolled in the temporary assistance for 
needy families program. American Journal of Public Health, 92(9), 1458-
1461.
Yeung, W. J., Linver, M. R., & Brooks-Gunn, J. (2002). How money matters for 
young children's development: Parental investment and family processes. 
Child Development, 73(6), 1861-1879.
Zarit, S. H. (1989). Do we need another 'stress and caregiving' study? Gerontologist, 
29(2), 11-12.
213
Zhou, Q., Eisenberg, N., Losoya, S. H., Fabes, R. A., Reiser, M., Guthrie, I. K., 
Murphy, B. C., Cumberland, A. J., & Shepard, S. A.  (2002). The relations of 
parental warmth and positive expressiveness to children?s empathy-related 
responding and social functioning: A longitudinal study. Child Development, 
73(3), 893-916.
Zimmerman, M. C., Ramirez-Valles, J., Zapert, K. M., & Maton, K. I. (2000). A 
longitudinal study of stress-buffering effects of urban African-American male 
adolescent problem behaviors and mental health. Journal of Community 
Psychology, 28(1), 17-33.